VFD overvoltage fault ? 3

[nmark]

Hello,
I have (2) vfd's controlling 2 motors(conveyors) with a input voltage of 460 volts in real life I measure 494 volts at the input. The drive manufacturer says voltage input in good for +10%. 460*1.10=506
But over the past 2 weeks both drive's have faulted on a overvoltage fault numerous times.The fault has occured when the drive is running at a constant speed. I do not have a line reactors or chokes installed.
Both of the drives, motors are new . In addition to the new conveyor motors at the other end of the plant are 2 new 1000HP 4160 volt compressors DOL starting. We are going to monitor the voltage for spikes etc.. later this week.But while I wait any suggestions?

 

[ozmosis]

Sundays are typically a day for OV trips as generally other loads are off and the voltage tends to creep even higher.
Monitoring the DC bus on the drive display is ok but will not be fast enough to capture transients or spikes on the network. The spikes will not need to be too big wither if the AC is 500V. you will need a scope on the DC link to capture what is happening on the DC link when it trips.
Are the motors driving the gearboxes via belts? sometimes if you get slackness in belts the slip can cause an OV trip if the DC link is already high.

 

[Skogsgurra]

Exactly my thoughts, Patrick. Sundays can be a lot higher because other loads are off. Have had cases with bad PF in induction motors due to that. So bad that thermal protection tripped.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[DickDV]

Sounds much more likely to be power problems than regen.

In view of that, it seems to me that two possibilities exist. One would be to drop the incoming voltage either by transformer or by adding at least 5% reactance. The second would be to change the drive to one build to international voltages which typically are 380-10% to 500+10%. These are much more tolerant of voltage variations than drives built to 460V +/-10% or even 15%

 

[jraef]

Huh.
In all the years I have used those "name sounds similar to Yamaha" drives (if I'm interpreting that correctly), for some reason I never noticed that their input voltage specs are 380 - 460V +10%, but I just checked and they are! Wow, the details we can miss sometimes when we are blinded by technology I guess. I cut my drive teeth on that brand 20+ years ago, but have never had that problem. Even as a competitor today I still have respect for their products. Was I just lucky all those years?

 

[LionelHutz]

I agree with the others. Sunday will have the lowest demand on the power system in your area so the voltage will be the highest. So, it looks like a source problem.

I would still have a DB resistor in this application. It should not have to be very large because you should just need it to dump some energy when any temporary over-voltages appear on the buss.

You have never answered if the conveyor can drive the motor? The drive may have that decel feature but it likely will not allow the motor to run over 60hz. So, if the conveyor drives the motor over 60hz the drive will hold the frequency there and trip.

Another thing, saying your operators are not changing the frequency doesn't mean there isn't a problem with the analog circuit changing itself.

 

[DickDV]

With regard to Lionel's post, it is also possible when running in vector mode to have the speed loop tuning so overtuned that the oscillations cause regen.

Seems unlikely here but, absent anything more obvious, it wouldn't hurt to check. Of course, if running open loop scalar V/Hz, that wouldn't be a possible cause.

 

[jraef]

The drive may have that decel feature but it likely will not allow the motor to run over 60hz. So, if the conveyor drives the motor over 60hz the drive will hold the frequency there and trip.

In the "sounds like Yamaha" drives, the stall prevention feature is active all the time, not just when accelerating. It will drive the output reference frequency down, overriding any commanded speed, in order to reduce current (which is how it detects a potential stall). That's why I postulated that the compressors starting DOL on the system may have dropped the voltage, which would make the conveyor motor current rise for a given speed command. But if the stall protection kicks in and lowers the speed command, and then the mass on the conveyor wants to keep moving, you can end up in regen without having to go above 60Hz. All it needs is for the motor to be overhauled beyond whatever the output frequency is at any given moment.

I had that problem on roll handling conveyors at a paper mill. They used vacuum grapplers to handle the rolls, which took 2 x 400HP vacuum pumps. When they cycled on and off while the paper rolls were moving, the conveyor drives would trip on OV. We just disabled the stall prevention, I don't really like that feature much anyway. It's one of those things that tends to create more problems than it solves.

 

[nmark]

I have a power analyzer on the line side of the drive.I haven't downloaded yet to get all the gory details but when 600HP 4160v compressor turns on a -40volt line dip occurs the same occurs with a 1000hp turning on so far no tripping on a OV fault.
The line feeding the drive is seeing peeks of 690v to 710v. But at sometime in the wee hours last night 1 drive faulted twice on OL.After investigating wiring methods which are not up to par looking at installing load reactors between drive and motor.Will try and go about upgrading wiring also( having motor wires installed in own conduit instead of PVC conduit with multiple motors).Half the length is in pvc the other half is metallic wireway with everything but the kitchen sink in it.Motor wire lengths are less than 150 feet from drives.

