Jogging problem <solved, noise issues>

[mario.morosavljevic]

Mario...
A couple things to try..

First, set all your inputs to "unused" except for the estop and set all to "NO" test motion, report. then:

1) Set parameter 968 to 12 to allow a longer pulse in the step & direction signal. If that doesn't cure the issue:

2) Connect your step and direction signals to the DB25 header.

If those steps we already tried that. That was first Mr. Keith suggestion we tried both already.

[mario.morosavljevic]

Problem solved.
It wasn't a problem in either settings, hardware wiring nor DB25 connection. We called our good friend electrical engineer who's fixing literally everything that runs electrons through it (from motherboards to boat autopilots).
The problem was electrical noises on analog 10V output on Acorn which confused it during work.

Thank you all for help, time and effort to solve a problem,
Mario M.

[cnckeith]

I would like to ask you what exactly is parameter 968 which we corrected earlier because we would like to try to get it back to the default value?

Thanks once again for your help, time and effort and I hope I can get in touch with you in the future if there will be any questions,

968 is the step frequency. you should leave it set to 12. 12 = 100,000 steps per second. this is the setting that works best with leadshine drives.
if you run the wizard again and press "write settings to cnc control", you will have to reset 968=12 (this will not be this way in the upcoming v4.14 release)

glad you got it running, post a fresh report.zip and photos of your machine and we can help you optimize things.

show your electrical engineer this post as well.
viewtopic.php?f=63&t=2188

[mario.morosavljevic]

I would like to ask you what exactly is parameter 968 which we corrected earlier because we would like to try to get it back to the default value?

Thanks once again for your help, time and effort and I hope I can get in touch with you in the future if there will be any questions,

968 is the step frequency. you should leave it set to 12. 12 = 100,000 steps per second. this is the setting that works best with leadshine drives.
if you run the wizard again and press "write settings to cnc control", you will have to reset 968=12 (this will not be this way in the upcoming v4.14 release)

glad you got it running, post a fresh report.zip and photos of your machine and we can help you optimize things.

show your electrical engineer this post as well.
viewtopic.php?f=63&t=2188

Thank you very much and I will explain our solution, attach report.zip and photos tomorrow, it's late here and it's been a long day. We didn't use a zener diode.
We're glad to be part of such a supportive community as this one. Wish you all a blast day,
Mario M.

[mario.morosavljevic]

Greetings,
as we promised I'll attach photos, report.zip and explain how we fixed it.

1) Problem solved by nullifying noises on the 10V analog output. In parallel between (+) and (-), we added a 2.2μF capacitor. We did the same on both power supplies (24V and 5V). On 5V PS, we added 1000μF and on 24V PS 470μF capacitors. That's how we got a stable system (As you can see on VFD now we have 0000.00 input literally no noises at all). Apologies for a very messy electrical closet but we just fixed it and we did seriously a lot of rewiring in past few days, didn't have time yet to give it a makeup.

Our friend electrical engineer replied to you:
"Capacitors can annulate noises over and under referent voltage where Zener diode can cut noises only over referent voltage. With Zener diode, you won't get accurate 10V output."

2) We calibrated all axes and we're very very surprised how accurate Acorn and CNC12 are. Congratulation guys on making such a great system. Our Y axis is accurate in 0.01mm (0.000394 in) error on 2m (6ft 6.74in) of distance.


3) Next step is to configure M6 tool change macro with moveable and fixed plates. We already have them wired and made just not configured and no macro yet.

Kind regards,
Mario M.

[cnckeith]

thread on custom macros for tool changes.
viewtopic.php?f=63&t=1493

[cnckeith]

just checking the latest report.zip.. still something not right with steps per rev on Z axis, why is it at 1000?

[mario.morosavljevic]

Greetings,

1) Thanks for a link to a thread with a macro.

2) On X and Y axes we're using 3200 micro steps after calibration because displacement for 1 micro step is 0.01mm (0.0103...). So it's as round as it can be. For Z axis we used 1000 because according to calculation displacement is 0.005 mm per micro step and we tested it out, the machine is working just as good as with higher micro-steps. We'll listen to your advice and go over 1600 so we'll put 2000 and then displacement will be 0.0025mm which is still ok but in reality that precision doesn't really mean much. On the other side, it will be in the desired torque/micro-step ratio. By increasing micro-steps torque is decreasing so we went with that assumption in our calculations.

Do you think 2000 is fine for acorn and cnc12?

We don't have micro-step resolution same on all axes because on X and Y we're using rack and pinion setup and Z axis has ballscrew.

3) We're not sure what exactly "ProbeDetect" and "ProbeTripped" means? How do we need to set it up for probe touch plates? We're using both fixed and moveable plates. Fixed for TLO and moveable for Z0 height.

Thanks and have a nice day,
Mario M.

[cnckeith]

3) We're not sure what exactly "ProbeDetect" and "ProbeTripped" means? How do we need to set it up for probe touch plates? We're using both fixed and moveable plates. Fixed for TLO and moveable for Z0 height.

Look in the Wizard I/O dropdown selection for Inputs. ProbeDetect is used to indicate to CNC12 that a probe is plugged in. Probetripped is used to indicate to CNC12 that a probe has been triggered. set up those inputs and wire the probe according to the Acorn schematics on same subject found in the Acorn schematics zip file.