Dunno if Tim and Bob (his brother in law) used it, but one of the local
engine shops (Davenport Machine Service) designed and built their own
inertial dyno. Basically you design a large billet to take up to whatever
maximum rotational speed you need, and to have a large moment of inertia.
Connect that to the engine through an electric clutch mechanism. Spin the
billet up, and then engage the clutch to start the engine. Then after the
engine is at test temperature, you bring the RPM up to the low end of the
band you want to characterize, then put it wide open. Then the torque output
is calculated from the differential of the RPM, multiplied by the moment of
inertia of the billet. The inertia of the billet was initially calculated so
that a 2000 HP engine would do a dyno run of about 5 seconds, or something
like that; I forget the exact numbers. Then he developed a gearing system so
he could do piddly 250 HP wannabe-engines.
Because this system doesn't use any kind of force-balance or friction based
load, it's very accurate and repeatable. After selling a number of these
(including a full computer control system) to several engine builders in
north america, he made a couple of miniature versions to do model airplane
The only downside is that you can't do steady-state tests.
> -----Original Message-----
> From: Bob Palmer [SMTP:email@example.com]
> Sent: Wednesday, March 29, 2000 3:08 PM
> To: Ronak, TP (Timothy); firstname.lastname@example.org
> Cc: 'Tiger News Group List'
> Subject: RE: "Stroker" Motor
> Let's see, 485 ft-lbs @ 6200 rpm, that's 573 HP! Was this where you got
> peak torque or peak HP? (Peak HP I assume.) Was that a 350 cu in engine?
> I should correct a misunderstanding regarding conclusion #5, which I
> have anticipated. The area referred to is the integrated torque over one
> full cycle, i.e., two rotations of the engine. Changing rod length changes
> the shape of the curve because of the different rod/crank angles, but the
> integral remains constant. This neglects frictional effects, which of
> course are greater with the short rods. You talk about running your engine
> between 5,500 and 7,400 rpm and maximizing the area under the torque
> I think what you need to maximize is the area under the horsepower curve.
> Dyno guys like to talk torque, but it's really horsepower that does the
> job. The concept of getting the maximum integrated effect is important
> I'm not sure what to make of the 1 jet size difference. What we'd really
> like to know is a comparison of the volumetric efficiency. Generally, a
> bigger jet size would suggest better breathing or longer cam duration (but
> the cam's the same in this case).
> As far a scientific method, yes we'd have trouble publishing your results
> as is. One thing not to overlook is the repeatability of the dynamometer.
> This can't be taken for granted and, in fact, I understand it's a real
> important issue with professional engine builders to keep their dyno
> accurately calibrated. Of course, they need a lot better than the ca. 8%
> accuracy we're talking about in your case.
> OK, back to work now!