Parker 590+ 4Q 110A - problem with the table feed during processing

[pareto86]

I have a problem with the table feed while machining. Speed ​​feedback is taken from armature voltage measure. Settings: Nom motor volts, armature current and field current are written from the engine and are set to - 220V , 46A and 3.1A . Main curr . limit at 120%. The drive has been calibrated and carried autotune . If there is no load table goes well over the entire length , but when there is resistance of the machined detail, then at some point table stops. while it stops, there is no alarm on the drive, in diagnostic- > current fbk, the value is 25A , diagnostic- > speed feedback - 6%.
Table weight 5 ton and detail is about the same.
We are working currently at 7-8% speed as it doesn't stop the table.

We have used the same drive on another machine , where the feedback is also armature voltage and there is no problem as above. The question is - should we use tachometer , or is there any parameter in the drive , which will help us to solve the problem?

 

[Skogsgurra]

Even if the autotune was carried out, it is possible that the IR compensation can be increased. Find the IR parameter (can also be called "load compensation" or something like that, but I think Parker says IR Compensation) and increase it until tabele starts shaking. Then reduce until table stable and test if that helps.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

I should do the shaky thing with an empty table, right? I just read the manual and there is a parameter called IR compensation. It's set to 0% by default. The speed drops by approx 7% under the load (from 56 to 52 mm/min). Should I put this value, or go with the higher one, let's say 20%

 

[Skogsgurra]

If drop is 7%, then I would try a little less than the indicated drop. Make IR Compensation 6%. That should help a lot.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

Thanks alot. I'll try to do that and see if that helps. Hope the table won't stop anymore. What about setup parameters::speed loop::spd.prop.gain and spd.int.time? Are they worth looking at aswell? I'm going to work on Monday, so I want to prepare as many possible solution as I can.

 

[Skogsgurra]

Autotune should have taken care of that already. Don't touch them.
Wonder why the IR Compensation wasn't set by Autotuning. It should, in my opinion.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

Tbh I'm not sure if it's set to 0, but according to manual it should be. Autotune sets int gain, prop gain and discontinuity values so I guess IR comp is still 0%. I'll make sure on Monday. I'll post the result as soon as I check it. Fingers crossed!

 

[Skogsgurra]

Good observations. Good thinking.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

Hello again, I'm 1 day late, cause we needed to do some testings. IR compensation worked surprisingly well so it was that simple... Unfortunately that wasn't even mention in polish language manual, so it seems for the future I have to read english manuals. Anyway, thanks for help!

 

[Skogsgurra]

I am also late. Did a job in Trondheim, Norway. Eleven hours trip to get there, two hours on site (1 h job, 1 h coffee-drinking) and eleven hours back. So I feel exhausted now.

Glad it worked. This is not only valid for the Parker drives. It is universal and can be used regardless of drive manufacturer. Polish? Is that your native language?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

Oh, I believe traveling that far might be exhausting. I've been to Norway few years ago - worked as a picker. I traveled from Bergen to Fresvik over 3 hours and I really enjoyed the view and the stay! Yes, I'm from Poland. It seems we have poor translators or they don't think we've got to know everything

We encountered today another problem with the machine, but I guess it's mechanical defect of the machine, or the thing we mount detail to the table. The feed without processing is going very smooth, but if spindle goes into the material, at about 50% table starts to stop for like few miliseconds. It stops more often over 75%. It looks like the machine choke. It's very old (late 60s) machine, and motors are powered through the amplidynes(before we replaced it with parker drive). We had similar problems with amplidyne - feed rate was floating, sometimes stoping. Everyone said it's electrical thing, so we decided to go for parker drive. Now feed rate is fixed, but machine still chokes. Have you encountered something similar? Is it really electrical thingy, or more probably mechanical?

 

[Skogsgurra]

Amplidynes! The last time I saw them was in the old Centurion tanks. They were used for aiming and to keep barrel steadily on target when firing under advancement. Long time ago.

There are several things that can cause this. The first thing I would check is if there are any extra windings in the motor or in the circuit. You should not have any other winding than the excitation winding and the armature winding. Interpoles are OK, but remove any series winding (or make sure it is connected the right way, better remove it).

There is also your current controller. Remember that your motor is an old one that was built for pure DC (which is what the amplidyne delivered) and that usually means that armature time constant can be rather long. If you don't adjust the controller accordingly, it takes some time for the current control to "understand" that more torque is needed. That usually causes the "hesitation" that you describe.

You then need to adjust the current controller so you have a crisp and stable step response. Check the (English) manual to see how to proceed. Warning 1: make sure that the motor shaft is locked before doing it. Warning 2: do not pass high current through the commutator for more than a few seconds (it doesn't like it). Warning 3: disconnect excitation winding before doing any of the above. Warning 4: Make sure you re-apply excitation before running the motor again. If you don't, it can/will accelerate to a speed where it kills itself. Use a fast recorder or oscilloscope to observe current rise-time. Try to get close to around 40 ms. That isn't extremely fast, but fast enough. You need to be somewhat careful with the old motors. They were not built for very fast torque changes.

I am visiting the Erzgebirge/Praha region a few weeks from now. I could, perhaps, be of some assistance. Are you close to there?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[pareto86]

My place is like 300km from Prague, so it's quite a distance. Thanks anyway! This machine is very crucial for production, so we've got to deal with it asap.

About the amplidynes - there are still 2 plano milling machines in our factory which use them! Actually I'm doing renovation of one of these - going to replace all telephone relays logic with the PLC as well as all 50 years old electronic inside.

As for the motor - there's just excitation and armature winding, both directly connected to Parker. In Parker drive there's autotune which sets the settings you mentioned automatically. I done it right after I mounted the device and set the basic values. That went well without errors, so I guess it's fine(not sure though). I have done the same procedure on different machine and there are no troubles with "hesitation". I'm going to attach tachogenerator on Sunday, and if that won't help I'll replace the motor. It'll be a little problematic, but might be worth trying in order to eliminate something at least.