Dynamic stress measurement with strain gauges, without stop the machin

[fer72]

Hi,
I want to measure dynamic stress using strain gauges in a heat exchanger (60ºC), but we can't stop the process to glue the gauges correctly. Has anyone ever glue a gauge in a working machine?, how were the results?.

 

[IRstuff]

Stress relative to what? Assuming you could install while the machine is running, you're only going to be able measure the stress relative to the condition at which the gauge was installed, which seems less than useful.

TTFN

 

[MikeHalloran]

You will be able to measure the change in strain from the installation condition, from which you might be able to infer a change in stress. But you won't know what the baseline strain actually is, until you shut down the process. If your gages survive that long, you may be able to back- calibrate your data at that time.

Perhaps we could be more helpful if you told us what problem you are actually trying to solve.





Mike Halloran
Pembroke Pines, FL, USA

 

[sms]

Since you are measuring dynamic (peak to peak) cyclic strain you should be ok, but as mentioned above, if the dynamic strain is low, that won't really tell you anything, you really need to shut it down and cool it off prior to making the measurement so you can get the total strain.

When I have had dynamics issues I have installed strain guages on running equipment and it worked fine.

-The future's so bright I gotta wear shades!

 

[fer72]

The problem with the heat exchanger is that the flow rate was increased, and due to this change a baffle inside the exchanger looks to be in resonance (check by means of vibration measurements and CFD and FEM calculations) been the vibrations levels in the outer shell higher than those measured before the change. So the customer wants to know the safety factor of the installation working in these conditions.
Our intention is to make a detailed ODS, make FEM model with the new data to calculate static stress component, and using the vibration data and the adjusted FEM model consider the dynamic stress component, and check this result using strain gauges.
In my opinion the gauge will not be glued correctly due to the different curing time from the outside to the center. The gauge will have parts in tension and others in compression as a result of the vibration, and the results may not be accurate.

Thanks for your answer.

 

[MikeHalloran]

Strain gages are usually attached with CA. On a shell surface at 60C, that's going to cure _very_ fast, so you may not have time to apply a pressure pad to develop the ideal thin glue line.

I think I understand your concern about the CA curing progressively and locking in different amounts of strain along the gage's length because of the vibration. I've read a fair amount of stuff about strain gages, and I can't recall that being mentioned.

You may get better results with a slow curing epoxy adhesive. Put a film of Saran over the gage, then a silicone pad, then a rigid pad, and clamp the whole mess on the shell while the epoxy cures. With any luck, the clamped gage will dampen the vibration locally.

OR, much better, ask the strain gage manufacturer for help.

Given the level of fuss that's required to produce a good strain gage installation on a static object at normal temperatures, don't get your hopes up about accuracy or repeatability. You might do better with an accelerometer.



Mike Halloran
Pembroke Pines, FL, USA

 

[mtnengr]

You might want to look into the possibility of using holgraphic methods: holgraphic interferometric or morie methods.

Using Google or Dogpile, I came up with the site

http://www.stressoptics.com.

I am sure there are others. The most important thing is to have a clear shot of the surface and since your situation is dynamic, you must consider a high intensity strobe. I believe this is a research project and not an off-the-shelf item! Establishing a surface displacement contour is a considerable distance away from establishing a stress contour.

 

[PulseMaster]

Depending on the amount of strain happening in the material, you might not get a good bond. I've had the best luck with weldable gauges(they come with their own metal backing and are often used for high temp). You put them on with a little spot welder. To repeat what the previous replies said, if you have a high static stress, the dynamic won't be very usefull. Although, you should be able to get a good answer from FEA for the static, and a decent number from the field test for the dynamic such that you can do fatigue calcs with a reasonable amount of surety.