Torque Limit

[Guardiano]

Hi,

We had a breakage on a shaft and we are now having an issue with our insurance company. The SWL of the shaft was 2400 kNm with a safety margin of 2. The load is a stone crusher which is operated through a gearbox and the output shaft drives the load. The nominal torque on the low speed side of the gearbox is 2200 kNm and the torque limit on the VSD was set to 2000 k Nm. The VSD was set to operate on a constant torque curve and the mode of operation was the direct torque control one. The insurance company is claiming that the VSD could have given 250-300 % of the maximum torque during the starting phase of the drive, particularly during seizing period, and this was the cause of the breakage. They have difficulties to accept that the torque limit set on the drive would prevent the output torque to raise above 2000 k Nm. How accurate is the torque accuracy on a VSD ? I know that DTC technology claims the best accuracy amongst open loop drives.

Thanks.

Guardiano

 

[jraef]

It's very accurate. But in DTC and SVC drives, everything is based on a motor model comparison against current measurements, so at the very instant you start, there is nothing to compare to. It will usually catch up within a few milliseconds but if you had a situation where the drive was stopped and restarted immediately before the motor magnetic field collapses, it can possibly create a huge torque spike. That by the way is the case regardless of having a VFD in place or not and in fact the VFD helps, but it is not immune. This is one of the best arguments for 3 wire control, or at least some sort of de-bounce in the command circuit.

Also in a lot of cases we program a delay on tripping to allow for starting. Did you check the programming?

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[jraef]

Thirdly, you mention a crusher. What type? I have seen, on jaw crushers, a situation where users attempt to clear a jam by purposefully starting and stopping the motor quickly in order to "rock" the jaw back and forth to increase the crushing pressure, usually with total disregard for the mechanical and electrical consequences, because they don't want to get out of their comfy air conditioned cab to clear it manually. If Joe Bubba operator figured out the torque limit in the drive, he likely took it as a personal challenge to find the way to get around it. And did!

"Dear future generations: Please accept our apologies. We were rolling drunk on petroleum."
— Kilgore Trout (via Kurt Vonnegut)
For the best use of Eng-Tips, please click here -> faq731-376

 

[desertfox]

Hi Guardiano

Any chance of a picture of the broken shaft?
Just looking at the failed shaft can tell us a lot about the failure, also how long as the motor been in service?

desertfox

 

[Guardiano]

Thanks for the replies. Ok, jraef I understand that the drive do modelling of the motor, what bother me is the following. During initial starting, would the torque limit work to limit the torque to the set value or could the shaft be subjected to transient torque above the set value? How reliable is the DTC to achieve this performance?

Thanks.

Guardiano