I'm hoping one of the drive gurus here can give me some advice on an issue I'm having.

I have an ABB ACS800 running a 75HP, 1775 RPM, 460v/60 HZ, 84.1 FLA motor that drives a conveyor belt that delivers pine wood chips to a digester.

This drive is tripping on Overcurrent several times per shift. The motor and gearbox/reducer have been changed twice. Still having this nuisance trip, all the time.

I would say the cable from drive to motor is around 250-350 feet. There is no motor terminator nor line/load reactors on this drive. The correct cable was used, thankfully.

Our "drive group" here at work say that the lack of reactor/filter could not possibly have anything to do with this problem, and I think it could be a factor, so I wanted to throw this out there for you fellows and ask for advice.

I went through the drive parameters, and the "Load Current Limit" is set to 500%, Load Torque Limit is 300%. Do those numbers seem right to y'all? Our mechanics have checked bearing temps, rollers, and all that stuff and say there are no mechanical issues.

Any idea would could be causing this trip? Our DCS monitors the current so we can trend it, and we aren't seeing the spike...but our trending program (PI) is on a 1 minute polling time and it's easy to miss a fast spike when it's only grabbing data once a minute.

Thanks in advance for any assistance you can offer.

There are several flavors of "over current" trips in almost all VFDs, in the ACS800 especially. Do you know the EXACT fault description or code? For example there are differences between a hardware current trip and a software current trip, usually the hardware trip is a fixed value in the firmware that protects the drive components, the software trip is based on the limits that you program into the drive. So if for example you have the software set to 500%, but the drive is only capable of delivering 200% for 3 seconds, you might not be expecting a trip at 300% current, but you will get the hardware over current trip in 3 seconds, because you are attempting to tell the drive to do something it is not capable of doing. It's not like a starter, the components have limits.

So why would they allow you to set the software limit at 500% if the drive is not capable? Good question. Some mfrs don't, some do, because if you NEEDED that kind or torque and current, you can over size the drive by 250% for example. That way you can get the necessary torque from the motor, even if it needs 500% FLA to produce it, but you are no longer exceeding the hardware capabilities of the drive.

I'm not saying this is what's going on, I'm pointing out that a simplistic "over current" trip is not as simplistic as you might have thought were it not a VFD, you will need more detailed info.

By the way, there will be a fault history inside of the drive, with time stamps on every fault. That can be VERY useful in troubleshooting something like this. For example if you have a power monitor for the plant and you compare the trip times of the drive to voltage dips from the utility and see a correlation, there's your culprit.

 

It's a 100HP drive, so it's a little oversized. I cannot seem to locate the fault queue in this drive. I'm more accustomed to Powerflex and Reliance, we have a ton of those and only 2 ABB drives in my area. Could you let me know how to access the queue?

Am I barking up the wrong tree with thinking we need a load reactor/filter? Our drive guys claim that reactors are for saving the drive windings and nothing else, but I thought there was more to it than that.

When I reset the fault, I'm pretty sure it just says Overcurrent, but I've seen the HW Overcurrent fault you are talking about on Powerflex drives. Next time it trips I will see if there's a fault code, if I cannot find the fault queue before then.

We have our own power plant here, and the "drive room" where this drive is located contains another 12 drives, and all are on the same bus from our powerhouse, so I would think voltage drops would affect more than just this one drive. But I can look into it.

Any other things I should check out? My other techs just claim operations is overloading the belt, but I don't think so since the motor starts right up with no issues after being reset. I would imagine an overloaded belt plus locked rotor torque (start-up inrush) would trip the drive back out instantly. The drive group just says "it's not the drive." The mechanics say "it's not our problem" so I'm going this alone and really appreciate the help.

