PDA

View Full Version : BUILD LOG: Small rigid router - Design advise needed



mitchejc
24-08-2014, 09:48 PM
Hi, this is my first stab at the design after having a good look at some of the great builds here. Any advice or comments would be greatly appreciated! Requirements for the machine as following
Exclusively for machining aluminium
Cutting area: 600mm x 450mm x 150mm.
Speed is not a great issue
I got hiwin HGR15 rails and blocks and Chinese 1605 ballskrews and ballnuts so I'll use those.
Since aluminium is not cheap here I'm going for a welded steel design and then alu on Z because there's a bit of machining required there.


Some basic dimensions for reference.
Distance between gantry side rails: 800mm
Distance between top and bottom rails on gantry: 300mm
Z plate with 250mm
Box section used is 160x80x3mm and 120x80x3mm
10mm 10mm sheet for flat parts to be laser cut
Y carrage and Z assembly is 25mm alu with a 10mm plate at the back


A few questions:
Will the long belts (+-650mm on Z and 550mm on Y) cause any issues? HTD 5mm pitch x 15mm width with 20tooth pulleys
The gantry is very rigid and HEAVY but I'm worried about ressonance. Can one fill it with something that's not going to add a lot more weight that will help with resonance.
Does the 3 rails on the gantry make sense? I've got 1 on top and two at the bottom.


Please comment if you spot anything that's not going to work or things I can improve without adding major cost

http://www.mycncuk.com/attachment.php?attachmentid=13201&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13202&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13203&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13204&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13205&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=13206&stc=1

Clive S
24-08-2014, 10:53 PM
You could shorten the z belt by moving the motor and turning it the other way up so that the pulley if facing down. ..Clive

mitchejc
25-08-2014, 05:45 AM
Thanks Clive, I'll give that a go then I can also move my Y stepper up a little bit to shorten that belt.

Blackrat
25-08-2014, 07:29 AM
ja ja wat se jy :P

why even go with belts ? are you going to have a step down ratio ?

i wouldnt bother with the 3rd rail, i think it will just make it harder to clock up ...

what or how are the two bottom rectangle tubes fastened ?

looking good :D

EddyCurrent
25-08-2014, 09:09 AM
Looks pretty good but to make life easier later I would draw in your energy chain, drag chain, whatever you call it because it could affect some aspects of the design such as , mounting brackets, length of travel in X direction. Also consider limit switches and homing switches locations.

mitchejc
25-08-2014, 09:44 AM
:-)

No, its just 1:1 ratio. My understanding is that belt drive will make the machine a little smoother. X and Y would be possible to do direct drive but Z will be difficult as I'm trying to keep it as flat as possible.

Not sure but I'm thinking of welding on tabs or a piece of flat bar on both sides that will allow me to bolt it to the table.



ja ja wat se jy :P

why even go with belts ? are you going to have a step down ratio ?

i wouldnt bother with the 3rd rail, i think it will just make it harder to clock up ...

what or how are the two bottom rectangle tubes fastened ?

looking good :D

mitchejc
25-08-2014, 09:49 AM
Thanks for the advice EddyCurrent. I will start adding those as I can see it might be an issue with the design as it is now, especially the drag chain running on or behind the gantry.


Looks pretty good but to make life easier later I would draw in your energy chain, drag chain, whatever you call it because it could affect some aspects of the design such as , mounting brackets, length of travel in X direction. Also consider limit switches and homing switches locations.

Blackrat
25-08-2014, 11:13 AM
belts aint gonna make it any smoother .... id run a coupler on the two axis' , so much easier and cheaper

im not quite with you on bolting it to a table ? you HAVE to join either sides , no ways you going to cut ally decently if they are seperate from each other, not to mention the nightmare of trying to get them parallel by clamping to a table

Clive S
25-08-2014, 11:29 AM
I am no expert but if you follow this forum and take notice of the Master's on here, coupling with belts helps to eliminate resonance which can stop a motor dead in its tracks. Try running a stepper with a loose mounting to see the effect. ..Clive

mitchejc
25-08-2014, 11:56 AM
Thanks for the feedback Blackrat. I also like the simplicity of direct drive but quite a few people here believe belt-drive is the better route. I don't know how much better?

The table/base will be a very sturdy steel frame so the two sides are build separately but when bolted down they are connected via the base or am I maybe misunderstanding your question. I guess I can also weld them to the table but that won't be much stronger than having several bolts on each side of both or will it? I thought bolting them would make the alignment easier.


belts aint gonna make it any smoother .... id run a coupler on the two axis' , so much easier and cheaper

im not quite with you on bolting it to a table ? you HAVE to join either sides , no ways you going to cut ally decently if they are seperate from each other, not to mention the nightmare of trying to get them parallel by clamping to a table

EddyCurrent
25-08-2014, 12:00 PM
With regard to the base, you need to post some drawings I think. With anything, you need good solid foundations to build from.

Blackrat
25-08-2014, 12:05 PM
The table/base will be a very sturdy steel frame so the two sides are build separately but when bolted down they are connected via the base

thats not going to work

no ways you gonna bolt it together so that the rails come out within a decent tolerance ....
have a read about epoxy leveling a frame if you havnt already ?

mitchejc
25-08-2014, 02:48 PM
Yep, you are right, I'll do the drawing for the base and post it as its obviously an integral part of the design.


With regard to the base, you need to post some drawings I think. With anything, you need good solid foundations to build from.

mitchejc
25-08-2014, 03:11 PM
I'll try and get the drawings done, but the picture I have in mind is basically a 4 legged steel tube structure made from 80x80 tubing with lots of support under the table and space inside where I can sandbag the whole thing to add lots of weight to the base. I then epoxy level the top of the steel base to get that flat, bolt 2 layers of good quality birch ply + a 5mm steel sheet on top. Then bolt the gantry sides to that surface and epoxy level them before the rails go on. Does it sound practical or are there an easier way to build a sturdy base. I started off with the idea to do a concrete slab on top of a steel structure but there's just to many issues with that approach.



thats not going to work

no ways you gonna bolt it together so that the rails come out within a decent tolerance ....
have a read about epoxy leveling a frame if you havnt already ?

mitchejc
25-08-2014, 10:15 PM
Ok this is roughly what I have in mind for the base. Idea is epoxy level the top of the base just to get it +- flat and level. Then put 2 layers of 16mm sealed birch ply on top and then a 5mm steel plate and secure that to the base with bolts. Apart from using bolts I really don't see another option of how to secure the machine to the table? I'm referring to the red parts in the second picture. I really don't want that to be the weak point so I'm open to suggestions. Am I approaching this the wrong way to wanting to build the router and base separately? Any advice would be greatly appreciated!
1321813219

Boyan Silyavski
02-09-2014, 02:17 PM
IMHO:

- no need for such a beast table.

Use 100x100x4 or 100x100x3 and make something similar like i did, worked quite well. Its simpler, cheaper and easier to make. At the back you can see the same design but with legs. See post #53 (http://www.mycncuk.com/threads/6457-Sturdy-and-Fast-all-Steel-CNC-my-first-build?p=51273#post51273)



-Z belt length bothers me. See how i did it. post #109 (http://www.mycncuk.com/threads/6457-Sturdy-and-Fast-all-Steel-CNC-my-first-build?p=59627#post59627)

JAZZCNC
02-09-2014, 09:57 PM
thats not going to work

no ways you gonna bolt it together so that the rails come out within a decent tolerance ....
have a read about epoxy leveling a frame if you havnt already ?

Depends how it's done. Epoxy leveling isn't the be all and end all solution. I've said this many many times but for a good DIY machine you need to build in as many oppertunity's for adjustment as possible and bolting allows plenty of scope for shimming etc to get machine accurate.
If done correctly then it's actually better than epoxy because you can fine tune to very high tolerence.! Only thing is it takes much more time and patience.


belts aint gonna make it any smoother .... id run a coupler on the two axis' , so much easier and cheaper

Actually can make a huge difference esp on all steel machine that's running analog drives with limited resonance handling. This is actually the main reason to use Belts and the ratio option and flexabilty is just a bonus.

Often folks running direct drive don't realise they have resonance problems and just accept that the speeds and performance they are getting is just the max level they can tune to or achive from there motor/drive setup. When often the reality is that the same motors/drives when run on belts and unaffected by resonance will achieve much higher and smoother performance.

mitchejc
03-09-2014, 06:49 PM
@ silyavski (http://www.mycncuk.com/members/10304-silyavski)
Thank very much for the reply, I was hoping you would have a look at my design. Your build really inspired me to go the metal route!

Point 1: Agree 100%. After calculating the cost for that brute of a table I realized I have to scale it down a bit so I'll do that
Point 2: The belt length is also bothering me. I really like the way you did yours but I have a bit of a problem because I already have a 350mm long ball skrew for my z and also my steppers which are very long so there's just no way I can do it the same as you did yours. My compromise at this stage is to make the gantry top narrower to move the stepper closer to the ballskrew or go with the long 650mm belt. Which one do you think is the better option? I think my wife will poison me if one more parcel with cnc parts gets delivered here so getting a longer ballskrew is not an option :friendly_wink:




IMHO:

- no need for such a beast table.

Use 100x100x4 or 100x100x3 and make something similar like i did, worked quite well. Its simpler, cheaper and easier to make. At the back you can see the same design but with legs. See post #53 (http://www.mycncuk.com/threads/6457-Sturdy-and-Fast-all-Steel-CNC-my-first-build?p=51273#post51273)



-Z belt length bothers me. See how i did it. post #109 (http://www.mycncuk.com/threads/6457-Sturdy-and-Fast-all-Steel-CNC-my-first-build?p=59627#post59627)

mitchejc
03-09-2014, 07:30 PM
@JAZZCNC
Thanks for the great input. I'm going for the belt option. I think I got decent steppers and drivers but I'll do anything that might deduce resonance and I like the possibilities that the belts give e.g. change the ratio if I needed. Not sure with a ballskrew system but I know from first hand experience with rack an pinion that belts also offer little bit of safety margin with strong steppers and would jump teeth or snap when things go horribly wrong.

