What's wrong with this picture? 2

[oldfieldguy]

I have a s****ns CGP 9000 HP 1800 RPM 4160-volt motor in service driving a centrifugal natural gas compressor through a step-up gearbox. The motor is fed by a s****ns VFD.

Last year the motor was taken off-line due to high vibration.

Investigation revealed that the cooling fan, internally mounted, had torn and distorted. It looks amazingly simlar to the picture below. The manufacturer replaced it under warranty.

Now, six months and 4500 hours later, we had another vibration incident and shut the motor down. We opened it up and, you guessed ! Fan failure! Second attachment is this year's failure, in situ.

The first failure was with multiple thousands of hours of run time. Our VFD can do 105% RPM, but we seldom push it that hard. The service is not severe. Loads are steady, and process records at the time of the failures indicated no shocks or unusual conditions on the system.

I am looking for answers. Several people have said it looks like an overspeed issue, but this is an induction motor hard-coupled to a step-up gearbox and a centrifugal compressor, and any overspeed on the motor would have had dire consequences on the compressor. Process records do not show any speed variations at the time of the vibration alarm initiation.





old field guy

 

[oldfieldguy]

Here's the fan from last year's failure. Ignore the rust. It spent a year in the dumpster before we retrieved it for investigation.

At the time of this failure, the fiberglass air diverter inside the motor was shredded. I thought that this failure preceded the fan failure, and that the fan tore up when it ingested fiberglass bits. That diverter WAS held in position by nylon hardware. S****ns replaced the fiberglass, installed the new parts with stainless steel hardware, and replaced the failed fan.

Now I'm thinking that the fan failed first, the elevated vibration was more than the nylon bolts holding the shroud together could stand, and it disintegrated as a consequence of the fan failure.

The manufacturer says they have hundreds, if not thousands of these motors in service with no similar failure. I've had the same motor fail twice.

Did I win the lottery?

old field guy

 

[cranky108]

If you won the lottery, then I must have won it several times over. I suspect the manufacturer, who may be facing several of these failures, will use a delaying tactic to keep you from putting real pressure on them. This is a tactic used by several electrical manufacturners until people started talking on industry forums.

I'm not in much of a position to help as I don't work with motors. But good luck.

 

[dpc]

Obviously, you have seriously abused and mis-used this finely engineered piece of equipment.

It's pretty tough to figure out the root cause sometimes, but it makes sense that the fan failed first. What really annoys me about most of the major equipment manufacturers these days is that they will ALWAYS start out by saying that this has never happened before in the history of the world, so it must be something you did. Eventually after enough fuss is raised and you are accidentally allowed to talk with an engineer who doesn't have a handler present, you learn it is a known issue - but you're still SOL because it is out of warranty. It used to be fairly common to get service advisory when a problem was discovered with a particular product. It's very rare these days, except for a couple of companies. Unfortunately, one of the worst offenders just purchased the WORST offender, IMO, so that bodes really well for the expected level of customer service from the new mega-company.

Sorry for the rant - I guess I really can't help you much, but I feel your pain!

 

[oldfieldguy]

dpc-

Allow me to slide off to the side of this conversation.

When we were going back and forth over this failure back in May of last year, I posited the question about how many and where these 'hundreds' might be and how they track things. He told me, presumably with a straight face, that they had no way to track them.

I told my boss who was sitting in on the conversation that I can spend fifteen thousand dollars on a Toyota that is one of a couple hundred thousand and ten years later if Toyota does a recall, I get a letter in the mail. So S****ns can't keep track of a few hundred motors they sell at a quarter million dollars apiece? Please!!!!!

I have made the promise to do serious harm to the next person within my organization who buys big power equipment with that logo on it.

Enough for the rant.

Solutions? Ideas?

old field guy

 

[jraef]

OFG,
Did anyone inspect the motor bearings and races for evidence of fluting? My thought is, maybe the bearings are being damaged by shaft currents from running on the VFD, which increases the vibration in the motor. The fan may be just the weakest link that fails first.


"Will work for salami"

 

[macmckim]

I have seen similar failures on large VFD duty gas compressors (other brands, not brand S)
in the five cases that I saw - all failures were due to "fan resonances", IE critical frequency in the fan. All motors were driven from VFD. Between the motor/gearbox/compressor, when the motor speed is decreased from its primary speed, a vibration/ critical frequency builds up in the system and fans fail. Overspeed was not an issue with the compressor train.
The solution was for the mtr mfg to build an heavy gage steel fan.

Macmckim

 

[itsmoked]

I think macmckim has nailed it.

That relatively large complex fan could have several resonance points. Your VFD running that motor closed loop with some flow or pressure transducer could easily be visiting the fan's problem speeds. While a heavier gauge fan would change those speed points it won't get rid of them.

You can have the system run at some speed point for years without reaching a harmful speed point then because your end user's demands change the motor finds itself frequently in the resonance zones and hence your lottery number start winning all the time...

Since a heavier built fan of the same style isn't a guarantee, nor something you're likely to get out of a big company (that-knows-it-all) anyway, I'd suggest two solutions:

1) Instrument the machine temporarily to discover the resonance speed(s) and then use the VFD's "forbidden" zones to prevent all further visitations to those speeds.

or

2) Do a complete end run. Remove the imbecilic fan altogether and put a standalone blower on the machine to provide optimized air flow that you could even tie into the motor's thermal sensors to provide exactly what the motor needs instead of the, "this should be more than enough", but resonate fan the maker stuck on there. You might save some energy that would even payback or prevent another costly down time.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

I have been thinking about this and came to the same conclusion as macmckim and 'smoked. Itsmoked expressed my thoughts very well. I like both of his proposed solutions.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[DRWeig]

I'll just throw in that I've seen similar failure mode on a large (100 HP) HVAC fan. I think that normally on those, the motor or fan bearings go first and add to the resonant vibration.

Good one, macmckim. I like itsmoked's suggestion.

Best to you,

Goober Dave

Haven't see the forum policies? Do so now: Forum Policies

 

[oldfieldguy]

Jraef & DRweig-

We haven't taken the motor apart yet this time because we're waiting for S****ns to get back with us on a fix. In the last instance, the bearings were disassembled and no evidence of damage was found.

Because of the VFD and the ability to operate at low RPM where self-cooling might be an issue, we do have an external blower ducted into the motor. It is brought on line at the inception of high temperature as detected by RTD.

My own engineering half-way suggested doing away with the internal fan and relying on the external blower for cooling. I am somewhat reluctant. The external blower is almost never needed under normal operation. If we run it full time with the main drive, then we get to play with what I deemed a poor installation from the standpoint of maintenance of that blower.

I am suggesting to the S****mns engineering bunch that we have a possible critical frequency issue. I will have to get the production folks to pull up some historical data on RPM and load to offer as backup.

thanks to all-

old field guy

 

[Marke]

I have seen a similar failure on large fans that had severe resonance issues which came to light when the VFD was run in torque mode. Switching the VFD to V/Hz mode and the problems disappeared. It apppeard that there was speed instability in the loop and this was causing torque rippple which was exciting the resonance in the fan.

Mark Empson
Advanced Motor Control Ltd