Predictive maintenance with torque wrenches VS vibration

[MatheusMarkies]

I have a dynamic torque sensor that can measure the torque my electric motor is exerting. I also have a vibration and temperature sensor.
My problem is this: I want to know what the advantages of the torque sensor would be for predictive maintenance. Basically, I'd like to know what kind of problems could be identified by torque analysis that couldn't be identified by vibration?
The first advantage I see would be the fact that the torque wrench can be used for motors that run at low speed. But what are the other advantages?

 

[Andrew ONeill]

MY thoughts
I'm assuming this is actually a torque arm / physical thing, and not a calculated value via a VFD / VSD / VFVV Drive
Things it will probably tell you;

1) Long term efficiency trend, i.e power output vs electrical power input, if the efficacy drops off, could be a sign of some sort of problem
2) Broken shaft, major failure, it'll tell you that pretty quickly
3) you "could" possible use it as a pseudo vibration sensor, but it would need a very high sample rate (at least several sample per rotation of the shaft), not sure if anyone does this / probably a waste of time as you already have a vibration sensor
4) Motor demand side load, to see if the driven machine is having problems (torque sensors are often used to monitor agitators as an example)

and that's about it in my experience



Andrew O'Neill
Specialist Mechanical Engineer
Australia

 

[dgeesaman]

It all comes down to the reasonably expected failure modes.

In my industry, we have customers (and sales reps) who believe vibration can detect anything. It cannot.

Would the torque sensor be expected to read back excess torque (jammed / seized drive)? Insufficient torque (broken coupling?)?

Need to know a lot more context to comment further.

 

[MatheusMarkies]

Well, I'm creating a document in which I list some of the advantages of torque analysis.

My aim is to complement vibration analysis with torque and also to prove how relevant a torque wrench would be for analyzing the performance of a machine.

https://drive.google.com/file/d/1hktKzHZtIizmE8upr8AaYxzSmB0q1mmP/view?usp=drive_link

 

[IRstuff]

Having recently experienced a fan motor failure, I would think the torque sensor would have been able to see the imminent bearing failure because they started seizing from the degradation of the lubricant over time.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[Gr8blu]

If the torque measurement device truly is a "torque wrench", it won't be able to do anything except indicate which engineered joints need tightening - and only after the device is applied to a component of each joint with an attempt to "tighten". If the device is a load cell, it will measure torque as a linear force applied in a plane - which means it will indicate deviation in the magnetic and/or electrical field strength but will be unable (in most instances) to tell you whether the culprit is the rotating or stationary part. A load cell that was set up correctly might also indicate a stall condition. The last thing a properly set up loadcell type torque device might tell you would be a "loss of torque" condition, which would equate to an open circuit condition in one of the windings.




Converting energy to motion for more than half a century

 

[MintJulep]

Basically, I'd like to know what kind of problems could be identified by torque analysis that couldn't be identified by vibration?

Then start with a list of problems that could be detected by any of the methods (vibration, temperature, torque).

Ask yourself how using the three methods in combination might improve the detection vs. any single method.

 

[mfgenggear]

Would this be static torque vs dynamic torque.
And why the two don't mix.
Especially when there is high torque with high rpm. And could get dangerous. Or is this off track..
Vibration in my world of gears, bearings and shaft is a very bad issue.
A study of initial vibration and vibration during many cycle. Can detect a failure. Add the noise factor. With bearings about to fail. Or are failing. Torque I believe would be to late.

 

[Orange_kun]

We use in-line torque sensors between motor and the device under test. It gives us characterization data for the device, and also serves as an e-stop triggering sensor. The devices vary in terms of vibration levels and required torque input, but any changes in torque once things are running is a great indicator that something is going wrong inside.