I've thought long about Blackrat's suggestion not to bolt the sides down but there's much weaker points in the design so I'm pretty sure that not the weakest link. The other thing is I'm really a novice welder so I'm not very confident I'll be able to build the base and gantry sides as one unit to any reasonable tolerances. With the bolt on plan I do limit that factor a little as I have the opportunity to shim an adjust a little like you suggested.

I wish I could hear silyaski machine take an aggressive cut into a piece of alu then maybe I can put my mind at rest about the steel structure resonance :-) Has anybody tried to fill the steel sections of the gantry etc with high density polyurethane foam and would that make any difference with resonance? I'm referring to the two part stuff that can be pored thru relatively small holes and then it foams into a plastic like solid to fill the entire void.

Boyan Silyavski
03-09-2014, 07:51 PM
@ silyavski (http://www.mycncuk.com/members/10304-silyavski)

oint 2: The belt length is also bothering me. I really like the way you did yours but I have a bit of a problem because I already have a 350mm long ball skrew for my z and also my steppers which are very long so there's just no way I can do it the same as you did yours. My compromise at this stage is to make the gantry top narrower to move the stepper closer to the ballskrew or go with the long 650mm belt. Which one do you think is the better option? I think my wife will poison me if one more parcel with cnc parts gets delivered here so getting a longer ballskrew is not an option :friendly_wink:


Then flip the motor. Look at the z i am doing right now.
Note that:
-the plate is reinforced against vibration with the small triangular pieces
-longer towards the front plate - to provide dust protection
-longer towards the back - to mount the cable chain there

132901329113292


I wish I could hear silyaski machine take an aggressive cut into a piece of alu then maybe I can put my mind at rest about the steel structure resonance :-) Has anybody tried to fill the steel sections of the gantry etc with high density polyurethane foam and would that make any difference with resonance? I'm referring to the two part stuff that can be pored thru relatively small holes and then it foams into a plastic like solid to fill the entire void.

Dont worry about that. It does not vibrate at all. I see your design ok. Not quite the expert though. use common sense and reinforce with 10mm steel rib plates where necessary against bend and twist

Important part of my design are the stair like pieces which reinforce in all directions and don't transmit resonance. To understand better what i am saying, this design is better than just sticking two profiles together.

Split and conquer! That's how you fight resonance.

Soon will make some aluminum pieces for my machine i am building right now, so will show a video about the aluminum, though my friends machine/ the yellow one from the build / has only 0.8kw spindle so it could not be appreciated very much, but there is no resonance, glass like finish.

Boyan Silyavski
03-09-2014, 08:05 PM
Just saw your gantry. what size profile you use?

I don't like that back plate there, only adds weight and no gain at all against twist nor against bend in the middle.

Look how i do it on my build right now. Profile 100x100x4 up and down and in between 60x60x3 or 60x60x4 , 2 pieces soldered together, . The Z is 3mm from gantry. Just welded today my 1800mm gantry and confirm that is stiff as it could be. Will measure the bend tomorrow.

1329313294

mitchejc
03-09-2014, 08:47 PM
The top and bottom is 160x80x3 each with a 120x80x3 between them as per below picture. The main benefit is a very simple construction and less parts but the down side is those three sections are quite heavy. Because I'll have to buy 6m of each section the plan was to use the same for the sides that bolt to the table.

I agree the back plate does not offer much apart from completing the box instead of just a U so it might stiffen things a little but its probably not worth the weight and alu cost. Wow I like your gantry and ESPECIALLY the new Z design! Now you got me going back to the drawing board with mine :-) Will you please send me the link to the thread with your new design.

13295

Boyan Silyavski
03-09-2014, 09:44 PM
Post #56 (http://www.mycncuk.com/threads/6619-Quite-an-Unusual-one?p=60714#post60714)

Ignore the double ball screw, i will mount at least once a vibrating hammer so that's why. Same goes with the triple bearing blocks. 2 are enough. The 2 square bars on the Z plate can easily be substituted by structure of 20mm plate, but i don't have the time to fiddle with it so i will use solid alu bars.

Today just soldered many things. Still no time to update the thread, but one of the things i most like about this design are the gantry legs, save some money on plates, cutting and welding. Cheap profile i mean instead of laser cut plates. As i changed some bits of the design, the gantry legs will have a plate below slightly larger or access hole for mounting the bearing blocks. But i wanted it strong left right, that's the change from the initial plan. look below:



13297

mitchejc
03-09-2014, 10:40 PM
Thanks for taking the time to reply. I really like your design and I can see you put a lot of thinking into making the construction easier without compromising strength. The simplicity of the two solid bars on the top and bottom is a great plan. I'm going to keep a very keen eye on your build to see how you do things. Time for me to start changing my drawing to incorporate some of your new ideas. The only thing I will do different on mine is to not have the rotating ball nuts on the x. Not because I don't think its a good idea but its my first serious build and I'll rather avoid that bit of complexity and go for the stationary x motors on mine.
How are you mounting the spindle inside the z box? Are you making a custom mount or will you just be modifying those standard cast spindle mounts?

Boyan Silyavski
04-09-2014, 12:38 AM
How are you mounting the spindle inside the z box? Are you making a custom mount or will you just be modifying those standard cast spindle mounts?

13300 13301




The simplicity of the two solid bars on the top and bottom is a great plan.


if speaking about the gantry, today while soldering changed the idea. As it was impossible to weld it well from all sides. So did not use the square tubing for support. Just made more ribs. Anyway i have bought 6m for the 2 plates, so i had enough of it. Instead used the square tube pieces/60x60/ to align the ribs. Its ridiculously strong. the gantry i mean. but the welding goes ridiculously slow. Great care should be taken. First spot weld, then short passes, then wait, then short passes... Still have to finish it tomorrow.

13302 133031330413305

JAZZCNC
04-09-2014, 12:50 AM
Has anybody tried to fill the steel sections of the gantry etc with high density polyurethane foam and would that make any difference with resonance? I'm referring to the two part stuff that can be pored thru relatively small holes and then it foams into a plastic like solid to fill the entire void.

No it doesn't work very well there's not enough density. The best success i've had is with kiln dried sand. Very cheap and simple with good results.

mitchejc
06-09-2014, 06:21 PM
Thanks for the photo's Sylafski. VERY nice!! That gantry looks extremely solid.

mitchejc
06-09-2014, 06:36 PM
@JazzCNC

Thanks Jazz, the sand makes sense. I'm bit worried about adding more weight to the gantry as I estimate it's already +- 80-90kg Z & spindle included but I guess a 20-30kg of sand is not going to hurt or will it? My plan is to run 8nm easyservos, will that work or do I need to back off a bit on the gantry weight?

Tom J
25-10-2014, 11:43 AM
hi
Are you going to use servos or motors with encoders, if not, than you should change design for one motor on X but with two screw on each side. Loosing steps issue - big problem when you mill harder stuff like alloy.
I had the same problem - its complicate design, but for long term your gantry always run smoothly, without jamming on one side

mitchejc
08-11-2014, 05:27 PM
Tom, thanks for the advise. Yep, I have EasyServos which are basically steppers with encoders but if I have any issues with racking I'll certainly go the single motor route.

mitchejc
08-11-2014, 06:06 PM
Thanks for the advice so far. Muchappreciated! Plan is still to build a base with aflat top and then bolt my machine on top of that. I'm basically done building the base and I'll post a few pics soon so its now time to start focusing on the actual machine.


I really had a hard time deciding onthe gantry design. I know the box section designs are proven and works verygreat but if possible I would like to try something a little different with a welded structure made from round tubes, angle section and flat cross supports. Tubes will be sand filled. I'm hoping to get away without epoxy leveling so the idea is to have the X and Y rails sit on 10mm thick ally plates bolted on top of the steel that I can hopefully machine flat and parallel. If that fails I'll go the epoxy route.


Below is what I have in mind. Pleaselet me know if you spot any fundamental issues or things I canimprove. Steel parts are blue and alu parts are grey

http://www.mycncuk.com/attachment.php?attachmentid=13816&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13817&stc=1http://www.mycncuk.com/attachment.php?attachmentid=13818&stc=1

The core of the gantry looks like this consisting of 76mmx5mm round section, 50mmx5mm angle section , 100mmx10mm flat bar and several 10mm and 6mm thick braces
http://www.mycncuk.com/attachment.php?attachmentid=13819&stc=1

Boyan Silyavski
10-11-2014, 01:02 AM
What are these 3 small diameter bars along the sides? mounting threaded bars with nuts to hold all together when welding?

What do you want to hear that may be you don't know-beautiful to look at but a waste of material with no further benefit, even worsening some strength points. Unnecessary heavy/plates/ where not needed / table structure/ but nevertheless pleasing to the eye .

What about the orange material, what is it made from? The sides are weaker than box section so if this is weak all will creep.

Basically the "fundamental" issue i see is this:
Instead of strong all supported rails supported by strong all supported steel box structure, supported by itself and gravity on a not so strong table, you are trying to do exactly the opposite- make very strong table, weaker structure and even its weaker semi supported just below the rails themselves. And that for at least 2 times the weight /$$$/ of the material needed.

That is why when i design something i design it like this- bench top, even if its 3x2m steel cnc. Why? Because that's the main structure that will do the job. All else is secondary-serves to lift it to some level or to put something between the floor and the "real structure".

So why not forget all and design a good strong benchtop CNC and then design separate table for it...


When i finished the yellow machine from my first build i said to my self one thing- on the next one i will do all possible to avoid laser cutting and massive laser cut part list and i see here lasercut steel that weights a lot and $$$ to cut.


Plus actually similar sized steel tubes are not stronger or more ridgid than box section:

100x100x4 moment of inertia 236cm4 , torsion inertia 353cm4 Shear modulus/rigidity 36.86cm3
100x4 tube 139cm4 278cm4 55.33cm3


Your mistake here is that yes, 200mm x4 steel tube is much more stronger and rigid than 2 x 100x100x4, but imitating it with 3 tubes and a lot of steel ribs does not do the same job . The questions with 200x4 tube is how to fix the rails and how to deal with overhang/the bulge of the tube/ mantaining at the same time the spread between the rails.



So, box profile structure beats all in simplicity and strength, especially if made stair like with short pieces in perpendicular direction. Sides and gantry.

