CNC Router V1.0 (For Cutting Multiple Materials) [Archive]

biketrialsdave

02-01-2013, 07:56 PM

Hi all,

I figured it was about time I started a proper build log since starting a thread in the new members section. Before I get into the details, a little background of myself. By day I am an Electronics Engineer and by night I am a keen hobbyist! I have built many projects, re-built a couple of motorbikes and fix my car on a regular basis (its french ;) ) but building a CNC router is something I've wanted to do for ages. Now I have some spare time (girlfriend depending!) and a little spare cash (girlfriend depending ;) ) I feel I can finally go ahead with one.

So here goes...I guess the first question to ask myself is what I want the machine to cut. To that I answer: Wood (MDF), Plastics (Acrylic, Delrin etc) but also Aluminium (basic plates/small parts).

I have kicked off a design in solidworks and have a VERY basic outline (not done any detail yet) of what I am looking to achieve:

7845

Along the way I have already made some decisions on parts, as follows:

20mm Hiwin Rail/Carriages all round.
RM1605 Ballscrews.
FF/FK12 Ballscrew End Supports.
80x80 Profile 8 Extrusion for the Y axis supports.
Chinese 2.2kW Spindle
Nema 23 Motors.
THK KR33A Z axis (Purchased this cheaply off ebay)
Approximately an A3 build dimension.

Like everyone, I have a TON of questions to follow but will post those up separately when I get time to compile them all.

As always, I'm open to opinions/criticism! It may well be that the whole design changes before I start ordering parts.

Best Regards,

Dave

routercnc

02-01-2013, 10:39 PM

Hi Dave,

On the basis that it is broadly similar to my machine, but with slightly better components and smaller footprint (and assuming that the sides are 15-20mm aluminium plate?), I would expect this to perform very well for MDF and plastics. Aluminium would be best cut in the middle of the bed since you only have a central ballscrew (like my machine). Even so this would not suit regular aluminium cutting.

Adding a plate across the back of the gantry, between the 2 cross members, would be a useful addition.

I know it is only A3 size, but if you ever think you would upgrade to twin ballscrew in the future then you could increase the width of the end plates and add ballscrew mounting holes so it is ready to go. You would also need more holes in the gantry sides to mount the ballscrew.

biketrialsdave

02-01-2013, 10:45 PM

Thanks for the feedback. Twin Y axis (I assume you mean Y?) ballscrews was on my list, I will probably add the mounting holes in so I can add them in the future. All plate will be 20mm thick I think.

Been doing some more work...

7853

At the moment I only have 2x Hiwin carriages on the X axis. Would there be much benefit/need to double up on these? I.e. two carriages per rail? If so then I will need to increase the X axis length to get a decent travel. I am leading towards this anyway...

Dave

Jonathan

02-01-2013, 10:51 PM

Would there be much benefit/need to double up on these? I.e. two carriages per rail?

Definately. The rigidity will be greatly reduced if you don't use two, since with only one on each rail the axis can more easily pivot (or 'rack') about the center point, i.e. the ballnut, which on the gantry will cause the cutter to deflect substantially.

biketrialsdave

02-01-2013, 10:58 PM

Will add some more carriages in tomorrow :) thanks!

Wobblycogs

04-01-2013, 05:54 PM

I believe routercnc meant twin X-axis ball screws rather than Y (where Y is the gantry). On a small design like this you'd still be able to use a single motor to drive both X-axis screws if you go down that route.

biketrialsdave

04-01-2013, 07:11 PM

As in the following?

7859

martin54

04-01-2013, 07:27 PM

Wobblycogs

04-01-2013, 07:33 PM

Normally the longest and lowest down axis is called X and the gantry Y with the Z going up and down like this:
7860
I've seen one or two designs where the gantry is wider than the bed is long and in that situation it's a little less clear which is X and Y. I believe our American cousins also refer to Y as X.

JAZZCNC

05-01-2013, 12:29 AM

No the other way round lol, from what I understand there is no standard but on this forum they refer to the X axis as the back & forward axis, (longest axis on you build) & the Y axis as the left & right movement. Z is as you have it up & down.
Reason for the twin ballscrews is to prevent the gantry from twisting as it moves backwards & forwards.

Depends if you view machine from front or side. Really your just using the Cartesian coordinate system which is X Axis horizontal or left/right and Y Vertical or Forward/backwards.

If you stand in front of machine like milling machines user's do then the moving table is X axis and it goes Left/right and Y axis goes away and towards you.

Router users often load and stand at the side of machine which is often the longest Axis. So setup machine to work from the side and match the coordinate system of left/right is the X axis and again the y axis goes away and towards you along the gantry.

When I first put my machine vertical the hardest part or strangest part was getting my head around the fact it was still setup to be used and viewed from side and know I was stood in front of it. Kept thinking the code was wrong because I was expecting it to move L & R for X when it went UP/down instead.?? Very confusing.!! For it all to make sense all I had to do was tip my head to the side and things looked right.!

To match the CAD/CAM coordinate system and still look right in Mach I should set it up so that the X axis runs along the gantry left/right and the Y axis vertical Up/down.
I haven't done it yet but I should really has it's a simple has swapping the motor inputs so X axis uses the Y axis input and vise versa.

Ricardoco

05-01-2013, 12:33 AM

biketrialsdave

06-01-2013, 11:46 AM

Ok, so I think I'll add the option for twin ballscrews on the Y axis, I mean X, or Y... :playful:

Swarfing

06-01-2013, 04:03 PM

Stand to the side of your machine and make single finger gun with your left hand, Thumb pointing up = z, pointing finger straight out = Y, trigger (middle) finger pointing right = X. If you always shoot the axis's you can not get it wrong even if you can not remember which hand

biketrialsdave

06-01-2013, 08:40 PM

Twin Y axis (I'm sticking with calling it Y for now!) ballscrews installed. Also doubled up on the X axis Hiwin carriages and increased the X axis width, I have 343mm of travel now - pretty much what I was after. LOADS more to do!

7889

Dave

biketrialsdave

10-01-2013, 08:31 PM

Question: How should I go about limit and home switches?

My current thought is to only have home switches and use software limits. Is this something that can be done in Mach 3? In fact I am considering using Inductive Proximity Sensors for good repeatability. Any opinions on this kind of setup?

Dave

biketrialsdave

13-01-2013, 10:35 PM

Just to re-iterate, I would use 3 (x,y,z) proximity sensors for homing and software limits for emergency stops...

JAZZCNC

13-01-2013, 10:52 PM

Just to re-iterate, I would use 3 (x,y,z) proximity sensors for homing and software limits for emergency stops...

Question: How should I go about limit and home switches?

My current thought is to only have home switches and use software limits. Is this something that can be done in Mach 3? In fact I am considering using Inductive Proximity Sensors for good repeatability. Any opinions on this kind of setup?

Dave

On a small machine like this then I'd suggest you use Limit switches has it will be from one side to the next in blink of an eye.

Personally I use separate limits and home switch's on machines I build. Has well preferring them that way for flexibility of moving my home position around if needed, It's also due to the way I use either purpose safety relays or Standard relays to run my control box.

Having a safe control box means Limits should not be under software control and this means they can't be used for home switch's has they would trip the system every time you homed the machine. Being safe doesn't cost much money compared to surgery for replacing fingers so for £20-30 more doing it right and correct is a no brainier to me.!!

Proximity switch's are fine so long has they are accurate enough. Some require very close fitments of the pickup to achieve good accurecy and this can be a problem with damage depending on what your cutting.?

biketrialsdave

14-01-2013, 05:08 PM

What configuration of switches do you use then? 6x Limit switches + 3x Home switches? How do you resolve the problem of hitting limit switches when homing?

Dave

biketrialsdave

14-01-2013, 05:38 PM

OK, so I warned of a whole heap of questions when I first started this project. Here are a few of the ones at the top of my list... If anyone would be kind enough to answer just one that would be very much appreciated! I hope they make sense...

Should I rebate/slot the aluminium plate where the aluminium extrusion butts up against it? Would there be anything to benefit alignment strength wise? I have done this with the extrusion piece connecting the gantry sides at the bottom but was thinking of the main X and Y extrusions.
This is a general tolerance question. Should I perform a worst-case tolerance stack-up when calculating tolerances for holes etc? A particular question I have regarding this is the diameter of the bolt holes. I want a close-fit to aid with alignment. EDIT: So for example an M8 bolt, what clearance should I include and what is a realistic tolerance if I have the aluminium plate machined manually or on a CNC mill? I'm going to use the ANSI dimensions for hole clearances etc. Should I go for close or normal fit for bolt clearance?
Following on from the above question...Another particular example of the tolerance question is where the Hiwin carriages are recessed into the gantry 'uprights'. The carriage blocks are 44mm high (I cannot see a tolerance in the datasheet...). Lets assume this tolerance is +/-0.1mm. Should I specify the slot in the upright to be 44.2mm +/-0.1mm? This would ensure the part would definitely fit, however if the carriage was only 39.9mm then this fit may not be tight?
Would my design benefit from an Aluminium plate across the back of the two extrusions on the X axis? If so, should I mount this to the two gantry 'uprights' as well as the extrusions? Would 10mm thickness suffice?
In my current design I have the Y axis hiwin carriages mounted to the lower portion of the 80x80 extrusion profiles. Would there be much difference/benefit in mounting these to the upper portion?
How do you determine how heigh to have your Z axis? What I mean by this is do you design around a typical flute length and add some? Should the Z axis be able to go through the bed?

Cheers,

Dave

JAZZCNC

15-01-2013, 01:06 AM

What configuration of switches do you use then? 6x Limit switches + 3x Home switches? How do you resolve the problem of hitting limit switches when homing?

Dave
Yep 6 + 3.!!
Limit switches are just that the Limit of travel, or just before it. HOME switches are usually just in front of the Limit switches, usually on left side but can be anywhere you like really.? By this I mean you could put the HOME switch in the middle of each axis then the centre of machine would be your X0,Y0 machine coordinate. Depending which side you place work material will determine if the Work coordinates are on the positive or negative side of MACHINE coordinates.
By far the most common and less confusing way is to mount the HOME switches on the extreme left. This way your WORK coordinates will always be Positive in relation to MACHINE coordinates.

If your using Micro switches and separate home switches then the switch needs to work on a ramp trigger.! By this I mean the HOME switch gets triggered by riding up a ramp or bump which it can ride over without damaging the switch. When things are correct and working fine and the switches are positioned on the left side then you'll never normally pass the home switch unless there's an issue and if something does go wrong and passes it then not far away will be the Limit switches.

OK, so I warned of a whole heap of questions when I first started this project. Here are a few of the ones at the top of my list... If anyone would be kind enough to answer just one that would be very much appreciated! I hope they make sense...

Should I rebate/slot the aluminium plate where the aluminium extrusion butts up against it? Would there be anything to benefit alignment strength wise? I have done this with the extrusion piece connecting the gantry sides at the bottom but was thinking of the main X and Y extrusions.

Yes it helps with keeping things aligned and registered. If you look at the pics below you'll see the pocketed end plates where profile bolt. The pockets give the profile a register to sit against keeping rails aligned perfectly parallel to each other. A close fit to the profile also resists any twist.

This is a general tolerance question. Should I perform a worst-case tolerance stack-up when calculating tolerances for holes etc? A particular question I have regarding this is the diameter of the bolt holes. I want a close-fit to aid with alignment. EDIT: So for example an M8 bolt, what clearance should I include and what is a realistic tolerance if I have the aluminium plate machined manually or on a CNC mill? I'm going to use the ANSI dimensions for hole clearances etc. Should I go for close or normal fit for bolt clearance?

Problem with close fit is that unless everything is machined absolutely perfect and to the same tolerances then the potential for miss alignment greatly increases. Being realistic and considering the DIY factor then you want some wiggle room so leave at least normal amount of hole clearance.

Following on from the above question...Another particular example of the tolerance question is where the Hiwin carriages are recessed into the gantry 'uprights'. The carriage blocks are 44mm high (I cannot see a tolerance in the datasheet...). Lets assume this tolerance is +/-0.1mm. Should I specify the slot in the upright to be 44.2mm +/-0.1mm? This would ensure the part would definitely fit, however if the carriage was only 39.9mm then this fit may not be tight?