 

Overcurrent (2310)

 

Don't be too quick to rule out mechanical issues, I have had bearings get too hot, cause motor to pull enough to trip O/L's, by the time someone sees machine is down and resets, bearings have cooled down and turn fine. if you can, make sure nothing is dry for yourself. Also, overcurrent is not the same thing as overloads. You may have a intermittent short, megger the conducters and motor. That's a crazy long distance between motor and drive, I would have installed the drive closer, and while I probably would use a load reactor myself, not sure if that is the issue here. You said you have two drives , maybe you could run conductors from that drive and see if motor still trips, so you know your drive is good. Good luck, let us know when you figure it out. Just remember, if it was easy anybody could do it, they would'nt need you.

 

When I get an overcurrent fault, I always pull the T leads off the drive and meg all the way through to the motor. It only takes a few minutes to rule that out.

Here's the manual:

http://www05.abb.com/global/scot/sc...eritydisplay/6f96ed70e1b467a9c12578f80034ed25/$file/en_acs800_standard_fw_l.pdf

The fault code:

OVERCURRENT
(2310)​

3.05 FW 1 bit 1
​

Output current exceeds trip limit. Check motor load.
Check acceleration time.
Check motor and motor cable (including
phasing).
Check that there are no power factor correction
capacitors or surge absorbers in motor cable.​

Check encoder cable (including phasing).

 

It's been megged multiple times. This has been an issue for two years now, and everyone just points fingers and says it's not their problem.

The drive group set up all kinds of monitoring equipment on it at one point, and couldn't figure it out.

The mechanics have walked the belt, while it's running, and checked bearing temps and all that multiple times.

I can't use the other drive, as it's running another belt, and it's a much smaller drive.

There are about 10 Reliance drives (DC of course) running off of the same bus that trip/amp up regularly (rotary chip feeders) so maybe they are causing a voltage dip. I'm going to look into that tomorrow.

 

Have you checked the accel/decel times? Is your trip occurring when the belt is emptied quickly or loaded quickly? That might slow things down and you could be seeing that.. Maybe try lengthening those times?

 

I would install a motor terminator.

 

If the issue is load related, or system dynamics related, you may be able to take care of it by decreasing the current limit, if it is at the default 150% of the drives rating, it's not hard to cause an Overcurrent with the acc/decel time or rapid heavy loading. I always set the current limit to 100% of the motors nameplate rating(if system dynamic allow), this will usually prevent this. Look in group 20. The 800s current limit is very fast reacting, if set right you can stall a motor and not result in an Overcurrent trip. Also make sure you are running this in DTC mode with a standard auto-tune done(motor uncoupled). If the drive is in scalar mode, you are not getting near the torque response it is capable of.


Sent from my iPad using electriciantalk.com

 

How long is the interval of tripping after the system has cooled and restarted?

 

To see the fault que hit the ACT (actual) button on the display, then hit the double arrow buttons


Sent from my iPad using electriciantalk.com

 

At that distance you are approaching the maximum recommended, a DV/DT filter may be in order, but I would try lowering the CL first (20.03), also ensure over voltage control is enabled(20.05).


Sent from my iPad using electriciantalk.com

 

An after thought, if someone were to disable ground fault protection in group 30, Overcurrent would be the next trip in line to catch an issue.

 

Now there's an idea!

One time I got a call from a remote site with mobile crushing equipment that included a vibratory feeder that was run from an ABB ACS400 VFD, and the complaint was that people were getting a "tingle" when they touched the machine steel while standing on the ground. The VFD was not showing a GF, but when I went over to the motor (40HP) I could HEAR the arc tick taking place inside of the peckerhead! I shut everything down and sure enough, the blue wire nuts they had used to make up the motor connections had rubbed against the walls of the peckerhead until they wore through the plastic and one was shorting to the steel. Why didn't the VFD pick it up? Turned out it HAD, but it was annoying someone that it kept tripping off, but since they couldn't figure out how to disable the GF protection, they cut the ground wire so the VFD had no ground reference point back from the machine... problem solved.

 

Ouch,lol

 

i would have found out who cut the ground wire and gently planted a size 12 boot up their @$$
then i would let their bosses have them

good find JRaef!:thumbsup:

Well, I knew who it was. It was the same guy who would later be signing the check for my bill! So I kept my boot firmly on the ground.

