Can't get Abs Pos to count positive with CW rotation

[martyscncgarage]

Tom makes some good points. What drives and motors are you using??
You are getting a drive fault on WHICH axis?

Marty

[kevincnc]

CSS is working perfectly so far.

I think the early drive faults were when I was messing with the VFD decel, and the last ones were on the Z when the tapping move started.

DC Servos are 850 oz-in DC, tuned pretty tight- https://www.cnc4pc.com/nema34-850-oz-in ... esign.html
Drives- https://www.machdrives.com/brb.aspx
Belt reduction is about 2.5:1 driving 20mm X 5mm ballscrews.

If you watch the tapping video at 400 RPM, the acceleration looks pretty slow and smooth. It doesn't seem to me that going to 500 RPM should make it fail. I can try something like 1" for parameter 240.

Hopefully it's encoder noise, I'll pull the cable out and away from everything. I'll see if it works at higher speed on single point threading also. I'd hate to spend $150 on an encoder and modify the mount & pulley if that's not the problem.


Thanks
Kevin

[tblough]

It not how it "looks". You've slowed down the acel and decel factors so of course it looks smooth. However any mismatch in the commanded velocity of the Z axis caused by the slow acel/decel will result in "drunken" threads where the pitch varies.

Parameter 240 is not in inches. It's in turns. A setting of 1 allows the spindle to make a whole revolution while the axis accelerates or decelerates to get to the correct speed.

[martyscncgarage]

Run an external thread and see how it goes.
Since you have DC Brush servo drives, and you got a Z axis fault, that seems suspect. The last error log I posted showed drive faults.
Post a fresh report.

Does the software for your drives allow you to see following error? Have you tuned the drives? Followed the drive manufacturers process?

I don't recommend DC Brush servos with so many other drive alternatives now.

Marty

[kevincnc]

It not how it "looks". You've slowed down the acel and decel factors so of course it looks smooth. However any mismatch in the commanded velocity of the Z axis caused by the slow acel/decel will result in "drunken" threads where the pitch varies.

Parameter 240 is not in inches. It's in turns. A setting of 1 allows the spindle to make a whole revolution while the axis accelerates or decelerates to get to the correct speed.

Okay that makes a lot more sense, thanks. So since I'm starting .100" away, 1 rev would be .050" and it should be up to speed halfway there as long as it can accelerate fast enough.

Without changing anything else, I increased the RPM to 600 and it ran successfully. Then it faulted a few more times, and then ran successfully again. It took a few more tries faulting, but I was able to get a video of it working.

Every time it faults, it goes backwards and looks like this-

That would seem to me to mean that it could only be the encoder itself or noise. If so it doesn't show up on the RPM display. I never saw it fault at 400 RPM, but saw both a fault and a successful cycle at 475 RPM also.

Run an external thread and see how it goes.
Since you have DC Brush servo drives, and you got a Z axis fault, that seems suspect. The last error log I posted showed drive faults.
Post a fresh report.

Does the software for your drives allow you to see following error? Have you tuned the drives? Followed the drive manufacturers process?

I don't recommend DC Brush servos with so many other drive alternatives now.

Marty

I ran an internal 16 TPI thread and just noticed that you said external. I was using it around 250 RPM previously to thread my 5C drawbar. I ran it at 2000 RPM a few times with no problem, but now realize that 600 RPM may still fault. I can try it again, but it seems that I need to work on the encoder.

With the Drive OK input wired per the standard schematic, you can't tell the difference between a VDF fault and a servo drive fault, correct? I assumed that the drive was getting a following error but when I looked last time I'm pretty sure the AMP LED was lit up. I think it could fault either way trying to follow too quick of a move.

I know what you mean about the DC brush servos. I already had one and a couple of old Gecko drives (I though) and really wasn't sure if this machine would be worth putting a bunch of money into. You can hear how bad the Chinese double-nut rolled ballscrews sound. The machine works better than I expected, so I already got some nice ground screws for it, and will probably put AC brushless servos on. I can only accelerate the Z about 180 percent faster than it's going before it starts to wiggle the machine, but now realize the value of being able to home to the index pulse, especially for a lathe.

After buying a second DC motor, I found that I only had one Gecko drive, and bought the two Machdrives instead. The software is actually pretty nice with on-screen scopes, and their tuning procedure is detailed, and they are well tuned. They're about the best I've ever seen for DC servos in fact, they are perfectly silent when holding. I could plug in and graph the error and see fault history, and will do that if it becomes necessary.

Tomorrow I'll pull the encoder wire out so it's away from all the other wires and see what happens unless anyone has something else to try. The report attached was taken right after the fault video above.

One last thought- is there a filter setting in the Acorn for the encoder that might clean it up? It seems like I remember something like that helping with an MPG once.

Kevin

[cncsnw]

One last thought- is there a filter setting in the Acorn for the encoder that might clean it up? It seems like I remember something like that helping with an MPG once.

Try setting Parameter 323 = 16

[tblough]

Okay that makes a lot more sense, thanks. So since I'm starting .100" away, 1 rev would be .050" and it should be up to speed halfway there as long as it can accelerate fast enough.

