HV DC LINK TRIP IN VFD 1

[Electricrifying]

I am facing problem with my 690V,1500kw vfd unit which is 7 years old(make:Nelco),It is a 12 pulse rectifier unit driving a PD pump(Make:Geho) for dense slurry disposal with output pressure upto 160bar.It has functioned well till three months before the vfd started tripping on HV DC LINK BUS alarm.Incoming voltages has been checked found within limits,The output reactors are also found healthy, On calling the service engineer team, they changed few speed and current loop parameters and the drive started functioning well again but since last one month the tripping has again started (without any change in the load).On checking in oscilloscope we found that at times for afraction of second the rpm suddenly shoots up/down(about30rpm deflection) and the dc voltage at the instant ahoots up above the limit(max dc limit:1100VDC),leading to tripping of the drive. Plz help.

 

[itsmoked]

It can be "shooting up" because of two reasons:

The load is greatly reduced momentarily perhaps by clogs that restrict flow then pass as the pressure rises.

Or, the control part of the VFD is starting to get flaky and is actually briefly accelerating the motor in error. As soon as it starts accelerating the electronic/electrical issue stops and the drive instantly returns the output frequency to the correct speed. The instantaneous speed reduction causes the rotational energy of the pump and motor to briefly re-generate. The generation shoves energy (charge) into the DC link capacitors and the result is over-voltage on the link. (Fault)

This progressive failure can be temporarily mitigated thru changing operating parameters to make the drive less responsive. As the failure progresses even that method will begin to fail.

Typically this type of failure is caused by either contamination on the circuitry or age-out of some component or components, capacitors. You should have someone competent go over the control circuitry looking for contamination and distressed looking parts. Depending on your situation you might want to consider getting a spare control board on-hand.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[Electricrifying]

How to check capacitors(995mfarad) as on physical inspection those looks good and we have total of 6 such vfd, and problem is coming in three of them.

 

[davidbeach]

Assuming that “PD” pump is positive displacement, a clog would increase the motor load rather than decrease it.

 

[waross]

Hi David. I think Keith meant that a clog would result in increased torque/frequency. When the clog clears there may be an overshoot in speed.


Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

I too think that is a likely scenario. But the first thing I would check is the speed command signal, because that is potentially simpler to fix. If you are using an analog speed command and there is something wrong with the signal, noise, inconsistency etc., you can get a similar effect. For example if you are using a 4-20mA signal at 20mA so the pump is at full speed and a wire is loose, so occasionally the signal goes to 0mA, that is telling the drive to stop. But then the loose wire reconnects and it tells the drive to go to full speed again. If you are only looking st the output of the drive, all you see is a drop, then a spike in current as the drive attempts to perform exactly as it is being told. The same could be true if you are using serial comma to give a speed value and the number is changing rapidly because of unfiltered noise on the line. Check all of that first.

If the speed command signal is rock steady, then pursue the other mechanical angle as Keith suggested. It could be that something is in your fluid that is causing a jam, it could also be a bearing that is about to fail, something like that in the pump or even inside of the motor.

As a general rule, "flaky" parts in a VFD tend to be consistently bad once they go bad, not intermittent like this. They are also usually annunciated in the VFD HIM module (although I admit to knowing zero about that brand of VFD).


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[Electricrifying]

All speed signal and reference cable has been checked, and pd is piston diaphragm pump, one kind of positive displacement pump, How to nullify this regenerative effect in vfd, wihout connecting a braking resistor or chopper.

 

[waross]

Check the DC link capacitors.
If there is a PID controller associated with this drive dial in a little more proportional band.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

Age-related reduction in DC link capacitance would cause the effect you're seeing. For a given amount of energy regenerated into the DC link, a smaller capacitance will see a larger rise in voltage.

How old is the drive, how warm is the environment the drive is housed in, and how hard does it work in relation to its capability? All are factors in capacitor aging.

 

[Electricrifying]

The drive is 7 years old and the issue for 3 drives (total installed qty. :06),the drive is houaed indoor in clean environment qith avg room temp. 26 degrees)

 

[waross]

A leaking or slow closing discharge valve may be allowing the discharge pressure to give the motor an over-speed "Kick" as the piston starts the down-stroke.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[ScottyUK]

At 7 years old the caps will be tired. Certainly a possibility to explore. Have you verified the bus-overvoltage independently of the drive? I appreciate you won't be using a multimeter for this measurement!

 

[waross]

Scotty; Is a little kick back combined with old caps a reasonable suggestion?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Ed B]

We usually see it when the braking circuit goes bad. Either the resistor but most likely the switching device in series with the resistor. Fairly common failure.

 

[ScottyUK]

Hi Bill,

Yeah, something needs to provide a source of energy for regeneration to cause the DC link voltage to rise. The load may have changed slightly to cause some regeneration where previously there was none, or the load has always done this but as the effective capacitance on the DC link reduces the change in voltage for a given amount of regeneration increases (E= 1/2 CV²).

 

[itsmoked]

The OP does state there is a speed change associated with the fault, so there is the potential for an energy transfer.

for a fraction of second the rpm suddenly shoots up/down(about30rpm deflection)

Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

I support this as the most probable cause;

or the load has always done this but as the effective capacitance on the DC link reduces, the change in voltage for a given amount of regeneration increases (E= 1/2 CV²).

Thanks Scotty

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[John2025]

I don't work in the MW world but, is an 1100VDC limit for a 690VAC drive right? It seems a little low to me. My 480VAC drive has an 840 limit on the DC bus.

 

[Electricrifying]

The capacitors are rated 1100V so the upper limit is set.

 

[John2025]

Hmm, that still seems low to me. I wonder if drives rated for 690 are really running on the edge. How about your actual input voltage? Here in the US my 480 is really more like 495. Maybe you're just running out of capacity between the 1100 volt caps and a slight over-voltage on the supply?