Centfugal pump application

[Feg]

Centfugal pump application, the motors are 250kw 355 frame 4pl 420 amp B3 foot mounted with flex coupling. The motors are powered by VSD and are supposed to run at 48hz (according to pump co). Problem here is that 4 motors have failed. The original TECO motors were running for approx 1 year, the surface temp of the motor was 72 degs in the middle with the DE bearing around the same. The motors failed and the rewind shop said "winding condition poor due to motor operating at elevated temperatures over long periods.resulted in motor phase to phase insulation and slot insulation becoming brittle(ie.thermal ageing)".

The motors were changed for a different brand (Dutchi), one has just failed again, all 3 coils are down to earth. They are not black but a dark brown. This motor lasted just a week.

It was noticed then that the motors were drawing nameplate current at 48Hz and the opperator was over speeding the motor when he needed it and was drawing at least 10% over nameplate current maybe more.

I am wondering why the VSD did not protect it, as it is set up at FLC, with 110% overload. This overload can be 1 in 10 minutes according to the VSD maker. My feeling is the motors are too small. I would welcome your opinions.

F.E.G

 

[7anoter4]

If the motor is designed according to IEC then IEC 34-12 Standard states as a minimum frequency for unlimited time operation 49 Hz.
You have to check the supply voltage at the motor terminals [shall be not more than 400*1.03=412 V if the frequency drop to 49].
You have to check the current and to limit it to rated [not more]. The ambient air could be more than maximum permitted [usually 40oC] and the position of the motor against a wall or other object which could avoid free cooling air circulation may affect the long term overheating.
If the altitude from sea level is more than 1.5 km the cooling effect of the air is reduced then the current should be reduced accordingly.
It would be better to check the winding temperature and to limit 10-15 degrees under the maximum permissible according to insulation class [A=100oC, B=130.F=155 and H=180oC].

 

[waross]

It was noticed then that the motors were drawing nameplate current at 48Hz and the opperator was over speeding the motor when he needed it and was drawing at least 10% over nameplate current maybe more.

If the operator has a valid reason to over speed, consider a larger motor and possibly a larger VFD.
If the operator DOES NOT have a valid reason to over speed, consider changing the operator, or changing the programming to prevent any over speed.

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

 

[edison123]

Motor rated freqency ? Cooling fan set up ?

 

[Feg]

Sorry, 50Hz. std axial fan on the back of the motor.

 

[jraef]

Wow,
48 out of 50Hz: not a factor, shouldn't need extra cooling.
6 meter lead length: not a factor.
Motor rated for VFD use: motor capability not a factor.
Changed brands of motor: motor quality probably not a factor.

I am leaning towards Waross' assessment. Methinks the term "when he needed it" is being misrepresented to you, probably the "10% over nameplate, maybe more" as well.



"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376

 

[Feg]

The issue i have is why the vsd did not protect the motor, as the thermistors are connected straight back to the vsd.

 

[edison123]

I would also say it's the overload that is frying the windings.

Have the thermistors been checked for proper functioning ? They are notoriously unreliable. I prefer PT 100 RTD's even in random wound stators.

 

[jraef]

Thermistors are not analog protection devices, even though some people promote them as such. They are threshold devices. First off, are you sure your VFD is programmed to read the thermistors? Most come with that function disabled as standard, you have to tell the drive in programming that thermistors are connected and then you have to tell them what the threshold level is. Then are you sure they are wired correctly? Unlike RTDs, thermistors must be wired in series with each other. If you connected them in parallel, the resistance value becomes meaningless.


"If I had eight hours to chop down a tree, I'd spend six sharpening my axe." -- Abraham Lincoln
For the best use of Eng-Tips, please click here -> faq731-376