PDA

View Full Version : Newbie CNC Build



JunkieHobbo
07-03-2012, 07:00 PM
Hi Guys,

Long time lurker here, I have finally got to the point where I want to start my CNC Build. So before I start I would like to get some advise from the already experienced members of the Forum. I have many questions about the Build and also the Electronics.

First the Build:

The CNC will be made out of Steel as that is what I have access to a fairly low cost, I was looking at Aluminium Extrusion but the Price was Prohibitive. Also the other reasons I decided on Steel is for the rigidity as one of the materials I would like to work on is Aluminium. The downside of going with Steel that I see is that it is not as flexible as the Aluminium Extrusion is, if you want to modify something in your build, are there any other Pros/Cons ?

I did find this CNC (http://www.youtube.com/watch?v=AFFWJnhTuQk&feature=BFa&list=FL7FAfZ_I0MXI6z_-Xl1O4Vw&lf=mh_lolz) on Youtube (http://www.youtube.com/watch?v=AFFWJnhTuQk&feature=BFa&list=FL7FAfZ_I0MXI6z_-Xl1O4Vw&lf=mh_lolz). It is made of Steel and seems to have a good design, I would like to copy this design, is there anything in the design I should look out for?

I have a fairly small space to put my CNC, so the first thing I have nailed down is the size of the overall Base, 1200mm x 1000mm. With this size I feel I can comfortably work on 1/8 Sheets of Wood/Acrylic or Aluminium Blocks.

I have started the design process, here is the Base. I will post more pics later on as I get time to draw the parts.

5431

Now to the Electronics:

My Goal with this would be to have a Accuracy and Speed but I am fairly sure I cannot have both, so I would like to be more on the Accuracy side than on the Speed side.

I have been looking at various electronics and Motors and I am not really sure what to get to achieve my Goals.

So far I have gathered that:
- There is no good or bad PSU, just get one that is suited to run your Steppers/Drivers etc...
- Digital Drivers and better than Analogue Drivers
- Servos are better than Steppers

Then again with all this above I could be Horribly wrong, so if I am, Please do not sugar coat it!

Thanks in advance for the Guidance, I am sure there will be more questions!

Web Goblin
07-03-2012, 09:29 PM
The machine on youtube is very basic in its design with bearing running on what looks like steel box section. This wont give you high accuracy. the box section wont be perfect for starters and also with bearings and running surfaces you tend to get crap between them and this gives problems. What you end up with is stuff ground into the running surface and this makes the bearing jump or stick when it tries to pass over it, I have seen this happen many times on the profiling machine I work on, you need to keep them very clean. Also if you want to cut ali I think the design would be too weak for it unless you took very light cuts. I think it would need some serious beefing up. Have a look through the forum builds and Im sure you will find something there to suit your needs that you can copy and modify to suit. Both accuracy and speed are obtainable relatively cheaply and stepper motors will suit your needs nicely when added to some linear bearing and slides and some ballscrews. Servos are better but cost loads more and the controller for them would also be more expensive. I use USBCNC myself and I like it but there are others out there for you to try. As for PSUs there are good one and bad ones. Buy a decent quality one is my advice.
Apart from that welcome to the forum.

Ian

JunkieHobbo
08-03-2012, 11:41 AM
Hi Ian,

Thanks for the Reply.

I like the design that JAZZCNC is working on, a picture here: http://www.mycncuk.com/forums/showthread.php/4242-CNC-Novice-Looking-to-build-little-help-Design-Sourcing-and-General-Help

Would this be a more solid design with the high sides? Any tips for designing a Strong machine for working on Aluminium?

The USBCNC looks interesting, have you had any issues with it? I was looking at GeckoDrives, what you think of them?

Going back to the Drawing board for now.

luke11cnc
08-03-2012, 12:20 PM
Hello Junkie

welcome to the forum
First what is your budget for the build.

James

JunkieHobbo
08-03-2012, 12:35 PM
I am looking at around €1000 for the Build Excluding the Spindle.

Web Goblin
08-03-2012, 01:26 PM
The machine JAZZ is working on would be fine for ali I would think. If you think the side are a little too high for you then you could follow the same idea and lower the frame height a bit to suit. I like USBCNC and so far havent had any real problems with it once I got a bit further up the learning curve. Its getting better with each new update. Dont know about the Gecko drives as I havent used them.

