Medium Voltage SMC-Flex Soft Starter - Pulsating Motor 1

[PLCTech105]

All,

We have 400A medium voltage (2,300V) SMC-Flex soft starter on a 330 A Motor.
About 1 year ago we noticed a significant pulsating at a consistent frequency (approx. once per second)... never faster or slower only louder or softer.
When the load is removed and the motor is uncoupled the pulsating goes away.
The cables from the SMC to the motor were verified with a Hi-Pot test.
We had a power quality study done.
- Voltage varies from Leg C at 2,354V to Leg A at 2,462
- Harmonics are almost non-existent
- Phasor on leg B was 2 degrees off
- Amperage has a swing of 10 amps while pulsing

The process has no identified issues. The pump impeller is good and new, suction, and discharge are all working as they should be.
The outlet flow matches with the pump curve for amperage and PSI.

The motor is physically moving when examined with a strobe gun at the same pulse rate and gets hotter the more flow increases by opening the valving down stream.

The motor was removed and replaced 2 times and immediately the pulsating returns so we know the motor itself is not the issue. This unit has been in stalled for about 15 years and until last year this had never occurred.

At this point we are not even sure what we could check next. Has anyone seen an issue like this?!?

 

[waross]

What has been changed in the plant.
Look especially at anything that someone says;
"That can't be the cause.
A WAG, (Wild ASSet Guess).
Look at reflected waves.
The simplest case, is a long straight discharge line from the pump, leading to a discontinuity. Most often a 90 degree elbow.
Each impeller blade passing the discharge sends a pressure pulse down the discharge line.
Part of that pressure pulse is reflected back.
When the transit time is such that the reflected wave arrives as an impeller blade is passing the discharge the pressure waves will tend to add.
Each iteration adds a little more until the pressure breaks something.

Changed piping?
Added valve or changed.
Suction head or discharge head changed?
Has the impeller been changed?
If the new impeller delivers only a few GPM more or less than the old one, it may have changed the loading on the motor slightly so that the rotational speed has moved into the forbidden zone.

How to check?
Do you have any way to throttle the flow a little.
A little less flow should let the motor turn a little faster and hopefully get it away from a forbidden speed.
[Anecdote Alert]
The first time I encountered this it was a 60 HP pump with a long straight discharge line.
The pumping rate was speed controlled by a VFD.
The PID controller inadvertently ran the pump at a forbidden frequency.
The reflected pressure grew until a piece of casting about the size of a large mas hand was blown out of the pump scroll case.
The first guess was a faulty casting and the pump was replaced.
A week or so later, it happened again.
This time it was correctly diagnosed as a forbidden frequency.
The forbidden frequency was programmed into the VFD and there were no more issues.[/AA]
The moral?
Don't underestimate the power of reflected waves.
In defence this was a long time ago and VFDs being PID controlled was a first.
This was a very first part of the learning curve of VFD applications.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[FacEngrPE]

Do you "run" in bypass?
When in bypass is the pulsing present?
This question is based on what today's SMC Flex starters do today. 15 years ago available features might not have included this combination.

https://www.rockwellautomation.com/...-or-not-to-bypass--is-that-the-question-.html

 

[PLCTech105]

waross,

Thank you for the detailed response.
We are currently looking into the impeller as it was changed and the size increased by the OEM to increase flow. During our next outage we are planning to return the impeller to the previous size.
There were no other changes to the system during this time frame.
Since it is on a soft start and not a VFD we cannot change the speed but we have varied the flow using valves on the outlet.
The only change that we see doing this method is the intensity of the pulsating but no change to the frequency.

Regards,

If you don't have time to DO IT RIGHT,
you must have the time to DO IT OVER!

 

[PLCTech105]

FacEngPE,

The SMC portion of this system was upgraded in 2015 so it's newer.
We do run in bypass once the motor is up to speed. This soft start has two large vacuum contactors. With starting and running coils for each.
Once the motor starts the Bypass contactor closes putting the system on full line power.

We have had the Manufacturer out to PM the system but we are still not sure if it is working as designed.

I have attached the drawings if this helps.

Regards,

If you don't have time to DO IT RIGHT,
you must have the time to DO IT OVER!

