buildlog.net

I got the board fully tested outside of a laser. I figured out my problem with the micro stepping. When I bought the rotary DIP switch, I noticed they had a complimentary mode version. This inverts all the switch functions. At the time, I made a mental note that it might come in handy one day.

It turns out I accidentally bought that one. So my intension of making the the worlds simplest micro stepping selector, actually created about the most confusing one. It still allows you to select any mode, but sort of in an upside down backward way. All I have to do is buy the correct part. Fortunately I only installed them in one board.

The fan works great. The drivers barely get warm at 1.70amps. Everything else works fine. My next step is to install it into my laser and test it.

I did a blog post on it. There is a video there too. (Sorry about the horrible lighting)

http://www.buildlog.net/blog/

I am testing in the laser now and I am having some electrical noise issues. The Pololu drivers appear to be extremely sensitive to noise. The motors would jump a little whenever the laser was fired with any decent amount of power. At first I thought the noise was coming through the connections to the laser power supply. I totally disconnected it and made a manual system to fire the laser. The jumping was still there.

I moved the parallel port cable away from the power supply and high voltage cable. That did not help. My HV wire was pretty long, because I did not want to cut it. I trimmed it by about half and looped it through a ferrite bead twice. That helped a lot, but there was still some jumping. I then ran the 24V power supply off a long extension cord to another room. That fixed it completely. I reattached the laser P/S to the interface board and the jumping is back at a very small level.

I'll try a few more filtering tricks tomorrow, but I don't have much of the right supplies, like more beads and disk capacitors. I know some real experts in this area I can talk to on Monday.

I just got my Pololu driver board wired up tonight and I tried the X axis only. I am testing it using Mach3. I am seeing the same kinds of instability but I am not even firing a laser yet. When I try to move the carrier a few inches in moves in 1 inch "hops". When it gets to the end of the run it is not quiet. It seems to be jittering a bit. I suspect electrical noise.

Tomorrow, I am going to run a wire between the PC and laser chassis. I also need to figure out how to set the current for the X axis.

I agree with Bart that the Pololu parts seem sensitive to noise.

You are going to need to get the drives in a metal box. That and possibly opto isolate the inputs, or at least use a line driver. That should clean up any noise coming down the signal line. I dont think ferrites will help. Dont want to put caps on the signal lines, it will mess up the pulse form.

I have also used Schmitt triggers before. They can add a some amount of noise immunity. They are a lot faster than optos, but less effective for noise.

The steppers are still jumping with zero connections other than power so I don't want to get carried away with optos at this time. I tried separating the board from the laser power supply and at only about 6 inches further away, the jumping almost entirely stopped. I am think about adding a metal shelf above the current location. A thin steel shelf should have good shielding.

I also have some parts on order to play with like bypass caps and some optos and schmitt triggers.

Here is a proposed location. A shelf would go under and extend a few inches past. Mounting metal to metal to metal should help too.

Hi just a thought, do the inputs need pulling down to ground or pulling up to 5v depending on how you are driving them. Sometimes this can help if inputs are very sensitive. I usually use 10k resistors for this on ttl.

I pull them down manually whenever they are disconnected from control with a test connector. When they are hooked to the PC or controller they should be strongly driven to one logic level or the other. It might be a good idea to prevent issues if people have a loose connection. The motors could run wild even without EMI issues.

Since the noise was present with only power connected, I felt signal isolation would not fix the problem. The laser power supply was practically touching the 24V supply, so noise was probably getting in due to that. I built a simple steel shelf for the laser power supply. I deliberately left the wiring a little messy temporarily to really test it. The motors are completely stable now. No amount of pulsing of the laser appears to influence the motors. I need to clean up the wiring and do some more testing, but I am off to Wisconsin for the weekend.

I actually like the final result. It keeps the laser power supply right near the tube. I will probably add a horizontal member to pick up the outside edge. The shelf is a little bigger than it needs to be too.

Bart,

You've now got me wondering if using this type of shelf will get the p/s and tube close enough to allow the stock "short" wire on the power supply to reach post on the laser tube without modification.

Brian