Author Topic: Looking for transients in 600V 3 phase system, need opinions  (Read 3270 times)

0 Members and 1 Guest are viewing this topic.

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Not sure this is the right place for this...

Having an issue with a VFD repeatedly blowing it's input side guts all over the control cabinet and everyone I have asked is as dumbfounded as I am as to why.

60Hz, 610V - 612V measured phase to phase.  VFD is rated for 500-600V +10% / -15%

Power comes from electrical room through a disconnect, circuit breaker, and line reactor to the VFD.  The motor connected is a 1.5Hp conveyor belt unit, draws 0.3 A when running. 
A pair of soft starters for 30hp pumps are also tied into the disconnect and seperately fused, zero trouble with those bad boys.

I'm suspecting transient spiking/surging in the building but do not have access to a power quality analyzer such as the Fluke 435.  I was contemplating using my DS1054 and a B&K PR2000B high voltage probe, setting up a pass/fail envelope test and letting it run.

Can this be done safely or am I suicidal for even thinking of doing this?
Any other economical ways of measuring/logging this?

Thanks
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9320
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #1 on: July 03, 2019, 11:35:38 pm »
How far is the motor from the VFD? How far is the VFD from the source? I'm assuming it's very long since I see little other reason to use such a high voltage at that low power level. Could be standing wave issues.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Online langwadt

  • Super Contributor
  • ***
  • Posts: 4857
  • Country: dk
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #2 on: July 03, 2019, 11:41:38 pm »
a few small transformers in series so they run at a bit lower than rated voltage ought to give a reasonably linear isolation
 

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3362
  • Country: au
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #3 on: July 04, 2019, 01:23:28 am »
Power comes from electrical room through a disconnect, circuit breaker, and line reactor to the VFD.
If the VFD suddenly stops pulling current, could the line reactor be giving the VFD input an inductive kick?
 

Offline T3sl4co1l

  • Super Contributor
  • ***
  • Posts: 22436
  • Country: us
  • Expert, Analog Electronics, PCB Layout, EMC
    • Seven Transistor Labs
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #4 on: July 04, 2019, 03:48:47 am »
What regulations is the VFD rated for?  In particular, high-line and surge ratings would be of interest?

Tim
Seven Transistor Labs, LLC
Electronic design, from concept to prototype.
Bringing a project to life?  Send me a message!
 

Offline capt bullshot

  • Super Contributor
  • ***
  • Posts: 3033
  • Country: de
    • Mostly useless stuff, but nice to have: wunderkis.de
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #5 on: July 04, 2019, 05:52:52 am »
I was contemplating using my DS1054 and a B&K PR2000B high voltage probe, setting up a pass/fail envelope test and letting it run.

Can this be done safely or am I suicidal for even thinking of doing this?
Any other economical ways of measuring/logging this?

No, don't use this kind of probe for this kind of circuit. You'd want a differential probe rated at 600Vrms CAT IV here.

Quote
Any other economical ways of measuring/logging this?
If you know what you're doing, one could build a series connected resistor string to safely divide the voltage down to the scope's input level, and use two channels in subtract mode.

Otherwise, no.

Your symptom description clearly points to line overvoltage / surge, but I've also seen input rectifiers been destroyed by (otherwise mostly harmless) bursts. Once a single rectifier diode fails, the vast amount of available energy at the input takes care of blowing the stuff up.

One reason for surge / burst might me other consumers (the 30hp pumps you mentioned) connected to the same circuit. Not while they're running, but rather at turn-on or turn-off.
Safety devices hinder evolution
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #6 on: July 04, 2019, 07:46:04 pm »
How far is the motor from the VFD? How far is the VFD from the source?
Source to Panel is about 50m, VFD to Motor is 5m.   Voltage is to keep a single voltage in control cabinet, 600VAC is for two 30hp pumps.

Power comes from electrical room through a disconnect, circuit breaker, and line reactor to the VFD.
If the VFD suddenly stops pulling current, could the line reactor be giving the VFD input an inductive kick?
Possible I guess however when motor is stopped, VFD still drawing small current for itself.  Seems unlikely though.

What regulations is the VFD rated for?  In particular, high-line and surge ratings would be of interest?

Tim
Unit is AutomationDirect GS2-53P0, there isn't much more in the manual, just the 5kA @600VAC short circuit withstand rating.

I was contemplating using my DS1054 and a B&K PR2000B high voltage probe, setting up a pass/fail envelope test and letting it run.

Can this be done safely or am I suicidal for even thinking of doing this?
Any other economical ways of measuring/logging this?