Doesn't matter if the recess or slot is wider than the carriages. What matters is the slot has a Machined reference edge which the Carriage Reference edge sits and registers too. Again look at the picture of the Gantry uprights (bits8) the long slot with 4 holes is for the Carriages and the top edge registers against the carriage Ref edge but the slot is 1mm wider than the carriages. Having the slot a tight fit would make no difference other than making it harder to fit.!!

Would my design benefit from an Aluminium plate across the back of the two extrusions on the X axis? If so, should I mount this to the two gantry 'uprights' as well as the extrusions? Would 10mm thickness suffice?

Yes & Yes.!

In my current design I have the Y axis hiwin carriages mounted to the lower portion of the 80x80 extrusion profiles. Would there be much difference/benefit in mounting these to the upper portion?

Slightly better in the top has the distance from carriage mounting to top off gantry is reduced so having a similar affect to shortening the gantry height.

How do you determine how heigh to have your Z axis? What I mean by this is do you design around a typical flute length and add some? Should the Z axis be able to go through the bed?

Think more about drill lengths than End mills has they tend to be longer. Just workout the thickest material your likely to use then allow for either a longish drill bit or long end mill.
Regards cutting thru the bed then that depends, you want it to at least cut a few mill into the bed for surfacing.
Something to think about and I often do on wood routers is allow the spindle to pass the end of the bed, which often unless designed not to they often do anyway. Then make the Z axis with a little more travel so it can machine down past the bed.? This way you can machine into the edges of panels etc or cut over height material by clamping to end of machine.

biketrialsdave

15-01-2013, 10:28 PM

Thank you very much! I will digest this information tomorrow...hopefully get let out of Jury Service early!

Dave

biketrialsdave

16-01-2013, 06:21 PM

Yep 6 + 3.!!
Doesn't matter if the recess or slot is wider than the carriages. What matters is the slot has a Machined reference edge which the Carriage Reference edge sits and registers too. Again look at the picture of the Gantry uprights (bits8) the long slot with 4 holes is for the Carriages and the top edge registers against the carriage Ref edge but the slot is 1mm wider than the carriages. Having the slot a tight fit would make no difference other than making it harder to fit.!!

If I have read this correct, the top edge "registers" with the carriage (i.e. points it in the right direction?) but does not actually have a close/touch fit with the carriage?

Another question regarding clearances... I am in the process of adding milled out sections in the end plates to locate the extrusions. Should I allow an additional margin (say 0.5mm??) for the extrusion? Or should I design an 80x80 slot with 4mm radius corners as per the extrusion?

Thanks,

Dave

biketrialsdave

16-01-2013, 10:10 PM

Little update... It doesnt look like much has changed but I've added a lot of minor details. Currently ALL aluminium plate is 20mm thick. Does anyone think this is overkill for the plate which attaches to the X axis carriages (on top of gantry) and the plate which attaches the motor mount to the Z axis?

My useable bed dimension is now 343mm x 434mm. I think that's fairly reasonable for what I will be cutting.

7969

Dave

JAZZCNC

17-01-2013, 04:39 PM

If I have read this correct, the top edge "registers" with the carriage (i.e. points it in the right direction?) but does not actually have a close/touch fit with the carriage?

Yes the top edge sits directly on the carriages, the edge acts has the register. If you have them already look at the carriages and at least one edge will be machined, this the the reference edge.
The rails will also have a Machined Ref edge, exactly the same applies with the rails regards Referencing against an edge thou in your case using profile you will only be able to do this if you machine the profile or have a separate plate machined which bolts to the profile.

Another question regarding clearances... I am in the process of adding milled out sections in the end plates to locate the extrusions. Should I allow an additional margin (say 0.5mm??) for the extrusion? Or should I design an 80x80 slot with 4mm radius corners as per the extrusion?

The only sure way is to measure the profile your using then machine to light interference fit. Wouldn't suggest you have them machined without the profile first if your wanting close fit.
Same goes for anything you want machined really, don't trust Cad or dimensional drawings where fit is important has there is always some slight difference waiting to bite you. Can't beat having the part in your hands and I won't machine anything were tight or close tolerance is important without parts in hand.

Currently ALL aluminium plate is 20mm thick. Does anyone think this is overkill for the plate which attaches to the X axis carriages (on top of gantry) and the plate which attaches the motor mount to the Z axis?

Nope it's perfect and will help reduce resonance. . . .Certainly wouldn't go below 15-16mm.

Can I suggest you change the position of the BK/BF bearing blocks and put them on the outsides of the gantry and End plates, or at least the motor side bearing.?
The way you have them on the inside will mean they are trapped between the gantry side and end plates which will make fitting very awkward and inflexible, will also mean they need to be done at the same time has bolting profile together.:thumbdown:

Put them on the outside at the motor end and you can insert the screw thru the ends/sides after building frame. Notice you have the motor mounts has closed blocks like this 7970 so just machine the inside to clear the bearing block and fitting will be much easier and very neat, The bearing will also be protected from crap inside the motor mount so that's another plus.!!

biketrialsdave

28-01-2013, 09:22 PM

Quick question, do end mills tend to come in 'standard' lengths? I am still struggling to calculate my Z axis height. I only have 90mm of travel to play with on the KR33 so it's fairly critical.

On a side note, I have fairly drastically reduced the size of the plate which attaches the motor mount to the KR33. The idea being that the motor mount is almost exactly centralised on the two KR33 carriages to reduce canter-levering effects. Good idea?

Dave

8061

biketrialsdave

03-02-2013, 09:20 PM

Bump :highly_amused:

JAZZCNC

03-02-2013, 10:11 PM

BUMP BUMP.!!

Thought I'd replyed to this question.?? Obviously not but did mean too.!

Look more towards drills has they tend to be the longest things you'll put in the spindle.
Endmills can be various lengths depending on flute length. Common 6mm typical standard endmill will have a flute length around 12-15mm and stick out around 18-20mm. Overall length roughly 55mm.

biketrialsdave

16-02-2013, 08:22 PM

Thanks JAZZ, I will do some calculations.

There is one thing that has been bugging me for a long time regarding setting up material to be cut. Here goes!....

1. Lets say I have placed an aluminium billet roughly in the centre of my bed and I want to cut a simple motor mount out of it. Lets assume that the billet is slightly bigger than the size of the motor mount. How do I align the router up so that the flute is in the right place to start cutting?! I assume that if the billet is larger than the part then you would manually put it in the corner somewhere?

2. Now here comes the real question. Lets assume I have finished milling one side of the mount and I need to turn it over and mill the other side. How do I ensure that the mount is in exactly the same place once flipped?! Or would you put it in a new position and somewhere align the cutter in exactly the right place to start cutting?

:hypnotysed:

I hope I am not missing something obvious here! :)

Cheers,

Dave

m_c

16-02-2013, 09:04 PM

1. if the billet is oversize, just touch of roughly from the relevant sides. Touching of can be as simple as with the cutter spinning, just jogging up until the cutter marks the material, then hitting the relevant zero button (or entering size if not working of zero)

2. Depends on tools at hand, and how many you'll be doing.
For any more than a few, you really need to make up some kind of jig capable of accurate mounting.
However for one offs, you really need some kind of additional tool. For cnc, a touch probe is the best suited, however there's nothing stopping you from using old manual techniques, such as edgefinders, dial gauges, bit of paper stuck on, and jogging up to where you need to be.

biketrialsdave

16-02-2013, 09:42 PM

Thank you, that confirms what I suspected!

On a slightly different topic, where is a good place to buy a set of end mills? Just after standard metric sizes...if such a thing exists! It will also help me work out my Z axis height.

Cheers,

Dave

m_c

16-02-2013, 11:35 PM

Do a search on here, as it's been discussed before.

Endmills I get from ArcEuroTrade when I'm ordering other stuff, but I'm using a milling machine and don't use that many non-insert endmills. What I do are mostly 12mm, with the occasional 6 or 8mm. Router needs are very different.

JAZZCNC

17-02-2013, 03:19 PM

M_C answered most of it but here's couple more suggestions.

1. A: Good Habbit to get into starts at the Cad when you define the part. I always place it 2 or 3mm in from X0Y0 then when you touch off the billet side you know you'll always be inside it and won't run off the edge.
B:Another way is to draw the outline of the billet in cad and have X0Y0 on the corner and work piece inside it.

2. Easy way for one-off's if working from square/rectangular billet is to use the billet corner has X0Y0 like in option B: above then when you flip just make sure the edge is parallel to Axis and pick up X0Y0 from corner again.
Problem comes when you can't work from square/rec stock or if need to relocate already completely machined parts.?
In this case then it's often better to place X0 Y0 in cad on a hole centre or other defining feature that's easily picked up off using touch probe or other manual technique.

Regards the touch probe then doesn't need to be any fancy or expensive tool just piece of wire with plate on the end of a known thickness and using a spare input setup in the control has probe input. Then put plate between tool and work piece.
So long has the tool diameter is known you can just use what ever tool is in the spindle for quick touch off, if you want very accurate position then best to use accurately known shaft or tool shank.
(Tip: If your working with Ali or metals and can isolate the work piece from the main frame or Ground then you don't even need to put the plate between the tool and work piece so can enter just tool radius into DRO for offset. Just touch probe wire to work piece.)

This can then be used in few ways either by using the G31 probe command directly and telling it the axis direction and distance to travel looking for the edge.
Then when it's touched off you zero the Axis DRO's and type the radius of the cutter + plate thickness directly into the Axis DRO to give correct offset from edge.

To find hole centres you'd do this for each X,Y axis if you know the hole radius then type that minus the cutter radius + plate thickness has either positive or negative value depending on which side of hole you probed has the new offset.

With that said the best way and easiest way when it comes to probing is to use pre designed scripts with probe commands etc for each type of probe routine you want to do which does all the offsets etc for you then add a button to your screen set so it's a one click affair.! . . . OR . . .
By far the best and easier way if you use Mach3 is just buy Gerry's 2010 screen set which has all the probe routines you'll need and more besides. The CNC Woodworker - Mach3 2010 (http://home.comcast.net/~cncwoodworker/2010.html)
With this screen set and simple touch plate on wire you'll very accurately find most locations and will be the best £15 you ever spent.

digitaldodo

18-02-2013, 05:07 PM

Looks to be a nice build project. I will be watching your WIP with interest as the size and style are similar to the one I am thinking about trying to design and then build. I fear it will take me some time to get my project off the ground though.

Andy

biketrialsdave

23-06-2014, 09:16 PM

Wow, over a year has passed since I last touched this project! (too many other projects getting in the way...)

I have decided to sell my laser cutter and put the money towards funding this new project....

I have been doing quite a bit of thinking and have concluded that if I'm going to do this I may as well do it properly! Browsing the forum I came across a design by JAZZCNC which looks much more substantial. In fact it may look a little overkill for what I need. I may be aiming to use more aluminium profile to keep costs down. Although I really have no idea as to how much those ally plates (gantry in the link below) would cost to have machined?

http://www.mycncuk.com/threads/6274-1rst-build-time-to-do-next-step/page2

JAZZCNC

23-06-2014, 10:32 PM

In fact it may look a little overkill for what I need. I may be aiming to use more aluminium profile to keep costs down. Although I really have no idea as to how much those ally plates (gantry in the link below) would cost to have machined?

Aluminium profile doesn't always keep the cost down by the time you factor other things like fixings etc which you will need to take advantage of profile.
Granted the design of mine you have been looking at does require quite a bit of machine work and really you need a machine to build this one but the aluminium it's self won't cost much more than Profile by the time you have factored in all the hidden items profile needs and it's much stronger and neater.

To be honest how far you want or need to go depends on the type of work your planning on using the machine to do. If your mostly using it for Router type duty's IE: Wood, plastics etc then you don't need to go this far but if your wanting to cut aluminium with any decent depth of cut and finish quality then this is a minimum. Now this machine was designed from the begining as an all round machine which can do a decent job of all tasks but Make no mistakes this does come at a price and slightly makes it master of None.
If you want the best machine without having to go silly on design and work required then your much better targeting the main use and build to that. This way you get a machine that is the best it can be and give great results at sensible money and time frame to build.

If you are prepared to learn welding or can weld then Buildng from Steel will save you lots of money and is quite forgiving. It makes possible for very strong machine chassis cheaply. This then allows you to spend money where it matters most and that is on correct linear components and quality electronics.
There will be some Aluminium parts required like Z axis but the major chassis parts like frame and Gantry can be made from steel if done correctly.

biketrialsdave

24-06-2014, 07:27 PM

Thank you for another informative post!