And what happens when the spindle drops 5 rpm for half a turn halfway through the thread and then bounces back 5 rpm over in the next half turn? Those 2 or more threads in the middle will not be at the same pitch as the rest of the thread.

I'll say this one last time. Either your encoder is causing the rpm to be reported wrong; your encoder is good and your VFD can't hold a constant speed; or your axis drive doesn't have enough power to keep up with the rpm fluctuations.

[martyscncgarage]

First,
Separate the DRIVE and VFD fault lines. Use DRIVE FAULT input for DRIVES. Move the VFD fault to its own input, SPINDLE OK. Then you can tell which is your fault

No filter settings for the analog spindle output.

Here is your error log. Notice all the limit tripped messages and their time stamps. If you can't explain them, you have noise in your system, that needs to be resolved:
(1) 02-24-2020 16:53:11 CNC12 Lathe v4.20 (revision 4.20.1) Jan 17 2020, 09:50:27
(4) 02-24-2020 16:53:12 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 16:56:07 4032 Reset Cleared
(1) 02-24-2020 16:56:09 407 Z+ limit (#50002) tripped
(1) 02-24-2020 16:56:11 407 X- limit (#50001) tripped
(1) 02-24-2020 16:56:21 407 Z+ limit (#50002) tripped
(1) 02-24-2020 16:56:22 407 X- limit (#50001) tripped
(4) 02-24-2020 16:56:41 999 Parm #240 modified: 0.30000 -> 1.00000
(1) 02-24-2020 16:57:34 438 Spindle slave position error
(4) 02-24-2020 16:57:51 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 16:57:52 4032 Reset Cleared
(1) 02-24-2020 16:58:18 407 Z+ limit (#50002) tripped
(1) 02-24-2020 16:58:38 407 X- limit (#50001) tripped
(1) 02-24-2020 17:02:26 438 Spindle slave position error
(4) 02-24-2020 17:06:50 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:06:51 4032 Reset Cleared
(1) 02-24-2020 17:07:10 407 Z+ limit (#50002) tripped
(1) 02-24-2020 17:07:24 407 X- limit (#50001) tripped
(4) 02-24-2020 17:11:03 9031 DRIVE FAULT!
(4) 02-24-2020 17:11:11 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:11:12 4032 Reset Cleared
(1) 02-24-2020 17:11:14 406 Emergency stop detected
(4) 02-24-2020 17:11:17 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:11:18 4032 Reset Cleared
(1) 02-24-2020 17:13:19 438 Spindle slave position error
(4) 02-24-2020 17:57:48 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:57:51 4032 Reset Cleared
(1) 02-24-2020 17:58:00 407 Z+ limit (#50002) tripped
(1) 02-24-2020 17:58:14 407 X- limit (#50001) tripped
(1) 02-24-2020 17:58:40 438 Spindle slave position error
(4) 02-24-2020 17:58:44 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:58:45 4032 Reset Cleared
(1) 02-24-2020 17:58:57 407 Z+ limit (#50002) tripped
(1) 02-24-2020 17:59:17 407 X- limit (#50001) tripped
(1) 02-24-2020 17:59:38 438 Spindle slave position error
(4) 02-24-2020 17:59:55 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 17:59:56 4032 Reset Cleared
(1) 02-24-2020 18:00:07 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:00:28 407 X- limit (#50001) tripped
(1) 02-24-2020 18:02:10 438 Spindle slave position error
(4) 02-24-2020 18:02:16 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:02:17 4032 Reset Cleared
(1) 02-24-2020 18:02:28 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:02:49 407 X- limit (#50001) tripped
(1) 02-24-2020 18:03:04 438 Spindle slave position error
(4) 02-24-2020 18:06:04 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:06:05 4032 Reset Cleared
(1) 02-24-2020 18:06:17 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:06:38 407 X- limit (#50001) tripped
(4) 02-24-2020 18:13:10 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:13:56 4032 Reset Cleared
(1) 02-24-2020 18:14:13 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:14:21 407 X- limit (#50001) tripped
(1) 02-24-2020 18:30:58 438 Spindle slave position error
(4) 02-24-2020 18:31:20 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:31:21 4032 Reset Cleared
(1) 02-24-2020 18:31:45 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:32:05 407 X- limit (#50001) tripped
(1) 02-24-2020 18:32:51 438 Spindle slave position error
(4) 02-24-2020 18:33:01 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:33:02 4032 Reset Cleared
(4) 02-24-2020 18:33:04 9031 DRIVE FAULT!
(4) 02-24-2020 18:33:05 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:33:07 4032 Reset Cleared
(4) 02-24-2020 18:33:13 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:33:20 4032 Reset Cleared
(1) 02-24-2020 18:33:44 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:34:05 407 X- limit (#50001) tripped
(1) 02-24-2020 18:34:16 438 Spindle slave position error
(4) 02-24-2020 18:34:22 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:34:23 4032 Reset Cleared
(1) 02-24-2020 18:34:47 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:35:07 407 X- limit (#50001) tripped
(1) 02-24-2020 18:36:07 438 Spindle slave position error
(4) 02-24-2020 18:36:42 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:36:43 4032 Reset Cleared
(1) 02-24-2020 18:37:08 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:37:29 407 X- limit (#50001) tripped
(1) 02-24-2020 18:37:39 438 Spindle slave position error
(4) 02-24-2020 18:38:31 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:38:32 4032 Reset Cleared
(1) 02-24-2020 18:39:05 407 Z+ limit (#50002) tripped
(1) 02-24-2020 18:39:25 407 X- limit (#50001) tripped
(4) 02-24-2020 18:55:09 999 Parm #240 modified: 1.00000 -> 5.00000
(1) 02-24-2020 18:55:34 438 Spindle slave position error
(4) 02-24-2020 18:56:49 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 18:58:34 4032 Reset Cleared
(1) 02-24-2020 18:59:45 407 Z+ limit (#50002) tripped
(1) 02-24-2020 19:00:05 407 X- limit (#50001) tripped
(1) 02-24-2020 19:04:17 438 Spindle slave position error
(4) 02-24-2020 19:04:45 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 19:04:48 4032 Reset Cleared
(1) 02-24-2020 19:05:15 407 Z+ limit (#50002) tripped
(1) 02-24-2020 19:05:36 407 X- limit (#50001) tripped
(1) 02-24-2020 19:05:52 438 Spindle slave position error
(4) 02-24-2020 19:05:59 9033 Reset Initiated, Press Reset to Clear
(4) 02-24-2020 19:05:59 4032 Reset Cleared
(1) 02-24-2020 19:06:24 407 Z+ limit (#50002) tripped
(1) 02-24-2020 19:06:44 407 X- limit (#50001) tripped
(4) 02-24-2020 19:07:50 9033 Reset Initiated, Press Reset to Clear
(1) 02-24-2020 19:17:01 Creating report...