 

[waross]

Since it is on a soft start and not a VFD we cannot change the speed but we have varied the flow using valves on the outlet.
The only change that we see doing this method is the intensity of the pulsating but no change to the frequency.

Changing the flow chnges the speed slightly, but in your case not enough.
I think that this is a strong indication of a forbidden frequency but not 100% conclusive.
To check for forbidden frequency with calculations:
You need to know;
The speed of a pressure wave in the fluid.
The time for a pressure wave to reach the first discontinuity, usually a 90 degree elbow or a control valve, and return to the pump. Call it Tr.
The time for each impeller blade to pass a given point. Call it Ti.
If Tr is a multiple of Ti then a forbidden frequency is probable.
Given the cost of an impeller that size and the time to change it, suggests that there were significant production advantages motivating the impeller change.
With that in mind, you may want to consider a piping change rather than an impeller change so that you may retain the added flow rate produced by the newer impeller.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jraef]

Once the bypass contactor closes, the soft start portion of this Soft Starter is no longer involved in any way… UNLESS there is a problem in the motor circuit.

If the motor circuit current begins to exceed the rating of the bypass contactor, the controls will switch BACK to the SCRs in order to current limit, then when the current drops, it will re-close the bypass contactor. Vacuum contactors only move about 1/8 inch, so they don’t make a lot of noise with the door closed, so you may not hear it happening. But that could account for the pulsing. The current rises, the bypass contactor opens, the SCRs current limit, the current drops, the bypass closes again and the current rises again. But because that current rise is so low and slow, it is not being detected as a Short Circuit by the controller. That gives rise to the theory below.

So then the question becomes, why is your motor current surging? You already replaced the motor so that’s not it, but if the motor leads are “leaking”, that could do it. You say that you performed a hi-pot test, but how was that done? A typical hi-pot test just applies a voltage and insures that the insulation dielectric is intact “enough”. But you should be able to also perform what’s called a “tip-up” test where you monitor the rise in voltage AND current with the tester to see that they increase in synch. If the current rises faster than the voltage, you have a small leak. The 10A might be the result of your sampling rate of your meter, but being that the contactor is rated 400A and the motor FLA is already 330A, it wouldn’t have to increase much to exceed the rating of the contactor, meaning the “leak” might only be a few volts, so it might fall into an error margin as “nominal”. That’s why the tip-up test might show it instead.

If you have a way to monitor the motor voltage during these pulsations, you would see the voltage dropping when the SCRs go into current limit. I don’t think that the old SMC-Flex monitored the output voltage, only the line voltage, but I will check for you later today when I am back in my office.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[jraef]

Follow up: The SMC only looks at line voltages. So to see this phenomenon, you would have to have something else connected to the load side.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[PLCTech105]

We have a power quality analyzer connected to the Load side of the SMC and it has been recording data for about 3 weeks.

We have not seen any voltage issues such as described that correspond with the pulsations of the motor.

I am attaching a snapshot of the voltage readings that shows all three phases are about 40 volts off from each other. This is quite odd that they are so perfectly spaced and I was wondering if anyone has seen this before and what it may indicate. The voltage readings on the line side all appear to be within 1-2 volts. I understand that the bypass should be line voltage but is the SMC causing this?

Regards,

If you don't have the time to DO IT RIGHT,
you must have the time to DO IT OVER!

 

[jraef]

Again, once the bypass contactor is closed, this is no different than any Across-the-Line (DOL) starter, so it is not "doing" anything to the power circuit. All it is doing at that point is monitoring it.

That is a very ugly waveform, you have something else going on here...


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[PLCTech105]

jraef,

I completely again on both accounts. The waveform is something very abnormal. We have several 3rd parties trying to identify this issue.

I also agree that the bypass should be straight line power but at this point we are open to any possibility. We just completed a pump study from another company who used equipment to read flow, speed, amperage, and vibration to create a detailed analysis on the pumps operational status and they found nothing abnormal there either.

This is a very troubling issue ... this is our 3rd motor rebuild in the last year and if this condition remains this motor will be replaced within the next month or two as well.


If you don't have the time to DO IT RIGHT,
you must have the time to DO IT OVER!