I have just picked up some used THK HSR15 linear rails (4x 815mm Long, 2 blocks each) for a very reasonable price. I would of preferred 20mm but these should be up to the job...

biketrialsdave

29-06-2014, 06:47 PM

Some progress on the new design...

Many more questions to follow! =)

1268012681

Lee Roberts

29-06-2014, 09:57 PM

Hi dave, put some additional ali profile in between the current lengths, then lock it all together from the back so its more like one solid mass.

.Me

biketrialsdave

29-06-2014, 10:02 PM

Good suggestion, had something like that in mind. As the design currently is there is a 120mm gap between the top and bottom profile pieces on the gantry. That would nicely fit another piece of profile and completely enclose the back (although then I wouldn't have easy access). Either that or perhaps some vertical pieces of profile between the two horizontal pieces maybe?

12682

biketrialsdave

30-06-2014, 07:06 PM

12693

A axis added - some detail missing.

biketrialsdave

01-07-2014, 10:10 AM

Is anyone able to give me a rough order of magnitude for the sort of price I will be paying for having the custom aluminium parts (gantry side etc) machined? I really don't have too much of an idea right now!

Cheers,

Dave

Wobblycogs

01-07-2014, 10:24 AM

Just my 2p worth... I'd say you have a lot of designing still to do. My guess is that you will replace some of that extrusion with plate and you'll probably add more (machined) plate into the design. Any guess at a machining cost at this stage would be unrealistic I think. You might also like to consider whether your design will require machining on both sides of any plate as that will increase the cost due to the set up time needed to machine one side, flip the plate and align it accurately for machining on the other side. If you have a look at my design I've specifically made it so that it can all be machined from one side with the exception of some counterbores that I'll do myself. Lets put it this way, I spent a lot more on parts than I did / will on machining costs.

biketrialsdave

01-07-2014, 09:01 PM

Thanks for the help, I will try and keep all parts machined one side only.

Slowly but surely I'm starting to make some progress...

12714

biketrialsdave

03-07-2014, 07:57 AM

My thread moved?!

biketrialsdave

05-07-2014, 07:32 PM

Small update...

12733

Now I've mounted the FF blocks on the outside of the plates I don't think I'll get an off-the-shelf motor mount which will fit over them? So I might need to use pulleys?

Wobblycogs

05-07-2014, 08:29 PM

Pulleys are the way forward, they give you flexibility in the final machine and generally make the design simpler. It's slightly more expensive to use pulleys rather than a direct connection and there are more moving parts which means more points of failure but I think that's a small price to pay.

One thing I would say about your design is start pricing it up. I'm guessing those corner connection blocks are the 90x90 ones - they are quite expensive if you need a load of them and they don't result in a machine that is all that stiff (they are great for display stands, not so great for machines). Once you start getting into the nitty gritty of placing components and pricing up I think you'll begin to consider using aluminium plate and box section more.

Looking good though.

biketrialsdave

05-07-2014, 10:37 PM

You are very right. I have been keeping an eye on the cost and realising that it soon adds up. The reasons why I am not leaning towards a steel box section frame; the first is I don't have a MIG welder! The second is that even if I could justify purchasing one then I cannot see how people manage to weld everything so that it is all square and flat? In my mind that would be quite a hard thing to achieve?

Wobblycogs

05-07-2014, 10:54 PM

I was meaning aluminium box section, you can get some really large sizes if you shop around. It's more expensive than a frame made from steel box sections but it's easy to work with and the supplier will cut it to length normally (I used aluminium warehouse iirc).

My first CNC design included a steel frame and I actually rushed out and bought a halfway decent MIG welder. I never actually built the frame in the end though, doh! Anyway, you don't need a fancy MIG to weld the steel frame together - a cheap and nasty stick welder from ebay will do the job, there's one up for £26 at the moment. The trouble with MIG is unless you go for one with some grunt it won't handle the 4mm thick box you'd want to build the frame from. As for keeping it square and level if you look around you'd be amazed what can be done with epoxy resin. Basically you weld up the frame as square as you can (lots of small welds + clamps) and then pour very thin epoxy over where you'll mount the x-axis rails. The epoxy will self level and should leave you with two sides that are in exactly the same plane. Check out this thread (http://www.mycncuk.com/threads/2148-Comments-sought-on-new-build-A-CNC-Router-for-RC-Gliders-and-Planes?p=59700#post59700) as Neil will be doing another epoxy pour any time soon (actually the technique in the first pour was fine he just fluffed the mixture).

JAZZCNC

06-07-2014, 12:12 AM

Dave don't be fooled into thinking profile is flat or level because it often isn't and to make this worse your design makes it so your pretty much guaranteed you'll have issues with the rails binding.? . . .Your relying completely on the quality of cut on those uprights to keep the rails aligned and 90deg to the base. The 90Deg brackets are not accurate enough or strong enough to hold everything square. The fact your using 45x90 orientated in it's narrow section when the Y axis forces will be trying to rock the profile doesn't help and this is compounded by the fact the 90deg brackets don't hold it in that direction either.

Sorry to say this but while you think the frame is strong it's actually weaker than you think because it's not braced or held the directions that matter and your under sizing the profile. To make profile work for this design you'll need 90x90 for all the base and much more bracing in correct locations using plates not those poxy 90deg brackets.

For this reason Steel is much cheaper, stronger and more accurate if used with the epoxy method.

biketrialsdave

06-07-2014, 10:52 AM

JAZZCNC

06-07-2014, 11:32 AM

Hmm maybe it's time for yet another re-think then...

I don't want to go OTT with the design but obviously it needs to be useful. What are you opinions on the gantry design as it stands so far? Are there any other alternatives to a steel box section base?

Thanks for all the help,

Dave

Dave don't get me wrong I'm not saying profile won't work because it works great and I use it all the time but it does need to be supported and braced correctly and by this I mean not using in the normal manner profile is used IE: using corner brackets etc. I'm also not saying your design is flimsy or weak just not has strong as you may be thinking. It will certainly have some chatter in harder materials as the frame vibrates.
For this high side design to work best with profile, esp when on it's thin edge you need to brace the hell out of it using plates.

Your gantry is ok-ish but I would have a bit more strength in the bottom profile as this takes the most the cutting forces. To be honest I'd have the L shape layout of the profile at the bottom but with lower profile laid flat.

I understand your reluctance to weld but it really isn't difficult or needs to be expensive. Don't need a Mig and like as been said cheap stick will work fine.
If you don't want to weld then how about finding someone who can weld and have them do it for you.? If you cut and preped all the steel it wouldn't take someone long to weld that simple frame.! . . 2 hours Max and I'm allowing for a very very slow welder there.!! If all steel was cut and preped it would literially take me 30-40mins tops just for the welding.

biketrialsdave

06-07-2014, 06:23 PM

Hmm well I do have an arc welder somewhere in the shed. Not touched it for years and was never any good at it! May need to get practicing...

JAZZCNC

06-07-2014, 07:05 PM

Hmm well I do have an arc welder somewhere in the shed. Not touched it for years and was never any good at it! May need to get practicing...

Get it out and get practicing. Don't use old rods and buy good quality rods not some cheap rubbish from car boot as they make all the difference. Keep them dry and stick in oven on low heat for 15 minutes before starting. Welding new metal like Box section is a piece of piss with a stick welder. Also Just realise you don't need Xray quality welds and Just like Snow makes every body's gardens look the same so does a good grinder and paint on pigeon shit welds. . .:encouragement:

biketrialsdave

06-07-2014, 07:11 PM

JAZZCNC

06-07-2014, 07:57 PM

Just pulled it out the garage, its the one in the link below. Very cheap and cheerful. Only a 100A output.. What sort of thickness steel box section would be best to use? 3/4mm?

http://www.metals4u.co.uk/welding/welders/silverline-677293-arc-welder-100a/detail.asp?prd_id=3124

4mm will be better than 3mm but either will work. With 100A you'll be limited to 2.5mm rods which will be better suited to 3mm.

biketrialsdave

06-07-2014, 08:55 PM

Clive S

06-07-2014, 09:54 PM

I will find all my gear and get hold of some scrap metal sometime this week! :) It's always good when you find a reason to use an untouched tool! Out of interest, how are people bolting their linear rails to box section? Bolt all the way through both sides?You can't bolt all the way through as it would squash the box section without spacer tubes I used 5mm box section so just drilled and tapped it but others have glued a steel plate inside the box section with gorilla glue and then drilled and tapped. As Dean says you don't have to be super accurate because you can use epoxy on the top with a moat between the rails so that it levels out on the same plane.
When and if you come to this step just ask as quite a few people on here have used it. I used the Wests System which is ultra slow curing ie 15 -20 hours and so levels out very well ..Clive

GEOFFREY

06-07-2014, 11:07 PM

Why is it not possible just to drill a bigger hole in the bottom face of the tube and nut up to the underside of the top face? G.

JAZZCNC

06-07-2014, 11:36 PM

Why is it not possible just to drill a bigger hole in the bottom face of the tube and nut up to the underside of the top face? G.

It is possible Geoffrey but to be honest it's not needed. Mostly I use 4 or 5mm without any problems but in the past I've fitted rails to 3mm box with no issues. The fine pitch gives enough threads engaged and because of the number of bolts you don't need to swing on them to tighten down.
I just use spring washers with loctite on threads and never have any trouble. The rails on my machine are into 3mm box and have been on for over 6yrs without any bolts coming loose so it works well enough.

Wobblycogs

07-07-2014, 11:18 AM

When I started building my first frame I (stupidly) decided to bolt it together, I was worried but stripping threads etc so I did some experimenting...

The test piece was 4mm steel box section drilled and tapped for use with regular M6 socket screws. With a screwdriver and a hex head I couldn't do the bolts up tight enough to strip or damage the thread in the box section. With a regular Allen key I generally couldn't damage the thread, after repeated re-tightening as tight as I could do by hand I did manage to make a couple of the test holes a bit loose. With a cheater bar on the Allen key I could strip the threads and using a drill driver without torque limiting I could strip the threads. My conclusion was 4mm box + M6 was more than strong enough for what I was doing.

I also experimented with some 3mm box + M6 and concluded it wasn't good enough. You can get a thread in it but it's clearly not much good. M5 worked and M4 (which is what I suspect your rails will be using) was fine.

biketrialsdave

07-07-2014, 12:31 PM

Yet again, thank you for all the help! I think I may end up going for something like 60 x 40 x 3mm box section as this is within the capabilities of my welder and isn't too expensive.

Clive S

07-07-2014, 07:59 PM

Yet again, thank you for all the help! I think I may end up going for something like 60 x 40 x 3mm box section as this is within the capabilities of my welder and isn't too expensive.If you can go with 4mm all the better for drilling a tapping. ..Clive

biketrialsdave

07-07-2014, 08:24 PM

biketrialsdave

07-07-2014, 09:11 PM

Started work on the base...

12740

JAZZCNC

07-07-2014, 09:15 PM

Hmm if I can get away with 100A/2.5mm electrodes then maybe. Although it is an added cost...

The rail I have takes M4 bolts so thats 4.28 threads for 3mm and 5.71 for 4mm... I think the rule is 1x the diameter?

Listen trust me it will be fine. One more added point My machine was mounted vertical for well over a year and other than hanging from the ceiling you couldn't put more stress on the bolts has they had all the weight of the gantry bearing down on them and they didn't even flinch. Go for it won't be problem.!

JAZZCNC

07-07-2014, 09:19 PM

Started work on the base...

12740

Better just put some triangle plates on the front uprights and your sorted.

Still need to sort that gantry thou and put the bottom profile flat with vertical one above it.! . . . Hard to see from drawing but what size profile is it.?

biketrialsdave

07-07-2014, 09:23 PM

3mm it is!

Yes, still a bit of work to do on the gantry (motor mounting etc). The profile is 80x40. If I can mount the motor on the outside of the gantry end plates then I was going to put another piece of horizontal profile in to enclose the back completely.

While I'm here, what brands of welding rods do you guys recommend? There's plenty of "Super 6" ones on Ebay (6013) for a good price?