Please excuse me about the colors,but it helps me better express my thoughts and organize them. I tried not to make all in red and bold :hysterical:

Blackrat
10-11-2014, 07:12 AM
Have you priced up laser cutting yet ?

One large dia tube is going to be much stiffer

mitchejc
10-11-2014, 09:22 PM
Silyavski, thanks for the detailed reply, I do value your input. Like many others on here I'm no mechanical engineer and I go with the standard LAR approach which does not always yield scientific correct results, so please bear with me.


No,those thin parts are old fishing rods that I want to glue gun in there just to stiffen up those unsupported sides a little. Yep, its threaded rod just for for alignment when welding :-)


The orange part is 3 x layers of 18mm quality birch plywood glued together and bolted to the base.Considering the thickness of the wood and close spacing of the steel supports under the plywood, and big surface area of the X-axis sides, I think this will be more than strong enough to bolt my X sides to without, right? There's various good reasons why I 'd prefer NOT to weld the X sides directly to the base.


I totally agree, the base structure can be build stronger for less money, but its already build, so I guess that ship has sailed :-)


I sort of agree about the X sides that's not fully supported. I thought if the rail sits on top of 10mmalu on top of 10mm steel flat bar thats on top of the ribs spaced200mm apart it would be good enough? I'll fix that and make it stronger.


As drawn the steel parts of the gantry weighs +- 60kg if I remove the 750 x 100 x 10mm piece of flat bar which does not add much apart from weight. That gives me +-100kg including z and motors which is getting a bit on the heavy side. I can shave off weight by replacing some 10mm ribs with 6mm. I also agree, in this case the round tubes does nothing apart from cosmetics (which is also important to me) but does complicate some things a little bit. I'll get rid of them and go for something more practical. Would that be a step in the right direction or is the whole steel section and rib design just a bad idea?


I love welding but I hate cutting and grinding steel, and accurately marking and drilling big holes in steel plates is not much fun either, so the laser or hi-def plasma approach just seemed more suitable. If I may ask why are so against the laser approach apart from the additional cost?


I'm glad you did not do everything in bold red as that might have pushed me over the edge an got me to build the whole damn thing from polystirene and cold glue :cheerful:


Thanks again for helping me to get this right!

mitchejc
10-11-2014, 09:36 PM
Have you priced up laser cutting yet ?

One large dia tube is going to be much stiffer

Nope, I have not. I did get some parts laser cut about 6 or 7 years ago and at the time the price seamed very reasonable. I hope I'm not in for a nasty surprise. Until I get a design past the cnc police there's no point in costing the cutting:cheerful:

Yep, the big dia tube is much stiffer but there's just no practical way I've seen to make it work for a gantry router due to the reasons Silyavski listed above.

Boyan Silyavski
10-11-2014, 11:05 PM
I see.

-IMO is very bad idea to lay the machine on anything else but steel. I will advice against it. I mean the orange part as you have drawn it is bad idea.

-I agree the design should please your eye. I wouldn't do it that way at all as my moto is "maximum efficiency, minimum effort" . I mean i advice against using steel tubes at all. Though as i see it the gantry could be left as it is if you insist on keeping the design, just take care the plate bellow the rails to be truly all supported, not hanging in the air.

-The sides definitely could not be as they are now


I mean, change everything, but if you insist keep the gantry :hysterical:

Sven
11-11-2014, 07:55 AM
I haven't read this thread for some time now and I am surprised at the design of the router as it is now. It looks really nice.

Am I correct in that you will have pieces cut by laser as part of the build?

If so, why not build a steel torsion box from nothing but laser cut sections?
That would save a hUge amount of time, and be stiffer too.

mitchejc
11-11-2014, 08:04 PM
Yep Sven, the idea is to have the parts laser cut. Interesting idea. I'm sure what you are suggesting would be possible and EXTREMELY stiff. Its amazing how stiff some of the light torsion box or honeycomb composite structures are like on the photo below

http://www.singcore.com/images/Metal-Strong2.jpg

Sven
11-11-2014, 08:30 PM
I'm currently in contact with a guy who is building an extreme fixed gantry router. He can calculate this stuff in to extremes.
He suggested to me that a steel torsion box is nearly as stiff as a solid steel one and when designed right wil only need spot welding to keep it together, but can even be glued if you have the right technology available.

mitchejc
21-11-2014, 07:53 PM
Ok, now its down to bartering. A friend needed some aluminium house numbers urgently and he had something I wanted, so I cut the numbers Wednesday night and swapped this
http://www.mycncuk.com/attachment.php?attachmentid=13892&stc=1
for this
http://www.mycncuk.com/attachment.php?attachmentid=13893&stc=1

Great deal, don't you think? :-)

Back to plan A of using I-beam for the gantry sides. The beam I got was a bit higher (254x148) than I hoped for but I'll make it work. I guess for me DIY CNC is more about what I have/can get/can build rather than building the ultimate design from the best choice of material. I also listened to the advice and went with a slightly more practical approach and used 76x76 square tubing, and some 10mm and 5mm plate and also bolt or weld the X sides directly on the base. Gantry weight is +- 60kg excluding Z. I know Y rails top bottom is the better approach but I'm thinking its going to much easier for me to get them parallel and on the same plane if I put it on the front. I'm hoping the laser cutting is not going to be too expensive otherwise I'll have to go for plan B with the gantry. See pics below.
Again please comment if you see any obvious issues.
http://www.mycncuk.com/attachment.php?attachmentid=13894&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=13895&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=13896&stc=1

Boyan Silyavski
23-11-2014, 04:34 AM
Looks awsome. Will need a lot of patience welding.

mitchejc
23-11-2014, 07:17 AM
Thanks Silyavski, yep that's going to be many many hours of welding. I'm a bit worried that the two square tubes on the front will bend a little during welding as its only welded on three sides but I'm hoping it will be ok. Maybe I'll only weld them top and bottom and not the back to prevent distortion. Do you think I need some ribs on the inside of the I-beam to prevent flex in the Y direction or is it going to be stiff enough with just the bearing and motor mount plates welded in?

JAZZCNC
23-11-2014, 11:40 AM
Your making life much harder and expensive than it needs to be with all those rib plates.!!. .:dejection:

Two large thick walled pieces of box section arranged in a L shape with a few triangle braces at the rear will easily match your gantry in strength and be much cheaper and simpler to achive. It will be much easier to achive accurecy as well because with that thickness of steel on those square tubes then heat from welding will almost certainly have an affect either pushing or pulling on the boxsection with so many braces to weld, even on just 2 sides.
If you used a Tig welder you may have chance but even then I'd be surprised if it doesn't.! You will certainly have to take it very very slow keeping welds short and leaving plenty of time between to stand any chance.

Your design will be strong but no stronger than more conventional setup and 3 times harder to achive and probably 3 times more expensive.?

Also Please don't tell me this is just for cutting wood with occasional aluminium use.?. . . .:smiley_simmons:

mitchejc
23-11-2014, 04:06 PM
Thanks for the input Jazz. I totally agree with the proven L shape being stronger per $. I'm not too worried about setting it up as the whole thing can be bolted together and adjusted before welding BUT I only have ARC. A bit of bowing of those tubes is not going to be the end of the world but if I get twist its going to be a throwaway so I'm seriously reconsidering my options. I can't get any square tube with walls thicker than than 3.5mm at my local steel shop so that's another curve ball I have to try and dodge with whichever design I'm going.

>>Also Please don't tell me this is just for cutting wood with occasional aluminium use.?

Yes, but I also want it gentle enough to cut little harts on pink mash mellows and then maybe the odd piece of brass in between :-)

JAZZCNC
23-11-2014, 11:07 PM
A bit of bowing of those tubes is not going to be the end of the world but if I get twist its going to be a throwaway so I'm seriously reconsidering my options.

Any movement of those tubes is bad news that will affect accurecy and take plenty of work to correct. This is the problem with such ridged design and complex setup that as limited scope to adjust out any build error.!!

Cutting marsh mellow can be harder than you realise..:ghost:

mitchejc
25-11-2014, 06:48 AM
Jazz I think if I use old bra straps to hold down those marsh mellows they are not going to be that difficult to cut ;-)

I'm pondering my choices here wrt to the gantry... my heart says go with the ribs and my mind says listen to these experienced gents :-) Blackrat on here got me into contact with a local supplier of the right epoxy for leveling so that will make things a little easier whichever way I go.

Boyan Silyavski
25-11-2014, 04:47 PM
OK, now you have a nice picture. We could all agree it's the coolest design we have seen for a while.

Arc welder? Yeah I arc welded half of the first machine I made. But from now I am telling you it would be a mess with this design. Read my build where I changed at the last moment my gantry design cause I couldn't weld it as I designed it first. Luckily for the better.
And I have arc, mig, tag.

You should listen to your heart but also to your brain and your pocket.

At the end it should be doable in real life. That's the important thing.

mitchejc
25-11-2014, 06:30 PM
Thanks for the good advice Silyavski. Fortunately I'm not building the machine to any deadline so I do have some time to rethink the design and if I have to redo the gantry design its ok, that's what I will do. You are right, the pocket is often where bad choices hurts the most :-)

mitchejc
16-12-2014, 08:02 AM
Ok, I spend about three nights trying to redesign my gantry and every time I ended up starting to put ribs in :-) Got a quote on having the parts laser cut for the original design and it came to the equivalent of GBP120-00 so I decided to follow my hart and take a chance against good advice.

Ok, two weeks later and I can now say I FULLY understand what Jazz and sylyavski was trying to tell me and building this gantry design turned out to be a HUGE amount of work!

mitchejc
16-12-2014, 09:15 AM
How do I convert this to a build thread?

