SSD/Parker 590 thyristor drive Missing Pulses 2

[Skogsgurra]

I had a case yesterday where a Parker Hannifin (formerly Schackleton) drive has started tripping with message "Missing Pulse".

Managed to catch a trip. The recording below shows armature current (blue) and corresponding Iact. The Iact has a lot of ripple in it and it looks like it could be the +/-12 V supply that causes this. But there is no documentation and we all have a feeling that we could benefit from hardware schematics.

Questions:

1. Anyone had this with "mature" SSD drives? Tips?
2. Anyone has a copy of the drawings available? Links? Scans?

Gunnar Englund

http://www.gke.org

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

 

[iop95]

Check page 7-13 in this doc.

https://www.sdsdrives.com/downloads..._trips_and_fault_finding_sds_quick_guides.pdf

 

[Skogsgurra]

Thanks iop, but that one doesn't say much about the hardware.
We do not want to risk the drive or motor and SSD doesn't seem to care. So a classical analogue repair seems to be the best we can do.

The next best thing is to adjust the Discontinuous parameter so supervision gets active somewhat higher up in current. That will, in its turn, probably necessitate a current controller P gain adjustment and, to guard against collector damages in the motor, add a 50/100 Hz detector that warns if there should be any REAL missing pulses.

Still interested in hardware diagrams.

Gunnar Englund

http://www.gke.org

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

 

[iop95]

As I saw, "Missing pulse" is a "resulting" condition alarm and not a real gate drive pulse monitoring.
They do that from Ia, Ua, If and Uf.

In your drive Ia is measured by LEM or transformer current sensor? What is frequency for IAFBK in above picture?
What model drive is it?

Check this doc/page 5-6

http://manuali.eltex.biz/Manuali_Parker/590 DRV Digital_HA352494.pdf

They provide more tips.

 

[ScottyUK]

Hi Gunnar,

A possible aside - if that current transducer is a LEM type then try adding an electrostatic shield around the conductor at the location where the LEM is positioned. I prefer to use adhesive copper foil with a drain wire soldered to the foil and connected to earth, but at a push I've used aluminium foil and a croc clip for a temporary arrangement. Watch the clearances to any live metal! The HF noise should drop away signifcantly and leave the true current waveform. The noise problem in the output is caused by dv/dt coupling from the conductor to the transducer as the conductor suddently shifts in potential relative to earth during commutation.

 

[Skogsgurra]

I think you are right, Scotty.

If you look at the IAFBK signal just after the trip, it goes from noisy to quiet when armature current drops to zero. That is something that I ought to have noticed, myself.

Anyhow, the pump people decided to invest in a new drive. So everyone is happy!

Gunnar Englund

http://www.gke.org

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

 

[ScottyUK]

Yeah, I spotted the abrupt change in waveform character after the drive shut down and that made me think of the dv/dt coupling effect. You can possibly guess I have previously been led on a wild goose chase by LEM tranducers and this problem. In my case it was transducers used as part of the over-current protection for the static rectifier feeding the field of a brushless generator. Many people just blindly trusted the transducers without considering whether the enormous di/dt reported by the transducer was even possible and where the energy infeed to produce such a rapid change might come from. I maintain a healthy wariness of Hall effect transducers - they are good, but not infallible.