12741

Lee Roberts

07-07-2014, 11:00 PM

Dave,

See screen shot, i had 5 mins spare. The red and green sections would fix to your side plates and then to each other, the profile is 80x80, the plate is 20mm, note the datum for the rails on the front plate.

.Me

biketrialsdave

08-07-2014, 08:05 PM

How did you manage to edit that, Lee? Thank you! My only concern with that design is that I thought the rails were better off not being on there side?

Lee Roberts

08-07-2014, 08:27 PM

How did you manage to edit that, Lee? Thank you! My only concern with that design is that I thought the rails were better off not being on there side?
Hi mate, I pulled your image into the software I use for Web Dev then did some cut and paste magic with what was already there :).

For these rails it's not as important, open round type bearings, then yes that would be correct. You could still mount them at the top and bottom, the choice is yours really but the objective is to get and keep the bearing units as close to the cutter as you can, that being said the the difference here would probably be negligible. I did a few version's of the mock up and one had the rails as suggested, if you prefer that design add plates to the top and bottom, as well as the back, I forgot to say that the back plate didn't need to be 20 mil though, also going with the top/bottom arrangement also means that the spindle plate/assembly stands further away from the bearings due to the lead screw interference with the ali profile.

Lee Roberts

09-07-2014, 01:06 PM

Dave slapped this together for you quickly so you can see approximately what you may end up with, without needing to get down and dirty in CAD.

Nothing is to scale obviously as it is just a drawing, note the X Axis screw mount position and that it would probably need moving back (away from spindle) to allow for clearance of the nut and mount (at the spindle plate), side plates would need to be bigger and other plates also bigger, bigger side plates are not a bad thing outside of additional cost because you really don’t want to go any less then 2-250mm centres for your rail bearings, both designs pretty much force you in that direction anyway so I didn’t mention it before but I thought that was a little bit ignorant of me, so here it is.

You could close things up with datum’s and pockets for the rails and bearings and so on, keep an eye on the proportions as well, if you look at the original image I’m working from, the gantry looks small next to the spindle, if something looks small, it probably is...resize to match the proportions if you can.

Also, others will probably say it’s not as important as getting other things right, and they are right but try and keep everything lined up and parallel, so your rails, lead screw and nuts. On the Y Axis (bed) the nut could be moved back towards the spindle in my previous image, vertically aligning it with the rails and lead screw, the middle of that plate “looks” nice but would it be better for the loads, if they are all inline?, they all come together to work as one.

Just my thoughts...What do I know...hope it helps!

.Me

JAZZCNC

09-07-2014, 11:35 PM

No need for that box design around gantry it add's very little but expense and more potential for binding. Also with this design the ballnut won't fit between the Z axis backplate and gantry. Those wide top n bottom plates with rails sat back just make long levers acting upon the bearings which will could cause binding.!! . . . . Not a design I'd go for.!

Your design in post #67 will be ok. Don't waste money or time on the box around gantry setup no point or bennifit IMO unless bracing sides.!

Jonathan

10-07-2014, 09:22 AM

The significant advantage from putting the rails one the top and front of the top and bottom beams respectively (as Jazz suggested earlier), is that you can make the rails a bit longer and thus increase the spacing of the bearing blocks a bit without widening the machine or loosing travel. I'd certainly recommend that since the stiffness due to the bearings is proportional to the bearing spacing squared (assuming what they're mounted to is rigid etc), so every little helps.

biketrialsdave

12-07-2014, 10:28 AM

Not much progress but changed the "L" shape on the gantry as JAZZ suggested. I'm now using 2 pieces of 80x120 and 2 pieces of 40x40 (mainly for filling the gap). This used to be 60x30 but thought I better bulk it up. Also the height (total height 350mm) and width (now 175mm) of the gantry side plates has increased a little to accommodate. Still undecided with what to do at the back of the gantry... I need to leave enough room to mount the motor on the inside (using pulleys).

12764

biketrialsdave

13-07-2014, 12:06 PM

Ahhhh! really not sure which direction to go with the gantry design. It just does't look right now. Maybe I should mount the motor on the outside of the plate and the fixed (FK12) ballscrew mount on the inside of the gantry plate? (I.e. direct connection, no pulley).

Who knew this would be so hard!

Wobblycogs

13-07-2014, 05:46 PM

Lol, I know exactly what you mean. I must have gone through a dozen different designs before I settled on what I was going to build. The most disheartening bit is when you think you've got it done then notice there's a collision or something doesn't quite line up correctly.

I'd be interested to see a front view as your design you seems to have a full height side plate on the gantry. Do your x-axis screws stick out a long way?

biketrialsdave

13-07-2014, 07:41 PM

"I'd be interested to see a front view as your design you seems to have a full height side plate on the gantry. Do your x-axis screws stick out a long way?"

Not quite sure what you mean by that?

Here are some options for the gantry..really don't know which route to go down now...Option A would allow an internally mounted motor (without increasing the side plate width). However I am also leaning towards option B. No matter which option I go with I still need to do something about the big gap where swarf could get to the ballscrew.

12769

Wobblycogs

13-07-2014, 08:15 PM

Option A isn't going to be very strong, it has almost no vertical bracing. Essentially all the stiffness would come from the bolts through the side plates into the profile.

Option B is ok but Option C is going to be a bit stronger. If you want the space internally to mount the stepper then you could go with option A but put a 10 or 15mm plate across the back (I assume the sloped side is the front). You should really have a look through some of Jazz's threads if you want to see a good design that keeps everything internal, the design I'm thinking of has a 20mm plate a the front of the gantry with a slot machined out of it for the screw connection. It served as inspiration for my design.

If you want to do a quick experiment get a couple of lengths of 2x4 timber and nail them to some ply in roughly the design of option A, it'll take 5 minutes to build and you'll immediately see why it won't work as a gantry as it'll wobble around like a jelly. Grab yourself a bit of hardboard and nail it across the back, you could probably stand on it them.

JAZZCNC

13-07-2014, 09:06 PM

If you want a simple design that works is strong and neat then try this setup. This machine design with higher sides is just for cutting woods and soft materials but you can use the same "L" arrangement without the high sides. Before anyone says it the ballscrew at the rear makes no noticable difference to cutting and the advantages of keeping chips away from screw are clear to see has is the neatness.

If you want stronger than this then look at the All aluminium gantry but it's much more expensive, thou it is very strong and again neat with good protection to components.

biketrialsdave

13-07-2014, 09:22 PM

JAZZCNC

13-07-2014, 09:55 PM

Thanks Jazz! I had thought of something similar but was afraid monting the ballscrew like that would cause problems. Is the general idea of the gantry sides to keep them as short as possible? Well not the sides, but the distance between the top and bottom profile carriages on the X axis?

Yes if you want the best possible strength then you want short stubby sides but this is only really required if you want to cut hard materials most of the time. Gantry height just needs to be high enough to provide support to the Y/Z axis and resist Twisting/flexing. With this design the height of the gantry supports and takes the full load of the Z axis bearings with just the ballscrew and motor above the gantry.
The red machine in the pictures is twin ballscrew driven and this cuts aluminium no problem with resoanable quality of finish but it's not designed to do this all the time and it would eventually take it's toll on the machine, where has with woods etc it will happily cut 24/7.

With regards to the ballscrew at the rear then IME it hardly makes any difference unless wanting to cut hard materials with deep DOC. Which in this case then you'd want a stronger gantry design. For all wood and lighter aluminium use then it's no problem at all, if it was I wouldn't be using this design.!

To be honest if your wanting to cut all materials then you have to lose something somewhere. If you build for cutting the hardest materials then you need strength and weight, this comes at the cost of speed unless you increase the strength of motors etc and this costs cash.
This is wasteful if your only ever going to cut hard materials 5% of the time. So IMO you get a much better machine if you target the main use and design for the optimum doing this.! . . . Jack of all trades works but truely is master of none and you have to accept this other wise you'll be sadly dissapointed.!

biketrialsdave

14-07-2014, 09:46 PM

Finally feeling like I may be going in the right direction now! (*awaits comments saying I'm not* =P).

There's just enough room to mount the X axis motor on the inside of the gantry.

12783
12784
12785
12786

JAZZCNC

14-07-2014, 11:10 PM

Won't need that top piece it's only adding weight and expense.

biketrialsdave

15-07-2014, 07:56 PM

Thought you might of said that! Now removed.

I am looking to mount the motor on the inside of the gantry but I'm going to need a long motor shaft or a pulley with a long hub length to get through the 20mm plate and out the other side with enough length. Any suggestions?

biketrialsdave

20-07-2014, 07:25 PM

It looks as if it will be easier to mount the X axis motor on the outside of the gantry plate (See below). If I wanted to mount it on the inside and still use the FK bearing block then the shaft would be too short to mount a pulley.

12853

I am now looking at the Y axis ballscrew mounts in a little more detail. In the screenshot below how is the bearing mount attached to the steel box section? (Does anyone have a link to this build thread btw?)

12854

I assume that there's another piece of welded steel plate on the end of the box section? As below?

12855

While I am on the subject of the "end plate", I assume some sort of adjustment will need to be built in to allow movement in the Z axis as if I use the epoxy method on the box section then it's going to add some height?

Dave

biketrialsdave

22-07-2014, 06:48 PM

Yet more questions... this time about welding and distortion.

What is the best technique to use to minimize distortion? I assume I should tack weld each side of the box section before doing full fillet welds? Do I need to do any kind of stress relief to the joint after? I am only practicing at the moment. Picked up an Evolution Rage 3-S to cut some box section and purchased an auto darkening helmet =)

I've also ordered one of these, which I hope will help:

12874

Dave

Lee Roberts

22-07-2014, 09:02 PM

Hi Dave,

Yea tack it all up in sections keeping any eye out for square, once your happy fill them in, alternate the areas your working on so they have time to cool down while you work another area, it's the heat build up your trying to avoid.

If you can clamp stuff down so it's less likely to move or distort, if you want to get really into it, do some googling on preheating, Jody over at http://www.weldingtipsandtricks.com has a pretty comprehensive website and youtube channel.

Where did you get the jig from?

.Me

Neale

22-07-2014, 09:16 PM

Speaking as a pretty second-rate welder, I found it really useful to build the bulk of the frame using just one spot of weld at each joint. I then used a series of Spanish windlasses, sash cramps, etc, to pull it all square before going round and building up the welds. Diagonal bracing followed. I cut all the box section as accurately as I could using an angle grinder in a cheap stand but it wasn't perfect. Your cutoff saw might do better. And if it goes pear-shaped (literally!) then there's always the angle grinder to cut the weld to start again - another reason not to weld too much before you're happy with the shape. The great thing about welding is that for something like this, it is very forgiving. I cut one piece on the wrong side of the line, but built back up with weld and ground back to dimension...

biketrialsdave

22-07-2014, 09:33 PM

Good tips! Hopefully I will get some time over the weekend to practice with my new tools :) Still got a LOT of design work to do...I'm thinking of moving away from the FF blocks + aluminium plate for the Y axis and maybe just using steel channel (or angle) welded to the box section and mounting BK blocks to it? One of my biggest concerns for the whole project is actually drilling accurate/straight holes! I have a pillar drill but it is pretty useless as there is a lot of slop in the chuck.

This is the jig I bought, got to be worth a punt at £21!:

http://www.ebay.co.uk/itm/111409678703?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2648

Lee Roberts

22-07-2014, 11:21 PM

Cheers Dave.

biketrialsdave

25-07-2014, 11:46 AM

Good tips! Hopefully I will get some time over the weekend to practice with my new tools :) Still got a LOT of design work to do...I'm thinking of moving away from the FF blocks + aluminium plate for the Y axis and maybe just using steel channel (or angle) welded to the box section and mounting BK blocks to it? One of my biggest concerns for the whole project is actually drilling accurate/straight holes! I have a pillar drill but it is pretty useless as there is a lot of slop in the chuck.

Anyone got any opinions on the above? I appear to be coming to a point where I'm not sure where to go next. Still not happy with the X motor on the outside of the gantry plate either :(

Wobblycogs

25-07-2014, 02:15 PM

I wouldn't worry too much about the stepper sticking out the side of the gantry (I'd have called that the Y stepper but there's no hard and fast rules) as it's just basically a metal lump there's not a huge amount to go wrong as you have it. If you can't design it such that it's internal maybe consider sticking a bent aluminium box around it. I managed to squeeze my Y stepper inside the gantry (see here (http://www.mycncuk.com/threads/6418-Wobblycogs-CNC-Mk2?p=48392#post48392)) but it's pretty tight and I expect to be replacing the belt more frequently that usual. That's not quite the final design that I've gone for but it shows the basic layout.

biketrialsdave

27-07-2014, 11:21 AM

I am now looking at the Y axis ballscrew mounts in a little more detail. In the screenshot below how is the bearing mount attached to the steel box section? (Does anyone have a link to this build thread btw?)

