VFD Noise Problem

[andrewc23]

I am a mechanical contractor with limited electrical expertise. We have (4) large 30,000 CFM air handling units. They have 6 supply fans and 4 return fans in an array. The supply fans are broken up and fed by two separate VFD's. Same for the return fans.

During start-up the drive tech was able to get each of the drives to operate individually. However, when he put the 2nd drive for each array, both drives went totally wonky. They wouldn't operate and it appeared to cycle from lots of different modes.

The 480V drives feed separately in conduit do a factory supplied junction box on the outside of the unit. From there, the electrician connected to factory run wiring to each fan array. These (6) wires are bundled together from the junction box to just before they split and go to each array. There is a thought that these cables being bundled is backfeeding noise into the drives.

One other note - there are control wires on each drive and also harmonic filters prior to the drives.


Any advice or insight would be greatly appreciated.

 

[kwired]

I am a mechanical contractor with limited electrical expertise. We have (4) large 30,000 CFM air handling units. They have 6 supply fans and 4 return fans in an array. The supply fans are broken up and fed by two separate VFD's. Same for the return fans.

During start-up the drive tech was able to get each of the drives to operate individually. However, when he put the 2nd drive for each array, both drives went totally wonky. They wouldn't operate and it appeared to cycle from lots of different modes.

The 480V drives feed separately in conduit do a factory supplied junction box on the outside of the unit. From there, the electrician connected to factory run wiring to each fan array. These (6) wires are bundled together from the junction box to just before they split and go to each array. There is a thought that these cables being bundled is backfeeding noise into the drives.

One other note - there are control wires on each drive and also harmonic filters prior to the drives.


Any advice or insight would be greatly appreciated.

Is control wiring bundled with any power leads - especially any output leads? Bundling all the power wiring together shouldn't cause any trouble but capacitive coupled noise on control leads can be a problem.

Are all the fans starting at same time? Is there by chance any fan(s) already free spinning (either direction) at the time you try to energize that motor? If so the drive needs to have on the fly starting methods enabled or it will give you troubles.

 

[andrewc23]

Controls

Controls

Is control wiring bundled with any power leads - especially any output leads? Bundling all the power wiring together shouldn't cause any trouble but capacitive coupled noise on control leads can be a problem.

Are all the fans starting at same time? Is there by chance any fan(s) already free spinning (either direction) at the time you try to energize that motor? If so the drive needs to have on the fly starting methods enabled or it will give you troubles.

The control wires (which currently aren't energized) are not in same raceway as power wiring. The fans also are not free spinning prior to energizing.

 

[ptonsparky]

Are the loads of each VFD bundled together or are they mixed with another drive?
Are the motors the same size? All 5hp, and not mixing 3s with 15s for example.

 

[andrewc23]

The wires feeding each fan array (3 separate fans daisy chained together) are bundled together and are fed by a junction box that houses the (2) separate VFD's. The wiring from drives to junction box are in separate conduits. Motors are same size that are feeding the 2 drives.

 

[Jraef]

The wires feeding each fan array (3 separate fans daisy chained together) are bundled together and are fed by a junction box that houses the (2) separate VFD's. The wiring from drives to junction box are in separate conduits. Motors are same size that are feeding the 2 drives.

I'm still confused. Motors don't feed drives, drives feed motors.
Please check the list here.

So you have an enclosure housing 2 VFDs. Y or N?
Each VFD has it's own harmonic filter of some sort AHEAD of it. Y or N?
You have 3 motors. Y or N?
One VFD is feeding one motor, the other is feeding two motors. Y or N?
Each motor is fed by a separate steel conduit. Y or N?
If not in steel conduit, you used VFD shielded power cable. Y or N?
The only place where the VFD output cables are in the same space is in the bottom of the enclosure housing the VFDs. Y or N?
When you run EITHER VFD alone, no problems. Y or N?
When you try to run BOTH VFDs at the same time, they "go wonky". Y or N?