No, don't use this kind of probe for this kind of circuit. You'd want a differential probe rated at 600Vrms CAT IV here.

Quote
Any other economical ways of measuring/logging this?
If you know what you're doing, one could build a series connected resistor string to safely divide the voltage down to the scope's input level, and use two channels in subtract mode.

Otherwise, no.

Your symptom description clearly points to line overvoltage / surge, but I've also seen input rectifiers been destroyed by (otherwise mostly harmless) bursts. Once a single rectifier diode fails, the vast amount of available energy at the input takes care of blowing the stuff up.

One reason for surge / burst might me other consumers (the 30hp pumps you mentioned) connected to the same circuit. Not while they're running, but rather at turn-on or turn-off.

Any additional way to protect this from these possible bursts?

I'm going to be doing an autopsy on the failed unit, try to find out exactly what the sequence of events was.

As for logging the line voltages, differential probes aren't cheap, I may just buy something like a Fluke 376FC and log a phase at a time.

Will keep you posted, Thanks for the help
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #7 on: July 04, 2019, 09:00:56 pm »
I've occasionally worked on 600 V 3P and found it's very unforgiving - when something goes wrong, it really goes wrong!

Is the site a standard vanilla commercial system or are there any special considerations?  Have any other pieces of equipment had trouble there?  Are there any big regenerating drives that dump power back on the line?  Do you get much lightning there?

Have you checked line to ground voltages as well?  I've encounted one system where a 600-208 3P transformer was used backwards to develop 600 V 3P for a particular machine.  The 600 V secondary is delta so it has no ground reference. If two or more machines are using this circuit, a line to ground short causes the circuit to be re-referenced and the two remaining phases are now 600 V to ground instead of 347V.  If there are MOVs or transorbs in the VFD or in the cabinet, they may not be able to handle this and expire, usually spectacularly.

Are there any MOVs or transorbs from any of the lines to ground?  They're generally OK on the input side (see above) but should never be used on the motor side.  One good transient on the line side forces the VFD to be a middleman and has to take up the voltage difference, something it's not designed to do.

Is the VFD wired and grounded correctly as per the manufacturer's specs? That is, the case grounded but everything else floating?

Is the motor OK? No leakage to ground?
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #8 on: July 04, 2019, 10:52:05 pm »
Yep.  We let the smoke out.

Opened up the VFD and found it uses a single module for both rectification and output, and the Rectifier side is blown all to hell. 

Additionally, on the circuit board, the trace on the MOV between L1 and L2 has been completely destroyed (TVR14112, 1100V varistor voltage)

Now I just have to find the source of the spike/surge.  there are several very large refrigeration units that i'm eyeing suspiciously....

Pics of carnage attached, blown trace circled
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #9 on: July 04, 2019, 11:02:53 pm »
Where is the line reactor again? How many mH?
I've seen locally generated transients (downstream of the reactor) from other gear on the same feeder cause problems, where it was nothing incoming.
Refrigeration compressors are the worst for making transients due to the back kick of the pump on an induction motor with a collapsing field.
This VFD might be acting as the surge suppressor for everything, so taking it out of service might cause other gear to fail or new problems.
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #10 on: July 04, 2019, 11:16:13 pm »
Line reactor is 6 inches above VFD.   Rated for 3.9A, 3%, 600V, 6.74mH.  Unit specified by VFD manufacturer
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline sibeen

  • Frequent Contributor
  • **
  • Posts: 271
  • Country: au
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #11 on: July 05, 2019, 12:33:19 am »
I'd be looking to hire a product that's dedicated to this sort of analysis. Dranetz, Hioki etc have units that do all the general electrical specs as well as capture transient over-voltages.
 

Offline AlfBaz

  • Super Contributor
  • ***
  • Posts: 2187
  • Country: au
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #12 on: July 05, 2019, 01:47:36 am »
Can the VFD handle the motors regen? You said it's a conveyor motor, does it have a contactor on the vfd's output, input or both? Perhaps your rundown params are too aggressive or the motor's model parameters are incorrect
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #13 on: July 05, 2019, 04:51:25 pm »
Can the VFD handle the motors regen? You said it's a conveyor motor, does it have a contactor on the vfd's output, input or both? Perhaps your rundown params are too aggressive or the motor's model parameters are incorrect
VFD is programmed to let motor coast to stop, belt moves very slowly, 10meters/min max, so no inertia to speak of.  Parameters have been checked repeatedly.