12854

I assume that there's another piece of welded steel plate on the end of the box section? As below?

12855

While I am on the subject of the "end plate", I assume some sort of adjustment will need to be built in to allow movement in the Z axis as if I use the epoxy method on the box section then it's going to add some height?

Dave

Is anyone able to answer any of the above? It would be great if there was a link for the red router listed?

Cheers,

Dave

JAZZCNC

27-07-2014, 07:01 PM

Is anyone able to answer any of the above? It would be great if there was a link for the red router listed?

Cheers,

Dave

Again it's one of mine Dave and no build thread as these where built by me for someone off forum.

biketrialsdave

27-07-2014, 07:08 PM

Again it's one of mine Dave and no build thread as these where built by me for someone off forum.

No problem JAZZ. I'm currently making some changes as we speak. A few questions are now arising:

1. What pulleys do most people go for? XL? Not sure on pitch etc.
2. Is there a way of extending motor shafts so that I can overcome this problem below?
12908

3. Is 10mm of motor mount adjustment enough for belt tensioning?

12907

Some progress on the Y axis too:

12909
12910
12911

Wobblycogs

27-07-2014, 07:54 PM

What most people do is rout out a recess for the stepper to mount in so you've got only maybe 5mm of material where the stepper sits (see here (http://www.mycncuk.com/threads/6418-Wobblycogs-CNC-Mk2?p=48784#post48784)). That way you generally get enough shaft to mount a pulley on directly without needing an extension.

I don't think 10mm of movement will be enough for tensioning, especially not on a long belt. I'd go for more like 25 or 30mm if you can.

JAZZCNC

27-07-2014, 08:03 PM

#1 Use 15mm wide 5mm pitch HTD pulleys and belts.

#2 Need extensions made on lathe but much easier to use BK/BF bearings inot profile and belt/pulley like shown in my picture on other thread.

#3 20mm slot.

biketrialsdave

28-07-2014, 08:44 PM

Tonight's updates:

12923

Still not 100% sure on the way I have attached the "end plates" as seen below. I.e. using 5mm steel plate welded to the box section and tapped with two M6 threads.

12924

12925

Wobblycogs

29-07-2014, 08:53 AM

The stepper in the first two pictures looks large for the size of machine you are building, are the 34s? I would have thought 23's would be enough for a machine of this size, especially if you have slaved motors.

The connection between the gantry screw and the top plate (third picture) looks like a weak point. I'd try and make it thick enough that it covers most of the body of the ballnut.

biketrialsdave

29-07-2014, 12:31 PM

They are 23's. I think the model I am using has a particularly long body.

I agree with the ballscrew mount. Could make it 20mm plate (from 15) or increase that joint some other way...

biketrialsdave

07-08-2014, 09:02 PM

Not touched the mechanical design for a while as I got distracted down the route of designing my own stepper driver...

So here's a small update. I'm feeling quite bad for how much this is looking like one of JAZZ's machines! Believe it or not I am not trying to copy you Clive!

I am still not happy with the way I have the Y axis plates attached to the steel box section. Think I need to come up with something better?...

12993

Clive S

07-08-2014, 09:37 PM

Believe it or not I am not trying to copy you Clive! If you was, I would be more than excited,:pride: but I don't see the resemblance to mine if it is me you are referring too. ..Clive

12994

biketrialsdave

07-08-2014, 09:52 PM

Oops! I meant Dean...it's been a long day.

Nice machine though! :)

JAZZCNC

09-08-2014, 10:29 AM

Oops! I meant Dean...it's been a long day.

Nice machine though! :)

No problem with copying my stuff and you wouldn't go far wrong copying clives either it's an excellent machine.!

Clive S

09-08-2014, 11:14 AM

No problem with copying my stuff and you wouldn't go far wrong copying clives either it's an excellent machine.!Your so kind. But I wouldn't go that far:tickled_pink: ..Clive

biketrialsdave

11-08-2014, 08:27 PM

I think the end is nearly in sight!! :excitement:

All steel plate is currently 3mm ("Y axis "end stops" and end plate attaching bed), should I make this thicker? Will there be any problems welding it to 3mm thick box section? I have a friend who works at a laser cutting place so I'm hoping to get those pieces cut there.

13008
13009
13010
13011
13012

biketrialsdave

05-09-2014, 01:31 PM

Any comments on the above would be appreciated =) I think I need to start looking at places to get the aluminium milled soon!

I've been a bit distracted with designing my own stepper drivers but going to put that to the side until I have the mechanical design all done.

EddyCurrent

05-09-2014, 03:59 PM

I think it would be better to extend the Z axis rear plate to meet up with the bottom plate, like this; it's not shown on this photo but my ball nut fits on the inside of that plate.

13312

3mm plate is too thin I reckon, I used 10mm flat bar for brackets, maybe 6mm would do, see what others say about it.

biketrialsdave

06-10-2014, 08:24 PM

Thanks for the input Eddy!

Well it's been a long time since my last post and since then there have been some large changes... Now, I am fully awaiting the groans and face palms when I say this....

I have decided to return to using aluminium profile for the frame. The reasons being:

1) My welding skills are not quite there. They aren't bad but I envisage a lot of trouble trying to get things square and aligned correctly.
2) I now have free access to a water jet which I plan to cut beefy steel brackets from to connect the profile together with.
3) I have re-evaluated what I will mainly be realistically cutting and steel may be OTT!

Anyway, enough excuses. Here's the latest design. I would really like to get the ball rolling with the custom aluminium parts. Am I best posting on the job request page and going from there?

http://www.mycncuk.com/attachment.php?attachmentid=13543&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=13544&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=13545&stc=1

Dave

biketrialsdave

12-10-2014, 10:09 PM

Job request can be found here: http://www.mycncuk.com/threads/7974-RFQ-Aluminium-parts-for-my-CNC-router

biketrialsdave

21-10-2015, 10:20 PM

JAZZCNC

22-10-2015, 02:31 PM

I really need to pick this project back up! Been having some thoughts recently and have come to the realisation that what I really need is a machine capable of cutting aluminium more than anything else. I'd really like to mill enclosures for my many electronics projects. Perhaps I need a re-think on the complete design!

The design you had wasn't too far away from cutting aluminium. It could benefit from a better/stronger base and the the linear stage for the Z axis isn't really up to aluminium but the general layout of high sides with gantry sat on rails is perfect for aluminium. Gantry design is nice and strong enough for aluminium.
If you sat that lot on nice piece of machine tooling plate or better still Granite base the frame would be sound for aluminum. Ditch the linear stage and put a conventional Z axis with decent rails etc and your sorted.

biketrialsdave

22-10-2015, 02:39 PM

Some encouraging words there, thanks Jazz! When you say "sit the lot on machine tooling plate or granite..." which bit are you referring to?

JAZZCNC

22-10-2015, 02:46 PM

Some encouraging words there, thanks Jazz! When you say "sit the lot on machine tooling plate or granite..." which bit are you referring to?

Your design doesn't show any base and looks like it sits directly on table.? So I mean put EVERYTHING on solid flat foundation.

biketrialsdave

22-10-2015, 02:50 PM

Ahh I see, yes your right. I'm not 100% happy with the useable z height I get either. I think I will redesign the z axis but may also have to increase the height of the gantry sides. I could do with finding a 'cheap' solid solution for the bed too...

biketrialsdave

22-10-2015, 02:51 PM

Not sure if to consider coolant at this point either!

JAZZCNC

22-10-2015, 02:59 PM

Not sure if to consider coolant at this point either!

Yes 100% consider it and also cable routing etc.

Cutting aluminium correctly will need a minimum of blown air and idealy minimum qty coolant ie: Fogless mister. Don't need full flood but better if can have it.

biketrialsdave

22-10-2015, 09:21 PM

I think an aluminium bed with t-track bolted to it looks like a good idea. Just need to find some for a reasonable cost!

biketrialsdave

25-10-2015, 09:41 PM

I may of found a cheaper solution. "MakerBeam" is a 10x10mm extrusion and relatively cheap. I was thinking something along the lines of this:

16364
16365

For the bed plate itself (shown in green). What would be the most cost-effective solution? The dimensions are 725x825mm. Steel or aluminium? What thickness would suffice?

biketrialsdave

27-10-2015, 08:20 PM

I might of solved my bed problems! Look what I found in the skip at work today (along with other goodies)... It's 6mm steel plate. The dimensions are perfect for the machine to sit on. The raised area in the middle also looks perfect for mounting a bed (aluminium plate or t-slot plate perhaps?). My only concern is how flat and level the plate is. I was thinking it doesn't matter too much if I skim the aluminium plate (or whatever) with the machine once it's built?16387

biketrialsdave

30-10-2015, 12:15 PM

:drunk:Decisions, decisions!

biketrialsdave

02-11-2015, 10:55 AM

routercnc

02-11-2015, 01:33 PM

My latest thought is to bolt down a 20mm aluminium plate to the top of the steel raised area (shown in photo). I will then use the machine to skim the surface flat. Then I will use the machine to drill a grid of holes which I can then tap (maybe the router could do this??) for holding jigs and fixtures.

A plate with holes in should work well. But I don't think the WC spindles have enough torque to tap directly? However, you can try thread milling using a special cutter (~£40), plus your CAM will need to be capable of a spiral cut with user parameters on the pitch:

Forward to 12:23 to see thread being cut:
https://www.youtube.com/watch?v=a43S2y7Ccy8

JAZZCNC

02-11-2015, 02:14 PM

My latest thought is to bolt down a 20mm aluminium plate to the top of the steel raised area (shown in photo). I will then use the machine to skim the surface flat. Then I will use the machine to drill a grid of holes which I can then tap (maybe the router could do this??) for holding jigs and fixtures.

Can't imagine that plate is flat enough to be useful and could cause you more trouble than it's worth.?

Personaly I would just buy Machine tooling plate that is ground perfectly flat. Put your Profile etc directly on this and then shim out any error using the nice flat surface for reference. No need to skim flat. Then you can drill tap matrix of holes. Forget tapping with router type spindle no where near enough torque or control to do this unless thread milling. It's nothing of job to do with spiral flute tap in a Cordless drill.

biketrialsdave

02-11-2015, 02:17 PM

Can't imagine that plate is flat enough to be useful and could cause you more trouble than it's worth.?

Personaly I would just buy Machine tooling plate that is ground perfectly flat. Put your Profile etc directly on this and then shim out any error using the nice flat surface for reference. No need to skim flat. Then you can drill tap matrix of holes. Forget tapping with router type spindle no where near enough torque or control to do this unless thread milling. It's nothing of job to do with spiral flute tap in a Cordless drill.

Just to clarify...use the machine tooling plate as the bed? The problem I see is they are very expensive? Any links to something appropriate?

Cheers,

Dave

JAZZCNC

02-11-2015, 03:11 PM

Just to clarify...use the machine tooling plate as the bed? The problem I see is they are very expensive? Any links to something appropriate?

Cheers,

Dave

Try these people. I deal with Lincoln branch but believe they have place in WH that will give you a quote and you'll find the price of Machine tooling plate isn't that much more than normal plate. Being Cast plate it's much more stable and being ground it's obviously flat so is worth the extra IME.!
http://www.ascmetals.com/

Here's Some I prepared earlier.!!

16414

biketrialsdave

02-11-2015, 03:50 PM

Thanks Jazz! Would I get away with 15mm?

I think I need to sit down and consider reducing the size of my machine. I don't envisage (famous last words) doing any large wood projects etc. A bed size of ~400mmx400mm would be more than big enough. Costs would be reduced across the board and I would have more room to manoeuvre around the machine in the shed! The biggest downside I see is that I already have THK linear rails I purchased a few years ago which would need cutting down. Seems a shame to cut them (also, I'm not sure how easy that would be!?).

JAZZCNC

02-11-2015, 04:07 PM

Yes some times best to sit back and re-think before jumping in.