Also:
You have separate OL relays for the two motors fed from one VFD. Y or N?
You have a load reactor on the VFD feeding two motors.Y or N?

The thread title says "noise", but that's not what you say later. Please describe "wonky" better.
The speed hunts around from what the drive is told to run at?
The motors make a lot of noise, as in DIFFERENT from when only one VFD is running?
The VFDs themselves are making strange noises, again DIFFERENT from when only one is running?
By "Noise" are you meaning ELECTRICAL noise, as in Common Mode Noise?

 

[andrewc23]

I'm still confused. Motors don't feed drives, drives feed motors.
Please check the list here.

So you have an enclosure housing 2 VFDs. Y or N? - no, two separate drives
Each VFD has it's own harmonic filter of some sort AHEAD of it. Y or N? yes
You have 3 motors. Y or N? no, 3 motors fed by one drive, 3 motors fed by the other
One VFD is feeding one motor, the other is feeding two motors. Y or N? no see above
Each motor is fed by a separate steel conduit. Y or N? no. cabling from all motors was factory run and is bundled. Pics attached
If not in steel conduit, you used VFD shielded power cable. Y or N? - conduit from drive to junction box on unit. No conduit from jb to motors
The only place where the VFD output cables are in the same space is in the bottom of the enclosure housing the VFDs. Y or N? don't understand
When you run EITHER VFD alone, no problems. Y or N? correct
When you try to run BOTH VFDs at the same time, they "go wonky". Y or N? correct

Also:
You have separate OL relays for the two motors fed from one VFD. Y or N? not sure, these motors were factory wired to jb.
You have a load reactor on the VFD feeding two motors.Y or N? harmonic filter? There is a filter for each drive.

The thread title says "noise", but that's not what you say later. Please describe "wonky" better. Drive starts behaving very erratic including won't operate motors.
The speed hunts around from what the drive is told to run at? Doesn't get that far
The motors make a lot of noise, as in DIFFERENT from when only one VFD is running? drive doesn't operate motors when both drives are engergized.
The VFDs themselves are making strange noises, again DIFFERENT from when only one is running? no strange noises. The "noise" I was referring to is line noise which is what they theorized is causing the issue.
By "Noise" are you meaning ELECTRICAL noise, as in Common Mode Noise?

thank you!

 

[Jraef]

If I'm seeing what I think I am seeing, this is a MAJOR F-Up!!!

It looks to me as though you have 2 VFDs. The output leads from each VFD are in their own conduit, but then they BOTH go into the SAME junction box, and inside of that box, BOTH outputs are tied together and feeding the 6 motors? That's what I am seeing in picture #667.

If that's the case, it's no WONDER this is messed up, and in fact you may have already damaged the VFD transistors!
http://forums.mikeholt.com/attachment.php?attachmentid=13120&d=1438887110
If this was factory pre-wired this way, it is a major mess and by the way, NOT code compliant.

If you connected the two VFD output together like that, you made that part of the mistake. Each VFD can control a set of three motors, or each VFD could (if sized right) control 6 motors. But you CANNOT tie the two VFD outputs together!

Even if you did that part of it, it still violates the NEC, which REQUIRES that each individual motor have it's own separate overload protection. A VFD provides OL protection for ONE motor, but it cannot separate that out for 3 different motors, it only sees ALL of the current.

 

[ptonsparky]

It does look wrong but It wasn't clear enough for me to make that call. Besides, that being such a major screw up, I had to doubt myself.

 

[kwired]

It does look wrong but It wasn't clear enough for me to make that call.

makes sense now that it was pointed out, but same for me - pictures were so small I didn't really know what I was looking at.

 

[Rampage_Rick]

The output leads from each VFD are in their own conduit, but then they BOTH go into the SAME junction box, and inside of that box, BOTH outputs are tied together and feeding the 6 motors? That's what I am seeing in picture #667.