We're hunting for an Industrial Electrician who would have the necessary equipment and are considering installing a SPD such as this: https://www.intermatic.com/en/surge-protection/hard-wired-surge/ag6503l3 in the panel as additional protection.
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7678
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #14 on: July 05, 2019, 07:08:13 pm »
Panel pic for all from your other post, but don't see the line reactor. It helps us to see what the panel looks like.

It's best to clamp surges at the source but we don't know what is generating them. I would rent some equipment to find out, it's cheaper than guessing.
Braking can kick up the DC bus voltage but is that where the 1100V MOV that blew is?
The SPD is just three MOV's inside, it clamps at 2,500V line-line which may be too high.
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #15 on: July 05, 2019, 07:16:58 pm »
Old picture, line reactor has been added, its located  between breaker and VFD.  I dont have current pic and machine is 3hrs away.   The MOV in question is located between L1 and L2 on input of VFD.
I am not braking the motor, letting it coast.

I have kicked this up the command chain to get someone with the proper gear / knowledge to solve this as I'm just a Electronics Engineering Technologist, not an Electrician.  Nonetheless, my boss wants answers from me, so here I am.
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2851
  • Country: us
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #16 on: July 05, 2019, 07:21:29 pm »
Power comes from electrical room through a disconnect, circuit breaker, and line reactor to the VFD.
If the VFD suddenly stops pulling current, could the line reactor be giving the VFD input an inductive kick?
YES, this is very possible.  But, it is not just when it stops drawing power, it is EVERY CYCLE of the line, it draws current for a moment on each phase, then stops as the sine wave begins to decrease.  As soon as the rectifier stops conducting, there will be a spike from the inductor.  You need some kind of circuit to absorb that spike to protect the rectifiers.

Is this model supposed to be run with a line reactor?  Does the manufacturer specify the inductance for a line reactor for this size of VFD?  Is the line reactor within that spec?

Presumably, for a 600 VAC VFD, they must be using 1200 V (or higher) rectifiers, so this must be a pretty bad spike to take out the rectifier section.

Jon
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #17 on: July 05, 2019, 07:26:17 pm »
Power comes from electrical room through a disconnect, circuit breaker, and line reactor to the VFD.
If the VFD suddenly stops pulling current, could the line reactor be giving the VFD input an inductive kick?
YES, this is very possible.  But, it is not just when it stops drawing power, it is EVERY CYCLE of the line, it draws current for a moment on each phase, then stops as the sine wave begins to decrease.  As soon as the rectifier stops conducting, there will be a spike from the inductor.  You need some kind of circuit to absorb that spike to protect the rectifiers.

Is this model supposed to be run with a line reactor?  Does the manufacturer specify the inductance for a line reactor for this size of VFD?  Is the line reactor within that spec?

Presumably, for a 600 VAC VFD, they must be using 1200 V (or higher) rectifiers, so this must be a pretty bad spike to take out the rectifier section.

Jon
Line reactor is 6 inches above VFD.   Rated for 3.9A, 3%, 600V, 6.74mH.  Unit specified by VFD manufacturer
Line Reactor was added after first failure in an attempt to solve the problem.
« Last Edit: July 05, 2019, 07:28:45 pm by Sterno »
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2851
  • Country: us
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #18 on: July 05, 2019, 07:31:34 pm »
Can the VFD handle the motors regen? You said it's a conveyor motor, does it have a contactor on the vfd's output, input or both? Perhaps your rundown params are too aggressive or the motor's model parameters are incorrect
Good point!  Conveyors are a special problem, as they have VAST inertia.  If asked to stop, the load on the conveyors can keep it moving for a long time.  If the VFD is set for dynamic braking, then it needs an EXTERNAL braking resistor to handle the dissipated energy.  It may also need to have the braking time adjusted so that the regenerated energy is kept in check.

What you can do is connect a DVM rated for the voltage (going to be close to 1000 V!!) to the braking resistor and then stand well clear while starting the conveyor and then stopping it.  You will see the DC link voltage in bursts across the braking resistor during the deceleration.  (Some VFDs also allow you to program the panel display to show DC link voltage - you can use this instead.)  You want to keep the DC link voltage well below the maximum for the VFD, to protect the semiconductors.  Actually, the VFD should fault out with an OV fault code if this happens, but maybe the micorcontroller is not programmed to be conservative.

Jon
 

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2851
  • Country: us
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #19 on: July 05, 2019, 07:36:12 pm »
Line Reactor was added after first failure in an attempt to solve the problem.
OK, has the unit blown again since the reactor was added?
I'd still try to do the DC link voltage reading while the conveyor is decelerating.  This is easy if available on VFD display.
Otherwise, it is tricky and a bit dangerous to get to this with a DVM, but many drives do have these available on the screw terminals inside. 

