VFD Faults

[wveng25]

We are having problems with a VFD, at a Pump Station, having "HW Overcurrent Faults" after the weekly generator exercise event. The odd thing about the fault it does not happen when the weekly exercising of the generator occurs, but it happens approximately 3 to 6 hours, or more after the generator runs. There appears to be some connection between the VFD faulting and the generator weekly exercise, but all parties are at a loss as to what it may be. The generator is a Kohler 100KW/125kVa natural gas fired. The service is 277/480 volt 3 phase. The VFD is a Allen Bradley Power Flex 400 and it is used for a Pump Station with 2- 24.8 HP, 460 volt submersible pumps. Has anyone ever run across this time of problem before or have suggestions for possible causes of this problem.

 

[waross]

Try exercising the generator on a different day to reduce the possibility of coincidence.

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

 

[itsmoked]

Does the generator test increase the water pressure on the pump inlets?
Has something reduced the pumps' working heads? More flow than in the past?

If the DC brake volts are adjusted too high or an improper DC boost setting can trigger that fault.

Any chance of the drive's room humidity rising dramatically after the generator has run?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[jraef]

The HW current fault means that the drive was being asked to provide more output current that it is designed to handle for longer than it can handle it. So for example, the PF400 is strictly a "Variable Torque" drive (in A-B parlance referred to as "Normal Duty"). That means it is designed to deliver 110% current for 60 seconds, or 150% current for 3 seconds in an I2t curve, or 180% instantaneously (100msec). If you attempt to exceed that, the drive will EITHER respond by reducing the commanded speed to reduce the load on itself, or if that is turned off in programming, it will fault. So we can assume it is turned off. Check the programming of A148, the OL Response function, if set to 0, that would allow it to trip if the current rises too fast. Also check A170, the Boost Select function. The default setting is a value of 4, which is for a VT load profile, but that includes what is called IR Compensation. That automatically allows an increase in the voltage boost at low speeds if the motor is cooler. Try setting that to a value of 5 which turns off the IR Comp, see if that makes a difference. Don't know why it would be related to the generator test, but I do know that IR Comp can sometimes make a drive trip itself under the right (wrong) circumstances.

That brings up this curious thing about your observation of it only being related to having been run from the generator 6 hours earlier. It's possible that this is coincidental, but it sounds as though you have observed it as a consistent event sequence. That raises more issues than it solves. I can't see it having anything to do with the VFD itself, although honestly, I kind of like Keith's idea on humidity. For example when the generator is exercised, the exhaust must be vented, so a lot more outside air is introduced to the room to replace it. Six hours later the pump station is being called for less duty and the VFDs shut down, cool off and the added room humidity condenses in and around the drives, current flows to ground and the drive is tripping on that current flow, or at least the sensors used to determine current flow are being shorted and giving a false signal. Remote, yes, but I think plausible.

So what else, if anything, is different about how it is run when in the testing of the generator takes place?


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[ScottyUK]

An engine cooling down releases a lot of heat to ambient. If the drive is in the same room as the engine, could the ambient temperature be rising over a few hours to the point where the drive is faulting? I don't know the Powerflex as well as Jeff, so possibly no link there.

 

[wveng25]

I should of mentioned that "HW Overcurrent Fault" also occurs at about the same interval after the generator shuts down when a power failure occurs and these are at random times. In addition the generator was out of service for about 4 weeks and no faults occurred. There is somehow a relationship with the generator and the " HW Overcurrent" faults occurring. Also the generator and VFD are not contained within a structure, but outdoors. The VFD is under a canopy to protect it from the weather, but is open on all sides. I don't believe the ambient humidity would be an issue. Also I know of no issues with the head changing to cause the faulting. We will check the programming and make the changes mentioned to the setting and see if this helps.

 

[davidbeach]

So, it sounds like "the generator" is a red herring. It sounds like the VFD isn't dealing well with the brief power outage associated with the transition between normal and generator. Could be the interruption itself or it could be some difference in voltage, phase angle, or something else between the two systems.

If it is just the interruption then you could cause it without the generator at all. Maybe a longer dead time would help.

 

[jraef]

6 hours from when the generator stopped running is a long time to associate any electrical issues with.

I'd be taking a close look at what happens in that pump station 6 hours after the generator shuts down. Sounds to me as though it might be something set on automatic, like maybe an engine block heater? Oil filter? A capacitor bank that is disabled while the generator is on and re-enabled 6 hours later? Battery charger?

Here's aanther issue though, are we assuming the VFD was actually running the pump when it faulted, or could it trip on HWOC while sitting idle?


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[melspuds]

I would hang some data loggers on the incoming 480v line and the drives output and see if the fault occurs during any incoming 480v line. To me this pumping station is remote. What are the other loads on the primary line, how close or far from the substation, are tap changers or cap banks being switched, other harmonics on the line, ect. The A-B Power Flex series are a very tough drive and the only one time I had a problem with one, was with harmonics being feed from a buss with large DC drives. We changed the 480v incoming line which took care of the problem. You might have to use an isolation transformer.
On a HW (high water) alarm,the pumping system normally starts the second pump. The second pump tries to start but may be spinning in the opposite direction due to first pump running and creating reverse flow down the second pump head if both pumps discharge into the same pipe or vessel. If that is the case, you need install or repair check valves at the pumps discharge or put the check valves at some other location on each of the pumps discharge lines and put an anti-backspin timer on each pump. It has been a few years since I have worked Power Flex drives, but I seem to remember there is such a timer in the drive setup.
ps: afterthoughts 1) Check all of the grd connections @ the service, drives, and pump motors.
2) Check you water level controls and wiring, such as floats, ect.

Hope this helps and good luck
Dave

 

[jraef]

"HW Overcurrent Faults" means Hard Ware Over Current, meaning the current attempted to rise faster and/or higher than the software programmed settings and reached the design limits of the hardware itself. It's the drive's last line of defense against a catastrophic failure.


"You measure the size of the accomplishment by the obstacles you had to overcome to reach your goals" -- Booker T. Washington

 

[melspuds]

jraef, you are correct. I forgot, I was thinking only about pump control systems. Sorry about that!
I most likely worked on at least 20 Power Flex Drives, which most were for pumps.
Dave

 

[powersoff]

Ran across this fault before. While VFD output was on, a downstream manual disconnect would intermittently(people) open and/or close.

Any components that could manually or automatically isolate input or output of VFD might need further investigation.