Motorcycle frame resonant freq. measurement

[saintz]

Hi All<br>Not sure if this is the exact forum for this question, but I'll toss it out anyway. If someone can help, I will be forevermore grateful. What I would like to know is if there is a way to measure or calculate the resonant frequency of a specific motorcycle frame and swingarm combo. I.E. at what engine speeds (frequency) will harmonic vibrations be introduced into the frame and thus felt by the rider? I am not an engineer, but it seems obvious that the rear fork assembly will act as a tuning fork at some engine speeds, and quite possibly some of the harmonics will be in a range noticeable to the operator. I ask this question because I have a new Harley Superglide sport, and everyone that I have contacted with the same model has complained about a pronounced vibration that came on at 2400 RPM and was gone by 3000 RPM. Thank you for considering this.<br>

 

[justapine]

Hello saintz, it is because the frame's natural frequency works on that range, 2400-3000rpm. It is the characteristic of your bike's frame. I did a study on this matter and I used software called STRAND6 to calculate the natural frequency and it was theoritical. I don't know if there is any way to measure it practically.

 

[GregLocock]

Yes you can measure it practically.

This is called Experimental Modal Analysis.

All vehicle manufacturers spend a lot of time and money on this. I'll describe how it was done in the good old days, which will give you an idea of how it is done these days, without all the boring signal processing.

Set up the structure hanging from elastic straps. Attach an electromagnetic shaker (like a loudspeaker coil) to one corner of it. Use the shaker to drive the structure at one frequency, at a constant force level. Measure the vibration with an accelerometer at each point of interest, in each direction. This will have a magnitude and a phase, relative to the driving force.

Plot all these out and you have a mode shape. Repeat for all the frequencies of interest. If you are lucky you can use a strobe to see what is going one, but this is usually impractical for a body or frame, as the amplitudes are too small to see. You can also use a laser and generate a hologram of the shape, which is a nice method.

I've skated over many of the practical difficulties, including how you define a 'frequency of interest'. These are the resonant frequencies, or natural frequencies.
Cheers

Greg Locock

 

[GregLocock]

Sorry, i re-read the first post and discovered that you want to know the resonant frequencies.

OK, method 1), using the setup in the previous post.

Put you accelerometer at the driving point on the structure, with its axis aligned with the force axis.

Use an oscilloscope to view the forec signal on the x axis, and the acceleration on the y axis. Vary the frequency

At resonance the two will be in 'quadrature' and you will see a circle on the oscilloscope. This will work quite well on a motorbike frame, I think.

2)Alternativley plot the ratio of the aceelerometer's amplitude to the force amplitude as you vary the frequency. Peaks in this function are at (well, near) the resonant frequencies.

We don't do it this way any more, but it is a perfectly good method.

There are an awful lot of details I have left out, and there are many pitfalls. In the case of your Harley there is a strong possibility that the engine is generating more vibration at 2400-3000 due to a resonance internal to the engine. We measure those as well.







Cheers

Greg Locock