I'm still a bit surprised to see the VFD with NO EXTERNAL braking resistor in a conveyor application.

Jon
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #20 on: July 05, 2019, 08:09:10 pm »
VFD does not brake, it is set to coast.
Gear reduction on conveyor is 780:1  Conveyor moves at 20 feet/min max.  Conveyor uses plastic links on a sliding surface, pretty high friction, conveyor stops very quickly.

Unit has been working fine with line reactor, that said, refrigeration tech hasn't been back to activate his end.

I have kicked this up the command chain to get someone with the proper gear / knowledge to solve this as I'm just a Electronics Engineering Technologist, not an Electrician.
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 9320
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #21 on: July 05, 2019, 08:55:19 pm »
Is a special order 600V VFD and motor really cheaper than an off the shelf 208V VFD and motor plus a step down transformer?
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline SternoTopic starter

  • Contributor
  • Posts: 20
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #22 on: July 05, 2019, 09:12:02 pm »
Is a special order 600V VFD and motor really cheaper than an off the shelf 208V VFD and motor plus a step down transformer?

2 kVA Step-down transformer costs more than the 600V VFD, it would likely increase total cost by $500. I have considered switching to that but it wouldn't address the root issue of the surge.

I should note that this isn't the first one of these we've built.  Five out in the wild as of this year, this is the only one to give us any electrical trouble.  Dozens of other small panels using same vendors VFDs, they've been trouble-free.

As stated above, now it's the Electrician's turn to find the surge.
Electronic Engineering Technologist, Graybeard
Mostly Industrial Controls; PLCs, VFDs along some microcontroller and analog stuff.
 

Offline duak

  • Super Contributor
  • ***
  • Posts: 1048
  • Country: ca
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #23 on: July 06, 2019, 03:09:48 am »
This is a low probability conjecture given that the failure looks like it was caused by a transient because it took out the traces to the MOV:

Suppose one of the AC line phases wasn't solidly connected into and out of the MCC or into the VFD or the MCC has a bad contact?  The VFD is basically then single-phasing with all the input power being handled by only two phases, ie., four of the six diodes in the module.  575 V VFDs are rarely rated for full power with single phase input so their input rectifiers aren't rated for the higher current.  Because the L1 and L2 traces are damaged maybe the rectifiers on those two phases overloaded, shorted out and then opened or tripped the MCC.  I'd use a clamp on meter to measure each of the line currents when the VFD is running at load.

I remember an MCC with a screw terminal that could be torqued down but the wire could be pulled out.  I think the setscrew had a damaged thread that didn't allow it to fully clamp the wire so it was a high resistance connection.  This would be the kind of subtle failure cause that would eat VFDs until discovered.

Most 3 phase only VFDs must be derated to 60% of full load when operated on single phase.  How loaded was the VFD in this case?  Since this is a conveyor load, there is a friction load at all speeds and during start up there is the inertial load to accelerate.

Also, when I look at the picture of the enclosure, I see the two modules to the left of the VFD with short wires to ground, whereas the VFD's ground wire looks like it heads off somewhere in the harness.  The GS2 manual calls for a resistance to ground of less than 0.1 ohms. (p.2-11, upper leftish)

When I look at the image of the module, I think see that at least one output phase has also been incinerated.  This makes me think that a transient passed thru the VFD and was ultimately returned to ground in the motor.  I would check the motor, motor cable and all connections for leakage to ground.  I had one CNC connector have its insert's plastic insulator burn between the phase connections because coolant had seeped in.  If the motor is old or is not rated for inverter duty this might be a consideration.

Best o' luck
« Last Edit: July 08, 2019, 04:47:33 am by duak »
 

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2851
  • Country: us
Re: Looking for transients in 600V 3 phase system, need opinions
« Reply #24 on: July 06, 2019, 03:27:06 am »
I should note that this isn't the first one of these we've built.  Five out in the wild as of this year, this is the only one to give us any electrical trouble.
Well, how many of these VFDs have blown?  Pretty crazy that you've built 5 units that were just fine, now they are blowing up!  Is there anything NEW in the plant that could be causing surges?  Might be some other gear in the plant that is using intermittent bursts of current and causing utility tap changers to step up the mains, then when the burst ends, you get high voltage for a few seconds.  It could, of course, even be some other plant on the same feeder.

Jon
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf