Surge Suppressors

Status
Not open for further replies.
C

Cow

Senior Member
Location
Eastern Oregon
Occupation
Electrician
I'm totally green when it comes to these things. I had a good farm customer of mine ask for some pricing for one. The story is he's replaced one well VFD three times this year. It's just a little 15-20HP Yaskawa. Supposedly, they just replaced it again yesterday and he's a little tired of it....

He said they had an underground ground fault on the direct buried wires feeding this wellhouse. When they powered it back up, the drive didn't come back online. I don't know what caused the drive to fail the other two times.

Is a surge suppressor anything that'd help this type of situation or am I just blowing my customers money even suggesting anything? The area they're in is known to have surges so it may not be a bad idea anyhow....but whether or not it'd help relieve some of the hit from a good ground fault, I don't know.

Any suggestions? I was looking at a Leviton 32277-DY3, but supposedly this requires a neutral. I just have straight 480+ground to this wellhouse. I called Leviton tech support and they mentioned ALL their surge suppressors require a neutral...

I'm thinking they're only option may be to pony up a few thousand and stick a big one right on their service with no additional protection 200+ feet away at their wellhouse.

Any ideas are appreciated...
 
My thought would be to see if you can find out just what on the drive is failing. Same part each time? Does the smoke look like it came out in the same place?:roll:

The drives I have worked with do a very good job detecting ground faults on the load side before it becomes a major issue. I don't know if this feature can be disabled or not. Looking at your post again it looks like the ground fault was on the line side of the drive. Again though I would think the parameter settings would help to protect the drive unless they have been turned off. If the same people are replacing the drive they may very well be repeating the error.

Do the suppressors actually need a Neutral or just a grounded conductor?
Is this a wye or delta service?
 
If they had a ground fault. The fault might have traveling back on the ground wire. Until the breaker trips the ground is energized. Some/Most drives like AB PowerFlex drives have capacitors that reference ground. They provide a grounded jumper for a grounded/ungrounded system. If you have a solidly grounded system then you should leave the jumper installed. If you have an unground, or corner grounded system you should remove the jumper. Perhaps while the ground was energized your capacitors or other circuitry failed from overvoltage. If that is the cause removing the jumper might be able to prevent the failure, but having a solid ground would be better. Just a wild guess.

Surge protectors are handy, but a surge protector won't protect from the ground fault you said they had.

http://samplecode.rockwellautomation.com/idc/groups/literature/documents/in/20a-in010_-en-p.pdf
 
101109-2139 EST

Cow:

You need to know what is failing in the drive. This you must determine before anything else because it sets your direction to solving the problem.

A VFD will have input rectifiers to convert from AC to DC followed by filter capacitors. If the failure is a result of input voltage spikes, then it is likely a diode failure.

A well designed drive probably has input MOVs as shown in the reference in a previous post. These also might fail on sustained high input voltage.

Assuming the failure is from a high voltage on the input side and there are adequate MOVs in the drive, then I have a suggestion that might save the drive.

In each hot input line put a negative temperature coefficient thermistor, and a very fast blow fuse. The fuse would be the minimum size that would not fail on initial application of power and/or full motor load in combination with the thermistor. In other words made as tight as possible for fast tripping.

A capacitor input filter draws very large input current on application of power. A negative temperature coefficient thermistor has a high room temperature resistance and this resistance drops substantially with temperature. When power is off for a short time, maybe a minute, then the resistance returns to a moderately high value. This lowers the initial surge current at application of power. As current flows thru the thermistor the resistance drops. This is a means of limiting peak inrush current thru the fast blow fuses.

Under normal operation the thermistor resistance is moderately low, but there is a continuous power loss. When a voltage transient occurs that is large enough to cause MOV conduction the intent is that the fast blow fuse will open quickly and hopefully prevent damage.

Another means to prevent blowing of the fuses on application of power is to have the contacts from a time relay relay short the fuses for a short time just after power application. This is possibly a better approach because one can put more conditions on the operation, whereas the thermistor requires the person applying power to understand the characteristics of the thermistor.