He paid for his mistake, literally. He had me drive from Seattle to almost the top of Mt. Hood in Oregon, about a six hour drive, stay over night, 3 meals including a KILLER steak dinner and $60/hour (this was in 1991) portal-to-portal, all told I charged him I think a little over $4k. Hopefully he also learned his lesson about how LUCKY he was that he didn't kill someone! I drove that point into his skull repeatedly while I was there.

 

I agree about making sure the ground fault detection is enabled. ACS800 ground fault detection is VERY good.

The problem with meggering it all and calling it good is that you really need a continuous fault in order for that assumption to be valid. Being that your fault is so intermittent it's probably a bad assumption. Making sure the GF is enabled will rule out an intermittent gore d fault at least.


Have you checked the harmonic content on the line side of the VFD, and checked the peak-peak voltage and or looked for reflections on the load side of the VFD output?

My other suspicion is with site-generated power (i.e. likely less "stiff" than typical utility power) plus multiple drives without filters etc the power quality might be horrendous. It would be worth checking.

 

I also thought of cable capacitance charging current, but you mentioned that they "used the right cable" so I moved on from that. But what if they didn't really use the "right" cable or they didn't properly terminate it? A really common mistake is that someone gets the shielded VFD power cable, but then installs it like shielded control cable and grounds only one end of the shield; POWER cables must have the shields grounded at BOTH ends. Doing it wrong basically renders it useless in helping to reduce cable capacitance interaction with adjacent cables and the subsequent possibility of charging current spikes. Another possibility I just came across is people ASSuming that just because a cable is "shielded", it is "VFD cable". Not so. Just as important as the shielding is the insulation material. Cheaper cables use PVC insulation, which is either not of uniform thickness, or much thicker to compensate, which INCREASES the potential for cable capacitance. VFD cable must use XLP (Cross Linked Polyethylene) insulation on the motor conductors. Do you have a brand / spec on the cable? Were the shields grounded at both ends?

 

Just an update. We increased the accel/decel times, still tripping several times a shift.

We have an analog current output signal going to the DCS, and it's not seeing more than 80A output, motor is 84.1 FLA. Not sure if I mentioned this before.

Thinking about changing the drive, not sure what else to do.

 

Did you decrease the current limit in group 20? Did you check to see if ground fault is enabled in group 30?

 

I've tried pretty much everything, finally enabled auto-reset for Overcurrent. Did that a few weeks ago and have been enjoying the peace and quiet of not having that nuisance call.

Now, in the last 2 days, it's gone down once on DC Undervoltage and once on DC Overvoltage. Any chance this could be related to the Overcurrent issue? Perhaps, as I mentioned before, a symptom of having a room full of drives...some without line/load reactors, most sharing the same group breaker?

 

Do you have any other drives you can run the motor off of temporarily? I'll rig almost anything together temporarily just to help narrow problems down. Insulated taps and SO cord ran to an adjacent MCC bucket or drive can be a lifesaver.

I've had drives throw up random codes or draw excessive current before, it usually means they're about to smoke.

 

At first thought I'm with you on the reactor but just for "****s and giggles", let's say that that is not the problem. AT the supply end of the load are any of your control wires running "parallel" to the 480? If so, try to reroute and approach the 480 and/or cross it at a right angle.

 

I dont think we have another drive, of this type and size, on hand at the moment. This is one of only 3 ABB drives in my area, everything else is Powerflex...not sure why we went with this drive for this application.

Well, I say everything else is powerflex. Most of the drives in this particular drive room are Reliance DC drives from the late 70s. We really need to do some upgrades.

I just find it odd that every drive aside from these 3 ABBs has either line/load reactors or an isolation transformer. I don't think this install was properly engineered.

EDIT: No, there are no control wires running parallel to 480v.

 

Do you have a copy of drive windows where you can send me a parameter list? Enabling Overcurrent auto reset is a bad plan. Are any of the output leads sharing conduits or raceways?

 

I was just informed that someone noticed a piece of the belt was separated, and was flopping around and getting hung on roller occasionally. Will see if that solves the prioblem