Anyway, here's a few pictures of the gantry build. Took 3 full days to assemble and weld. I certainly know my little red arc welder much better now :whistle:

http://www.mycncuk.com/attachment.php?attachmentid=14084&stc=1
Setting it up was relatively easy but took a lot of effort. Think 4x1m threaded rods + 88xnuts and you'll get the picture.

http://www.mycncuk.com/attachment.php?attachmentid=14085&stc=1
setup and squared.


http://www.mycncuk.com/attachment.php?attachmentid=14086&stc=1
test fit ballskrew before welding

http://www.mycncuk.com/attachment.php?attachmentid=14087&stc=1
Weld, weld, weld, weld... Remove flux, turn over, weld, weld weld...

http://www.mycncuk.com/attachment.php?attachmentid=14088&stc=1
Welding done apart from the 2 base plates which can only be done later. Very relieved it turned out pretty straight and no twist.
http://www.mycncuk.com/attachment.php?attachmentid=14089&stc=1

Would I recommend this approach for other first time builders? Well, if you are the kind of person that would describe three days of Chinese water torture as an interesting experience, then this is definitely the design you should go for...:biggrin:

Have to say its an amazing feeling to see the drawings turn into reality!

GEOFFREY
16-12-2014, 05:10 PM
I am afraid that I cannot comment on the design, but that looks AMAZING. Well done. G.

mitchejc
16-12-2014, 09:41 PM
Thanks for the kind words, Geoffrey. Yep, I also not 100% sure how good it will turn out but I'm putting all effort in and I hope to have it cutting in a few weeks time.

EddyCurrent
16-12-2014, 09:53 PM
+1 what Geoffrey said.

JAZZCNC
17-12-2014, 12:37 AM
Well done it's Well OTT and looks Cool but do Hope for your sake it is true and not got any twist because it will be a proper Ba@~?rd to tweak that bugger into shape. . LOL

I'd let it sit for quite a while to release stress's before trying to do any accurate setting up of it.! . . . That's a lot of welds in small area that will want to settle down and move.!

mitchejc
17-12-2014, 08:12 AM
Thanks for the advice Jazz. OTT... please see observation below:-)

The granite slab I welded on is pretty flat and judging from that the two areas where the 2 X base plates will be welded is very much on the same plane, so no significant twist. I would say definitely less than 0.5 mm which is great as I was prepared to deal with 2 or 3mm if I had to. My plan is to weld in the 10mm base plates and then let the gantry settle under its own weight on top of the two alu blocks that sit on the linear guides with metal epoxy in between and release agent on the alu blocks. That way I should get a very good mating surface there and not induce stress bolting it together.

Two observations that might help other newbies.
1) Parts looks a lot more bulky in real life than on your little cad drawing. If you go with 'looks about right' on your cad, it will translate to 'way over designed' in real life.

2) Holding the parts for welding with threaded rod works well but put washer and maybe copper slip on the nuts as they may bind due to movement during welding which makes them hard to remove. I had to cut one rod to release the longitudinal stress that caused two nut to lock up completely.

3) Early in the weld a thin tag weld broke in half with a very loud ping sound during cooling, and I thought this is the beginning of the end as the offset on the broken tag was quite obvious. I did the opposing welds and when that cooled the tag had moved back to almost its original position and you could barely see the hairline crack where it broke.

mitchejc
17-12-2014, 08:58 AM
How do I change this thread into a build log or should I just create a new thread under build logs to post my progress pictures?

EddyCurrent
17-12-2014, 09:47 AM
How do I change this thread into a build log or should I just create a new thread under build logs to post my progress pictures?

Send a PM to Lee Roberts and ask him to move the thread.

mitchejc
17-12-2014, 10:18 AM
Thx Eddy, will do.

JAZZCNC
17-12-2014, 11:51 AM
My plan is to weld in the 10mm base plates and then let the gantry settle under its own weight on top of the two alu blocks that sit on the linear guides with metal epoxy in between and release agent on the alu blocks. That way I should get a very good mating surface there and not induce stress bolting it together.

I use this method along with shimming all the time and it works great but don't use release agent your better with a thin film of plastic between, Cling film is what I use. Release agent doesn't always release good enough and can affect the epoxy curing, try a sample first if you do.

Boyan Silyavski
19-12-2014, 11:45 PM
Parts looks a lot more bulky in real life than on your little cad drawing. If you go with 'looks about right' on your cad, it will translate to 'way over designed' in real life.




That's true. Same happened here. For second time :butterfly:

I did not have internet for a week and see now that the build goes very well. I will call you if i need to design spaceship :triumphant:

mitchejc
22-12-2014, 06:07 AM
Hehe, yes silyavski call me for the help on the spaceship design just make sure you buy big rocket engines as the ship might be a bit on the heavy side :-)

mitchejc
22-12-2014, 06:25 AM
@Jazz
Thanks for the cling film plan, it works great and is a lot less effort than the release agent. Yesterday I was mating the surfaces where my x sides bolts to my frame and I already laid down the epoxy when I realised I don't have enough cling wrap left and had to make a plan in a hurry so I ripped a page out of a magazine that was lying there. That glossy paper worked remarkably well. The paper obviously stuck to the epoxy side but I removed it with a wet cloth afterwards. Funny thing is the ink remains so you end up with an inverse print on the epoxy surface of what was on the paper.

mitchejc
22-12-2014, 11:19 AM
I really want to put sand in the three sq tubes in the gantry but I'm getting worried about the weight of my y and z. I'm not actually at the point in the build where I'm going to put the sand in but I need to decide now if I'm going to as it affects other choices. I know its a bit late in the process but I'm a bit shocked after actually totaling the weight of my y and z components. Comes out to +- 128kg. The sand will add another odd 20kg's. I'm willing to sacrifice performance for cut quality but maybe I've taken it too far. Not sure if I completed everything correctly but for my X calcs the motor calc spreadsheet turns red (margin of 2.5x instead of 3x) when I fill in my specs but I don't think its accounting for the fact that I have two motors? Will I be ok with this or should I rather reduce weight and not put sand in?

mitchejc
26-12-2014, 09:16 PM
Sorry, realised above was a bit of a daft question. The spreadsheet does not allow for dual motor but obviously I can just half my weight for the calcs if have dual motors, so after redoing the calcs all looks fine and I'm doing to fill the sq tubes with sand.

mitchejc
28-12-2014, 08:14 PM
Some progress photos. The base was done a while back. I figured its a good place to start to learning welding.

Started with a flat work area and made a jig to get the legs square
http://www.mycncuk.com/attachment.php?attachmentid=14215&stc=1




welded lower frame
http://www.mycncuk.com/attachment.php?attachmentid=14217&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14219&stc=1

welded upper frame


http://www.mycncuk.com/attachment.php?attachmentid=14221&stc=1

lower and uper frame done
http://www.mycncuk.com/attachment.php?attachmentid=14222&stc=1

spot welded legs into place using the jig
http://www.mycncuk.com/attachment.php?attachmentid=14223&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14224&stc=1

welded in braces after the top and bottom frames were welded to the legs.
http://www.mycncuk.com/attachment.php?attachmentid=14225&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14226&stc=1

basically done apart for some diagonals
http://www.mycncuk.com/attachment.php?attachmentid=14227&stc=1

mitchejc
28-12-2014, 08:39 PM
Building the x sides with I-Beam and laser cut parts.

Cutting the i-beam. Thought this was going to be a huge issue because the beam I got is much larger than my cut-off saws capacity but it was done a few minutes later using two cuts and the angle grinder for the remaining bit in the middle.
http://www.mycncuk.com/attachment.php?attachmentid=14228&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14229&stc=1

I got all the holes lasered way oversize and made alu inserts for them. Just figured its a lot easier to ream alu bushes should I need some adjustment later to align things
http://www.mycncuk.com/attachment.php?attachmentid=14230&stc=1

welded in ball skrew mounts
http://www.mycncuk.com/attachment.php?attachmentid=14232&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14231&stc=1

welded in motor mount
http://www.mycncuk.com/attachment.php?attachmentid=14233&stc=1

some test fitting just to see how these parts goes together
http://www.mycncuk.com/attachment.php?attachmentid=14234&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14235&stc=1

I think using i-beam for the sides were a good plan and really easy to assemble and I can recommend this kind of approach for other newbies.

mitchejc
28-12-2014, 08:59 PM
More work on the x sides. Drilling large dia holes by hand in the thick I-beam turned out to be a mission so I made two mdf boxes and drilled it on my small bench drill.

http://www.mycncuk.com/attachment.php?attachmentid=14236&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14237&stc=1

For basic alignment of the x-sides I used the centres of by ball skrew mounts
http://www.mycncuk.com/attachment.php?attachmentid=14238&stc=1
Once aligned I welded in tabs to make re-alignment easy
http://www.mycncuk.com/attachment.php?attachmentid=14239&stc=1

Used steel epoxy putty as per Jazz's suggestion for shimming. Worked like a dream. I can't imaging doing a steel home build without this stuff as there's just no way a beginner is going to weld something of this scale together and end up with perfectly square mated surfaces.
http://www.mycncuk.com/attachment.php?attachmentid=14240&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14241&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14242&stc=1

x sides almost done.
http://www.mycncuk.com/attachment.php?attachmentid=14243&stc=1

GEOFFREY
29-12-2014, 01:26 AM
Looking great, and your "shed" is to die for. G.

Boyan Silyavski
29-12-2014, 09:53 AM
Looking great, and your "shed" is to die for. G.

I noticed that too. Lucky man. Great development!

mitchejc
29-12-2014, 12:06 PM
Thanks gents. The shed was actually build years ago as sort of an entertainment area which I started using as a workshop as there's just no space left in my garage. The tiled floor is a bit of a drawback because its not very flat and my wife still treats the area as hers so she does regular inspections for paint, glue and weld marks on the floor :-)

Have to say I did not realise that building the router was going to this much work! Everything is taking much longer than I thought but I'm enjoying every minute of it. Ahhh well ok not every minute but most of them.

mitchejc
31-12-2014, 07:14 AM
After Jazz's stern but true words about over design I do feel a little bit guilty to continue posting pictures but heck, if its not right please comment, so we can all learn.

Completed the alu base plates that joins the gantry to the x rails. I've made the plates adjustable in both y and z direction so I don't have issues putting it all together later. Had the spacing of the ball nut mount wrong on my cad drawing so I had to make little adapter plates and pockets.
http://www.mycncuk.com/attachment.php?attachmentid=14268&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14269&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14270&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14271&stc=1

Ran a belt sander with 80 grit belt over the alu which did give it a rather rather nice brushed kind of finish.
http://www.mycncuk.com/attachment.php?attachmentid=14272&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14273&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14274&stc=1

Done and all holes drilled and tapped. Thought these we going take a few hours but ended up more than a days work
http://www.mycncuk.com/attachment.php?attachmentid=14275&stc=1

Like most of my parts the actual size of these surprised me when seeing them in the flesh and I probably could have gotten away with much thinner material?

