A bit of an old thread, but fascinating work. I really hope you can work out the resonance this way as it would make things much simpler for the DIY engine builder!
This paper talks about creating an accurate but simple model to estimate inertia and torsional stiffness of the different crank throws based on a computer model of the crankshaft. That’s then reduced down to a simplified ‘equivalent model’ of rigid bodies with rotational inertia but no elastic deformation and a shaft segment with elastic deformation but no rotational inertia like so:
That came out pretty accurate to the measured model of a diesel agricultural engine. It all seems doable by someone with more CAD skills than me, so might be a method of verifying whether they match the resonance when struck.
Another idea would be to see if someone has a crankshaft where we know where the resonances lie, and measuring what happens when thy’re struck. For instance, I know that 4.0l AJ6 engines have harmonic issues at about 5850rpm, Alfa 2600 I6s at 7200rpm, BMC C-Series at 6200rpm (probably), M54B30s at 7300rpm and Chevy 292s at 5500rpm. You can work out the frequency of those by dividing the rpm by 60 to get Hz, and then timesing it by the number of cylinders (because even when piston isn’t firing there’s still a sharp spike in force as the piston changes direction at TDC).
Alternatively, if anyone knows what rpm the 2 1/2 engine starts coming apart we could compare directly to the measured figures here. Looking for misc weird things happening like cranks shearing, flywheel bolts working loose, oil pumps failing, timing jumping all over the place, severe increase in harshness. That sort of thing. If they occur somewhere around 3750rpm or 6790rpm that would be interesting.
The other thing that’s worth mentioning is that you’d probably need to repeat the measurement with the flywheel and non-damped pulley attached (or core of a damped pulley), plus bobweights approximating the rotating weight of the conrods. The resonant frequency exhibited in engines is a function of the inertia of the whole rotating assembly, not just the crank. I expect this also includes the clutch basket too, although the springs in that might make it difficult to measure the resonance by striking. Hopefully that would only change it slightly.
So, perhaps somewhere lower than 6790rpm might be more likely to correlate with observed harmonic issues.
So, latest interesting piece of research. Turns out there are two different natural frequencies for crankshafts (or anything really). The first set is longitudinal/transverse natural frequency (so bobbing up and down on a spring, or wobbling back and forth on a stick). This is a function of gravity, mass, and resting deflection (which itself is a function of weight and spring stiffness). The second set is torsional natural frequency (twisting along an axis), which is a function of torsional rigidity and moment of inertia.
As it’s the latter we’re interested in, the question becomes ‘does striking a suspended crankshaft excite the torsional frequencies or just the longitudinal/transverse ones, or does it produce both?’.
If it does produce both, then there might be some guessing as to which peaks represent torsional vibration and which represent longitudinal/transverse. As you had differences in the amplitude of different frequencies depending on where and how the crank was struck, I wonder if that might be a way of finding out what’s what (although I’m not sure how!).
Answering the original question:- no. The reason is that when in the engine, there are engine bearings bolting to the assembly at various points, then there is a big flywheel bolted to one end and a damper bolted to the other end. It’d make a totally different vibration pattern when such constrained, so the note would be different.
For argument’s sake, if there are seven bearings at even distances from each other along the crankshaft, then you might expect a somewhat damped frequency six times your ring to be produced. There would then be harmonics added on top.
In practice, the bearings are probably deliberately unevenly spaced from each other to disprupt the formation of a nice even wave at the resonant rpm and thus this spacing will produce a resonance pattern which is smeared out rather than a distinctive sharp peak. The harmonic balancer would then be designed to either push that shallow peak out beyond the working rpm or to further spread it out and flatten the response.
Yep. About 15—20 years ago I wrote a short technical article on crankshaft and torsional resonance that was published in the XK Gazette. Also, as noted above: Crankshaft Resonance - #12 by Mike_S
@Mike_S I don’t suppose you have a copy of that do you? Or know what issue.
I suppose the question would become if striking a crankshaft imparts enough energy in the right direction to excite a measurable torsional vibration, or whether all of the peaks we see above are lateral.
@MarekH The damping effect of the bearings would be true of lateral vibrations, but they do nothing for torsional vibrations. You’re right that the flywheel and harmonic damper would affect these though. These would need to be attached to get a reading (and potentially the clutch basket, although that being sprung I’m not sure how much it would impact things).
In all likelihood it’s not possible to identify torsional harmonics by striking a crankshaft, but I’d still like to see it tested a little further!
You would have to define “better”. There’s always a cost-benefit consideration. Some beady-eyed purchasing manager looking over your shoulder asking "Do we REALLY need THAT? The old rubber insulated inertia ring is dirt cheap and is not likely to fail under warranty (or anywhere close to it). It does what it’s supposed to do for not much money. If you can rationalize that you intend to operate outside of the original factory constraints, then you may be able to justify one of those expensive aftermarket dampers. In the design world, you are almost always operating within some kind of cost restraint. If you are operating in the context of critical infrastructure or aviation (not me, I did chemical process equipment) I would guess that the budget generally bears what the desired mean time between failure dictates. And then they replace it when it hits its half life. I’ll bet the cost budget for an OEM damper is just a few dollars.
Correct, no 2-1/2s, and it was only added to the 3-1/2 engines beginning with S.1201 about mid '47.
I didn’t find any mention of it in the Service Bulletins.
The old rubber insulated inertia ring is dirt cheap and is not likely to fail under warranty (or anywhere close to it). It does what it’s supposed to do for not much money. If you can rationalize that you intend to operate outside of the original factory constraints, then you may be able to justify one of those expensive aftermarket dampers. 
In the design world, you are almost always operating within some kind of cost restraint. If you are operating in the context of critical infrastructure or aviation (not me, I did chemical process equipment) I would guess that the budget generally bears what the desired mean time between failure dictates. And then they replace it when it hits its half life. I’ll bet the cost budget for an OEM damper is just a few dollars.