Ask your self what's the main purpose of the machine going to be.? Be realistic and honest with your self. If it's mostly wood then you don't need go to the trouble of expensive plate.
Your original design sat on nice flat stable surface will work ok for most woods, plastics etc. If it's aluminium then yes the strength and accurecy is needed so the costs rise.

Regards cutting rails then it's always something to avoid really purely on being a waste. The actual cutting is easy. All you need is angle grinder with 1mm thin cutting discs and you'll whiz thru them.

Been said many many times on forum with lots of examples and this is think towards Steel if you want cheap and strong. But slight warning here.!! IME trying to make something from the scrap bin work, like what you found, always ends in a compromise of some kind and in the grand scheme often doesn't save you much money long term so be weary of this appealing route.!! (That said Steel offcuts etc are ok just avoid structures for bases etc)

biketrialsdave

02-11-2015, 05:44 PM

Agreed. I've always struggled with being a perfectionist and over-designing things!

While I have your very valuable attention, do I need to go with "Heavy" extrusion over "light"?

biketrialsdave

17-11-2015, 05:26 PM

Booked tomorrow off work so hoping to spend some time on this now I've got some other projects out the way.

I've been looking at stepper drivers too. Think I'm going with the AM882/EM806. It looks like Aliexpress is the place to get them, just need to find a "reputable" seller... On the subject of buyign things, I can't seem to find the bloke in China who was the go to person for ballscrews etc? His eBay shop was something like "linearmotionbearings2008" ???

Clive S

17-11-2015, 06:02 PM

You could also try Fred at BST http://es.aliexpress.com/store/314742 email him direct for the best price

Chia website is http://stores.ebay.com/linearmotionbearings

JAZZCNC

17-11-2015, 07:37 PM

Go with BST slightly better quality and great to deal with.

biketrialsdave

18-11-2015, 09:21 PM

JAZZCNC

18-11-2015, 10:06 PM

Thanks guys!

Quick question (I'll post an update later)....I'm going for HTD 5mm pitch pulleys (as suggested), what ratio should I start with for x and y axis? Also, I assume I want a larger diameter pulley for more teeth/belt contact as opposed to a smaller one?

If using typical router setup with 10mm pitch screws and steppers then you don't have a ratio so it's 1:1.
Common pulley size is 20T. Don't go too small because there's not enough teeth engaged. Don't go too large because you increase screw inertia which affects acceleration and power required.

The main reason for using pulleys is to lessen resonance which affects steppers. The abilty to apply a ratio is just a bonus.
One advantage if usng say 10mm pitch is that you could apply 2:1 ratio and increase torque and resolution if ever required. Or go other way and apply ratio if you ever needed more speed but with 10mm pitch you won't anyway thats why 1:1 ratio is often used.

Same with Z axis keep 1:1 ratio but use a 5mm pitch screw for better resolution/torque.

biketrialsdave

18-11-2015, 10:22 PM

Just the information I was after, all makes perfect sense. I can't thank you enough. Will have to send you something your way when I'm done!

JAZZCNC

18-11-2015, 10:58 PM

Just the information I was after, all makes perfect sense. I can't thank you enough. Will have to send you something your way when I'm done!

Don't be daft it's what we do and why we are here to help each other.!. . . . just post pics and vids. .

biketrialsdave

19-11-2015, 10:59 PM

Still tons of work to go but a few pics of the latest design... I went back and reduced the overall size fairly substantially as I realised I didn't need it. I'm a lot happier with the general direction now.

165701657116572

routercnc

19-11-2015, 11:20 PM

Coming together but . . .

the spindle hangs a long way out in front of the gantry - what's happening with the Z axis? Can you show a side view of the Z axis ? Is it profile rails on Z - looks like they are very close together and possible you are not pocketing the ballscrew etc. therefore giving large overhang.

Stiffness drops off significantly as you move out from the gantry so worth getting it more compact if possible.

JAZZCNC

19-11-2015, 11:25 PM

Dave drop that Linear stage for the Z axis it's massive weak spot on machine.

I'd also look to beef those plates holding motors etc but other than those 2 things it's coming together nice. .:applause:

Edit: didn't see Rout's post same thing really.

One other thing regards the Drop bracket on the Y axis at rear it needs more support from a brace.

biketrialsdave

19-11-2015, 11:28 PM

I know what you're saying, that is a concern of mine. At the moment I'm using a thk kr33 assembly as I bought it a while back. However, I am planning to use the excess thk linear rail that I'll have left in the future. Perhaps I should design that in now...

JAZZCNC

19-11-2015, 11:43 PM

Perhaps I should design that in now...

No perhaps about it Dave needs to be done. The Z axis is KEY area don't matter how well built rest of machine if Z is shite or weak.!!

biketrialsdave

20-11-2015, 12:25 AM

No perhaps about it Dave needs to be done. The Z axis is KEY area don't matter how well built rest of machine if Z is shite or weak.!!

OK, I'll get onto that tomorrow! I assume 2 linear rails with the ballscrew in-between them sitting in a slot to reduce the overall depth? I'm using THK HSR15 carriages for the other axis, I assume these will be OK for the Z too?

The drop brackets on the rear are 5mm steel at the moment. I can increase this and also add some kind of right angle bracket perhaps.

cropwell

20-11-2015, 09:18 AM

Lee - How do I delete a reply I made ?

Neale

20-11-2015, 11:33 AM

OK, I'll get onto that tomorrow! I assume 2 linear rails with the ballscrew in-between them sitting in a slot to reduce the overall depth? I'm using THK HSR15 carriages for the other axis, I assume these will be OK for the Z too?

Even if you don't really need them for strength, this is one place where "20" rails and carriages can be useful to give just a bit more clearance between Z plates for the ballscrew and nut. If you have to buy them and don't have something in stock already, this might be something to consider.

JAZZCNC

20-11-2015, 02:47 PM

OK, I'll get onto that tomorrow! I assume 2 linear rails with the ballscrew in-between them sitting in a slot to reduce the overall depth? I'm using THK HSR15 carriages for the other axis, I assume these will be OK for the Z too?

Yes more than enough regards forces etc but has pointed out by neale 20mm can work better regards spacing etc. Personally I use 20mm on all axis for simplicity and hardly any difference in price.

The drop brackets on the rear are 5mm steel at the moment. I can increase this and also add some kind of right angle bracket perhaps.

Yes you need more bracing to stop deflection. 5mm steel isn't really upto to it without being braced.

routercnc

20-11-2015, 09:00 PM

I have 15mm rails on Z and they work ok but they use M3 bolts on the rail and carriages which look very small compared to M5 on the 20mm rails on my Y axis. When cutting I can feel vibration across them especially the top ones and for this type of machine I would go with 20mm on Z.

biketrialsdave

21-11-2015, 04:19 PM

Thanks for all the feedback guys. I think I'll go with Hiwin 20mm for the Z axis. What would be the "minimum" recommended centre distance between the two rails? I want to keep this as small as possible to maximise the effective Y axis length but without compromising rigidity. At the moment my whole Z axis assembly is 130mm wide..

Lee Roberts

21-11-2015, 10:30 PM

routercnc

21-11-2015, 11:14 PM

My Z axis plate is 147mm wide and the Z rail centres are only 115mm apart. I did this to maximise Y travel and placed my Y axis bearings pretty much back-to-back.

This works fine for me, but it can give you more stability and less likely to rack if the Y bearings are spread further apart, but you are trading Y travel. If you do spread the Y axis bearings out further, closer to Lee's figure, then I don't think you need to follow suit with the Z rails and can keep them fairly compact.

16583

16584

16585

biketrialsdave

22-11-2015, 01:45 PM

Your machine looks great, thanks for all the advice.

I've increased the plate width to 170mm now. The z rail centres are 118mm apart at the moment. Do you think the carriages need to be separated more in the z axis? (They are 119mm centres) I'm trying to reduce this length as much as possible in order to reduce the total height of the z axis plate (if that makes sense.

Also, is it worthwhile pocketing the carriages slightly like I've done elsewhere?

16587

JAZZCNC

22-11-2015, 02:16 PM

My rear plates are between 180-200 wide which I find is good balance between strength and space saving for a small machine. On larger machine I will try to make wide has possible to get most ridgidty. Carriage separation I make wide has possible and take the bearings to the edges of plates. Vertical Separation is roughly half the rails length.

Regards the pockets then it depends on the material your using.? If your using Ground cast plate then no because it defeats the object of the plate being ground. If just rolled Plate then yes it needs to be machined because it's often not flat.
If using Ground plate and you want registers for the bearings then Mill some Aluminium blocks or dowels into the plate inline with the edge but slightly overhanging the bearing side then secure with loctite.
Then Machine away the overhanging part to give your reference edge but set cutting tool flush with ground surface so not cut away. In practice it's actually much quicker than buying rolled plate and machining away pockets and more accurate.
Personally I tell people not to bother unless they are sure the whole machine is accurate from base upwards because they make adjustment much harder and in flexible.
In an idle world then yes you'd have them but the reality is that building in your Man cave then in-accurecy is sure fire going to be there so any thing to help adjust error out is welcome.
Also in DIY setting then it's unlikely you'll be chasing Um accurecy and neither will you be stripping it down every other week. So provided you take care when setting up and use good quality fasteners, tighten them up properly and use retainer then you'll not have any troubles and chasing away error will be so much easier.

One potential Gotcha I spotted on your drawing. The holes in the Top and bottom plate which fasten them to the rear plate may interfear with the bolts on the Gantry bearings.? Easy done so check it.!

biketrialsdave

22-11-2015, 06:33 PM

Some progress. Am I going in the right direction? My only concern is I've had to mill a channel to allow the BF12 mount to not interfere with the plate holding the motor. Where the slot is the thickness of the plate is down to 10mm. However, its quite a narrow channel...

165881658916590

JAZZCNC

22-11-2015, 06:59 PM

Split the difference by machining both sides like this.
16591

biketrialsdave

27-11-2015, 12:06 PM

I've been busy all this week so not had time to work on this. However, I've sneakily booked Monday off so I can get some more time on it! Instead of constantly asking questions, I've tried to think of all the main concerns I have at the moment. I would be very greatful if anyone can answer a few!

1. I need to work out how I am going to attach the extrusion (bed) to the aluminium tooling plate. I think I have two options: a) Either drill the tooling plate and insert bolts from the underside which screw into t-nuts in the extrusion. b) Or use aluminium corner brackets on the inside of the extrusion?

2. I think I need to assess how I am connecting the extruded pieces to eachother (for the bed frame). The X axis extrusion is actually comprised of a 45x45 piece and a 45x90 piece ontop of eachother. Unfortunately I cant seem to find a single 45x135 piece? At the moment they are only attached together by the steel end plates. Do I also need some internal corner brackets? I was also wondering if to add some plates bolted to the side of the 2 extrusion pieces?

3. How should I re-inforce the steel end plates? Would it be a case of simply increasing the plate thickness from 5mm? The end plates don't hold the ballscrew bearing houses (these are attached to the extrusion). The plates at the back of the machine do, however, hold the stepper motors.

4. Is it possible/recommended to add a vaccum to the z axis for swarf? Is this only to be used for wood? I am also going to add brackets for a coolant hose.

5. I need to consider draining the coolant. Could this simply be a hole cut into the tooling plate with a pipe which goes through the table beneath into a bucket? My only concern is I don't have a "fall" for the water.

Hopefully those all make sense!

Thanks again,

Dave

JAZZCNC

27-11-2015, 04:31 PM

#1 Drill the Plate and screw from under side.

#2 Drill thru the upper extrusion and bolt into lower T-slots. Can add some Blocks into the slots to stop them sliding.

#3 If just bracing then 5mm steel will be just about ok, But personally I'd use 9.5(3/8")mm Aluminium so don't have hassle of painting etc plus steel on ali is not good mix.

#4 Can do it but it's difficult and not very effective unless high vacuum. Much better with Blown air and minimal Mist to clear/Lube chips.

#5 Simple holes at one end and angle the machine on bench using feet.

biketrialsdave

28-11-2015, 02:24 PM

Hmm I may have to re-consider the steel then. The cost of getting all this aluminium machined is going to be a bit eye watering!