 

[nmark]

The wiring from the analog ouput at the plc is shielded cable in its own conduit (18 gauge) less than 50 foot away from drive. The HMI communicates with the plc via ethernet.

 

[itsmoked]

Jeff I'm not getting how stall prevention is 'realized' by an internally generated 'slow down'? Would you want to increase the drive 'signal' to prevent stalling?

I guess we can stop asking nmark about overriding. He ain't gonna tell us....

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[nmark]

Sorry Itsmoked I googled overriding conveyor and I still don't understand the term. Would someone care to enlighten me please?

 

[davidbeach]

I think they mean what I know as overhauling, but in any case the question is "Can the load on the conveyor drive the system?" Traction elevators are a very common example of an overhauling system when going up with the counterweight heavier than the cabin or going down with the cabin heavier than the counterweight.

 

[nmark]

Thank You David.No the load is not overriding/overhauling conveyor.

 

[itsmoked]

Booyah!

That's good, and removes a likely cause of your over-voltage errors.

Thanks nmark.

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[jraef]

Well a traction elevator is one example of an overhauling load, but that is an intentional one. What I was referring to is an unintentional overhaul.

What I mean by that is, your conveyor is undoubtedly moving a load (otherwise, why have it?). That load represents a mass and we all know that a mass in motion wants to stay in motion. So your VFD is running the conveyor at say, 40Hz which represents 80FPM just to keep it simple. Suddenly the voltage drops and the stall prevention feature changes the VFD output speed reference to 25Hz, without a change in your external speed command reference. The mass on the conveyor is still moving at 80FPM and even though the speed is decaying, will take a certain amount of time to slow down to where the VFD speed is telling it to be. During that time, the mass on the conveyor is "pulling" the conveyor belt along with it as the speed decays, spinning the motor at a speed greater than the frequency output of the VFD. So it becomes an external driving force on the motor shaft spinning the motor at super-synchronous speed. All the while, the VFD output is keeping the windings excited. AC motor + excitation current + external driving force at super-synchronous speed = AC induction generator!

I still think you should try just disabling that stall prevention feature. That's why I don't like it. As itsmoked's question points out, it seems almost counterintuitive to do it this way. But that feature is intended to be used primarily when power constrains dictate that you CANNOT increase current as the motor begins to slip more, so it pulls back the speed command in an attempt to decrease the load on the motor. That works on some types of loads, but NOT inertia loads like a conveyor can be.

OK, I'm done. I wanted to be clear about what I have experienced but I could very well be all wet here anyway, so no more soap box. I promise...

 

[itsmoked]

No soap box about it!

So anti-stall should really be called the reduce-speed-to-match available-voltage feature.


Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[ozmosis]

We seem to have gone from voltage related trips to current related trips(over-load). I'm trying to think if stall could create an over-voltage situation and I would probably say not (but I don't have the experience of the yamaha-thingymagig) so I'm thinking the OL and OV are now quite different problems.

 

[LionelHutz]

Hey Jeff, I wasn't really referring to the stall prevention at all. Most drives can be set to different types of acceleration and deceleration control to protect the motor and/or drive. Another example is the deceleration control where the drive quits decelerating if the deceleration is going to cause an overvoltage trip. A third example would be torque control which controls the ramp time based on the a motor output torque limit.

What I was referring to is that if you set the maximum frequency to 60hz then this is usually a hard fixed limit. If the motor is forced into generating past 60hz the drive needs to have the capability of getting rid of the regen energy or it will trip. The drive will not allow the frequency above 60hz to protect itself from the trip.

Now, in this case, the conveyors are not supposed to be able to drive the motors so it likely does not apply. But, it can apply in some applications (pump jack).

So, it would seem we have a case of too much supply voltage.

nmark - The reactor needs to be between the source and the VFD. I still don't like this solution because the reactor has little voltage drop at low load or when the VFD is off so you'll likely still get trips at those times. The solution is that you need to lower your plant voltage. If the voltage is rising on Sundays you need to see if the utility will do something about it.

Now, having said that, I have seen a VFD trip on overvoltage due to a crappy job on the VFD to motor wiring. I believe there was ringing on the buss that was affecting the buss voltage sensing circuit. When it did happen the drive was tripping constantly though which doesn't seem to be what's happening in your case. I really think it was just luck the drive did not fail in that case I saw.

 

[Laplacian]

Out of curiosity, what is the parameter for "Input Voltage Setting" set to? Is it the default 460V, or have you changed it to match your nominal input voltage of 496V?

This parameter is the reference for the drive protective features such as overvoltage, brake resistor turn-on, stall prevention etc..

(This is right out of the manual for one of the drive series of a company that has the same first and last letters mentioned above)

 

[nmark]

Thanks to all who have responded.
Currently the stall parameter is disabled and the input voltage parameter is set to 500volts.
So far only 1 fault on overload 3rd shift on Wednesday and no faults over the weekend.