Thread: A New CNC Build

G'day Geoff

I know it is hard to resist the urge to buy motors and electronics first.

Get going on your build knowing that you will be using Nema 23 frame size motors and leave the other decisions till you are well progressed with the rest.

As things go along, as a natural part of the process, you will come to have a much better idea of what to get, than you do now.

Just my opinion of course.

Greg

Hi Greg

Your opinion is valued....thanks.

And, I'm slowly learniing that you're quite right. One step at a time is the way to go.

My main incentive to jump ahead was the cost savings on the G251 Gecko drivers whilst they are on sale. But, even paying the normal price may be less costly in the end than if I make a mistake now and end up with gear that isn't satisfactory.

Cheers

Geoff

I might contradict myself a bit here.

If the saving on the G251s is good, they are very likely just what you need for almost any of the Nema 23 motors.

For some people I would recommend they go with the G540 but with your electronics interest and experience the extra complexity of setting up the G251s won't be a problem.

Greg

Hi Geoff, just jumping on board to see how you go with your build. Looks like you are off to a great start, I will help if I can.

You are lucky to have Rod close enough to call on, he is certainly one of the most helpful people I have known, I am sure he gets sick of us saying that but too bad

Cheers.

Russell.

I don't do anymore than anyone else on here. I have recieved a lot of help and generousity so what goes round comes round.
Anyway I know a bloke up the top pointy end there who dishes out a fair bit of support and good advice as well.

Finally got some shed time today so I got my hands dirty and started my build.

Got all the cutting done for the machine’s base and then graded all the steel to determine which way round each piece should be, and spent a lot of time with my straight edges and marking where I might need to shim to get it absolutely flat.

I’m amazed at just how “unstraight” a supposedly straight piece of steel is. There’s going to be a shim or two used. Nothing too drastic, I’m only talking fractions of a mm, but I want the finished product to be right.

What surprised me most was the supposed 90 degree angle of my 100 x 100 x 6 angle iron. It’s actually about 92 degrees, and in my view that doesn’t say much about our steel manufacturing standards. Fortunately, it’s not overly important to my build as it’s not being used in a critical area, but it shouldn’t be that far out.

I’ve only started assembling one side of the base so far and will hopefully get the other side finished during the evenings next week. I have to say though, I’m so over tapping threads already, and I’ve only just started. There’ll be dozens, if not a hundred or so more holes to thread before this project is finished.

I’ve attached a couple of pictures. No1 shows the cut steel after grading. The little black crosses show the low spots that’ll need to be shimmed. There’s quite a few. Picture No 2 shows the base laid out upside down. Pictures 3 and 4 show how I’ve joined it together so far using 40 x 6 flat bar inserted into the main rail supports since the 2mm walls of the supports are not thick enough to take a thread. I’ve just gone straight through the bottom of the rail supports and tapped into the flat bar.

Cheers

I’ve got a question which has arisen from the assembly of my base.

Just how do you tell when the base is absolutely flat? How do you determine when the main rail supports are in the same plane using just gear available in the average shed?

The only method I can think of is using straight edges across the diagonals to make sure there’s no twisting, but, even so, if the work area you’re assembling the base on is less than perfect, those imperfections could be replicated in the base.

Using straight edges is only going to be as accurate as your eyesight will allow so how can you get your base flat to within say 0.2mm?

What other checks should you do, and how?

Cheers

Hi Geoff,

Looking good there. I like the steel flat inside the tube to give you extra tapping depth.
Steel and aluminum angle is rarely at an exact 90 degree and you just have to work with it. Shims are good but I would not get too carried away just yet. As an indicator you have taken more care than I would have at this stage.

You can still have an twist in your table top if you just go by a straight edge on the diagonals. First step I do is use a 1200 level down both sides and this eliminates the twist. Then use your straight edge down the table top - i.e. at right angles to your bearers. You can use a feeler gauge between the straight edge and bearers to measure the variation and this will help with the size of shim you need. Don't get too fussy as things will move until you have the machine finished and mounted on the stand. Learn to trust you eye and sight along your rails looking for daylight or other signs that show things are not quite right. Confirm this by measurment and make adjustments if you need to. Look at your connections as this is also a good indicator of how your machine is going.

Once the machine is finished use a dial gauge or two (you can borrow mine) and fix it to the Z axis and measure across and down your table. This will be the final test and adjustment and you will achieve better than your 0.2mm. Typically I can get the table less than 0.1mm across the full cutting envelope.

Looking at your photos I can see you have taken a lot of care in assembly so I expect that you are not too far away from achieving your goals there.

Steel angle is a big shocker, Aluminium is useually pretty good. I will never work with Steel angle again, unless I ended up with some ???? grade or something.

