PDA

View Full Version : VFD generates overvoltage alarm at startup.


JesperMP
August 5th, 2004, 09:17 AM
Hi, heres one for the drives experts:

My VFD (Siemens MM430) generates a fault at startup (F002 - Overvoltage). The fault does not happen every time, only approx every second startup.

The application is a large balanced vibating conveyor with two drive stations. The conveyor is agitated by the motors directly via excentric shafts (hope you understand what I mean).

I think that the fault is caused by the motors who generate the overvoltage when the conveyor/motors alternates between driving/regenerating during the very initial movements.

Are there any setings that can help avoiding this fault ?
The VFD is not equipped with a braking resistor. Would it help if a braking resistor is installed ?

NB: Stopping is done with coast-to-stop. Thats why braking resistor was thought not to be necessary.

Peter Nachtwey
August 5th, 2004, 09:59 AM
Reduce the acceleration rate. Just like accelerating a car, it takes more power to accelerate than anywhere else in the motion profile.

JesperMP
August 5th, 2004, 10:14 AM
Peter,
will try your suggestion.

Have put the startup ramp to 2 seconds. Longer times tended to generate overcurrent faults (due to hard start ?).
maybe some experimenting with this time will do the trick.
I'll let you know how it goes.

TRACY H
August 5th, 2004, 10:57 AM
Saw an application recently where a vibrating conveyor
was feeding a crusher. It started across the line and
then the drive took over the feeder was very hard to
start full of rocks. Just a thought. :huh:

jrwb4gbm
August 5th, 2004, 12:42 PM
Icrease the decel. time also, if possible.Installing the braking resistor will help. Remember to change the parameters for using the braking resistor. Installing a buck-boost transformer to lower the incoming line voltage, within the low voltage limits of course, may also help.

JesperMP
August 5th, 2004, 02:48 PM
Originally posted by TRACY H
It started across the line and then the drive took over the feeder was very hard to start full of rocks.
Yeah, I have thought of changing the whole thing to direct startup, and changing to the VFD for maintained operation.
But I need to be sure that it cant be solved in an easier way before changing the whole setup.

Originally posted by jrwb4gbm
Icrease the decel.
Nah, that cant be it. Remember it happens at startup, and it stops by coast-to-stop anyway.

Thanks for your thoughts :yeah:

Tom Jenkins
August 5th, 2004, 03:43 PM
1) In general over current trips occur on starting if the acceleration time is too short. Over voltage trips generally occur on stopping if the deceleration time is to short. If you are using eccentrics to drive the vibrator the torque varies and may even be negative as the eccentric is moving down. It is possible that the motor over-speeds, the drive tries to decelerate the load, and an over voltage trip results.

2) Try putting a big flywheel on the motor or the reducer shaft. There is a reason they put them on steam engines! It reduces torque fluctuations.

Bitmore
August 5th, 2004, 11:30 PM
generates a fault at startup ,,,,,,,,

That VFD has "S" curve parameters. If it were me, I would try adjusting high initial current to start the load, and a linear acceleration. I think the over voltage, if it is a DC buss fault, is due to the directly coupled eccentricity and the mass.
This adjustment would give your motor the added torque for starting, and the best part, high initial current discharges the buss!
The trick is to make it "look" like a line voltage start, then tweek the V/F curve to optimize the start.

bitmore

post script:
Tom are you OK?

JesperMP
August 6th, 2004, 06:24 AM
The fault is almost cetainly caused by the mass of the conveyor.

Originally posted by Tom Jenkins
If you are using eccentrics to drive the vibrator the torque varies and may even be negative as the eccentric is moving down

Actualy the conveyor is driven directly via exentrically linked shafts. Not via a rotating mass.
Even so, I think that the problem is similar - the conveyor changes between being driven and driving the motors during the startup phase.
(We DO have another conveyor with a rotating mass (and stating by VFD), but that one is starting up OK).

When we observe the DC link voltage during startup, it just barely touches the trip level. Sometimes it trips, sometimes not.
We have just had a Siemens person on site to help us with the problem.
He do say that there are two possible remedies:
1. To install an optional brake resistor with integral chopper unit.
2. To install a line choke.
We have ordered both these parts allready.

Tom Jenkins
August 6th, 2004, 01:53 PM
Originally posted by Bitmore
generates a fault at startup ,,,,,,,,

bitmore

post script:
Tom are you OK?