JAZZCNC
31-12-2014, 11:29 AM
After Jazz's stern but true words about over design I do feel a little bit guilty to continue posting pictures but heck, if its not right please comment, so we can all learn.

NO NO please don't stop posting and my post wasn't aimed to discourage you or anyone else from building like this if they have capabiltys or means.
It soley to let those sitting on the side lines watching and wondering if this is the level they will have to build too know that it really really isn't.

You Sir are doing a marvelous job and for a first build I take my hat to you, thou I think you may have bushel in your thatch loose for doing it like this. . .Lol

Keep it up and don't let my rantings sway you it's the last thing I want.! . . . Just Like I don't want Newbies swayed away.!

mitchejc
31-12-2014, 02:25 PM
Thanks for encouragement Jazz! The reason I felt a bit guilty was because I understood what you meant with your post e.g. I've seen this with air rifles as well were a few guys buy and post about extremely high end target rifles even thought they don't know which end should point forward and other newbies think that's what you need to have fun. After bonding the house and downgrading the car to buy a big $$$ gun they realize they actually had more fun shooting beer cans with the little budget plinker rifle they had. Now, I'm that guy that does not know sh%t but I'm still posting my stuff... :-) When all is done and dusted I'll post my "lessons learned" which will hopefully be helpful to other first timers.

Anyway, thanks, I only got great info and advice on here and I'm having a BALL trying to hack this beast together :peaceful:

>>> bushel in your thatch loose
Hehe, yes I get that often from friends. Why is someone considered nuts just because the they have the dedication and commitment of a suicide bomber towards their hobbies:highly_amused:

mitchejc
11-01-2015, 09:25 PM
A few more pics of progress on X.
Went with the Silyavski 8mm epoxy dams.
http://www.mycncuk.com/attachment.php?attachmentid=14368&stc=1

Thought I'd cast a T shape on my base to eventually use as reference to do initial setup of the gantry and Z. Things got out of hand as per usual and I had spare epoxy and did the whole base. Used modeling clay for the dams. VERY bad plan, don't do it. It works nicely to build the the dams but its a total bitc$%$ to remove the later.
http://www.mycncuk.com/attachment.php?attachmentid=14370&stc=1
base done
http://www.mycncuk.com/attachment.php?attachmentid=14371&stc=1
I made a little tent using a blanket and plastic sheet and stuck a little fan heater under there to get the temp up to about 50degC for proper post curing. You guys in the cold places might find that handy.
http://www.mycncuk.com/attachment.php?attachmentid=14372&stc=1

X sides painted light gray. The base is just white primer and is not going to be painted as it will be fully enclosed anyway.
X epoxy done and fully cured.
http://www.mycncuk.com/attachment.php?attachmentid=14373&stc=1

mitchejc
11-01-2015, 09:37 PM
Finished the last work on the gantry before it goes on the machine.

Did the last bit of welding for the bottom plates and epoxy dams
http://www.mycncuk.com/attachment.php?attachmentid=14374&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14375&stc=1
When I did this it was raining everyday here in sunny South Africa so I had no choice but to spray inside. This reminded me why I hate to spray indoors. Managed to get it sprayed but also the floor, myself, my tool cupboards and everything else in the near vicinity :-)
http://www.mycncuk.com/attachment.php?attachmentid=14376&stc=1
starting to make things ready for the epoxy. Did not take photos of the epoxy process but you you know the drill.
http://www.mycncuk.com/attachment.php?attachmentid=14377&stc=1

mitchejc
11-01-2015, 10:02 PM
Some work I've done on the Z parts during the last few weeks while I was waiting for epoxy to cure and paint to dry etc.


Here's some of the laser cut steel parts. Z will be steel and front plate from alu. I wanted my Y and Z carriages on top of each other in sort of a + configuration. I did not want any welding on the actual plates so all the parts just bolts to the plates. This might not make sense but I'll post more pics to hopefully explain it a bit better.
http://www.mycncuk.com/attachment.php?attachmentid=14378&stc=1
Welding the z ball skrew mount
http://www.mycncuk.com/attachment.php?attachmentid=14380&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14381&stc=1
After some serious welding, grinding and sanding
http://www.mycncuk.com/attachment.php?attachmentid=14383&stc=1
Managed to get it a little out of square to I'll have to epoxy shim my fixed bearing side a little
http://www.mycncuk.com/attachment.php?attachmentid=14385&stc=1

welding part of the z stepper mount
http://www.mycncuk.com/attachment.php?attachmentid=14379&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14386&stc=1

NB. Newbies please take note!! These leather gloves were new when I started this project, which should give you an indication of the amount of work that goes into building a CNC. ;-)
http://www.mycncuk.com/attachment.php?attachmentid=14388&stc=1
but I found this helps a lot, before, after and during the process:friendly_wink:
http://www.mycncuk.com/attachment.php?attachmentid=14384&stc=1

EddyCurrent
11-01-2015, 10:18 PM
Can't wait to see it all put together, it's a work of art.

mitchejc
11-01-2015, 11:19 PM
Most of the parts are now made and I've started assembling things. Still lots of drilling and tapping to do though.

Other newbies take note: When you ever post a picture like this and don't realize you have "slightly" over-designed your little 600x500mm DYI router, then sir, you are an idiot :wink: Ok, its not really that heavy but its more than uncomfortable for one person to lift.
http://www.mycncuk.com/attachment.php?attachmentid=14389&stc=1

setting up x rails
http://www.mycncuk.com/attachment.php?attachmentid=14390&stc=1

http://www.mycncuk.com/attachment.php?attachmentid=14391&stc=1

Made these little adjusters front and back of both gantry bases (stainless allencap in pic below) to help adjusting things before I epoxy putty mated the gantry to the alu base plates. I'll use them again for final epoxy shimming if required once the z is on.
http://www.mycncuk.com/attachment.php?attachmentid=14392&stc=1

Aligning and measuring stuff. Really pleased with how things are turning out and I'm cautiously optimistic I'll get the accuracy I was hoping for.
http://www.mycncuk.com/attachment.php?attachmentid=14393&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14395&stc=1

No mater how well you think things thru beforehand there's always going to be that one hole you have to drill in an impossible place. No turning back now so just have to make a plan...
http://www.mycncuk.com/attachment.php?attachmentid=14396&stc=1

X and Y rails and X ball skrews fitted. Have to compete the Z back plate before I can fit Y ball skrew
http://www.mycncuk.com/attachment.php?attachmentid=14397&stc=1

Test fitting parts of the z back plate. Still quite a bit work left on the z but I hope to complete that next weekend.
http://www.mycncuk.com/attachment.php?attachmentid=14398&stc=1

So close but so far....

GEOFFREY
12-01-2015, 12:29 AM
WOW. A little OTT perhaps, but awesome - absolutely mighty. I think that this might just qualify for the best homebrew ever. Geoff.

mitchejc
12-01-2015, 03:32 PM
Gents, thanks for the kind words. I really hope the end product is going to be be worth the effort that going in :-)

mitchejc
27-01-2015, 06:49 AM
Fitted and aligned the Y ball skrew and complete Z axis, so most of the mechanical parts are done now.

z front plate before all the drilling and tapping
http://www.mycncuk.com/attachment.php?attachmentid=14523&stc=1

Had to but join rails for my Z due to a change in plans. Was quite surprised as the carriages runs very smoothly over the joint with absolutely no click or resistance, so it seams feasible to join rails if one has to.
http://www.mycncuk.com/attachment.php?attachmentid=14524&stc=1

z stepper mount
http://www.mycncuk.com/attachment.php?attachmentid=14526&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=14525&stc=1

100mm spindle fitted just to see if all the holes line up etc.
http://www.mycncuk.com/attachment.php?attachmentid=14527&stc=1

Now time to start with the control box which is about where I run out of skill and creativety. Wish Eddy could pop over to my place to help a bit:-) I sort of know what needs to be connected but I cant figure out a good layout to do a neat job with routing all that wires etc.

EddyCurrent
27-01-2015, 11:05 AM
Wish Eddy could pop over to my place to help a bit:-)

Bit too far I think :highly_amused:
Here's a few things to keep in mind;

1. Use a proper metal electrical panel with backplate
2. The size of the panel is not the space you get, it's how big the backplate is and the distance it is from the inside of the door.
3. Two items are sort of fixed, terminals across the bottom or up one side, a door mounted isolator positioned along the edge of the door that opens.
4. Use CAD to draw rectangles representing the guts of the enclosure, i.e. backplate, terminal rail, door isolator, drivers, power supplies, bob + interfaces, fuses/circuit brreakers, relays, trunking, vfd, etc.
5. Just drag the rectangles about to get the best layout, the space you need is going to be bigger than you think.
6. Obey any rules regarding ventilation, and keep large transformers away from sensitive items like the bob
7. Use the cable trunking to route power cables in one and signal cables in the other.
8. The backplate will have an earth point, use this as your earth star point, you can use the DIN rail mounted earth terminals I used to make life easier.
9. Use proper terminating glands for the type of cable you will be using
10. Get a crimping tool and some bootlace ferrules, cable number markers make a better job too.
11. Use tri rated single core cables inside the panel, check the standard colours for panels used in your area, e.g. here it's black for AC power, blue for 24 DC, etc.
12. Make provision for cooling fans, ideally some blowing in and some sucking out. The ones sucking in should draw air through a replaceable filter, you need to keep a higher pressure inside
the panel so dirt is not sucked in through gaps.
13. Use DIN rail to mount everything.
14. Use quite deep trunking, it fills up quickly.
15. there must be more :eagerness:

mitchejc
27-01-2015, 01:37 PM
Bit too far I think :highly_amused:


Eddie, are you sure? Only an 11 hour flight I'll supply all the beer and sunshine you can tolerate :-)


Thanks for the very detailed reply, much appreciated. I'll go thru it with a fine comb tonight and google the stuff you are referring to that I'm not familiar with.