On a side note, what would you recommend as the minimum internal fillet radius? Or should I leave internal corners as right angles and assume it will be filleted to whatever when machined?

biketrialsdave

18-12-2015, 09:14 AM

Not much progress with the machine at the moment... I've gone off on a tangent with designing an Ethernet motion controller for Mach 3 (don't ask how that happened!)

On a side note, I also have a job request up for yet another project I'm currently working on: http://www.mycncuk.com/threads/9328-RFQ-Very-Simple-Machining-of-Two-Aluminium-Parts-Required

biketrialsdave

27-12-2015, 11:21 PM

Well the time has finally come... I THINK I'm finally finished! :eagerness:

I've decided to keep the steel end plates. I can get them cut for free (which is a big plus) and I'll paint them to help prevent bi-metallic corrosion.

16920169211692216923

routercnc

28-12-2015, 12:20 AM

Looking good now.

The X axis stepper motors have the adjustment slots horizontal - will this actually give you much belt length adjustment? Any reason the slots cannot be vertical?

biketrialsdave

30-12-2015, 12:01 AM

Looking good now.

The X axis stepper motors have the adjustment slots horizontal - will this actually give you much belt length adjustment? Any reason the slots cannot be vertical?

Good spot! I just checked and only have 3mm of belt length adjustment... I'll make the slots vertical!

Clive S

30-12-2015, 12:18 AM

Have you thought about just using one belt and one motor and drive, I use two motors but quite a few on here use one as it is easier to home and stops the problem of racking etc. Just a thought!

biketrialsdave

30-12-2015, 12:25 AM

Have you thought about just using one belt and one motor and drive, I use two motors but quite a few on here use one as it is easier to home and stops the problem of racking etc. Just a thought!

I haven't but I suppose I could easily use one keeping the same design. In fact, I'm not sure how I ended up with two motors in the first place...

Clive S

30-12-2015, 09:46 AM

I haven't but I suppose I could easily use one keeping the same design. In fact, I'm not sure how I ended up with two motors in the first place...There are pro's and con's for the two systems I don't think there is much difference in cost between the two.
.
With two motors you will need two homing switches and a controller that can do the slaving correctly and also have the alarm outputs on the drives working in case one motor stalls and then the other will rack the system.
.
Dean (Jazzcnc) usually uses one motor but I think on the machine he is now using it has two.
I use two motors and I am happy with it.

JAZZCNC

30-12-2015, 11:27 AM

Dean (Jazzcnc) usually uses one motor but I think on the machine he is now using it has two.
I use two motors and I am happy with it.

Yes it has but only because I inerited the machine it probably wouldn't if I built it my self.! . . . Thou It's just on the edge of too big for belts so maybe not.?

On large machines 2 motors make more sense because the belts get messy and custom lengths. At this size and with this design then I'd 100% be using belts.
It's only slightly cheaper in cost but much more reliable and stable in use. Can make full use of the machines potential and not worry about screws running out of sync with each other.

If using 2 motors then the ONLY way I'll use them is if the drives have Stall detect with fault output to E-stop the machine ie: EM806 or AM882 . Also external motion control card is prefered due to higher work load. (Personaly I won't use the Parallel port on any machine)

biketrialsdave

30-12-2015, 01:23 PM

Clive S

30-12-2015, 01:40 PM

Hmm maybe I will stick with one motor and use a single large belt. I don't think it will require much modification (whilst still having the potential to run 2 motors in the future). Just need to make sure I can tension the belt well. I'm thinking of a simple idler pulley which I can slide in a slot.You will need to make sure there are enough teeth engaged on the pulley with the belt. Also if you want to put material through the back of the machine the belt would have to be routed accordingly.

biketrialsdave

30-12-2015, 07:21 PM

Something like below? The bottom idler pulleys on both sides sit in slots for adjusting tension.

1695216953

cropwell

30-12-2015, 07:49 PM

You don't need to spend money on pulleys just to use them as tensioners.

JoeHarris

30-12-2015, 08:53 PM

What ability have you got for machining the aluminium? If only basic tools like me then you could probably lose some of the pockets. Would help with adjustment later too ?

JAZZCNC

30-12-2015, 08:59 PM

Too many idlers drop try to keep to a minimum you'll get away with just one. Belt will be above the profile but doesn't matter because you'll want to cover it up anyway.

16969

biketrialsdave

31-12-2015, 08:07 PM

I only went for more idlers to keep the belt beneath the extrusion...maybe I'll have a re-think.

On a side note, starting to look at my electronics setup. I think these motors look reasonable?: SY60STH86-3008*

biketrialsdave

01-01-2016, 12:08 AM

I've gone with your suggestion Dean but kept the slots in the plates so I've got flexibility in the future.

The cover needs work but looks like this now. It's starting to feel like the ends in sight!

1699616997

biketrialsdave

01-01-2016, 12:17 AM

What ability have you got for machining the aluminium? If only basic tools like me then you could probably lose some of the pockets. Would help with adjustment later too ?

Hi Joe,

I'm going to put an RFQ up to get all the pieces machined. If only I had a mill to start with!

JAZZCNC

01-01-2016, 12:48 AM

biketrialsdave

01-01-2016, 01:51 AM

Word of warning before commiting fully to this route.? Make sure you can buy a long enough belt of the shelf other wise it means having custom length belts made. Not a Big issue but still worth thinking about now.

The available belt lengths may also mean changing the location of the tensioners so work out the belt length needed for how you have it drawn now and see if belts are available that will work.

Oh Happy new year.!

Belt comes in at ~1.56m. Ive used this place a few times and they have belts up to 2.5m:
http://www.beltingonline.com/5mm-htd-timing-belts-4575?zenid=oak0bv86qnh7ifo3qcvkih53b6

Happy new year everyone! :)

JoeHarris

01-01-2016, 01:56 AM

Happy new year chaps

JoeHarris

01-01-2016, 09:50 AM

reading this and the recent thread on controllers is tempting me back to the 3 stepper and belt option! I can feel another redesign coming on for 2016. Good luck Looking forward to seeing this project become a reality...

JoeHarris

01-01-2016, 01:31 PM

Too many idlers drop try to keep to a minimum you'll get away with just one. Belt will be above the profile but doesn't matter because you'll want to cover it up anyway.

16969

Jazz, is there enough engagement on the top right pulley? What's the minimum? you could add one more like this?? I 17005

Jonathan

01-01-2016, 02:56 PM

Jazz, is there enough engagement on the top right pulley? What's the minimum? you could add one more like this?? I 17005

The spec. for HTD belts say make sure to have 5 teeth engaged. This is based on ensuring that with a sufficiently high load, the belt will snap instead of the teeth ripping off. This hardly matters to us, as we're not using the belts at close to their load ratings, however we do want to get good stiffness from the belts (especially with such a long best) so it seems a reasonable baseline to design for at least 5 teeth engaged.

biketrialsdave

01-01-2016, 11:36 PM

Thanks for the info. With a 20T pulley I should have around 5 teeth engaged with what looks like a quarter of the pulley.

On a side note, I am currently looking at my bed plate design. I have put in a recessed hole to insert a standard kitchen sink waste hole. Is this OK or is there a preferred method? Also wondering if to move it to the corner and tilt the machine towards it...

1703117032

biketrialsdave

10-01-2016, 11:48 PM

cropwell

11-01-2016, 12:59 AM

You could have a look at Estlcam, the demo is free, and if you like it, it is only 50 Euros or so to licence it. It has a lot of features and I have just done my first job with it. There are video tutorials on t'Interweb. There are some aspects I don't like - I found it easier to start again from the dxf file if I messed up a toolpath (rather than deleting the toolpath - which seemed impossible) Maybe I will get better at using it, but if I find foibles which need to be sorted, I will address them with the author (Christian Knuell). In my correspondance about licensing, he seems happy to help.

Cheers,

Rob

Clive S

11-01-2016, 10:07 AM

I've made some big progress over the past few weeks. Hopefully some photo's to come once a parcel arrives this week! :)

In the mean time, I'm starting to think about CAM software. I've looked at BobCad and CAMBAM. Not looking to spend an awful lot (for now). What would be everyone's recommendations for an easy to use package that will give me a good introduction to CAM?Fusion 360 has been banded abut on the forum recently it is a cad and cam package and I believe free for a year and then you only have to register for hobby use.

biketrialsdave

11-01-2016, 11:51 AM

Fusion 360 has been banded abut on the forum recently it is a cad and cam package and I believe free for a year and then you only have to register for hobby use.

Ooh this looks very interesting!

http://www.autodesk.com/products/fusion-360/overview

njhussey

11-01-2016, 12:40 PM

I've got CamBam and EstlCam and both have their merits for me. If you've already drawn your object as a dxf and it's only simple machining then EstlCam is great as you don't have to join all the different lines that make up your shape. If you want to just draw something simple and then do the CAM on it then CamBam is good. I don't do anything complicated so can't comment for 3D shapes but for simple things (parts for RC planes, plastic thrust bearings, simple plastic parts etc.) both programs are easy to use and good.

biketrialsdave

27-01-2016, 01:43 PM

I've not posted for a while but a lot has happened!

* I've received all the custom aluminium parts
* Ordered stepper motors and AM882's
* Ordered all ballscrews and linear rail for z-axis

Just need to make some more money before I can continue the spending! I'll put some photo's up soon...

JoeHarris

27-01-2016, 02:30 PM

biketrialsdave

29-01-2016, 09:11 AM

It certainly is like Christmas in my house at the moment...the girlfriend isn't too impressed with the number of boxes and smell of machine oil in my office though!

I am starting to get a little worried that my machine is going to be incredibly heavy. The parts for the Z-axis assembly alone weigh a ton! (Not literally, I am going to weigh them later). I just hope my 3Nm steppers are up to the job...yes, I know I should of done the calculations beforehand but I got carried away with buying shiny things!

JoeHarris

29-01-2016, 09:57 AM

njhussey

29-01-2016, 10:01 AM

biketrialsdave

29-01-2016, 05:39 PM

I've got a sodding great big 40mm front plate that my 2.2kw spindle hangs off and my 3Nm stepper whizzes it up and down no problem so you'll be fine!

That's re-assuring. I've just done some quick calcs and it looks like the steppers will be more than up to the job. Of course, it was an educated guess! :emmersed:

biketrialsdave

30-01-2016, 07:48 PM

More big steps today...cleared a big space in the shed and measured up ready to build my workbench/stand!

On a side note, what method does everyone recommend for finishing aluminium parts? Some of the larger parts I had machined for my router are not milled on both sides and so are dull and have the original "stock" finish (some have lettering etc). I want a finish which is durable and ideally helps oxidising. Anodising would be ideal but I think that's going to be a little price and not something I can do myself? So this leaves me with: Polishing (if so, what's the best way and will the finish oxidise over time?), spray painting, powder coating, other plating??

Thanks in advance!

JAZZCNC

30-01-2016, 08:06 PM

biketrialsdave

30-01-2016, 08:11 PM

It's bloody tool not a Pimp My Ride.!. . . :joker:. . . . . . Which is bit rich coming from a Bloke who's spent quarter the day Plastic coating parts for sexy little router he's finishing off. . .:hysterical:

Pimp my CNC!

:beguiled: I think I'll use some 1000grit sandpaper then see if I can replace the grinding wheel on my cheap Aldi bench grinder with a polishing one.

Wobblycogs

30-01-2016, 09:03 PM

I've polished a few bits of aluminium by starting with about 220 grit and a random orbit sander then going up the grits as high as possible before finishing by hand. I don't really think it's worth it though as aluminium is so soft it scratches again in five minutes. I suppose you could spray a lacquer on it maybe.

biketrialsdave

30-01-2016, 10:19 PM

Hmm I think I'll do a 'quick' polish but nothing overboard.

While I'm thinking about it, should I be using threadlock on all my parts? Or will torquing to the correct value be enough?

Also, do I need to remove all the machine oil from my new ballscrews before using them?

biketrialsdave

04-02-2016, 11:57 PM

Ahhhhh panic!!!

I thought things were going too well... I don't know how I've missed this but the 45 series rexroth extrusion has 10mm diameter holes for M12 screws on the ends. For some unknown reason I've designed my aluminium gantry sides for M6!!! I've seen you can get an M8 "cutting bush" which reduces the bore. Would my best option be to use these and get the side plates drilled out to M8?

=(

JoeHarris

05-02-2016, 12:03 AM

biketrialsdave

05-02-2016, 12:12 AM

It would be much worse the other way around! Why don't you just increase the size of the holes on your end plates to take m12? Or am I missing something?