Ian

JunkieHobbo
08-03-2012, 01:38 PM
Thanks for your input, I am hoping that there will be more from other users.

I have looked through the Forum and have seen a machine similar to JAZZCNC with lower sides, made by Jonathan. I have PM'd him to see if he can post more pics of it as it looks like it would do the Job.

What other designs do you think are suitable for the Job?

motoxy
08-03-2012, 02:14 PM
I am looking at around €1000 for the Build Excluding the Spindle.
Have a look around the forum. You may find that your budget is a bit low for cutting al. But you wil get value for money if you listen to our experts.

Bruce

kylelnsn
08-03-2012, 04:22 PM
Me and jazz are working together (Well he's the master so I am watching and listening and learning) to develop this machine, as my goal is to cut aluminium, we are bashing out a few details first and he's going to draw up a final design. And of course I will let you all know and keep you all updated!

JunkieHobbo
08-03-2012, 08:18 PM
@ motoxy: I have been and I am currently gather more information and then back to the drawing board. I know the Budget is tight, I do have an emergency fund if upgrades are required... ;)

@ kylelnsn: Would like to see the drawings once complete. Have you decided on the Electronics yet?

JAZZCNC
10-03-2012, 08:43 PM
Hi,

First Kyle's got it all wrong I'm no master..:redface: . . . But I have offerd advice with design thou struggling with finding time at the minute.!!

The machine(2 actually) I'm building which you referred to has high sides because the bed is adjustable. It will easily handle aluminium with bed set to it's higher settings keeping the Z axis extension short.
The adjustable bed as a few bennifits.!! It allows larger deeper material and/or the use of vices or 4th Axis plus allows flexabilty in material choice upto aluminium and even very light steel use with a decent spindle and correct cutters.
The high sides means the gantry sits direct onto the bearings giving maximum strength, no flexing flappy gantry sides.

These machines are a smaller version and losely based on my machine.They basicly incorparate improvements and lesson's learnt with my machine. They are an allround machine and very versatile which will do most jobs very very good. . . . BUT . . . If you want to just exclusively cut aluminium then I'd recommend you take a slight different route.

My machine almost exclusively cuts aluminium and without any trouble but if starting again to EXCLUSIVELY cut ALuminium then I'd drop the adjustable bed to gain Max frame strength, beef up the gantry and the Z axis and incorparate full flood cooling. . . . PLUS . . . I'd mount it verticle.!!! (See my post of mine roughly proped against wall cutting 10mm single pass)
While mine happily cuts at 1.5-2mm Depth of cut at 1000mm/min and will even rough upto 3mm(thou it moans a bit) just these few extra features will increase DOC and feeds plus increase finish quality.
My next machine will have all this plus an added unusual extra MAX Strength gantry/Z axis design which you'll just have to wait to see.!!

Regards good and bad PSU that depends.??
It's very important the PSU is sized correctly to the motors being used, unregulated supplys suit steppers better than regulated linear supply's and are the prefered choice because they handle the back EMF steppers produce when deaccelerating better.
Often it's cheaper and better to build a toroidal supply, doing this means you can size the voltage and amps pritty much exact to your motors ideal requirements.
It's a very common mistake for folks to buy the wrong size supply because they are confined to standard sizes available, often they buy too low voltage which is a big mistake. Steppers get there speed from voltage so it's extemely important to get this correct so in this regard yes there is good and bad.!!

I'll say to you what I say to everybody.!! Don't buy a thing untill you have fully settled on the design and know the materials and have an idea of weight's etc to be moved around. With £1000 budget you can't afford to buy a single thing wrong but it is just about do able.

Just ask if I can help and I'll try.

m.marino
10-03-2012, 09:56 PM
I would strongly follow that advice that Jazz is giving and don't forget to design space for the limit and home switches. Trust me on that one.

Michael

kylelnsn
10-03-2012, 11:52 PM
I stand by my first statement Jazz is a master, extremely knowledgable, and one of the most helpful people I have meet on a forum EVER!

As Marino said stick very closely to Jazz's advice and you won't go wrong!

(Looks for bowing down smiley)

motoxy
11-03-2012, 09:36 AM
don't forget to design space for the limit and home switches. Trust me on that one.