T0rnado69
28-01-2015, 02:56 AM
I'll supply all the beer and sunshine you can tolerate :-)

you got your ticket yet eddy :)

mitchejc
28-01-2015, 05:59 AM
Outch, got some prices on steel cabinets. My budget is experiencing a severe hangover to get the machine this far so are there perhaps any cheaper options that's feasible and then hopefully upgrade later? My other problem is that I don't have any tools to cut holes etc into a steel box so I have to factor that into the price. I understand why a steel cabinet is the best choice but can one get away with a wooden box with a metal back plate or something to that effect?

Blackrat
28-01-2015, 09:59 AM
lol i almost fell over at the prices ....

i went to visit a friend one day, and found an awesome box in his neighbors garage which he said i could have !!!

look into having a 2mm box laser cut, its still cheaper than buying one, you just have to draw it

+1 on the SIZE of the box , my box is quite big .... or so i thought lol

Clive S
28-01-2015, 10:09 AM
My other problem is that I don't have any tools to cut holes etc into a steel box so I have to factor that into the price. I see a pillar drill and hand electric drill in your shop. :highly_amused: 600 x 600 x 300 should more than big enough, You don't need to put the VFD inside the control box. Just how big a box was you considering? ..Clive

http://www.mycncuk.com/attachment.php?attachmentid=14535&stc=1

mitchejc
28-01-2015, 08:23 PM
@blackrat, maybe laser cutting sheet metal is a good plan as I know its going to cost way less than what these enclosures cost but I'm probably going to have a very hard time to arc weld it together. I've tried welding thinner material but it was hit and miss and I burned holes like you wont believe even with 2mm rods, so there's much to learn in that department.

Clive, now that's a great looking box! Those LCD displays really got my imagination going... Can I extend the offer for free beer and sunshine to you :-) I'm a bit scared of butchering an expensive box with what tools I have. I have not done the layout yet as suggested by Eddy but I was guessing maybe 600x400x300 as a ballpark? Content would be 1x PC ATX Powersupply, BOB, VFD(vfd face outside the box) 4 x Drivers and few relays. The driver are quite big at something like 180x160x60 if I recall correctly and I would like some spare space for future expansion.

Excuse my ignorance but what's those white square things in your box?

mekanik
28-01-2015, 08:58 PM
Try and get smaller rods, years ago before i got my TIG setup i was trying to patch up a car and only had a stick welder but got some rods from the Local BOC outlet, think they were about 1mm dia but had a substantial flux coating and was told it contained iron fillings, worked OK on thin plate so 2mm plate you should be fine if you could find some.
Know the feeling about the price of steel cabinets, Norweb insisted i had to buy a specific size/type of cabinet/kiosk. It cost 600GBP
Regards
Mike

Clive S
28-01-2015, 11:35 PM
@blackrat
Clive, now that's a great looking box! Those LCD displays really got my imagination going... Can I extend the offer for free beer and sunshine to you :-) I'm a bit scared of butchering an expensive box with what tools I have. I have not done the layout yet as suggested by Eddy but I was guessing maybe 600x400x300 as a ballpark? Content would be 1x PC ATX Powersupply, BOB, VFD(vfd face outside the box) 4 x Drivers and few relays. The driver are quite big at something like 180x160x60 if I recall correctly and I would like some spare space for future expansion.

Excuse my ignorance but what's those white square things in your box?I don't think you will get all that in that size box. I would think twice before putting the ATX inside the main control box as beside the board you would need a power supply, hard drive fan etc plus all the connecter to the monitor, keyboard mouse.
.
The LCD displays were from China about £2 each, the housing for them was done with a 3D printer. The white square things are the 5V and 24V power supply's there also another converter to step up the 5V to 12V to power the motion control board. 24V for the E-stop and 5V for the BOB.
.
I don't think many people on here put the VFD inside the control box. Remember most of these box housings have a bottom plate that you remove and make the holes for the connectors so if you f#%~k it up you just make a new plate.
.
Does the free offer include the flights. :yahoo: :beer: only kidding. ..Clive

EddyCurrent
29-01-2015, 12:05 AM
As I said earlier;

5. . . . , the space you need is going to be bigger than you think. :whistle:

If you don't do it on CAD then cut some cardboard shapes full size to represent the electrical items and shuffle them about.
I was going to say, if you look at standard panels that will give you some dimensions to aim for. It troubles me that after all that excellent mechanical work you want to scrimp on an enclosure.:sorrow:

TonyD
29-01-2015, 03:07 AM
Dont forget safety circuitry, ideally a Pilz, which you can get second hand off ebay. If you cant get a Pilz then folks make do with small relays.

mitchejc
29-01-2015, 06:13 AM
Gents, thanks for all the great info! I'll do the layout this weekend and figure out what size I actually need but I'm getting the feeling I'm in for a nasty surprise. I don't plan to put the PC inside the box and the PC power supply is only to power my bob, switches and relays. I'll read up on the Pilz thing to figure out what its for.

1) What's the reason for not putting the vfd in the box? Heat or electrical noise? Mine has lots of ventilation holes in it and I was thinking the best way to keep aluminium chips out of it was to put it inside my box but if that's a bad plan I need to rethink my strategy?

2) I got 4 x 80mm PC fans for the cooling. The uncool neon green ones with the LEDS but they were cheap:-) Was thinking 3 sucking air in thru a filter at the box bottom and one out at the top left. Would that provide enough airflow considering the climate is quite hot here and 35deg Cel. or higher is not that rare. I tried one just to get a feel and it did not exactly blow my hair back at 12V so I don't know.

I really don't want to scimp on the box but if I want see this thing cutting in the next few months I don't much of a choice. Its going to be incredibly hard to be patient while saving up for a box...

Boyan Silyavski
29-01-2015, 09:47 AM
-I do it like Eddy said, shuffling the pieces until i am happy.

-The VFD for a 2-3kw spindle is prohibitively big to put inside enclosure. Plus it could mean more thick cables and so on. Plus the cooling flow. Better put that somewhere you can see it all time.

-About the fans-make them all not to suck. This is wrong, it sucks :hysterical:. Make them take away the air and fit exchangeable air filter somehow. Cheap car motorcycle tuning filter or whatever similar. You don't want dust inside.

-I dont have PILZ, could somebody explain how, why?




Now just another idea:

I also hate spending money on expensive boxes. On any boxes. On my current build i decided to use whatever i had at hand. So i had some plastic sheet and 1mm aluminum with self adhesive backing. I fixed them together and used as legs the M5 allen screws for the Hiwin, as i bought a 500pieces box.

Most importantly i decided al the cables to go at the back and not be seen, plus the aluminum plate screens them additionally.


When i finish the machine i will build myself an enclosure . At least when i spend money will spend them properly and buy aluminum and make it beautifull or at least as i like.

As i left at the left/right side of the plate 15mm free, this will slide into the new box where cheap aluminum U profile will play the role of slide.

Some pictures to get an idea. Yeah, i was tired so some holes are not drilled in a line, i have to repair that. Money should be wisely spend on a DIY build. Buying the connectors and terminals separately and soldering them in my case resulted at least in 350eur savings.

Do you see the empty space at the right side of the drives? thats for eventual 5th 6th axis. The empty space bellow the BOB also is for whatever. When i started the build i did not know. Now i know for what it will be.
LEAVE EMPTY SPACES. YOU WILL NEED THEM LATER.


http://www.mycncuk.com/attachment.php?attachmentid=14538&stc=1 http://www.mycncuk.com/attachment.php?attachmentid=14539&stc=1http://www.mycncuk.com/attachment.php?attachmentid=14540&stc=1

EddyCurrent
29-01-2015, 10:00 AM
-About the fans-make them all not to suck. This is wrong, it sucks :hysterical:. Make them take away the air and fit exchangeable air filter somehow. Cheap car motorcycle tuning filter or whatever similar. You don't want dust inside.


Whatever you do with the fans there are two things that are a must, there is no other way;

1. Air entering the cabinet must be filtered.
2. The pressure inside the cabinet must be higher then the pressure outside.

My vfd is inside the panel, this is industry norm if done correctly and the manufacturers rules are obeyed. If you do put it in another enclosure then it will also need fans and filters.
Also you can either save money or do it right the choice is yours but only if the machine is for your own use.

This shows a typical layout of a bigger panel but you can see they keep the power control items away from the signal items.

https://www.youtube.com/watch?v=XY3gwpR67WU

Boyan Silyavski
29-01-2015, 10:14 AM
2. The pressure inside the cabinet must be higher then the pressure outside.


Eddie, i believe its the other way around. The pressure in the case should be Negative. Not Positive.

Anyway its relatively speaking, cause if a car filter lets 200HP engine breath easily without obstruction, it could provide enough airflow for some small ventilators :-)


Rule #2: Have More Exhaust Fans Than Intake Fans (Negative Pressure)
http://www.tomshardware.co.uk/faq/id-1858957/airflow-101-setting-fans-keeping-computer-cool.html


PS. Forgot to say before. Heat rises, hence the drives where i placed them, so they don't heat all the enclosure

EddyCurrent
29-01-2015, 10:44 AM
Eddie, i believe its the other way around. The pressure in the case should be Negative. Not Positive.

Well like I've said before, people can do what they want, I've worked in industrial electrics all my life and if any panels or switch rooms had been negative pressure they would have failed on a regular basis because of sucking in crap through every little gap. If you've ever worked in a Steel Works you would know what I mean.
If you can be sure the only way in for the air is via the filter then in this case it might be okay.

The very bottom line here for eaxmple;
"The only air introduced into the cabinet is filtered before it enters the Vortex Cooler. Vortex Enclosure Coolers maintain approximately 6” W.C. positive pressure inside the control cabinet."
http://www.vortexair.biz/Cooling/VortexCooler/vortexcooler.html

Here; http://www.nexflowair.com/panel-cooler.php; "Prevents dirt contamination by keeping enclosure at positive pressure"

And another thing, here in the UK it's not such a problem but in the OP's case where it can get very hot, it's no use blowing hot air into the panel and expecting it to be cooled, at some point a proper cooling system might need to be employed.

mitchejc
31-01-2015, 12:26 AM
Thanks for the help so far.

I'm not sure if this post is to try and validate my assumptions or a cry for help?