[martyscncgarage]

Please post pictures of your control cabinet wiring and make and model of your VFD, make and model of axis motors, what size power supply is driving your axis motors, big picture may help....

[kevincnc]

And what happens when the spindle drops 5 rpm for half a turn halfway through the thread and then bounces back 5 rpm over in the next half turn? Those 2 or more threads in the middle will not be at the same pitch as the rest of the thread.

I don't understand, why would it do that? The spindle speed is stable now on a Huanyang drive. By "bounce" if you mean that the encoder jumps 5 RPM instantaneously because of noise, I would expect a servo fault because it can't keep up. Once it starts tapping, I think it's fine and it tapped steel at 400 RPM at least 10 times, all with perfect threads. It never faults once the cycle starts. When it does fault, it's right at the start of the tapping, and it the z starts to go backwards when it faults. Maybe I'm still missing something.

Marty, I will separate the fault inputs, I just followed the standard wiring diagram. All of the tripped limits are from homing after an unsuccessful tapping attempt. I home after any servo fault because z position is not correct after the fault. I learned that the hard way.

I didn't mean a filter for the analog output, that is fine and stable now with the Huanyang VFD. I meant a filter for the encoder input. I'm sure that the Compumotor 6K6 motion controller that we used at work years ago had one, software controlled. We made a machine that used an MPG, and motion was jittery with it. We applied the software filter, and the jitter went away.

One last thought- is there a filter setting in the Acorn for the encoder that might clean it up? It seems like I remember something like that helping with an MPG once.

Try setting Parameter 323 = 16

That sounds like what I'm talking about, I'll try it.

Please post pictures of your control cabinet wiring and make and model of your VFD, make and model of axis motors, what size power supply is driving your axis motors, big picture may help....

The temporary VFD is a Huanyang 2.2kW until I get a better one. Until now I had trouble with the TECO and Yaskawa. This is the thread on that- viewtopic.php?f=60&t=3988&hilit=kevincnc

DC Servos are 850 peak oz-in DC, tuned pretty tight- https://www.cnc4pc.com/nema34-850-oz-in ... esign.html
Drives- https://www.machdrives.com/brb.aspx
Belt reduction is about 2.5:1 driving 20mm X 5mm ballscrews.
This is the power supply- https://www.galco.com/buy/International-Power/IP500U65, I'm pretty sure it has over 20A peak current capability.

Controls.JPG (42.39 KiB) Viewed 2934 times

Here are a couple old CAD models-

I'll ask again, is my z-axis accel/decel unusually slow at .3 sec? I'm away from the machine but I think my max speed is around 250 IPM. I get a following error if accel is much less than .25 sec, but can make it a lot quicker if I reduce the rapid speed if that's worth trying. I think my following error is set pretty tight and number of encoder counts before faulting could probably be increased. I think it allows position to get off by a few thou before faulting, I would have to check. I'm not sure how far out of position before faulting is reasonable.

If you watch the successful tapping at 600 RPM video, the acceleration at the start of tapping isn't jerky at all, in fact it looks to me that it accelerates a lot slower that it's capable of.

As you can all probably tell, the "cnc" in kevincnc isn't intended to suggest that I know what I'm doing, it's just my old ebay username.

Thanks for the help.

Kevin