Michael

And that really was said with feeling:lol:

bruce

JunkieHobbo
13-03-2012, 02:03 PM
Hi Guys,

Thanks for ALL the advice, after the whole weekend doing some research and messing with the designs and Ideas I have come to the point where I would like to know what hardware Sizes I should be going for.

I am looking at a design with a raised X-Axis, similar to JazzCNC design mentions earlier.

To complete my design work I need to know some things, only a few questions I promise!:whistling:

From my understanding Linear Profile Rails are the way to go as they are the "Most Stable/Accurate" option. What size would be best for the X-Axis and Y-Axis? Should I use them for the Z-Axis? Also how many bearing blocks per Axis? Is there a calculation to work it out?

Now onto the Subject of Ballscrews, again from research the "Most Stable/Accurate" option would be to have a Stepper running a Ballscrew on each side of the X-Axis, is this correct? Again, What Size ballscrew for each axis? Is there a Calculation to work it out?

@JAZZCNC: Some very good Advice, kept me busy for the weekend. Regarding the Budget, you are correct, I would rather Plan and Plan and not waste money as the budget is tight. When you say beef up the Y-Axis, could you elaborate a little?

@m.marino: Yes, some very Good Advice from JAZZCNC, regarding the Switches, I have added them to my notes as a Priority!

@kylelnsn: Yeah, JAZZCNC sure looks like he knows his stuff, not sure about all this Vertical talk though!

@motoxy: LOL! :lol::lol::lol:

Again, thanks all for your Help, Time and Patience!

JAZZCNC
13-03-2012, 08:46 PM
From my understanding Linear Profile Rails are the way to go as they are the "Most Stable/Accurate" option. What size would be best for the X-Axis and Y-Axis? Should I use them for the Z-Axis? Also how many bearing blocks per Axis? Is there a calculation to work it out?

15mm Profiled rail will be sufficient with 2 bearings per rail. Regards calculations etc you could do them but profiled rail & bearings, even 15mm bearings, are designed to hold a huge amount of weight far more than your's or most DIY machines will ever need so I wouldn't concern your self too much in this department.

With profiled rail there's different types of carriage, usually 2 lengths standard and long, then there's slim and wide type carriages.
Think of the bearings as a foot print the wider and longer the more planted the foot, if using 2 ballscrews to move the gantry the length can be reduced slightly because of it being driven and held both sides but longer and wider is better than fine n slim, trade off being extra length rails needed for same cutting area.

Use them on all Axis but with the Z axis mount the rails on the front plate and bearings on the rear plate, this will give the maximum strength when the Z axis is extended.



Now onto the Subject of Ballscrews, again from research the "Most Stable/Accurate" option would be to have a Stepper running a Ballscrew on each side of the X-Axis, is this correct? Again, What Size ballscrew for each axis? Is there a Calculation to work it out?

Yes driving the gantry with 2 ballscrews is much stronger than using a single screw because it reduces the racking or skewing affect when cutting at the outer edge's of the machine, more so the wider the machine.
Regards size etc then you have a few options.!! It's not rocket science but does revolve slightly round the type of use of the machine and the length.
Basicly you choose the screw by pitch and diameter and length.
If it's going to be used as a general purpose machine IE, wood, plastic's, soft metals then a good all round pitch is 10mm. This will give a good balance of speed and resolution.
If it's to be used for very detailed work Ie: Engraving etc then you will want a lower pitch like 5mm or less, the advantage being higher resolutuon. The trade off being slower rapid and cutting feeds.

The diameter is often determined by length and what's pitch is available in that diameter. Long screws 1300mm and above need to be thicker to handle the whipping affect unless some other method is used like rotataing the ballnut and keeping the screw fixed in tension at both ends.!
Again there's a trade off.!! . . Larger ballscrews weigh more and therefore take more energy/force to be accelerated they also produce more inertia which as to be de-accellerated the net affect is this requires larger motors which require larger drivers which require larger PSU's all this add's to the expense.

Depending where you buy the screws(Probably from china for cost reasons.!!) you'll find 10mm pitch is often only available in 16mm or 25mm dia for some reason.?? 20mm is nearly always 5mm pitch.?
For an all round machine Upto 1200-1300mm then 16mm Dia meter with 10mm pitch is a good choice.

