DC bus overvoltage at random times 1

[Smoem]

We're struggling with a Beijer BFI P2 90kW VFD right now. 3 phase 400V. It's driving a large rotating drum for washing fishing nets. This means it's seeing a very varying load through a gearbox at roughly a 240:1 ratio. It's a tried and tested application, which to my knowledge has never produced a DC overvoltage fault. Now one of our costumers are struggling with frequent alarms, 2-3 times every 5 hours of runtime. It seems to be happening at random, both during acceleration, switching directions, running steadily at 50hz, whenever. The problem started gradually happening this last month.
There's a decently oversized brake resistor connected, and it's in good shape. The HMI stores a log at 250ms intervals of the DC bus voltage during a trip, and it seems to just be a single sample of ~670V inbetween ~570V. I'm thinking theres a short spike that almost slips inbetween the samples for some reason. Had a multimeter across two input phases when it happened once, and there were no irregularities.

I've spent the day trying to replicate the fault, but it's nearly impossible. Very intermittent. The weird thing is that it's only happened at the end of a preprogrammed "cycle", where it spins one way for X minutes, and an extractor fan turns on the last Y minutes of the X minutes. The fan is also driven by a much smaller VFD in the same cabinet. I'm trying hard to figure out how the fan running could possibly cause spikes on the bigger VFD's DC bus, or if it's just coincidence. They are both fed by the same supply and share the same ground.

Appreciate any input.

 

[TugboatEng]

This occurs when the drive is regenerating or dynamic braking. Is the speed command changing or is the motor going into braking because it is being overrun by the load? I recently experienced this problem on one of my winches. In my case the over voltage occurred when the operator quickly reduced the speed control to zero. My solution was to increase the decel time. However, if your speed setting is constant you may consider increasing the regen power limit.

 

[Smoem]

The times I personally witnessed it tripping was:
Once on acceleration, it tripped while the load was unevenly distributed inside the drum and started over-hauling the drive faster than the drive accelerated.
And once while it was running at a constant speed with a very light load. At the end of the shift with 2 minutes left after 2.5 hours of steady operation, I noticed the DC bus readout started fluctuating randomly with every rotation of the output shaft. Still at constant speed. The VFD tripped after 1 or 2 minutes of this unstable voltage. Also some erratic "DC Ripple" logs, but I'm not quite sure what the mechanics behind that are.

The braking/regenerating in itself seems to work very well, the brake resistor barely gets hot. It's stopping completely and reversing direction every 25 minutes, but the faults dont seem to follow that interval.

 

[RedSnake]

It is as Tugs says.
It is because the moment of inertia of the drum and the load causes the speed to be higher then set speed and the motor becomes a generator instead, and starts to charge power back to the frequency converter.

You didn't say what kind of motor it has, or which kind of speed control is used, but whit a application like that when a accurate speed might not be the most important, I would choose something like >

 Sensorless Vector Speed Control
o This method provides increased starting torque compared to V/F mode, along with improved motor speed regulation with
changing load conditions. This method is suitable for more demanding applications.
Operating in Sensorless Vector Speed Control Mode
Beijer BFI-P2 can be programmed by the user to operate in Sensorless Vector mode, which provides enhanced low speed torque,
optimum motor speed regulation regardless of load and accurate control of the motor torque. In most applications, the default V/F
Speed Control mode will provide adequate performance, however if Sensorless Vector operation is required, use the following
procedure.

It could be a mechanical backlash in the gearbox or in the clutch or in the wedge on the motor shaft or whatever is used.

“Logic will get you from A to Z; imagination will get you everywhere.“
Albert Einstein

 

[TugboatEng]

To clarify what I meant about the regen power limit, that's how much power the VFD can send to the braking resistor. Anything above that has to get absorbed in the filter caps which is where the high bus voltage comes from. Since your resistor has excess capacity, start by utilizing it more.

Other options may include disabling regenerative braking in the run mode and only utilize it while stopping.

 

[LionelHutz]

The capacitors could be getting old and failing in the drive. Loss of bus capacity causes more ripple voltage on the bus and loss of bus stability.

 

[Smoem]

Thanks for all the responses, it turned out to be solveable with sensorless vector speed control. The load was chaotically varying after running for a long time, as the contents of the drum got bunched up more and more into a cohesive mass. It rode up as the drum rotated, and slammed down when it lost grip. VFD tried compensating too much too fast, while the shaft and drum was probably still oscillating from the sudden change in load. Some energy went to the brake resistor, but I dont think the transistor could open fast enough for the sudden peaking voltage. It seems to be running a lot more controlled now with the other control method.