I've gathered a lot of info on here and elsewhere and frankly, my head is spinning wrt to the controll box, wood or metal aside. I understand how to connect my drivers and VFD to my controller and I have confirmed my understanding by testing it without blowing anything up but if I look at the great control box projects and circuits on here, there's apparently a hell of lot more to it than that e.g. safety circuits and all sorts of contactors, relays etc. which has me questioning my plan. I'm trying to figure out whats really the bare essentials. I am guilty of some serious OTT stuff but when it comes to the electrics/electronics I just don't have the experience or knowledge to pull off anything elaborate.


My requirements are simple and I'm trying to achieve it in the simplest way possible. Please correct me if I'm missing essential safety requirements


1) I need an Estop that will let mach3 know its pressed and to disable my drivers but will leave the spindle running. My estop switch has a separate NC and a NO circuit so when pressed I'm thinking to use the normally open circuit to send a signal to mach3 and the NC circuit to disable my drivers.
No relays or anything fancy, just 1 x 5V circuit to my controller to let mach3 know and one 5V circuit running thru my 4 drivers enable pins in serial. With this approach I understand if I click estop in Mach3 it will stop sending instructions to my controller but it will not disable my drivers which is fine with me.


2) I'll operate my VFD manually wrt to run and stop and only use my controllers 0-10V output to change speed. Nothing else, no relays etc

3) Don't really have a plan what to do with the fault signals from my drivers? I need to disable all drivers if any one faults but not sure how to accomplish this in the most basic fashion? Maybe that's the stuff I see in other control boxes that I don't have the foggiest clue about :-)

4) Limits and Home switches. I actaully considered not fitting any as I've never run my old machine into the limits but my build is a lot smaller than that so its probably a bad idea. I'm thinking of fitting limit switches for Xmix,Xmax,Ymin,YMax and Zmax with one NC circuit running thru all the switches in serial. I'm not too concerned about homing as my controller do not have the capability for dual x axis home anyway and on my old machine I always cut from a G54,55... offset and I can't foresee me using the new one any differently. I'll configure mach3 to use the limit switches for homing for whatever its worth.


5) My control panel will consist of only the EStop switch and a 220v 24Amp circuit breaker that will cut all power to the machine (controller, drivers, vfd and waterpump) Guess which one I'll be going for when the sh%t really hits the fan :-)


6) My waterpump will run all the time when the main circuit breaker is on. In time I'll figure out how to use the build in relay in my VFD and use that to switch the waterpump on and off.


7) Apart from proper grounding of all the 220V stuff I have no additional safety anything, no contactors, no filters, no relays, no nothing.


Is this a realistic workable plan as a first stab at my control box or am being ignorant?


Sorry for the long winded post!

Boyan Silyavski
31-01-2015, 12:56 AM
what controller you are using? If it works without a plugin with mach3 it certainly can home double long axis separately.


It will be best to start stop the spindle with relay and control the RPM via 0-10V. What is the model of your VFD?

We will help here so you do it right. Its not difficult.

TonyD
31-01-2015, 01:49 AM
Mitchejc, its totally feasible just take your time.
When I started I looked into the electrical before the mechanicals (which was a mistake) and I think in the end had I not stopped I'd have had 3 iterations of my control box, the first one had no safety and Jazz and Andy intervened to help.

I would start with power, get the PSU's in and working, then get the safety system arranged to kill the power as needed with the combination of limits and estops. Then I'd go for the bob and make sure that the safety circuitry can send the right signals to that. From there I'd work on home and limits to the bob and then you've got the guts of a working control system, motors and drivers are usually a function of the right wires in the right place, not much scope for creativity or screwing that up, maybe there's some motor tuning via drivers but that’s icing once the rest is done, not saying its not important, just most likely done near the end.

The critical thing in IMO right now is figure out the safety and control (limits and home), there's lots of good advice on the forum, limits NC, estop the same etc.. Personally I found it challenging to get into all the relays and drew comfort from the reliability of the pilz, they are available second hard and if you can get one I don’t think you'll regret it. Even starting out with some power and a pilz with a reset button and the estop circuit will build your confidence a lot.

Lots of folks have great advice, some of it is a bit overwhelming when you're a beginner, I'd proceed as outlined and add the frills like spindle control, water pump, etc once you’ve got power, safety and comms to the PC via the bob sorted.

Ps. I needed a bigger box each time I added a new feature, either get one bigger than you need or figure out the basics without a box first.

EddyCurrent
31-01-2015, 01:11 PM
The panel for my own machine was called "OTT for DIY" at the time but I disagree, for me it was just normal. For example if you could easily produce mechanical items to a particular tolerance and finish, why would you choose to make one inferior ? I can see that cost is a factor but that does not mean you make it less safe, you just have to reduce the number of features while keeping the core safety aspects.

Lee Roberts
31-01-2015, 11:33 PM
I think the thing going amis here guys is the difference in cooling vs contamination free.

So, what you should have is a negatively pressurised enclosure but also have filters on the intake to stop contamination.

Cleaning a filter would typically be less evolved than cleaning a populated enclosure.

Like Eddy says, do what you want but put "safety" as close to the top of the priority list as you care for, heat control is one of the higher items on my safety check list ;-)

.Me

Boyan Silyavski
01-02-2015, 12:04 AM
Something like this is perfectly enough http://www.hobbyking.com/hobbyking/store/__18020__Hobbyking_Race_Air_Filter_1_5_Gas_Car_.ht ml
Most of these filters filters if they are a good brand, could filter up to 5 microns with excellent flow.

http://www.mycncuk.com/attachment.php?attachmentid=14568&stc=1

Definitely the enclosure has to be airtight apart from the filter , i thought that was clear, who wants to spend time cleaning enclosures. And a good thing to note is that when cleaning filter enclosure with a compressor, we Must be sure the air is dry. Spitting water drops on a couple of hundreds euro BOB could hurt.

But as i see it, there is not a law or something. I would say - negative pressure if all is in straight line and easy to cool. Intake ventilators to help the airflow if necessary, especially when there are some obstructions, otherwise-not. Hot things, air out- up in the enclosure, intake from bottom. All in straight line without obstruction if possible.

I would say also - forget 12v PC fans and similar. 220V - ~20-30w ventilator or 2. One all time, second when necessary. Little OTT here will not hurt.

mitchejc
01-02-2015, 06:32 PM
what controller you are using? If it works without a plugin with mach3 it certainly can home double long axis separately.


It will be best to start stop the spindle with relay and control the RPM via 0-10V. What is the model of your VFD?

We will help here so you do it right. Its not difficult.

Thanks silyavski. I'll be using a cheap controller I bought before I joined this group and I hope to upgrade to CSLABS CSMIO/IP-M later this year so no dual axis homing.

In bench testing I've been able to control my VFD speed from controller but I'm unable to figure out how to switch it on and off with my controller. My VFD is a Sunfar E300. The model number says E300-2S0022L. Attached is the manual

http://www.mycncuk.com/attachment.php?attachmentid=14570&stc=1&thumb=1

mitchejc
01-02-2015, 06:56 PM
Mitchejc, its totally feasible just take your time.
When I started I looked into the electrical before...

Tony, thanks for the info and reassurance. Your suggested approach makes sense and I'm going to follow that. I had a look on the PILZ website but there's a lot of stuff. Sorry for my ignorance but what does this thing do on a high level and where does it fit in? Which model number should I be looking for to get some docs?

mitchejc
01-02-2015, 08:08 PM
The panel for my own machine was called "OTT for DIY" at the time but I disagree, for me it was just normal. For example if you could easily produce mechanical items to a particular tolerance and finish, why would you choose to make one inferior ? I can see that cost is a factor but that does not mean you make it less safe, you just have to reduce the number of features while keeping the core safety aspects.

Thx Eddy, I get what you are saying. What is the absolute minimum core safety aspects considering I'm the only one who is ever going to be using this machine and I do have 100's of hours machining time on my old router so I've got a fair idea of what can go wrong from an operating point of view? In my mind estop is a nice to have as long as I have a nice big circuit breaker handy that will cut power to everything :-)

mitchejc
01-02-2015, 08:29 PM
Something like this is perfectly enough... Little OTT here will not hurt.

Thx, these type of filters are a good plan. I think you are right about the 12V PC fans, they are weak and I'll go for a bigger 220v fan or two. I'm not sure I'll be able get my box 100% air tight where the wires come in etc so it might be better for me to put my big fans on the bottom (cooling air input side) with filters to try and get a bit of positive pressure then I don't have to worry about small aluminium chips getting into my box.

Boyan Silyavski
02-02-2015, 06:32 AM
Thanks silyavski. I'll be using a cheap controller I bought before I joined this group and I hope to upgrade to CSLABS CSMIO/IP-M later this year so no dual axis homing.

In bench testing I've been able to control my VFD speed from controller but I'm unable to figure out how to switch it on and off with my controller. My VFD is a Sunfar E300. The model number says E300-2S0022L. Attached is the manual

http://www.mycncuk.com/attachment.php?attachmentid=14570&stc=1&thumb=1




These are the basic steps to make the VFD work with your spindle:

F0. 0 1: External input signal (0~10V / 0~20mA)


F0. 2 1st part of LED 1:Control by external terminal
F0. 2 3rd part of LED 1: Valid
F0. 2 4th part of LED: Self-startup when power-on 0: Prohibit




F0. 4 Upper freq [F0.3] ~ 1000 Hz 400/ if your spindle is 24000rpm, if not divide RPM/60


Check if these 2 are by default 10: if not set
F0.5 Acc time 10:
F0.6 Dec time 10:


Spindle speed control:
connect from BOB to terminals 10V/+/ and CM/common/
F1. 1 AI input upper volt [F1.0] ~ 10.0V , Leave that to 10 if you board outputs 0-10VDC for spindle control, change to 5 if the board outputs 0-5VDC
F1. 3 Max setting freq 400/if your spindle is 24000rpm




Protection:
connect terminals Ta Tb /NC normally closed/ to the ESTOP circuit, so that if the inverter trips it will stop the machine


When you are finished with all of the above:
F0. 10 Parameter read-in protection 2: Only allow to modify this parameter


How to turn it on off and the speed control:

not all the info is relevant to your case, so remember your settings from above, but read about mach 3 , relay assignment and checking the speeds. You will need a cheap tacho
If you dont have a relay output/alsmost impossible/ , you will need a cheap ??V relay/depends of voltage output of your board/ from ebay.

http://www.cjh.com.au//PWM%20Spindle%20Control%20using%20Mach3.pdf

Neale
02-02-2015, 09:53 AM
Sorry for my ignorance but what does this thing do on a high level and where does it fit in? Which model number should I be looking for to get some docs?
In crude terms, imagine two relays in parallel, both powered by a feed via the normally-closed estop switch. Each relay has a normally-open contact and these are wired in series and used to supply a control voltage to the external system - exactly what it does is up to you. So, in operation, both relays are activated, both relay contacts closed, and external system enabled. Hit the estop, the relays drop out, and the external system is disabled. Two relays mean that even if one contact sticks, the estop still works. For greater reliability, the two relays can be fed via a double-pole estop switch and two cables although I'm not sure if many people go this far. As it stands, though, the system would restart when the estop button is released or unlocked, so the safety relay box has a couple more relays which are associated with a reset or enable external switch which has to be pressed to get the main relays into their operational state. My Pilz relay actually has three sets of contacts so it can control three independent external systems/devices (maybe enable inputs on stepper drivers, ditto on VFD, and estop input to Mach3, for example). And that's one of the simpler Pilz devices! I bought mine as a new-but-obsolete version from eBay so the model number isn't very relevant (and I can't remember it off-hand).