For higher resolution but with less speed then 5mm pitch is a good choice, would will also be ok regards whipping upto around 1500-1600mm with 20mm due to less rotational speeds. (my machine use's 150mm 5mmpitch 20mm Dia with no whipp upto 12meter/min)

Another option and something I like and know Jonathan likes to use is connecting the motors to the screws via a short timing belt.
Doing this gives a few advantages.! . . It reduces resonance which direct drive can suffer from and it also allows for easy gearing to either increase speed or resolution/torque. Makes a machine that bit more versatile.


When you say beef up the Y-Axis, could you elaborate a little?

Yes very simple really.! When cutting metals, even soft metals like Ali or brass the machine needs to be very very strong if you want to have any kind of resonable depth of cut (DOC) and feed rate(FR) with a nice finish.
To be honest most half decent built machines can cut Ali or brass. . BUT . . There's cutting and there's cutting.!!
Lighter less ridged machines will have to take far less DOC and often go slower, thou less DOC does mean you can go faster on the FR but the machine becomes the restriction. Less ridged machines suffer from resonance which transfers to the finshed cut quality, the faster FR or deeper DOC the worse the finish due to resonance.

Basicly if you want to soley cut metals then build it strong and heavy in all departments, don't be affraid of weight, the Mass helps massively with quality of cutt.
Again the down side is more power and bigger components required to handle the weight etc. Plus an often over looked but again massively important area is the Z axis and Spindle used.??
No point building a strong framed machine with the best components etc if the Z axis is under built and flexing around or there's a weedy spindle attached that can't handle the DOC or FR.
If you want a decent machine just for Metal be prepared to build strong and spend money. . .OR . . . Some times easier/better to Convert a milling machine.



@kylelnsn: not sure about all this Vertical talk though!

Why would you say that.!! . . . When you have used a machine for a while you'll inderstand better and defiantly see/appreciate all the advantages.!
With a few weeks experience using it in the vertical position I can absolutly say 110% it's the best thing I've done to the machine since I first built it, and with the excepton of slightly more awkward clamping which I've fully adjusted to know i've made some little rest aids there's absolutly NO down side only big positives.
I would recommend anybody building from scratch to think long and hard and not be put off by the unusual positioning.!!. . . There are virtualy no down sides only positives..:dance:

JunkieHobbo
15-03-2012, 02:53 PM
@JAZZCNC: Thanks for the Great information and your time for posting.

I have started working on the Steel base Structure, I have the overall size but height will be adjusted/tweaked as soon as I get the machine completely drawn up.

Here is what I have so far, I am using 100mmx50mm Steel Box with 3mm wall, I am sure this will be more than strong enough.
5476

Is there anything the looks glaringly/obviously wrong?

I will continue to work on the Design and keep everyone posted.


Use them on all Axis but with the Z axis mount the rails on the front plate and bearings on the rear plate, this will give the maximum strength when the Z axis is extended.

I am not quiet following you here, could you show me an example of what you are talking about?


Another option and something I like and know Jonathan likes to use is connecting the motors to the screws via a short timing belt.

Does anyone have any pics of a similar setup?

On another note regarding the Stepper/Ballscrew on both sides of the X-Axis, what happens when 1 stepper falls behind or fails?

kylelnsn
15-03-2012, 04:30 PM
Why would you say that.!! . . . When you have used a machine for a while you'll inderstand better and defiantly see/appreciate all the advantages.!
With a few weeks experience using it in the vertical position I can absolutly say 110% it's the best thing I've done to the machine since I first built it, and with the excepton of slightly more awkward clamping which I've fully adjusted to know i've made some little rest aids there's absolutly NO down side only big positives.
I would recommend anybody building from scratch to think long and hard and not be put off by the unusual positioning.!!. . . There are virtualy no down sides only positives..:dance:

I didn't say that? Lol? Confused!

JAZZCNC
15-03-2012, 06:32 PM
Hi,

Please don't think I'm picking fault with what I'm about to say it's just my opinions mixed with experience.!

I'm currently building the machines that I think you've seen and based this design from. My frame is using 80x40x3mm and 40x40x3 I can tell you it's more than enough and very heavy so I would recommend you save a bit of money and go with 80x40 it will more than do what you want.