RHS or SHS is useually good to work with.

You could always use a digital level if you ground is 100% level to find any imperfections, not 100% sure how accurate they are as I only just saw one up on ebay. If they are as good as the cheap calipers they should be good to use (I find my $10 dick smith caliper pretty dam accurate)

BTW isnt the gecko sale ending on the 18th?

The base as close to level as possiblle is fine, as rod said you can shim it up accordingly.

My base bed is 10mm solid steel, weighs about 60kgs just for the one plate, and it is out of square with the router by 1.4mm across the y axis, not a big deal, simply do as I and most people do, place a sacrificial cutting board on your unit and then route it flat with the cnc router, means after routing the whole unit is perfectly flat to the router

Rod, Blueray, Simso

Thanks for your comments. I'm not worrying too much about the end accuracy yet, but just thinking ahead about the final setup.

I've managed to get just a little bit more time in the shed and tidied the base up a little bit. My intention was to drill and tap the X axis for the rails before dismantling it again for powdercoating. Managed to get it fairly straight and flat....close enough for now. I'm happy with it so far.

I've run into a problem though, and am not sure if it's normal, or if I should tackle my supplier.

As I mentioned earlier, I was going to drill and tap the holes to mount the X axis rails. This was the first time I had closely looked at the rails, having given them a cursory check for damage when they arrived from China. They looked OK.

Now though, on very close examination I have found that none of the four rails I received is straight. They are the supported rails (SBR20) 1000mm long. All of them have a slight curve varying between 0.5 and 0.8mm along their length. Two of them have an upwards curve making them slightly higher in the middle, and two of them have a lateral curve making them curve out sideways around the middle. Also, the aluminium mount or support doesn't sit flat on a flat surface, having what appears to be a slight twist.

I don't know if my expectations were too high, but I had a silly notion that they should actually be straight.

I'm wondering, when these rails are bolted down onto a strong flat frame, whether or not they will be able to be pulled straight, or whether just the act of bolting them down will straighten them up. I'm thinking 0.5 to 0.8mm is way too inaccurate for my machine if it can't be rectified.

Considering the cost of buying and getting these things here, I'm not a happy camper right now and feel like giving the supplier a serve....but I won't. I'll engage in reasonable dialogue first and see what happens.

Or, perhaps this is normal, and it's just a matter of adjustment. Whatever the case, for the time being, my build is at a standstill.

Any comments/guidance/words of wisdom appreciated.

Cheers

Hi Geoff,
I would expect straight rails as well but when I checked a set of 1300 rails (25mm diam) just then there is a slight woof if you put them back to back. Didn't measure it but I can see a bit of light through the middle. These rails are not expected to be self supporting so I would think you could easily pull them back to a frame. By the way by squeezing the rails with my hand I could close the gap on mine.

You can test how much flex is in them by clamping the ends and seeing how much flex is in the middle. You could also do a trial run with vice grips and see how close you can get. Wrap a bit of cloth on your jaws so you do not mark the surface. A clamp against the side of the rail and the frame should pull it sideways.

As for tolerances I would be aiming at 0.1 or less in both planes. If you have not had the frame machined flat where your rails are going to be mounted then you will have to work with a straight edge on the rails as you bolt it down.

I'm fairly sure you will be Ok just a little extra fiddling to get it there.

Just had another thought - you may want to check the tension of the bolts that go through the aluminum extrusion into the rails. Perhaps you have a few loose bolts there.

Thanks Rod

Having thought about it for a bit, I had a bit of a play with the rails and found that they are surprisingly flexible, and can quite easily be distorted, for want of a better word, back into something resembling straight.

It would be far better if they were straight in the first place, but with care, when mounting, they should be able to be shimmed and stressed into acceptable tolerances.

I don't know how one would get on with unsupported rails if they weren't exactly straight. No way they could be manipulated, being mounted only on the ends.

Next time I build a machine, I'll use the flat rails.

Cheers
Geoff

Hi Geoff
The flat rails are flexible too and rely on fastening the rail every 50mm or so and the wide rails have a double row of fixing holes. The rails are meant to be mounted on a machined surface and the mounting bolts all tensioned the same.

It is possible that the drilling and tapping on the hardened shaft relieves the surface and internal stresses and causes some distortion of the shaft. It might be that this is a factor of the supported rails however I would rather supported rails than end clamped rails on a machine.

From a home workshop perspective you can get them very close by using jigs, dial gauge, feeler gauges and straight edges. Brass shims, aluminum cans and paper make very good shim material. Alligning the rails is one of those jobs that the more you think about it then it becomes more difficult than it is. I'll try to remember to take a couple of photos of using jigs on my machine when I put it back together - painted it between showers on Sunday and paint should be hard enough to assemble this weekend.