Build log: my first steel build.

[eurikain]

Hi,

I finished modeling my machine, and am looking for advice on dimensions. I join to this post the Sketchup model so you can check it out

I am thinking:
- 12x12cm beams on the side
- 10x20cm gantry beam
- SBR20UU on X,Y,Z
- rack and pinion on X,Y
- leadscrew on Z
- 2.2kw chinese water cooled spindle with VFD

I already own (from a previously failed MDF build based on buildyourcnc.com'sorials):
- 3 x Nema 23 425oz
- 3 x Stepper Drivers CW230
- Power Supply 36V 9.7A
- 6 x limit switches
- 5 Axis Controller with parallel port
- PC with parallel port
- Mach 3
- E-STOP
- Bunch of electrical cable

I am looking forward for your advice !

Cheers !

Similar Threads:

• Once more I do a new all steel build..
• Diy steel router build
• My first build Steel & Ali
• Mint's Build Aluminum/Steel Build thread.
• New Machine Build- 2nd Build - Steel and Al 3'x5'

[eurikain]

I made a few changes on the model:
- simplified the gantry supports
- added the bed
- changed a few dimensions
- drilled a few holes

Still a way to go, but I'll start soon with the plans in Layout.

I would love some feedback from you guys, and anyone willing to modify the model and propose changes is welcome

Cheers !

[ger21]

If you post pictures, you'll get more replies. Many people don't have time to open your model to see what the machine looks like.

[eurikain]

Thanks for pointing that out I should have done that straight away but thought that a model would be better so you guys can play with it. So here are the pictures.

[wizard]

Doesn't look to bad at all. I might suggest larger cross beams to better support your X axis beams.

What materials do you expect to machine on the router?

[eurikain]

Thanks ! Great to see some positive feedback I just updated the model following your advice. Here is the updated picture They were 5x5cm beams, i made them 8x5cm beams

[eurikain]

What materials do you expect to machine on the router?

Sorry I forgot to say: foam, wood, acrylic (not so hard), also aluminium. I don't need anything above aluminium.
I'll be using a 2.2kW water cooled Chinese spindle with VFD, from eBay.

[ger21]

Most people use a spring loaded pinion to keep it in the rack. With a rigidly mounted pinion, backlash and/or binding will be an issue.
You also should be using a reduction, and not a direct drive.

[eurikain]

Most people use a spring loaded pinion to keep it in the rack. With a rigidly mounted pinion, backlash and/or binding will be an issue.
You also should be using a reduction, and not a direct drive.

That's why I thought of this setup:


You can tighten the whole stepper-pinion block against the rack using the large bar applied against it. After it is tightened, you can tighten the stepper in place. The racks won't move, but has to be perfectly straight and parallel with each other. I am trying to get a design as simple as possible, that's the goal of this whole thing I've seen those spring loaded mounts, but think they're a bit complicated.

[b33felix]

i agree eurikian, making it a fixed rack and pinion can be made but both the rack and the rail must be perfect paralel to each other, if not, you will ineed get backslash or worse things here's some info about them, http://www.atlantadrives.com/pdf/ads_rackmount.pdf and ATLANTA Drive Systems, Inc. Drive Systems Rack & Pinion Myths i want to believe they are not lying! ger, you've got much more knowledge than me, so please do correct me if im wrong.

[louieatienza]

i agree eurikian, making it a fixed rack and pinion can be made but both the rack and the rail must be perfect paralel to each other, if not, you will ineed get backslash or worse things here's some info about them, http://www.atlantadrives.com/pdf/ads_rackmount.pdf and ATLANTA Drive Systems, Inc. Drive Systems Rack & Pinion Myths i want to believe they are not lying! ger, you've got much more knowledge than me, so please do correct me if im wrong.

The Atlanta Gear system uses a spring loaded pinion.... Also they use a servo-grade gearhead for reduction. The high end portal machines use master-slave motors to introduce electronic backlash removal. Also you'd need a way to grind all mounting surfaces parallel and perpendicular to each other, as well as measuring equipment that will span the longer distances of the machines you're building. I agree that the ideal solution is to mount the rack perfectly, but practically it will be extremely difficult to do this with near zero backlash...

[b33felix]

have you seen dnelso plasma cnc post? http://www.cnczone.com/forums/plasma...uild-4x10.html maybe i am wrong idk.. do you think that router forces wont let do the same thing that he does? he makes it look really simple ..

[eurikain]

I understand the problem of backlash and binding, I know that the obvious solution would be to go with a spring loaded module, but I am just trying to find a simple design, without cutting too much on precision. I read the myths they on the link you provided They do say that forcing the pinion against the rack does not solve the issue. When I thought of this solution, I imagined it as a simpler version the spring loaded mouths they usually use. IMHO, the only difference with my idea and the spring loaded version is that SL can adjust to the racks being not perfectly parallel, but as the link said, it can happen that the spring doesn't even move.

As for the gear reduction transmission, I totally agree with you ger21, I will include one in the model.

Anyone interested to come up with a smart simple design for that R&P drive ? Somehow in the philosophy of the rest of the model

And as for the plasma build, I do not see any springs on the drives, do you ? Though it is a plasma cutter, so forces applied to the moving parts are constant, most likely no binding problem (correct me if I'm wrong). backlash is still an issus I suppose, but maybe not as big as with a router ?

