PRI Show

[fishracer]

I am new to this site and find it very informative.

I also recently attended the PRI Trade Show in Orlando and saw some interesting products while there. One company had a boring head that was programmable & expanded as it machined below the minor diameter. It appeared to be a high quality piece and is targeted to cylinder head technicians.

Another company had an inertia type engine dyno with a series of large diameter, heavy flywheels. It also had a Stuska absorber connected to the main centerline. I do not fully understand the concept and was hoping that someone here could explain it to me.

Thanks.

 

[NCShane]

Thats a pretty typical configuration for chassis dynos, but I haven't seen many engine brakes set up that way. The inertia can be adjusted by adding/removing flywheels for different size engines. This gives you a cheap and easy way to do some transient testing. The absorber is used for steady state operation (holding the engine at one speed once it has accelerated the mass).

Shane

 

[fishracer]

I do not clearly understand the purpose of performing a steady state test with the flywheel-mass (MOI) connected. Wouldn’t the power be the same with or w/o the MOI connected? [I can understand the acceleration curve being affected by a MOI change, but not power.] Is my thinking correct?

Your thoughts lead me to another question: How do the typical dyno companies compensate for engines with different inertia-effects (rotating/reciprocating weight)? Are there formulas involved?

BTW: I do not remember the name of the inertia-dyno company, however the rep claimed much closer repeatability figures than the normal +/- 0.5%. It seems like he spoke of around 0.25% or so. Why would this dyno be more repeatable than a traditional brake dyno?

 

[NCShane]

You are right, the inertia mass doesn't affect the steady state test. The only dyno that I am familiar with that uses a configuration like this is a dyno-jet chassis dyno. They use an inertia drum to measure hp on all of their units, and then add an eddy-current brake to the system if you want to do steady state testing.
I am guessing that this company is trying to give you some transient testing ability without the complexity and cost of an AC or DC dyno system.
The inertia effects for each engine are typically ignored for steady state systems (typical eddy current and water brake dynos). For transient testing with an AC or DC dyno, these effects can be measured by motoring the engine with the dyno (without combustion).

Shane

 

[fishracer]

Yes _ I am familiar with the Dyno-Jet chassis dyno. This is definitely not a chassis dyno, it is an engine-only dyno.

I should've given more information about my question concerning "inertia-effect" of each engine. I was speaking of the formula/procedure for calculating this effect during acceleration testing on the brake-style dyno's. This could lead to erroneous readings if not calculated corectly and I am trying to find out how each company figures this number.

Our company is looking for a dyno that will be used primarily for development testing. The repeatability claim of the inertia dyno has us giving it more consideration. The AC & DC dyno's are too expensive for us to think about and I am not sure they would be any better than the inertia dyno for our needs. (Perhaps not as good!)

Also; I found out that an engine builder named Ron Hutter has used on of these inertia dyno's for a few years now and likes it. Does anyone know Ron?

Thanks.

 

[fishracer]

I went to NC and saw the inertia dyno operate. It's ability to repeat was better than they claim and I have no idea why there isn't one of these machines in every NASCAR shop.

I had a couple of observations and would greatly appreciate input from members on here.

1) They use a "Lenco" case to engage and disengage the inertia mass and it seems to work fine. I know nothing about the Lenco box and would like some input from anyone who is familiar with them.

2) Concerning the software: The dyno repeats so well that it's almost a little hard to believe. I know that a fairly large amount of mathematical filtering is necessary for traditional dyno's to produce smooth & repeating power curves and asked the engineer about their filtering set-up. He claims that they do not filter elapsed time but do apply a light filter to engine rpm. (Due to engine torque spikes). Can anyone offer input on this? How much filtering is too much? (I would not want to "filter-away" good data in the interest of a pretty power curve!

Overall we left pretty impressed. Unless somwething is revealed that we are unaware of, we are probably going to purchase one of these machines.

Thanks.

 

[patprimmer]

In my opinion, any filtering is to much as the spikes can contain some very useful information.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[fishracer]

I agree! I've done quite a bit of research on this mathematical manipulation and the "amount" varies from one dyno company to the next.