Regards the frame layout I would change several things and add a few.! It's easier to show than explain so I've drawn a quick model to show it. Doing it like this gives more weld area and better stiffer bracing.
One thing I suggest you do is NOT weld the top piece which the rails sit on to the upright legs.? Better if you weld flat plates to the uprights and then bolt the top piece. This gives better control of getting the two sides parallel, basicly you can shim and adjust them into parallel planes.

RE Z axis. See pic with front plate missing (and only1 rail) this use's profiled linear rails but the principle is the same supported round rail.
As you can see the bearings bolt onto the rear plate that travels along the Y axis and the rails bolt onto the Z axis front plate which the spindle attachs too. The reason for doing it this way is to gain from the fact the flex of the front plate become less or more as the extension increases or decreases. Doing it the other way round the extension of the rail from the bearing is always the same distance whether extended fully in or out and with long Z axis this becomes a spring board at any z extension.!!

RE Timing belt. See updated model of current machine. The following will also answer your concern regards steppers falling behind
To be honest if your going to have a piece across the back like the model suggest's then I would urge you to connect the two ballscrews together with one timing belt and use just one motor to drive both. This address's several issue's and saves you some money, my own machine use's this setup so I can tell you 100% it works and works very very good so don't be put off with the use of timing belts. . . .Here's why.!!

First and foremost it completly eliminates any sync problems(falling behind) with the 2 screws and takes away any fear of racking or damage from one motor stalling and the other continuing which belive me when it happens, esp at high speeds, can do lots of damage. This is the main reason why I prefer this method.!
It allows higher rapid feed rates because you don't have to affectively de-tune your motors to be sure your in a safe non stalling tuning range like you have to do with slaved motors.
Because it use's belts it's easy to change ratio's for extra speed or torque/resolution, basicly depending if you gear up or down and the ratio you change one pulley for another and you have either doubled, tripled your speed or resolution/torque.
You only need one drive and motor to drive 2 screws which save a bit of money, thou you dohave the belts and pulleys to buy which lessons it a bit but your still in front and you save some input's and outputs on the parallel port.!!

Like I say don't be put off with belts they work fine and in over 3 years I've only broke 1 belt and that was my fault for not tightening a pulley after messing. The accurecy and repeatabilty of my machine is fantastic and in no way suffers thru using belts and I mainly cut aluminum with it. Just see my Aztec calender too see that and if any issues then no way would it do work to this level of detail.

Hope this helps.!

m.marino
15-03-2012, 06:35 PM
okay on X axis one or two ball screws? Also you might want to put a brace in half way down the X axis rail to help reduce any possible flex (depends on weight of gantry and such). Please remind me of dimensions as have been in the shop all day truing (tuning) the motors and travel. Therefore really not looking back in the thread for the info' (yes I am that mentally tired). Other then that looks like a good solid start.

Michael

Jonathan
15-03-2012, 07:00 PM
I have looked through the Forum and have seen a machine similar to JAZZCNC with lower sides, made by Jonathan. I have PM'd him to see if he can post more pics of it as it looks like it would do the Job.


I'm sorry I forgot to reply to your PM earlier about pictures of my machine and things. Had a bit more work than usual to do at Uni. Looks like Jazz has saved me the trouble of answering lots of questions :)
There's quite a few pictures in my build-log, especially towards the end:

http://www.mycncuk.com/forums/showthread.php/2288-1-7*0-74*0-4m-Mill-Router-building

(starts off embarrassingly bad then gets better!)

There are a few things I will change on the machine. Essentially just a stronger Y and Z-axis with profile rails, but like Jazz I have an 'unusual extra strength' design for that. Instead of having one piece across the gantry there will be two, with one Y-rail on each, and a box shaped the Z-axis operating between. Having said that it's still strong enough to cut aluminium at a decent speed 1.5mm DOC, or more with reduced finish which is not bad considering only round rails. I just want to see how much better I can make it...

If you need any more of specific parts just let me know and I'll take them.

I'd be inclined to add some triangles to that frame to stop it skewing.

C.AlveSilva
15-03-2012, 09:58 PM
Hello,

My name is Carlos and i'm form Portugal.

I want to built a CNC router to make 3D wooden molds.

I was thinking in the xx axis use racks instead of ball screws, (the racks in the place of the ball screws, put the xx motor in the middle of the yy axis and using a shaft with a pulley at each end).

In terms of accuracy, racks is worse than a ball screw??

I want the router to make molds, for it I need a lot of precision. (For this kind of work an error of 0.1 mm (0.0039 inches) is an huge error.)