[wizard]

W

Sorry I forgot to say: foam, wood, acrylic (not so hard), also aluminium. I don't need anything above aluminium.
I'll be using a 2.2kW water cooled Chinese spindle with VFD, from eBay.

Looking at your pictures many parts of your machine seem to be very light. In this case I was looking at the Y saddle and the Z assembly. Obviously this is a drawings scaled down to fit the forum but it just doesn't leave me with that wonderful feeling of robustness. Especially if you intend to machine aluminum.

As for the X / table assembly the idea behind heavier cross braces was two fold. One to prevent excessive deflection under load and the other to support the X axis beams when under lateral load. You might want to supply some outboard gussets to shore up the beams in the middle and ends.

In general it looks like you got the basic design down pretty good for a light weight machine. The problem then becomes is it too light weight for the type of aluminum machining you expect to do.

[ger21]

To be honest, spring loading the pinion into the rack is not the ideal solution, but it IS a simple solution to the problem. That's why it's the most common method you'll see on DIY machines.
Better methods, like sporing loaded split pinions, get very expensive, very fast.

[louieatienza]

I understand the problem of backlash and binding, I know that the obvious solution would be to go with a spring loaded module, but I am just trying to find a simple design, without cutting too much on precision. I read the myths they on the link you provided They do say that forcing the pinion against the rack does not solve the issue. When I thought of this solution, I imagined it as a simpler version the spring loaded mouths they usually use. IMHO, the only difference with my idea and the spring loaded version is that SL can adjust to the racks being not perfectly parallel, but as the link said, it can happen that the spring doesn't even move.

As for the gear reduction transmission, I totally agree with you ger21, I will include one in the model.

Anyone interested to come up with a smart simple design for that R&P drive ? Somehow in the philosophy of the rest of the model

And as for the plasma build, I do not see any springs on the drives, do you ? Though it is a plasma cutter, so forces applied to the moving parts are constant, most likely no binding problem (correct me if I'm wrong). backlash is still an issus I suppose, but maybe not as big as with a router ?

My friend has a commercial plasma table. No spring load of the pinion, but then again it uses rollers on hardened steel rails. There are no cutting forces to speak of.

Simpler doesn't always mean easier to implement. In the Atlanta solution it requires precision rack installation, servo grade gearbox (which has its own backlash), ISO style mount for pinion, which is spring loaded.. They also require constant lubrication, meaning protecting the components from contamination. The pinion obviously must be at a constant distance from the rack, but they should not be meshed tightly to allow clearance for lube. Atlanta makes racks that work with THK rail making alignment easier, but then you need THK profile rail.

[b33felix]

here's a commercial router that uses belt-gear reduction, with fixed pinion, of course they have machined mounting surfaces, which makes the mesh parallel to the guide and easier along the rack parallelism, plus really accurate rack dimensions. CNC Router Machines | 3 Axis & 5 Axis Systems | MultiCam multicam series 1000 look at the pics.

(BTW Im new to the site, and by all means i dont want to be rude or similar i'm just stating my personal point of view and trying to talk about it, if it helps at all..),

i believe that with a flat machined surface, and a way to adjust the spacing between rail and rack accurately, plus precision rack, making it a fixed pinion would be better, (you should be REALLY careful not to get something in between the rack and pinion, or it would cause something to break, or bend, or worse..) Actually, i'm planning to do it in my build, then i can tell you more about it with more insight on the practical matter. But, i will add the option to make it spring loaded too. who knows, maybe we are missing something here..

cheeers

[louieatienza]

here's a commercial router that uses belt-gear reduction, with fixed pinion, of course they have machined mounting surfaces, which makes the mesh parallel to the guide and easier along the rack parallelism, plus really accurate rack dimensions. CNC Router Machines | 3 Axis & 5 Axis Systems | MultiCam multicam series 1000 look at the pics.

(BTW Im new to the site, and by all means i dont want to be rude or similar i'm just stating my personal point of view and trying to talk about it, if it helps at all..),

i believe that with a flat machined surface, and a way to adjust the spacing between rail and rack accurately, plus precision rack, making it a fixed pinion would be better, (you should be REALLY careful not to get something in between the rack and pinion, or it would cause something to break, or bend, or worse..) Actually, i'm planning to do it in my build, then i can tell you more about it with more insight on the practical matter. But, i will add the option to make it spring loaded too. who knows, maybe we are missing something here..

cheeers

There's no argument that fixed pinion is the way to go. But if you don't have the means of machining those surfaces and accurately line bore the bolt holes, it becomes a lot harder. I can think of a few ways around it, however simple may not be the easiest to implement.

[eurikain]

Sorry for the delay, I'm moving in so don't always have the time to reply

So I've read all of your comments, and am going to redo the overall design to include:
- beefier / stronger grantry + Z axis as wizard suggested + go with 6mm thick extrusions instead of 4mm + thicker gantry support (as used on this build: BUILD LOG: 8x4 Wall Mounted; Upright ply cutter)
- spring loaded pinion drive with gear reduction as ger21 suggested
- no more holes in Sketchup (as this is a draft design, I don't want to waste my time drilling whole in Sketchup, it's a pain and most drilling plugins don't work on Mac OS X..)

I'll post pictures when I'm done

[eurikain]

Just a quick question. Would belt drive be an option for this build ?

What if I use AT/HTD 25-30mm width belt ? Would it be wide enough to avoid flex ?

Ballscrew is way out of price for me, rack and pinion is the current option I choose, but i am wondering if a belt drive would make it