So what is that to imply Greg? That link mentioned nothing about determining natural torsional frequency of a shaft? Especially a crankshaft where bearings and rod and reciprocating loads are going to modify it. Rapping something with a hammer and measuring the frequency may show what its vibrating at, but is that also it natural torsional frequency? You see alot of how to dampen torsional vibrations in text books, but not much in how to determine what the natural freq is. I guess it can be done by test looking for resonate periods.
What do you mean old school? For hand calc or test? There are equations for torsion frequency or you can do a modal test which is rapping it with a hammer and have a transducer in the torsion direction.
Also, what is your application that you need to know the Fn of the crankshaft?
Tobalcane
"If you avoid failure, you also avoid success."
“Luck is where preparation meets opportunity”
There are businesses who are founded upon just doing torsional vibration studies. A crankshaft with attached connecting rods/ pistons/crossheads with cyclic pressure events is not a trivial problem to solve.
There are methods to measure torsional vibration on a rotating shaft, such as with a rotary encoder and analysis software.
As noted above you didn't specify whether analytical or test, or modal. Your lazy question leads to long thread unneccessarily. What do you mean by old school?
On a running engine we used to use a toothed wheel, and inductive pickup, and a valve 'powered' FM demodulator. Plus ca change, still in use.
For a physical test in the lab as you say belt it with a hammer.
Analytical - Holzer.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
I did a fair amount of torsion speed measurement several years ago, using inductive probes, resampling, hilbert transforms, tooth-pass filtering and all those tools. But I'm not sure you could measure torsional vibration with these techniques. Speed variation through the cycle, yes. Gear rattle, probably yes. Clutch deflection, yes.
There's no particular reson why my proposed instrumentation could not have been developed in the 30s, you need a device that creates a voltage proportional to frequency, a coil, a magnet and a toothed wheel. And a CRO.
If you want to know how it was done during WW2 then you need to buy Practical Solution of Torsional Vibration by Kerr Wilson.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376