Would you, please, give me your opinion about the racks?

Thanks.

P.S.: Sorry for my English...

JunkieHobbo
20-03-2012, 04:56 PM
@JAZZCNC: No worries, I am really appreciative of your Help and anything that can help me make a better machine I will gladly accept, keep on picking! :tongue:

OK, I have made some changes as you suggested, I have gone with 80mmx40mmx3mm, every penny helps! Regarding the Flat plates, I get what you mean but how would you do it, would you mount a plate on the upright and also on the base of the 80x40 that will hold the rail, picture is worth a 1000 words.....:whistling:

I have noted down your points regarding the Z-Axis, once I get to the Z-Axis I will incorporate the suggestions.

Belts, I am convinced it is the way to go now, only 1 Stepper, I am assuming a larger stepper than Y&Z? Do you have some pics of your setup using the Belts, I have looked around and was not really able to find any good pics for reference to build on.

Here is an updated drawing, I added in some triangles as Jonathan suggested, I was just going to add some tabs to mount the panels to but Triangles will work and also add extra support.

5520

If the base is good I can then work on the X-Axis rails and Belt drive.

That is some seriously nice work there, hopefully once I get my machine up and running I will be able to produce something half as good!

@Jonathan: Thanks for the advise, I added some triangles they will have uses, one for support as you mentioned and secondly to hold my "Dirt Splash Panels" up, see the pic above. I might have some more questions as I progress but for now the pics will do great!

@m.marino: I will be going with 2 Ballscews on both sides of the X-Axis, and will be driven by one stepper and a belt. Added a Center brace as suggested by yourself and JAZZCNC's example. I have about 1200mm x 1000mm x 1000m Space to work with, so I want to build the Biggest/best possible machine to fit into that area.

@C.AlveSilva: I am by no means a Guru on this but from what I have researched and have learned from the guys on the Forum, Ballscrews are the way to go and Profile rails.

JAZZCNC
20-03-2012, 06:16 PM
Hello,

My name is Carlos and i'm form Portugal.

Would you, please, give me your opinion about the racks?

Sorry Carlos didn't see your post.

If you want to do mainly 3D work and fine tolorences then I suggest you use ballscrews on all Axis.

Rack is ok for general profiling and machining etc but to do really accurate work with lots of repeatabilty then requires Hi quality Rack & pinions and because often with R&P you have to gear then you'll also need Hi quality gearing units or Pulleys and belts.! . . . All these will cost more than cheap chinese ballscrews and possibly perform no better if not worse.!

So IMO ballscrews every time. Thou this does depend on length of machine and even then I'd still consider using them just with a rotating ballnut.!!

JAZZCNC
20-03-2012, 08:44 PM
Hi JunkieHobbo,

Couple of things regards the frame.! It's weight and the ease of build and adjustment.?
I know on my Cad model it looks like the frame is all welded but it's not and actually bolts together quite a bit. Really my models are not fully accurate models regards detail and more for me to visulise so don't pay massive attention to the detail. Thou they are to scale and dimensionally accurate regards materials and compnents which is also why I use them so I can judge component fit etc.

When you build this machine you need to incorparate as much adjustabilty as possible. Thats why the plates on the uprights for the top rail section to sit on.
On the machines I'm building now the main frame(excluding adjustable bed) is actually made up of 5 pieces bolted together (7 if you include top rail sections)
This gives adjustabilty in lots of ways and also helps with moving around.!!!!. . . . . It will be very very heavy so makes life much easier when in pieces.

Regards the plates there's several ways but this is how I do it.!. . . Basicly I machine flat plates with both holes and M12 nuts welded to them which are then weld onto the uprights, taking care to get them square as possible to the boxsection.
!!!!!!! (This next bit will be hard to explain so bear with me.!!!)
The plates are longer than the upright section for the middle uprights so can be bolted either side of uprights into the top rail section which has nuts welded into it. The end upright plates are only longer on one side so bolted on inside from below into top rail then another bolt accessed thru the open ends into plates which have M12 nuts welded to them.
These plates could be drilled and tapped but that would require thick plate so thats why I use welded M12 nuts. The same M12 nuts are welded into the top rail as the 3mm box section doesn't allow enough thread.
It may sound faffy and complicated but in reality it's not and doesn't take long if you have a decent drill press and can weld half decent.

