Ultrasonds in stressed bolts

[Handford]

Hi, I am developing a device that measures the tensional stress in the bolt by using ultrasonic resonances.

My problem is that I have carried out experiments and have observed an increase in resonant frequencies as the bolt becomes stressed. This does not seem to be consistant with the formula f=v/lambda as v decreases and lambda increases as the bolt is stressed. I realise this could be to do with transverse modes forming in the bolt which are dependend on tension, however in my setup there should only be longitudinal modes travelling up and down the bolt.

Another interesting factor I have observed is that the resonant peaks appear to get closer together for higher harmonics

If you have any possible reasons for this please submit them even if they are just areas I shold look into

Thank you

 

[btrueblood]

Can you calculate the approximate location (distance from transducer) of the peaks that you are seeing? Maybe tension is opening up a flaw? Another possibility is that the transducer coupling is being strained during bolt loading, resulting in artifacts in your data. A final thought is that somehow the supporting/loading structure is becoming a parasitic path as the bolt is tensioned, due to better contact between bolt and structure as load is applied...

 

[GregLocock]

Are you sure that you are not just measuring the widely accepted phenomenon that the joint is getting stiffer as the bolt tension is increased?

I'm a bit suspicious of your formula, why is lambda increasing (other than the bolt stretching which is not measurable I should think).

For that matter why does v decrease?



Cheers

Greg Locock