Clutch Slippage detection

[data74]

Hi Guys

I've got a machine with 3 * 37kW motors driven from one VSD driving one gear wheel, Each motor has an Individual MCB and an AB E3 O/L for protection, also before the gearbox is a fluid filled clutch. We tried to detect clutch slippage using proxies but they were not quick enough to detect an speed changes so we decided to compare currents from the E3....still this is not good enough anyone with ideas on how to this in code will be much appreciated.

 

[catserveng]

For industrial engines driving loads thru a right angle gear box with a clutch I use magetic pickups with 4-20mA transducers into the PLC, and have the PLC logic do a comparison of the speeds, usuually enable and disable with the clutch engagement signal. Has worked fine in that application. Last system was with an 1800 rpm gas engine driving thru a Philadelphia gear with a Western clutch, we used two Altronic digital tachometers with 4-20mA outputs going to an analog input card on a ControlLogix PLC based system. Hardest part was coming up with a chopper wheel for the right angle drive, but the gear vendor had a solution.

Hope that helps, Mike L.

 

[data74]

Unfortunatetly i can't install any hardware i.e tachometer/ encoder. The only option i'm left with is maniplating the code to be able to pick up the difference in the Individual line currents......I was thinking of using an S-curve or something like that??

 

[Skogsgurra]

If input and output shafts are accessible, you can use small magnets and (debounced) reed relay elements. Measure time between pulses with a timer or a recorder/oscilloscope.
I use this technique to measure slip in induction motors, belt transmissions and such things. Works very well.

Other technique is to put 3M reflex tape on the shafts and use a hand-held tachometer to measure the speeds.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[jraef]

Add the EC5 voltage transducer add-on module to the E3's (assuming they are E3+ units). That will allow you to monitor kW, not just current, which will be a direct reflection of shaft power and much more accurate. Then look for any precipitous change in shaft power. You can actually pre-program the E3+ relays using DeviceLogix to trigger an output based on that rate of change, leading or lagging, which might be faster than trying to communicate everything up to the PLC or PC and then doing the calcs.

"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

 

[data74]

Thanks for that Guys. Here is some more info open to discussion.
1. E3's are not accurate for frequencies less than 20HZ
2. Idea of using proxy switches has failed, they are too slow to pick up the speed differences
3. We still need to confirm the calibration of the curent measurement....compare to a known source
4. Try to use the S-Curve to monitor rate of change in the AB PF700 VFD
5. Use torque sensor for Torque monitoring eg (FUTEK)

more comments are welcome.

 

[Skogsgurra]

Did you actually try the proxy switch arrangement? If you are running 20 Hz, they will not at all be too slow. Works even at 60 Hz. There may be a delay, but that delay is constant and the front edge to front edge time reflects speed with good accuracy.

I have used this technique to measure slips below 0.1 percent.

Did you really try it?

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[data74]

Yes you are right, the proxies will pick up the Lower speeds better but at higher speeeds there were some significant discrepecies. Will send details of the original arrangement with proxies on each side of the clutch.

 

[WTRVS]

Peprl & Fusch (German) has a good solution for this problem:

http://www.pepperl-fuchs.com/global/en/13577.htm

This is a slip monitoring relay that uses normal inductive sensors to detect slip. This relay has alot of parameters that can be changed to make sure that it meets your requirements. It's easy to setup and it has digital outputs that can be read by a PLC. It can handle high rpm and the sensitivity can be adjusted easily.

Kind regards

 

[data74]

Thanks for the Info Forumees

 

[data74]

Why is it better to measure KW and not amps when comparing to similar motors driving the same load?
P= V*I*cos(phi) I thought that both would pick up the imbalances.

 

[Skogsgurra]

data74

Simply put, because cos(phi) changes a lot more than current when load changes. The reason, simply put again, is that the magnetizing current is more or less constant and masks the change in active, torque-producing) current.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.