how to measure belt slippage 1

[JIMGEN]

Does anyone know of a simple and relative inexpensive device to measure belt slippage between the driver and the driven ( v belts and pulleys)? I want to have an alarm or fault shutdown if the slippage increases to a preset limit. Generally we are dealing with 5V and 8V sheaves and usually 6-12 belts and the power transmitted is 200hp to 1000hp. It seems a manufacturer would provide such a device.

Thanks for any advice or comments.

 

[MintJulep]

Two tachometers and a bit of math.

 

[GregLocock]

That's the best way, but if you want to throw money at it you could use a laser doppler velocity gun, or if you want to start trying to infer results then measure belt temperature.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[IRstuff]

You could do an in-place measurement concept, like adding Hall sensors or encoders/resolvers run into a PIC or microcontroller or PC and monitor in real time.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[berkshire]

A piece of chalk and a stop watch. Presuming you are using matched set belts.
An odball belt and all bets are off.
B.E.

 

[amorrison]

Mark pulley and belts.
Get a strobe flash that can be activated by one of the markers - this will work during a full load test -slipage might only occur at full load.
Can a cell phone with an "app" be made to do this ? - along with complete recording video.

 

[Jboggs]

Or just switch to timing belts and be done with it.

 

[Tmoose]

Is the driver an induction motor, and not on a VFD? So the input speed doesn't vary much more than a couple of percent?
Then simply easuring driven speed would be pretty close.

Pick your desired output.

http://www.monarchinstrument.com/

Lots of choices, many < $500

 

[electricpete]

You may be able to infer speed from vibration sensors on each machine, also.

=====================================
(2B)+(2B)' ?

 

[ScottyUK]

Inductive pickup on each pulley to detect a flat or notch. You might even be able to use the shaft keyways if you're careful and there's a shaft protrusion. Connect each pickup to a pulse counter to obtain revs per second, multiply that value by 60 to get rpm. If you're really stuck for money use one counter and put a plug & socket on it.

 

[Compositepro]

This instrument will do everything you want, including alarm at a programmed slip.

http://www.redlion.net/products/groups/preset/gemini1&2/docs/gem1&2_manual.pdf

You will need to install a tach/encoder/mag pick-up on each shaft (available from this same company).

 

[dhengr]

Jimgen:
If your pants are down around your knees, you or either a teenager or your belt is obviously slipping. Otherwise, beer gut or up under your arm pits it’s a little tough to call. Suspenders might help.

 

[JIMGEN]

Tmoose,

The drivers are induction motors and is not on a VFD.(One motor has soft start) We have two motors on a common shaft and the end users (overseas) do not want to do proper routine maintenance to check belt tension so then they have burned the belts off one motor( that motor will continue to run) and then the second motor will shut down for over amp. So now the customer complains about burning the belts.

We want to compare the two speeds to determine slippage within predetermined limits for each motor and then have a fault to shutdown both motors if either motor runs with no load. We could take a look at amp draw but the motors sometimes are fully loaded and then within perhaps within even seconds will have no load.

Others have provided direction to suppliers and we will take a look.

Thanks

 

[Compositepro]

Greg, the instrument I suggested is not simply a pulse counter. It is fully programmable with correction factors for each shaft speed so it can display the true speed ratio between shafts or display 1.0000 when the ratio is whatever it is when there is no slip. Belt length is not a factor. It is so versatile that it can get complicated to program, hence the 60 page operation manual.

 

[GregLocock]

Sorry compositepro, that post was in response to an ill thought out post that got deleted (chalk marks on belt and roller)!

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[electricpete]

A traditional predictive maintenance routine could include periodic surveys of:
1 - bearing housing vibration - can find a variety of problems
2 - thermography - as Greg mentioned.
3 - Strobe inspection. Freeze the shaft and inspect. Freeze the belts and inspect. Look for flapping belts. Look for damaged belts.
All of above tools (and qualified people to use them) can be found at any large industrial facility. There are also contractors that specialize in this type of monitoring. It may seem that walking around is a heavy ongoing manpower cost compared to continous monitoring, but balance that against the advantage of having a knowledgeable person gather the data, stand in front of the machine to look/listen/smell/etc for other clues and think about the results.

Also, make sure qualified people are maintaining the machines.
Proper belt tension is of course important (too loose lessens belt life).
When belts are replaced, check the sheave grooves... if they are dished that will rob belt life.
Someone in a hurry may use a screwdriver to pry to belt onto the sheave without having to move the motor... may save some time getting the job done but you'll be back doing the job again very quickly.
I'm sure others can add to the list of "tips". The main point is to have the right guy doing the job, or send him to training, or hire him.

Sorry if all this is a little off topic, obvious, motherhood or whatever. Just seems like it should be mentioned.















=====================================
(2B)+(2B)' ?

 

[Tmoose]

"two motors on a common shaft..."

Both motors driving a common shaft via belts?
Kind of like this, but with belts in place of the gearboxes?

http://www.freepatentsonline.com/6820707-0-large.jpg.

"We want to compare the two speeds.."

So the only speed changes when one motor's belts are gone is one motor at full speed, and the other motor is at something greater than full load slip, and the driven shaft may be proportionally and incrementally slower, unless the surviving belts actually slip some more, too?

I guess rather than try to track rpm of at least one of the belts in each set I'd expect if monitoring both motor amps, which are approximately linearly related to speed in the operating range, it would show when one motor was doing even a little less (>rpm), and the other was doing more (<rpm). And provide that indication somewhat Before one set of belts was completely dead, so that would be the trigger for belt maintenance.

http://docs.engineeringtoolbox.com/documents/651/electric_motor_current_torque.png

 

[JIMGEN]

Yes, two 500hp motors on one shaft.

And when both are running the load can vary from no load to something approaching full load.

If I sense amp draw then I would need to look at the amp draw differential between the two motors. If the belts are gone or slipping excessively on motor A (the motor would be running at or near no load) while motor B would be carrying all of the load (which could be no load, partial load within motor B load rating or most likely
overload on motor B along with the thermals tripping or the breaker tripping.

It would be great to find a module that we could just hook the output of a CT for each motor and have that module control signal a fault to the starters as the control circuits for each motor are in series so if one motor shuts down then the other will also shut down.

Thanks

 

[ScottyUK]

Maybe something like Hydria's HPL-500 power monitoring relay? Have a look on Google. Needs a 3-phase voltage connection and one CT.