The frame pics below will give a better idea. I'm still building the frames but when get a chance then I'll post a few pics of the actuall steel work, when I'm further along I'll probably start a thread showing it(them) coming along.

Regards the belt drive then there's better ways than I've done it. Mine was rushed and thou it works very well it's not pritty so wouldn't do it that way again or recommend you follow it.
I've done a quick alteration of my frame and mock up to give a rough idea of one way, needs belt tensioner etc but should give you an idea.!! There's other ways if you look around and easily done with motor at same end in line with belt acting as the tensioner incorporation with an idler.

Hope this helps.

JunkieHobbo
21-03-2012, 09:42 PM
@JAZZCNC: Thanks again for the reply and your time.




When you build this machine you need to incorporate as much adjustabilty as possible.




I really like your way of thinking, that the machine should incorporate adjustability. I like the modular approach aswell, I do feel this would be a far easier build for me by doing it this way with minimal welding.

BUT

Just a quick question, I know another question! :whistling: Will the bolted sections hold up and keep structural stability as good as the welding would? I am asking because it is a CNC machine that works on precision and bolts are " not a fixed" part of the structure, so there is a chance for them to loosen, how would you solve this problem? Then again, I could be obviously missing something here as usual.

Regarding the Plates, the pics answered my Question 1000%, Thanks again!

JAZZCNC
21-03-2012, 11:12 PM
BUT

Just a quick question, I know another question! :whistling: Will the bolted sections hold up and keep structural stability as good as the welding would? I am asking because it is a CNC machine that works on precision and bolts are " not a fixed" part of the structure, so there is a chance for them to loosen, how would you solve this problem? Then again, I could be obviously missing something here as usual.

Well think about it Just about every CNC machine even those costing 10's thousands of pounds are bolted together in some form or another. . . . They are not welded in one piece.!!
The bolted sections will hold up fine so long as you use correct size and good quality bolts and tighten them correctly. If you want to ensure they don't move then use locking/locateing pins.!!. . but they don't move if good quality bolts with locking washers with a bit of thread lock for good measure.!

Swarfing
21-03-2012, 11:23 PM
Jazz is right on this very much so. My table is totally welded and it turned out fine? well not true, even with all the care in the world the bed still needed a lot taken off the surface to get a straight bed. In most cases that is fine but if i had used bolts i could have adjusted a lot of that out before surfacing. Just to add a bit more to the design try and include adjustable feet, can make a world of different to noise as well as helping adjust to unleveled floors. It aso makes it much easier if you ever needed to move it as well.

JunkieHobbo
22-03-2012, 01:51 AM
@JAZZCNC: OK, back to the drawing board as they say! I will look into using bolts as I mentioned in my previous post, it will make life ALOT easier for me if there is not alot of welding to do.

@2e0poz: I am going to see what I can draw up using Bolts and some welding, I have a feeling I am going to end up with something similar to JAZZCNC's design. Regarding the Feet, that was always in the Plan, just never showed it in the drawings, I learnt my lesson on that one already.

JAZZCNC
22-03-2012, 09:10 AM
it will make life ALOT easier for me if there is not alot of welding to do.

Actually doing it the way I have means in the terms of building it makes it harder and longer bolting to together, there's actually no less welding required.?
Welding takes minutes per joint.! . Cutting the plates, Drilling the plates, Drilling the matting sections and welding/grinding in the plates/nuts etc takes meny hours.

You have to take much care when cutting the steel that the ends are square so when you weld the plates they are square. If the plates are not perfectly sqaure then when you bolt the sides to the base pieces they wont be square and much shimming will be needed. The goal is to reduce shimming as much as possible.
The way I do it is to tack the plates top & bottom then bolt them together to check fit, If far off then grind either the top or bottom tacks off and adjust with No1 special re-alignment knocking stick.:naughty:.
Then when happy weld full joint. This gets me very close thou the plates can and do move with the heat from welding so need rechecking/adjustment, for the final adjustment and close fit I can just skim a little of the plates with grinder. After this I'm pritty much spot on square and if not then just minor shimming will be required.

