Help! 1 of 5 steppers working..

Hi everyone!

It's been a really long time, I just haven't had time to play with this CNC project for something close to a year due to family/renovation/interstate employment dramas....  Honestly, I'm basically relearning this stuff since my memory isn't as great as I'd like it to be. So please forgive my re-newbie status.

After much ado... I've started to build the actual machine, but before I got too far, I wanted to check if my previous work (control box, system config) was still ok....

I dusted off my control box and LinuxCNC computer, motors and cables... only to find things have indeed gone awry in my neglect.

The PC stopped booting... so I resurrected the hard disk, backed up the config files, replaced the hard disk with something a little less... beaten up.. then reinstalled. Ok, I'm an IT nutter, and installing an OS... that's my wheel house.. no dramas there. Still got some minor OS config issues to work out, but for the most part, it looks good. Talks to the Mesa card over ethernet, on 10.10.10.10 address, and clearly works with WiFi, since I'm using the system to create this post.

I got a little... not arrogant.. but perhaps a little less humble, when I thought... "Hey maybe this won't be so hard!" (in my naivety)

So getting encouraged, I fired up the control box... everything turns on as expected... all 5 CL86Y stepper drives in non-alarmed state, Mesa 7i76e card lit up. VFD responds to 10V analogue input. Great... although it's spinning backwards (I'd disconnected the spindle a year ago.. so I guess I wired it up incorrectly. So I swapped any two of the power wires around on the VFD  to shift the direction... (a little trick I learned building a spindle-driven router table in prep for this project) and everything is golden.

Did some basic jogging using the Gmocapy interface... I started with my 4th axis because that's what I landed on... it works beautifully. I tried some basic MDI commands... and noticed no other steppers respond...

So! 4 out of my 5 stepper motors (interestingly, all failing motors are connected to TB2 on my 7i76e) and they are completely unresponsive to Step/Dir signals. The fifth stepgen on TB3 works perfectly. No alarms, no errors.. as far as LinuxCNC is concerned, the motors are responding as they should... but they're not in my little garage-shaped reality.

Interestingly, all 5 motors are powered... leave the system alone long enough, they get warm... and I can't turn them by hand. If I force them with clamping pliers and an old towel/rag, they fight me, then alarm when I go to far.

So... Stepper Driver-Motor cables look good... at least in theory. Perhaps it's a Mesa-to-Driver issue?

So I checked each and every Step/Dir wire from the Mesa 7i76e stepgens to the driver... and they're ok. I've just finished checking encoder cable from the driver all the way to the socket that the motor plugs into, the encoders check out, (makes sense since drivers are all happy with "no alarms here"). I

So my current thoughts are:

1. Somehow all TB2 connections aren't working.
2. Is there some jumper, weird PNCConf setting I'm missing? When I use PNCConf... The I/O connector 1 (presumably TB2) Are configured as:

        0: X Axis StepGen
        1: X2 Tandem StepGen
        2: Y Axis StepGen
        3: Z Axis StepGen
        4: A Axis StepGen

Any guidance would be great. I've included my "SpindleGood" config files, so please let me know if I've missed anything you might need.

I truly hope you're all well, whether you can help me or not. I've just been beating my head up against this for a couple of days, and I'm really starting to run out of ideas.

Thanks for reading!
Hamish.

P.S. Any Forum users in Canberra? Love to see a working system, and chat about it!

 

First, yank out all 5 drive connectors on the Mesa and plug them one by one in the TB3, power up, test.
If all are good, check the ini file, edit with text editor, not with office writter stuff, look for step time and step space, set all to 5000, wire everything back and test again.
Be aware that the tandem axis/joints will not jog before homing and should show a warning about homing sequence set to -1 or -2

Hi Tommylight,

I hope you and your white "fluff ball" is well. I dimly recall you mentioning that you had one a while back.

Thanks for the response. I'll look into those suggestions tomorrow. I was also going to use the driver configuration adaptor+ software to see what the little in-built signal oscilloscope is reading.

-Another thought... maybe the FPGA's memory has been corrupted. I was thinking about reflashing it from the backup.. just to see if that does anything.

Will have a look, try each suggestion, and see what's going on. Will relay what I find here in a day or two.

I would not flash the Mesa board.
From experience with over 70 or 80 of Mesa boards, they do not fail, ever.
And since plenty of parts are working properly on the Mesa board, makes it very unlikely to be the issue.
-
The fluff ball is doing good but i had to throw her out of the shop, again, as it peed in the shop!
The first time it destroyed a new Lenovo laptop power supply, got banned for several months from entering the shop.

If no axis on TB2 work and you had a previously working hal/ini file setup,
that may mean than one of both of the step/dir driver chips for stepgens 0..3
have been damaged (these are U12 and U19)

This can happen is step/dir signals get accidentally connected to something
outside of the 0 to 5V range, say field power.

You can check for damage by measuring the output voltage on the step/dir pins (from + to -)
These should all read 5V when LinuxCNC is not running.

Hi PCW,

You must know the card well. I presume you work for Mesa, or have some serious electronics skills. Either case, I appreciate the information. So here's what I found....

I just checked, and the stepgen on TB3 it checks out to 5.04V, whereas the rest on TB2 range show voltages around the 0.45V mark (so less than 10% of intended). I was kinda hoping it'd be a jumper setting, but guess it's a chip issue as you say.

Honestly, I don't know how this could have happened in my absence. It was literally running until 1 minute before I packed it up, (Job offers saying "You start 7am tomorrow" do that) so the box was locked and disconnected from power. (I repurposed my 15A cables for my welder and other equipment that I took with me.. so if it got voltages higher than 5V.. I'm not sure where it came from). In any case, "Strange things are afoot..."

I looked at both U12 and U19 under a magnifying lamp... no sign of heat damage or cracked/cold solder joints.. not that it necessarily means much. I also checked the solder joints for TB2.. no obvious problems there either.

So... I presume I need a new 7i76? Given the short supply on Mesa's online store... I maybe outta luck here. Any chance of repairs? I'm not geared up for surface mount soldering though... my hands aren't THAT steady, nor is my sight THAT great.

Hmm, I think I'll email Mesa for advice.

Thanks for your help!
Hamish.