I think it's an illusion and there's one splice hidden behind the others. The brown and orange wires from the lower conduit feed to splices outside the frame, so obviously they're not spliced to the other brown and orange wires from the upper conduit. I zoomed in on the picture and it does appear that the one yellow wire goes in behind the other splice too.

Another possibility: Could one motor have 2 phases connected to one VFD and another phase connected to the second VFD? i see a lot of black wires there...

 

[andrewc23]

I apologize - this forum makes the pictures that small. The outputs are definetely not tied together. One drive per three wires.

The he thought is that because the six wires are bundled that you may be getting feedback from one set of three to the other.

 

[Jraef]

I apologize - this forum makes the pictures that small. The outputs are definetely not tied together. One drive per three wires.

The he thought is that because the six wires are bundled that you may be getting feedback from one set of three to the other.

OK, sorry, but it did look that way to me.

In general, a small distance and space like that j-box is not going to present much of an opportunity for crosstalk and induced voltages. But what is that last picture showing? Are those loomed bundles of the output wires after the j-box? If so, that would definitely be an issue. Those don't look to be shielded or in steel conduit.You should NEVER do that with the output conductors of VFDs.

 

[andrewc23]

That is in fact the wiring after the junction box that was completed at the factory. These are Trane Units, a top quality manufacturer. They claim this is how they do all units like this and they have never heard of a problem with this wiring. There is no conduit and it is some sort of webbing that ties the cables together.

 

[ptonsparky]

These are units that come from the mfg with VFDs installed? And their tech can't get them to run correctly? O bother.

 

[Jraef]

That is in fact the wiring after the junction box that was completed at the factory. These are Trane Units, a top quality manufacturer. They claim this is how they do all units like this and they have never heard of a problem with this wiring. There is no conduit and it is some sort of webbing that ties the cables together.

Well, I don't know what to say. If they say they have never had an issue with that, I'm shocked quite frankly. Not only is there going to be cross-contamination of those conductors, but they are ALL going to be somewhat powerful FM radio transmitters as well.

My suspicion is that in other instances they have been using shielded VFD cables for those motor connections, in which case that would be fine. But a laced bundle of loose wires? Horrid. Maybe some bean counter changed it because he was trying to save money by not using the shielded cable and never asked why? Who knows.

God luck with that mess...

 

[Besoeker]

My suspicion is that in other instances they have been using shielded VFD cables for those motor connections, in which case that would be fine. But a laced bundle of loose wires? Horrid. Maybe some bean counter changed it because he was trying to save money by not using the shielded cable and never asked why? Who knows.

God luck with that mess...

Not sure that an appeal to God is quite the answer..........

However,I agree with your point about shielded cables. Here in good old Britain we would have used a separate steel wire armoured (SWA) cable from each drive to each motor for power. And a separate SWA for field controls.
But what do I know....maybe, in 40 some years in the business you learn stuff.

The pics of the installation from the OP do look a bit messy though.

 

[GeorgeB]

If they say they have never had an issue with that, I'm shocked quite frankly.

Your shocked that they'd SAY they had never had a problem? I recall many conversations with OEMs when I, and others, had reported and identical (not this) issue, and that was ALWAYS the response. That was before the internet, but similar customers (at the time I worked for an electric utility) talked about problems among ourselves. Our officers discussed these things at the annual EPRI get-togethers between major utilities; we peons would occasionally get calls to furnish details.

Intentional ... I'd never say that. But communication between the field and offices is little better today than then.

 

[kwired]

Aren't the conductors all still inside the steel cabinet of the equipment though?

Why wouldn't that provide shielding, just like the other end is inside the drive cabinet?

 

[kwired]

I apologize - this forum makes the pictures that small. The outputs are definetely not tied together. One drive per three wires.

The he thought is that because the six wires are bundled that you may be getting feedback from one set of three to the other.

Uploading images to the forum doesn't work very well, to get good images to appear in posts you need to put them on a host site like photobucket and then link to them in your post via