Losing X Steps

[EricSchimel]

Ok all, I have a PRS Standard 4x8 circa 2008ish. It's got that black PC box for a controller and the round "Toroid" in the box for a power supply.

Here are pictures: https://goo.gl/photos/xe4BupgzYNjDrULT7

I've been having two issues lately that I believe are related:

1) Sometimes when I power up the machine my motors don't lock. Sometimes it's just the X axis, sometimes it's all 3 axes. As per ShopBot support, I checked my voltages on the toroid and the PC power supply and everything is within spec.

2) My last three sheets of plywood I've cut, using material, bits, speeds and even files I cut all the time have all suffered from lost steps. I am not silly enough to rule out operator mistake, but I've been cutting this way on this machine for over a year with no lost step issues.

The more I narrow the problem down, the more I believe that it's my X axis that's having issues. The last few jobs I've cut (and ruined) I've lost X steps only

Things I've checked:


My machine is plumb, square and level.
My axes aren't binding (checked my moving them all by hand
My pinions are tight and there is nearly 0 play in the axes when the motors are powered on and stationary.
Nothing has changed in my setup, no new computer, or anything.
My dust collector hose has a grounding wire all the way through it, and my dust collector itself is grounded to a steel framed building.
My PC box is grounded to the shopbot frame, and a multimeter test shows that ground is connected to the steel framed building too. For good measure I just added an extra ground directly from the PC box to the building.
I'm running factory set ramping speeds and cutting no more than .25" per pass at a max of 4 ips.


I've read :mad:other threads on here about issues like this. It seems a lot of people who have issues with my box have grounding issues. It seems like there's something worse going on here, since some of my axes don't lock on power up I'm wondering if the power supply in my box is getting flaky.

ShopBot tech is still trying to help me out, and their solution right now is to upgrade to an RBK box. I'm not against that, but I'd not like to have to spend $3k if there's some diagnosis and repair I can do on my own...

[danhamm]

Just a thought, if one of your filtering capacitors has failed or weakened this would cause a sine wave to the driver that could cause it to spin out and as your X axis has 2 motors it takes a larger current draw when it ramps up.

[jerry_stanek]

Won't Shopbot test it for you if you send it in

[EricSchimel]

@Danhamm Is there a way to test the capacitors?

@Jerry Stanek I'm sure they would, they have been a great help so far. The problem is that I may have two issues: 1: possibly a bad power supply, 2: interference.

The really frustrating thing is that is problem is intermittent and not easily repeatable. A potential bad power supply might be causing me to lose steps... Or interference? Who knows right now. At this point I wish there were some more definitive tests outside of "try this and see if it works" Frankly I'd rather the thing just break at this point so I could replace it, or some part of it to move on! :)

I'm trying to mitigate the interference problem right now by re-running all of my motor wiring to be 100% sure it's not getting static from my spindle power cable. I unplugged every cable and removed it from the machine. I figure this is a good chance to check each and every cable for damage, and while I think my cable routing was fine (and it's been unchanged for a year) I am neatening everything up and I'll make sure it's all run proper.

After I do this the only think for me to do next is mill something and see what's what. I can't get it to fail by air cutting. It's starting to get expensive milling stuff though!

[scottp55]

Eric,
Don't make your wires TOO neat!
When I first set up, I was a little too anal, and had everything bundled/coiled as neat as could be......Crosstalk! :(
Got messier, and happy for 2 years:)

[Gary Campbell]

Eric...
"1) Sometimes when I power up the machine my motors don't lock. Sometimes it's just the X axis, sometimes it's all 3 axes. As per ShopBot support, I checked my voltages on the toroid and the PC power supply and everything is within spec."

If accurate, this line from above should eliminate the interference portion of your theory. You can test for this issue, for hours if needed, without the spindle running. Doing the same test with the spindle on should change results
When you mention the voltages were within spec, did you measure them during the time you were experiencing a failure?
Have you tested free voltage (no drivers connected) at idle? under load?
Have you tried testing at half speed? Double?
Are you running the same files that caused the lost steps?
Are you changing one item at a time?
Are you logging the changes and results?
Have you changed motors? Drivers? Cables?

[EricSchimel]

Ok, a bit of an update:

I pulled every wire off of the machine. They admittedly were a bit of a mess, and got everything cleaned up. My spindle power cable was near the other cables a little, but no more than it says they can be in the ShopBot installation manual. Despite that, I re-wired it in a way that it's barely near any of the communication wires.

I also took the opportunity to group all the wires together so that I can diagnose things easier. Don't worry, I didn't mash them all together or anything, I just took them all out, inspected them all and re-ran everything nice and neat.