So as you can see it's not easier by any means.!! . . . BUT . . it should also show you how important I feel and know having this adjustment abilty matters to a accurate machine because NO WAY would I go to all this trouble if I could just weld the bugger up.:nope:

That said my way of building makes it harder because I didn't want lots of bolts and external plates showing so used the mix of welding and bolting.
If you wanted could build using just plates/bolts and very little to non welding if you wanted, it's just lots of plates,drilling and bolts because the plates would have to be both sides of each joint. . . . It would still be very strong and have lots of adjustment just not so good looking.!!

JunkieHobbo
02-04-2012, 06:01 PM
@JAZZCNC: OK, after a long busy week of research I could not come up with a better design that would meet my requirements so I am going to copy yours, if you don't mind of-course!

So I have gone and modified my last attempt to incorporate the adjazztability factor, This is what I have so far:
5629

There is 1 thing that I am not sure about or if it even matters, is there any calculation for the length/angle of the diagonal supports that connect to the sides or is it just "Aim for the Middle area"?

Again, are there any obvious things I have missed here before I move on with the drawing?

motoxy
02-04-2012, 07:31 PM
Jazz, looking at your base drawing what is the purpose of bolting the frame when only the connection between the rail support rhs and the frame will need adjusting to align the rails?

Bruce

JAZZCNC
02-04-2012, 11:46 PM
Jazz, looking at your base drawing what is the purpose of bolting the frame when only the connection between the rail support rhs and the frame will need adjusting to align the rails?

Bruce



Technically it doesn't need to Bruce but it does have a few benefits (Eh he Lee taught me to spell that correct last night. . :rofl:)

First the thing weighs a bloody ton so makes transporting and siteing that much easier. It also allows so much more adjustment in all area's and makes setting up that much easier.
If welded solid then your stuck with any errors without resorting to grinding or re-welding, it also very difficult making sure everything is square and without twist when welded solid and often the heat from welding can twist the frame. This doesn't tend to happen when done how I've done it and can easily be adjusted out or shimmed out when bolted together.

In the past I've done it both ways and this is by far the better way in my experience. Yes far more work but worth the effort IMO.

Junkie:

You need to add some triangle plates to the corner uprights where they meet the cross brace, like the pic below. (Take no notice of rest of frame it's not complete here, still needs corner braces amongst other things)
I can have mine going upwards because there's the adjustable bed which they don't come above but if your not having an adjustable bed then just flip them and take down towards bottom corners. They really stiffen the frame up and stop any sideways movement, they also make getting the sides square to the base very easy when initially setting up.
Don't be put off with bolting, the frame in the picture is unbelievably strong but yet very easy to setup due to all it's adjustment.

motoxy
03-04-2012, 07:42 AM
Okay point taken. Better go and redesign my base.

Bruce

JunkieHobbo
13-04-2012, 05:52 PM
Junkie:

You need to add some triangle plates to the corner uprights where they meet the cross brace, like the pic below. (Take no notice of rest of frame it's not complete here, still needs corner braces amongst other things)
I can have mine going upwards because there's the adjustable bed which they don't come above but if your not having an adjustable bed then just flip them and take down towards bottom corners. They really stiffen the frame up and stop any sideways movement, they also make getting the sides square to the base very easy when initially setting up.
Don't be put off with bolting, the frame in the picture is unbelievably strong but yet very easy to setup due to all it's adjustment.

OK, I made the adjustments as suggested.

5705

I also added the X-Axis Support with Profile rails and Bearings. It is coming together nicely now, if all is OK and I hav'nt missed anything glaringly obvious I will start to work on the X-Axis BallScrew Mounts.

Regarding the Ballscrews for the X-Axis, Are there any things I should watch out for? Is there a good way and a bad way to mount BallScrews? What is the optimum way of doing this using belts? Should this Plate supporting the Stepper and Ballscrew be Bolted or Welded?

I know, to many questions again...

JAZZCNC
13-04-2012, 07:41 PM
Coming along nice thou them sides look a little high if your not having an adjustable height bed. The Z axis will be quite long = flex = tool Chatter = Shite finish.!!

Regards the ballscrew mounts then again you need to try to build in has much adjustabilty for alignment as possible, you can weld support plates for the main ballscrew mount plates but would'nt weld the mounting plates them selfs better to bolt and allow adjustabilty.
It also depends on the type of BS mounts your going to use to which way or how you go about it.

Some more renders to give some idea how I'm doing it, hopefully be able to show machined parts soon they will be slightly different but close.