I could do this but M12 socket caps look massive and don't leave much material thickness. Perhaps I might have to compromise with M12 button head bolts and not countersink them?

JoeHarris

05-02-2016, 12:22 AM

What plate thickness have you got?

JoeHarris

05-02-2016, 12:24 AM

I have counterbored m12s into 20mm plate. The 8mm left between the head and the extrusion is plenty enough.

biketrialsdave

05-02-2016, 12:29 AM

JoeHarris

05-02-2016, 12:33 AM

I think that looks like the best solution. There 20mm so 8.4mm thickness left according to solidworks. Now the question is, do I attempt to drill them out and re-bore myself or send them back and get them re-done professionally? I'm worried about the drill bit wandering off from centre.

17511

Well that's up to you! My build has been decidedly diy so I'd have to do it myself! I would have thought the bit would find its centre in the existing hole...

JoeHarris

05-02-2016, 12:38 AM

17512 these were drilled out with several bits starting small and working up...... You would just be starting further down this process?

biketrialsdave

05-02-2016, 12:40 AM

They look very good! I assume you used a drill press? (mine is an el cheapo one and has a bit of play in the head). Did you use any kind of guide for the bit?

JoeHarris

05-02-2016, 12:42 AM

No, no guide and my press is only a table top Axminster thing so nothing pro! 17513

JAZZCNC

05-02-2016, 12:47 AM

I think that looks like the best solution. There 20mm so 8.4mm thickness left according to solidworks. Now the question is, do I attempt to drill them out and re-bore myself or send them back and get them re-done professionally? I'm worried about the drill bit wandering off from centre.

Minor detail not worth stressing over. Domed head look fine without being counter bored so don't bother just drill them out to 12mm.

17515

JoeHarris

05-02-2016, 12:49 AM

Wow Jazzy! Pimp my ride eh?!
Fair point but I have to say counterboring is strangely pleasurable!!

JAZZCNC

05-02-2016, 12:52 AM

Wow Jazzy! Pimp my ride eh?!
Fair point but I have to say counterboring is strangely pleasurable!!

Bling Bling.!!!!

Normally I do counter bore Joe because it's even easier when Cnc machine does it for you but In this case budget was tight and machine is small so 15mm plate was used so not much room to play with and it's only cosmetic anyway so no matter.!

JoeHarris

05-02-2016, 12:53 AM

Also - if you counterbore you could cap the holes with diamonds to complete the look!!

biketrialsdave

05-02-2016, 01:23 AM

I was wondering if a step drill (13mm max diameter) would be a better idea for enlarging as it would be self centering?

JoeHarris

05-02-2016, 07:26 PM

I was wondering if a step drill (13mm max diameter) would be a better idea for enlarging as it would be self centering?

Step drills are intended more for sheet metal. I would get a decent hss twist bit. You can pick up dormer bits on eBay for very little. http://www.ebay.co.uk/ulk/itm/140928964799

Lee Roberts

05-02-2016, 07:40 PM

I think you'll be fine, what may happen is you get chatter from the drill so may end up with the not so perfect edge.

Also you don't really need to buy proper counter bores, the right size drill bit used in the same way works great for pretty much the same effect, do you really need need flat bottom holes for the cap head to live in?

The timing is good on this one because I'm currently in the process of doing the same myself, stepping up in size, cheap drill press, simular sizes being dealt with, here's some pics:

http://uploads.tapatalk-cdn.com/20160205/79d20504b35fdf30b0ad7f86068b05ad.jpg

http://uploads.tapatalk-cdn.com/20160205/3d4e6bc0c491ef597cd13870c31fa3bf.jpg

http://uploads.tapatalk-cdn.com/20160205/4b498a39d859f16e144b0dc0fd0c129a.jpg

I'm stepping up to 15.5 and 20mm for the screw clearance hole (the very middle one) :). I can show the bigger holes if you really want but they came out pretty decent, don't use a step drill and if you don't mind go for quality drill bits.

If you do end up with a less desirable finish on the bore OD, you can always add a chamfer with a counter sink or again an over sized drill bit.

http://uploads.tapatalk-cdn.com/20160205/7f3cead5b955a7754893c8b5690b89f0.jpg

Edit: Just to add, I prefer to go with Dormer for drills and taps after using them after cheap crap and the difference is night and day, however even on eBay Dormer comes at a price, check out Osborn/Warwick/Heller/Goliath, look up Steve (cncpoorboy) on eBay he's got some smashing deals for what what looks to be old new stock, new stock and the stock is all those shiny metal things we're so attracted to.

.Me

JoeHarris

05-02-2016, 07:59 PM

Lee Roberts

05-02-2016, 08:11 PM

An m12 counterbore will cost the same if not less than the equivalent bit, so I would do the job properly... http://www.arceurotrade.co.uk/Catalogue/Cutting-Tools/Counterbores/HSS-Straight-Shank-Counterbores
I've added an edit Joe...

Define properly, plus a counterbore only dose one job, a drill offers you more options/value if you don't mind being a little creative with it.

Not getting into right or wrong as I think it's pretty obvious and such a simplistic thing being discussed, just saying what I did/do for these things.

.Me

JoeHarris

05-02-2016, 08:39 PM

biketrialsdave

05-02-2016, 08:42 PM

Fair enough! I guess you always have a choice whether to go Heath Robinson or cough up for the tool which has been designed for the job. It's no big deal. I've been milling aluminium with a router because I can't afford a mill! It depends where you draw the line. I have just developed an irrational love of counterbores since buying a set recently and want to spread the joy!!

Which set did you buy? There actually seems to be quite a variation in counterbore sizes which I find odd. Some that don't even fall under the category of normal or close fit...I wonder if the dimensions are just rounded on some websites...

JoeHarris

05-02-2016, 08:48 PM

Yes I found that. I bought my m12 from arc euro and a set of Ruko m3-m10 on eBay but only because I spotted a good deal, can be quite pricy otherwise.

biketrialsdave

10-02-2016, 10:44 AM

Think I have the bolt problem sorted now...will update soon. Now on to the next problem...

I was playing with my ball...screws last night before bed ;) I've noticed that the Z-axis ballnut (RM1605-C7) seems a little notchy and not smooth. As if there could be some swarf or something inside? The ballscrews are covered in a light machine oil (I assume to prevent oxidisation). Do I needto strip down and clean everything before use? Do the ballnuts require greasing first?

On a similar note, is there any specific preparation I need to do to my linear rails/bearings before use? (I need to assemble the HIWIN ones).

Cheers,

Dave

Davek0974

10-02-2016, 10:51 AM

On my HiWin, I followed the manual, I used light grease and worked it in slowly by running the carriage up and down by hand and giving a little grease each time, this is done on the rail of course ;)

biketrialsdave

15-02-2016, 12:45 PM

Hi all,

More progress has been made and my bank account is starting to hurt! I sorted the bolt problem, counter-bored the plate for M10 socket caps and then heli-coiled the extrusion to match.

I am taking photos as I go so I think I will do a big dump of information in one go rather than little updates every so often.

I have started to really think more about swarf and coolant removal during/after machining. Hopefully the paint diagram below will help explain what I'm thinking about doing. The gap will allow me to brush swarf straight off and into a bucket under the bench. Would the proposed design weaken the frame too much?

17605

EDIT: Infact, could I even remove that extrusion completely or is it going to put too much outwards pressure on the sides?

17606

biketrialsdave

17-02-2016, 10:27 PM

Very productive day today.

Sneaky peak...

1761717618176191762017621

sinnsvak

17-02-2016, 10:51 PM

Looking good!

Clive S

18-02-2016, 12:07 AM

Coming along nicely keep up the good work you are getting there:yahoo:

JAZZCNC

18-02-2016, 02:56 PM

Looking good Dave keep it rolling mate. .:applouse:

JoeHarris

18-02-2016, 08:31 PM

Nice work Dave. Don't tell me you didn't enjoy that counterboring?!

biketrialsdave

18-02-2016, 08:40 PM

It was very satisfying. Made a big mess though!

biketrialsdave

19-02-2016, 09:46 PM

Getting there slowly!

17636176371763817639

JAZZCNC

19-02-2016, 09:53 PM

Getting there slowly!

Like eating an Elephant Dave.? . . .Small bites at time then before you know it your eating desert. . :triumphant:

JoeHarris

19-02-2016, 09:56 PM

And desert is the electronics?!

biketrialsdave

19-02-2016, 09:59 PM

And desert is the electronics?!
Haha, well that's my day job so fingers crossed that will go OK! Been having thoughts of designing a BOB but need to not get side tracked!...

JAZZCNC

19-02-2016, 10:38 PM

And desert is the electronics?!

No frame is the Meat. Electronics are the Veg and Chocolate Chips are the desert.!!

Dave don't try to reinvent the wheel it's not worth the effort.!

routercnc

19-02-2016, 10:56 PM

Nice work Dave. Solid design and well made. You will be rewarded soon . . .

Lee Roberts

20-02-2016, 12:30 AM

Nice work Dave. Don't tell me you didn't enjoy that counterboring?!

lmao! ;-)

.Me

biketrialsdave

12-03-2016, 10:13 PM

I'm getting there! Will do a full write up at the end (if there is such a thing! ).

I need some help if I may. On the x axis (long axis) when the gantry approaches either end of the bed, the ballscrew gets tighter. It's not just one side, happens to both. Something must be bowed?179121791317914

biketrialsdave

12-03-2016, 10:22 PM

And could anyone explain what tramming is please?

routercnc

12-03-2016, 10:49 PM

If the ballscrews are bent then you will get a periodic tightening as you rotate it, especially at either end of travel.

If the ballscrews / mounts / nuts are all mis-aligned then it will have a constant tightening at either / both ends.

Assuming the ballscrew is fairly straight then looking at the design the ballscrew supports and ballnut bracket are probably not completely in line. It doesn't look like you can adjust in the plane away from the sides of the machine, only up and down.

Try moving the machine to the driven end, loosen the 6 bolts holding the ballnut to the ballnut bracket, and remove the floating end bearing. This should allow the ballnut to sit in an unloaded position.
Nip the bolts back up and move the gantry over to the other end. Gently slide the floating bearing back on - it should sit snug onto the mounting block on the side of the gantry. If it is out, fit a shim. If it is tight you need to machine a bit off the mounting block.

Tramming is related to getting the spindle motor normal to the bed so that the side of the cuts are 90deg to the top surface. You can get close by putting a bar in the spindle and using an engineers square to check the spindle is 90deg to the surface on all sides / angles. You can also get a twin DTI block which gives a much better indication of tram condition. Have a read . . .
http://www.cnccookbook.com/CCMillTramming.html

biketrialsdave

19-03-2016, 02:58 PM

I think I've solved the ballscrew problem, re-made one of the BF mounting blocks and shimmed it a little. How hard should it be to push the gantry back and forth on the machine? It takes quite a lot of force to get it going then it still requires a far amount of force to push it...

The mechanical build is nearly complete, just got the back gantry cover to finish!

179631796417965

Davek0974

19-03-2016, 03:15 PM

My one slides fairly easy with the belts off.

Did you try loosening the bolts on the BK & BF then moving the motion to the bK end and nipping them up then move to opposite end and nip the bolts up - there should be no tension on the screw when the bolts pull up.

This worked wonders on my build.

routercnc

19-03-2016, 05:42 PM

How hard should it be to push the gantry back and forth on the machine? It takes quite a lot of force to get it going then it still requires a far amount of force to push it...

If you mean with the ballnuts disconnected from the gantry then the gantry should move with one finger. A light shove should see it carry on for a bit and coast to a halt.

If it is hard to move and stops as soon as you stop pushing then either bearings are faulty (unlikely if new) or more likely is they are misaligned relative to each other.

They are very intolerant of misalignment in any plane. This is why epoxy is popular. One side could be higher than the other side, or they could be the same height but twisted, or they could not be parallel etc. etc. I does not take much to make them bind.

Did you set the first rail against a known straight edge?
Did you then use a DTI to set the other rail parallel to it?
Are they at the same level as each other (use straight edge across bearings and check with feeler gauge)

Loosen the bolts connecting the gantry to the bearings on one side, then see if it improves. If it does you have misalignment.

Time spent getting this smooth will pay dividends in the long run.