 

[PLCTech105]

jraef,

One more comment on the voltage issue in regards to the SMC circuit.
We have our own on site power generation turbines and we supplement power from Consumers Energy.

This issue is off one branch of the Power House and none of our other substation equipment is having this issue. As far as we can tell at this point this is the only motor/equipment effected on site.

On this branch we step down from 13.8KV to 2.4KV then there are two additional step down transformers from the 2.4KV to 480V for other equipment. The 480V equipment has no indications or power quality issues that can be seen. Only the One 1500HP 2,300V motor is affected and has this strange voltage issue.

I appreciate everyone's input as we try to resolve this issue.

Regards,





If you don't have the time to DO IT RIGHT,
you must have the time to DO IT OVER!

 

[waross]

Off the wall,but this may be worth consideration,if only to eliminate it as a possible cause.
1. Trouble shooting 101; What has changed?

The impeller was changed.​

2. Forbidden frequencies are a known issue with large pumps.

Is it possible that the new impeller is working at the edge of a forbidden frequency?
If an induced reflected wave condition was close to a multiple (or reciprocal) of the supply frequency, could the pulsing be a beat frequency?
Has the discharge pressure been checked with an instrument with a fast enough response to identify reflected waves?
What does the piping look like?
Is there a long straight section from the pump discharge?

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[Heaviside1925]

I am leaning towards waross approach. Motor detached - fine, power (though not pretty) - is not indicating a cause, this leads to some dynamic event causing this issue. As was noted, "what's changed?", the impeller, Occam's razor would point to the impeller, or a new dynamic introduced by it.
I would think this frequency should be showing itself or its harmonics somewhere. If the power has been analyzed, I assume an FFT was performed and if the pulsing is not showing itself in the power, then would contraindicate the cause be electrical or electromechanical. So, back to the impeller. It was changed to increase flow. Increasing flow would increase piping velocities as well as kinetic energy of the entire system. This could be a myriad of possible causes from pipe supports, to valves, to inline instrumentation, flow conditioners, pipe geometry changes.

Has the discharge pressure been checked with an instrument with a fast enough response to identify reflected waves?

This should provide relevant information and should be fairly easy to do.
Is the pulsing evident throughout the system or only at the pump?
If it's only at the pump, my first thought would be motor and pump mounts, base and foundation. I would first see if a dynamic analysis was performed when the mounting and foundation system were designed and what static and dynamic parameters were used. You may end up needing to beef up the foundation and/or install an engineered mounting system to dampen the frequency.

 

[waross]

The motor is physically moving when examined with a strobe gun at the same pulse rate

Reflected waves will do that also.
You may have to use various rated rupture disks to detect the reflected wave pressure transients.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jraef]

It’s a soft starter, not a VFD. A soft starter DOES NOT change the frequency. The frequency remains at whatever the line frequency is.

It might be load related, but again, if this is taking place AFTER the bypass contactor closes, then this is NOT A SOFT STARTER ISSUE!


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden

 

[waross]

Have you calculated the period for a sound wave to reach and return from the first elbow or valve, and compared that to the period for an impeller blade to pass the discharge outlet?
Look for a match or close ratio.


--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[PLCTech105]

All,

Thank you everyone who has participated in this thread.

I have learned so much through this process and all of your feedback!

We have been able to find the root cause of this issue and get the items resolved.

First, we found that one of the SCR's was letting through current and affecting the system. That was replaced and resolved.
Second, the voltage board on the SMC Flex was affecting the voltage to the motor causing as much as 80 volts between two phases (2400V A phase and 2480V C Phase). This board was replaced and this resolved the voltage issue.

At this point the motor was still reacting the same, with a constant pulse although at a reduced intensity and reduced load and temperature.
The motor shop that was rebuilding our spare was putting the motor through stress testing trying to break it and see what the issue was. That were using an across the line starter and hitting the motor with full voltage.

What they found when the motor finally showed the failure point was a manufacturing failure with the rotor bar shorting ring. This manufacturer had a very shallow trough and poorly brazed connections. The joints were separating within the shorting ring internally and this was not seen with the inspection process. The new rings were machined by our shop and done to todays standard. This resolved the issue.

Kind regards,

If you don't have the time to DO IT RIGHT,
you must have the time to DO IT OVER!