As to where it fits in - a normal estop setup can only switch one circuit. Generally, you want to switch more than that (as per comments above) so you are going to need some kind of relay setup or somesuch to handle multiple independent circuits, but for emergency use it needs to be reliable which a single relay might not be. Add in the need for an explicit reset switch so if estop is hit, then things really stay off, and it gets a bit more complicated still. The Pilz "relay" packages all that functionality into one box. You might want to trip the relay with the limit switches as well as estop but that's a decision for the machine builder.

EddyCurrent
02-02-2015, 11:12 AM
Safety relay info from a previous post.
http://www.mycncuk.com/threads/7723-Which-Type-of-Limit-Switch?p=60325#post60325

mitchejc
02-02-2015, 05:59 PM
@ Silyavski, thank you very much for taking the time to go thru my VFD manual to help me. I think I'll get it going now and I will also connect the VFD failure relay to my EStop circuit.

@Neale, Ahh, thx, now that makes a lot more sense to me. I can now see why one would use this. I guess emulating the functionality by using a few relays that drop out on estop would not be impossible but getting the reset functionality to work would be very difficult for me, so it would be a lot easier to just use a component. Had a look on RS-online but new they are expensive so I'll try and find a second hand or discontinued one from a seller thas will to send here. Just wondered why on earth they call some models a PNOZ, do they know what it sounds like when you pronounce it in english :-)

@Eddy, thanks that's a very useful link.

BTW you guys were right, I did a cad layout of my control box this weekend and its MUCH larger than I thought if I follow the component manufacture's guidelines for spacing between parts. I'm going to need a small lotto win to complete this machine if I want a nice PNOZ, a steel enclosure and a CSMIO/IP-M. Scratching around to see if I can't maybe liquidate some other redundant hobby stuff :-)

mitchejc
15-05-2015, 09:42 PM
Just thought I'd post an update. My build has been delayed a bit mostly due to lack of budget but I finally got going again. Got lucky and a friend contacted me to cut 6 x bench-rest rifle stocks which helped a bit for the hobby cash-flow. My stock making hobby is what what got me into cnc in the first place :-) Here's a pic of one getting cut
http://www.mycncuk.com/attachment.php?attachmentid=15330&stc=1



Anyway back to the topic. Got a nice 800x600x220mm steel cabinet for a very reasonable price (+- 70GBP). I was concerned about my ability to cut holes etc in the steel but got it done relatively painlessly using a Dremel and those small clip-in cutting disks.


http://www.mycncuk.com/attachment.php?attachmentid=15334&stc=1

I cut a alu panel to replace the steel one that came with the box.
http://www.mycncuk.com/attachment.php?attachmentid=15333&stc=1http://www.mycncuk.com/attachment.php?attachmentid=15341&stc=1


I made an air filter box with a cheap car filter to house my huge fan and keep stuff out of my control box.
http://www.mycncuk.com/attachment.php?attachmentid=15338&stc=1http://www.mycncuk.com/attachment.php?attachmentid=15339&stc=1http://www.mycncuk.com/attachment.php?attachmentid=15340&stc=1


I think I have figured out what I want to do with my wiring. I'm a bit further along but these were the last photos I took.
http://www.mycncuk.com/attachment.php?attachmentid=15336&stc=1http://www.mycncuk.com/attachment.php?attachmentid=15337&stc=1

mitchejc
14-06-2015, 08:24 PM
WHOOHAA, its running! :excitement:

Finally completed my control box. Decided to chuck 4 of the safety circuit relays and replaced it with an Arduino which resulted in a simpler overall setup. E-Stop button works as expected and any fault or position loss will stop all motion.
http://www.mycncuk.com/attachment.php?attachmentid=15522&stc=1

Steppers now mounted
http://www.mycncuk.com/attachment.php?attachmentid=15524&stc=1

Have to do the wiring of the machine now, fit cable chains, zero and limit switches etc but could not resist to try a few test cuts. No machine bed yet so I just clamped a vice on a few blocks of wood. Took a lot of effort to get this far but thinks are finally comming together and I'm absolutely over the moon with how the machine cuts. Part below was cut 15mm deep with a 6mm HSS endmill at 1000mm/min @ 3mmDOC and finishing cut at 500mm/min and 6mm DOC.
http://www.mycncuk.com/attachment.php?attachmentid=15521&stc=1
http://www.mycncuk.com/attachment.php?attachmentid=15523&stc=1

JAZZCNC
14-06-2015, 08:33 PM
Decided to chuck 4 of the safety circuit relays and replaced it with an Arduino which resulted in a simpler overall setup. E-Stop button works as expected and any fault or position loss will stop all motion.

Looking good but how do you work out that an Arduino can replace Safety Relay/s.?. . . . It's a Computer.!! . . . . . . In which case you may have just saved your self the hassle altogether and just used the I/O you had already.! (unless you didn't have enough I/O to start with.!)

mitchejc
14-06-2015, 10:06 PM
Thanks Jazz. Apologies, that came across wrong, ill try to explain. I wanted any fault signal to disable all movement and I could not figure out an easy way to combine the 4 stepper driver fault signals into one so I got them to switch 4 small relays and then wired the relay switches in serial so if any one opens it cuts my 5v enable circuit etc.

There are probably better solutions but I replaced those 4 relays with the arduino using 4 inputs and 1 output. Arduino Inputs connected to stepper driver fault outputs and the 1 output connected to all driver enables and spindle relay. The few lines of code check the inputs and if any change is detected it takes the output pin low disabling all drivers and switch the spindle off. Ive tested this to death and it works like a charm.

Not sure I understand what you mean about the spare IOs. I do have spare inputs available on my controller but i dont want to rely on mach3 alone to stop everything or is that not what you mean?

Lee Roberts
15-06-2015, 11:55 PM
I think what he's getting at is how u had it setup with the 4 relays, why the need for Arduino, are you then doing somthing else once a fault is detected?

Mechanical vs Digital I guess.

.Me

mitchejc
16-06-2015, 02:37 PM
Yep, Lee I had one relay per driver with the fault signal driving the relay so a fault on any driver would open that relay which would break my enable circuit which ran in series thru the 4 relays + spindle fault relay. So you are correct, I just replaced the simple logic that the electro-mechanical parts performed with digital.

Please keep in mind I'm very much a noob when it comes to this electrical and electronic stuff so how do you gents combine these sinking type stepper fault signals?

JAZZCNC
16-06-2015, 05:03 PM
Not sure I understand what you mean about the spare IOs. I do have spare inputs available on my controller but i dont want to rely on mach3 alone to stop everything or is that not what you mean?

What I mean is that there's no difference between Mach3 inputs and Arduino inputs Both are using and relying on Computers for safety.(BAD news) So if you had the spare inputs you may have well just used I/O on controller.

Regards the Drive fault signal and Relay then you only needed 1 relay and wire the drive fault signals in parallel so if any drive faulted it turned on the relay. The relay NO or NC contacts would be used to control other things like E-stop relay and send signals to control saying fault happened.!! . . . Much safer.

mitchejc
16-06-2015, 11:06 PM
Thanks for the advise, Jazz. Keep in mind we rely on computers for safety all the time, in fact in some cases we use computers to make mechanical things safer :-) I do not really agree with you comparing an arduino with a pc as they are worlds apart in terms of complexity and things that can go wrong etc but its not a topic I want to labour.

I believe what you are trying to get across is that the relay solution is ultimately more reliable and I'd still like to get that working if possible. I thought this was going to simple as pie and its not like I did not try and I spent more than a full day to get it working and failed to do so reliably. My challenge is the following: Fault output on the driver is a sinking type output. I connected it as per the manual with a simple little 5V circuit containing only a current limiting resistor and I got 0.7V on the driver fault output pin when in fault and 4.7V when not. I can swap these around in the driver setup when I set fault to "active high". I used this setup with a small relay board that basically contains a transistor and a diode but it switches even at 0.7V and its at that point where I started fiddling and burning time and eventually chucked in favor of the simple arduino setup which I got working in 30mins. So what's the right way to do this? Can you perhaps please help me out with a description or diagram or close-up pick of your fault signal setup if you are using Leadshine drivers?

JAZZCNC
16-06-2015, 11:35 PM
Thanks for the advise, Jazz. Keep in mind we rely on computers for safety all the time, in fact in some cases we use computers to make mechanical things safer :-) I do not really agree with you comparing an arduino with a pc as they are worlds apart in terms of complexity and things that can go wrong etc but its not a topic I want to labour.

Well if your so sure it's safe then you don't need any other system so why waste each others time.!. . . . Carry On.

mitchejc
17-06-2015, 08:12 AM
Jazz I dont mind spending more time in the name of improvement but you are right, why keep on fiddling if its tested to death and working. Im going to leave it as is and maybe revisit it when I fit a better controller at some point.