I haven't been OK for years! I'm not quite sure what specific aspect you are referring to though. Have I been too polite, or is my technical competence slipping? I'm not in a position to judge these things, but I gaurantee my wife can give an opinion!

rsdoran
August 6th, 2004, 10:50 PM
When in doubt...RTFM
http://www4.ad.siemens.de/dnl/Dc0NzQ4NQAA_14346754_HB/430_OPI_en_1202.pdf


F0002 OverVoltage OFF2
Possible Causes
! DC-link controller disabled (P1240 = 0)
! DC-link voltage (r0026) exceeds trip level (P2172)
! Overvoltage can be caused either by too high main supply voltage or if motor is in regenerative mode.
Regenerative mode can be caused by fast ramp downs or if the motor is driven from an active load.

Diagnose & Remedy
Check the following:
1. Supply voltage (P0210) must lie within limits indicated on rating plate
2. DC-link voltage controller must be enabled (P1240) and parameterized properly
3. Ramp-down time (P1121) must match inertia of load
4. Required braking power must lie within specified limits


I think Tom mentioned something about driven from active load...or similar. In this case maybe the decel time does apply.

JesperMP
August 7th, 2004, 03:21 AM
When I started to get the fault I scrutinised the manual with a magnifying glass. I certainly have pondered over the "remedies" mentioned in the manual.

The decel time (ramp downtime) has been increased allready. Made no difference.

And yes, I am certain that the problem is caused by the motors being driven by the load.
The Siemens person was quite sure that the resistor/dc chopper unit will solve the problems.
We shall see in a weeks time how it goes.

We DID discuss Toms idea of a big flywheel. Thats going to be the last resort.

kamenges
August 7th, 2004, 02:51 PM
I would have to believe that the chooper/resistor will solve your problem also. But I'm still a little fuzzy on why the problem is happening in the first place. The working theory seems to be that the load is overdriving the motor. But, while you don't say this directly, you seem to infer that once you get the motor up to speed you don't get overvoltage trips. Where does the overdriving load go once you are up to speed? I would think it is still there and would still cause you faults during normal run.

Keith

DickDV
August 7th, 2004, 09:27 PM
JesperMP, you need to pay special attention to Tom Jenkin's answer about regen on eccentric loads. That's the answer.

The trip on startup probably occurs on the first downstroke when the speed is too slow to permit any averaging of voltage on the DC bus.

If you could get up to speed, the stroking probably occurs fast enough to cause the DC bus caps to charge and discharge vigorously but being able to maintain the average voltage at a point below the trip point.

I can't say that Tom's suggestion of a flywheel is particularly practical, in most cases. It doesn't sound like it would be in your case.

I can offer two other possible answers. First, I have no doubt that adding a small amount of DC snubber braking to the drive would work fine. You will not need a chopper transistor or resistor anywhere near the full motor horsepower--maybe a third of motor hp would be enough. But the wattage on the resistor must be quite high due to the regen occurring on every stroke.

A second possible solution will only work if the drive and motor horsepower are higher than the actual continuous load horsepower by at least 20%. If that is the case, try limiting the maximum output voltage from the drive to first, 10% less than nameplate, and then, if that doesn't get rid of the fault, try reducing the output voltage another 10% and check for the fault. Of course, you may, in the process have reduced the motor torque to a level that will cause stalling when trying to get over the first lifting motion of the eccentrics in which case you will have to increase the voltage at least to the point where starting is successful. If the fault goes away at this level, you won't need to buy the snubber package. If the fault is still there, the snubber brake package is the answer.

The above "trick" sometimes works because the motor excitation voltage not only determines the driving torque but also limits the ability of the motor to regenerate when overdriven. I've used this on stamping presses occasionally when regen faults occur between strokes. Of course, that application is a bit easier because the start is into a simple rotating flywheel with no eccentricity until up to running speed.

Either way, its worth a try! Good luck!

JesperMP
August 19th, 2004, 02:42 AM
Here's the last chapter to this story:

The brake resistor with chopper has been installed, and the overvoltage alarm hasnt been observed since :)

It is very obvious when the chopper cuts in because of an audible "tick-tick-tick" noise from the unit. The "ticks" comes with the same frequency as the movements of the conveyor.
Its great with an audible feedback. Now you can HEAR whats going on. And it is now clear that it is exactly as we all suspected it to be.
In fact, I am amased that the drive can even operate let alone start up with the severe conditions it was working under without the braking unit.

Thanks again to all that participated.

CaseyK
August 19th, 2004, 03:16 AM
I would have opted for the big flywheel. They are fun to watch.

Bitmore
August 19th, 2004, 07:23 PM
KC I wish you were in 2M or 73CM talking distance, You know of the west link? We need an east link.

For those who don't know what the west link is... It is a link of 144 Mhtz repeaters between Texas and California, hams have been talking on this 2000 mile link with FM radio for about 20 years. I don't want to blow my horn to loud but you guys out there would not be talking on the internet if it wasn't for us hams and packet radio,,, we HAMS invented IP!