I powered up the machine, everything worked perfect.

I was about to start cutting again when I noticed that my input 2 wouldn't go off. A bit of investigation later and I noticed my X axis proximity switch light was off, even though it wasn't anywhere near the bolt. I wiggled the wire at the base of the prox switch and the light flickered back on. It appears that where the wire is broken internally. It's possible that the prox switch was flickering when I was cutting.

Could that have caused the problems I've been having?

Gary, I've been trying to re-produce the no locking on startup problem, and I have yet to be able to do that. I'm considering wiring up my multimeter to the toroid voltage so I can watch it was I cut, maybe that will give me some clues. The dead prox switch put the kibosh on cutting (and testing) much today.

[bleeth]

Yes that could and probably is causing that issue. Hopefully SB will sell you one to replace it. In the meantime try turning it off and running some files in scrap to test.
When mine was using that controller the deal with once in a while a motor not powering up on startup was an issue also. It was very rare and I never did go after a cause, I just verified that they were good as part of my routine, and if not did a power off and restart.

[EricSchimel]

UPDATE!

I've finally been able to reproduce the problem. I was just about to leave the shop and I thought I'd try powering off, and powering on my box one last time to see if I could get it to a state where the motors were not locked. Sure enough I was able to do it.

Here's a video of the problem of it: https://goo.gl/photos/47kFsSwSUa1xx7tD7

So after that I moved around the location of two of my stepper drivers, see the results here: https://goo.gl/photos/x8pmnHb8AYWNwHcm7

(If you watched the video, skip this paragraph)

I found that if I power up my box, sometimes my steppers don't lock. To reproduce the problem if I power the box down, but I don't let it power down all the way and I turn it back on, my X2 and Y stepper drivers show a fault. If I try and jog Y, it doesn't physically move at all, however the screen shows numbers changing in the Y direction. My X1 motor doesn't lock at all, however my X2 motor locks.

I took the stepper drivers from Y and Z and switched their places (remember I've been able to reproduce a Y fault) Sure enough, when I do the power down/power up routine I now show a fault on Z and not on Y.

My conclusion

Please feel free to poke holes in my theory but it looks like I've got two bad/flakey stepper drivers. Could I also have a compromised power supply that's burning them out? Maybe something else, like a bad motor? Everything else seems to be functioning fine... My motors seem to be working fine, and every time I check the power supply it's putting out the exact voltage it's supposed to be... Maybe these drivers just die sometimes and I lost two at the same time?

For reference these are the stepper drivers I https://goo.gl/photos/bjui4orUfjuJUTD4A

[jerry_stanek]

I would check with Gecko and see if they will replace them for free.

[EricSchimel]

Ok everyone, I thought I'd update you all on what i believe ended up being the problem:

First of all, I'd like to give a huge thanks to the ShopBot support people and Gary Campbell. ShotBot as usual was awesome in helping me out, and Gary stepped up and really helped me troubleshoot some really advanced stuff.

On my machine my X 0 location is about 3 inches in front of my stop blocks. So basically I can move my machine -3 inches on X to touch my stop blocks. I almost always park my machine at 0,0 when I turn it off.

Like many of us, when I power my machine on I pull it tight to the stop blocks to square it and I power it up, wait for the steppers to lock and release the gantry. When I started noticing this problem I was getting random driver faults (read above for the gory details). What I noticed by paying very close attention to the lights on my drivers is that when I move the gantry with the machine off, they start to light up. This makes sense as electric motors (stepppers) make electricity when they are moved by an outside force.

The drivers have built in protection to some weird voltage conditions. What was happening is that I was pulling my gantry back, generating at times some odd voltages and kicking on my box. If I did it just right (usually when I did it quickly) I'd cause the drivers to go into some sort of partial protection mode (amber light). They were still locked, but not green like they're supposed to be. When I would start to cut it's likely that because they were under load, they would lose steps because of being in this mode.

In some cases I was able to get not only my X, but other drivers to fail completely to the point of not even locking up at all (red light on the driver)

According to Gary, my stop blocks are farther away from 0 than a lot of other people, so that's probably why a lot of people haven't seen this.

So the solution? Now I pull my machine tight to the stop blocks, take a deep breath and power it on. It's been all green lights and no lost steps since then. That's a great solution now, and in the near future I'm going to move my stop blocks much closer to X0 so there's no chance of this happening, even if I'm in a hurry.

Again, huge thanks to Gary and ShopBot on helping me out. Hopefully if anyone else experiences this this will save them a lot of frustration.