.
 
ptonsparky said:
My thought would be to see if you can find out just what on the drive is failing. Same part each time? Does the smoke look like it came out in the same place?:roll:

If the same people are replacing the drive they may very well be repeating the error.

Do the suppressors actually need a Neutral or just a grounded conductor?
Is this a wye or delta service?
Click to expand...

I was actually thinking the same thing about seeing if the same part was smoking in the drive everytime, but I have no way of knowing since I'm not the one changing the drives, the pump company is.

It's fed from a 480 wye service. I asked Leviton tech support twice to be sure, and they said they do in fact need a neutral. From the research I've done, it seems most surge suppressors bleed the overcurrent from a surge onto the neutral. If it bled it to ground, it'd probably be the same as a ground fault I suppose, which got me into this situation in the first place?

StephenSDH said:
If they had a ground fault. The fault might have traveling back on the ground wire. Until the breaker trips the ground is energized. Some/Most drives like AB PowerFlex drives have capacitors that reference ground. They provide a grounded jumper for a grounded/ungrounded system. If you have a solidly grounded system then you should leave the jumper installed. If you have an unground, or corner grounded system you should remove the jumper. Perhaps while the ground was energized your capacitors or other circuitry failed from overvoltage. If that is the cause removing the jumper might be able to prevent the failure, but having a solid ground would be better. Just a wild guess.

Surge protectors are handy, but a surge protector won't protect from the ground fault you said they had.

http://samplecode.rockwellautomation.com/idc/groups/literature/documents/in/20a-in010_-en-p.pdf
Click to expand...

If I can't protect them from a ground fault, I'll just have to let them know that upfront. I still wouldn't mind finding a decent quality surge suppressor though, just to give them some options...

gar said:
101109-2139 EST

Cow:

You need to know what is failing in the drive. This you must determine before anything else because it sets your direction to solving the problem.
Click to expand...

I wish I knew, but I don't. This is far from an ideal situation.

Do you know of anything mass produced similar to what you suggested? I'd like to avoid having to make something myself with thermistors, time relays, fuses, etc.

I appreciate all you guys taking time to respond and giving me some ideas. I'll keep doing more research on surge suppressors.....
 
101110-2329 EST

Cow:

No I do not have a suggestion for a commercial product.

Of greatest importance is knowing what is failing. Without this type information it is extremely difficult to provide any sort of useful protection.

If the drive has no MOVs internally then it might be worth while to add external transient limiting. But it is just shooting in the dark.

Trial and error troubleshooting is a very bad approach.

Who is paying for these failed drives? What is the cost of failed components and service calls? What are the other costs as a result of the failures?

.
 
If you have not TVSS installed ahead of this drive, I highly recommend that you do so. This may not solve your original problem but will prevent other surges and lightning problems in the future. I also recommend installing a high speed power monitoring device such as a Drantez or RPM meter. A PQ expert can help you determine what the problem is.
 
I've had a few cases where MOV's failed on VFD's. And when they failed they took some electronic components with them. What sense does it make to put the MOV right on the board next to the components it is protecting if the MOV failure is going to take out what it is protecting?
 
If it were me, I would be installing a Line Reactor on that drive if it doesn't already have one. A Line Reactor slows down the rate of rise of a fault and can allow the VFD's internal protection to react and act without destroying themselves in the process. People don't like to add them because of the added cost and space, I call them "cheap insurance" and after the first failure, I definitely would have added them.

This is a very decent White Paper on the subject put out by my local PoCo, PG&E. Good reading.
 
We use line reactors on all of our drives. Its like a sponge and soaks up the disturbances. Also TVSSs on power.
Drives are a commodity item. The best drives are AB and Danfoss.
 
I gave the customer a price on a TCI KDR series line reactor as well as an LEA SP200 TVSS last week.

We'll see...;)
 
Is your supply 480V 3Ph 3 Wire delta, or is it really 480Y/277 only the neutral was not pulled?

Most drives, for at least the past 5yrs, do not like like not being connected to a wye system. In fact several of them make you remove an internal ground connnection, while others won't approve the installation at all.


OOPS, just read all of the posts and saw where you said WYE.
 
Last edited:
Status
Not open for further replies.
Top