Ideally there would be no filtering, however I have yet to see any raw data from a dyno test that was legible enough to decipher. There are even a few companies that do not provide the ability to view the raw data. The reasons for filtering are complex and also at times seem a bit controversial. At any rate - I realize the need for filtering and know that the best situation involves the least amount of filtering to produce a legible curve.

 

[SomptingGuy]

In my opinion, any filtering is to much as the spikes can contain some very useful information.

True throughout testing. Just give me the raw data and calibrations and I'll choose my own assumptions when processing. Raw data is really cool stuff.

 

[stout1]

Back in the day when I used to be an engine builder we used a water break dyno by DTS for our circle track programs (from my experience this is what most engine shops have and all nascar shops do have a dyno that build engines). The centrifical dynos as you are talking about we used on our pulling and drag racing engines and there was a couple reasons for this. 1-these larger engines had too much power for the water break to handle (water break are usually only good to about 1500 hp) 2-water break dynos have a very long cycle time per pull (if I recall about 15 sec.) and our big engines don't have that long of a life expectancy (they would need a freshen up after a full dyno session). I thought that the centrifical dynos only had a 2-3 sec. pull time.

Michael

 

[NCShane]

fishracer,
where in NC is this dyno located? Just curious.

 

[fishracer]

Shane: Their shop is in Concord, N.C. Their website is

http://www.revolutiondyno.com

(The intake manifold comparison shown on the website is interesting.)

SomptingGuy: That's an interesting reply. When viewing all of the raw dyno data from various companies - I've never been able to draw reasonable conclusions without some "smoothing" effect. Do you have your own program and use ascii data?

stout1: Sorry, but I am not familiar with a centrifical dyno. The inertia dyno I saw uses a series of large flywheels to simulate the vehicle's inertia. It also uses a gear-reduction unit. They tell me that Ron Hutter tests 1700+ HP Pro Stock engines on one of the same machines. I saw them test a NASCAR Busch engine on it and they told me that it can simulate the straightaway of any NASCAR track.

 

[stout1]

Centifical/inertia dyno's are the same thing, that's just what they were called where I worked. Interesting, me and Ron have the same last name, the storie that I've been told is that there were 5 brothers that came over from Germany together and 1 of them got lost in New York (the rest ended up in the mid-west). Maybe he's my long lost cousin. lol

Also you can never simulate what an engine will do on the track, at best what it does on the dyno is an estimate. I have seen engines with identical power and torque curves act very differently on the track. On the other hand the more ontrack testing of dynoed engines that is done the smaller that your error will be, but it will still be there.

Michael

 

[fishracer]

I made the same claim (that "you can never simulate what an engine will do at the track...") to the engineer with the inertia dyno company. He politely informed me that I was wrong, that their inertia dyno can exactly simulate race track performance. He explained that they configure the machine to match the vehicle inertia on a chosen race track, they then accelerate the engine from a beginning rpm to an ending rpm - and measure the elapsed time between the two rpm's. This E.T. becomes the engine's baseline and the engine builder works on the engine with the goal of reducing this E.T. If/when the E.T. is reduced, the engine is performing better.

The concept makes sense to me. Is there anything I am missing here guys?

 

[NCShane]

Well, it seems like the "inertia" of the car will increase as it accelerates, since the drag/downforce will increase at higher speed. Even if you factor the aero drag effects into the initail mass, they don't have any way to dynamically adjust it for speed.
Most of the cup guys that I know are using A/C dynos for track simulation these days. I'm not sure how much they worry about trying to simulate drag, etc. but I'm sure they will be looking into ways to increase the correlation between the dyno and the track as NASCAR continues to restrict their on track test days.

 

[fishracer]

You're right Shane about aero-load and other frictional loads affecting the inertia-effect.

I guess my question then becomes: If a test "window" is taken from on-track Pi-data and the dyno is physically configured to match those figures, how far off would you be?

Ex: Let's use hypothetical information from say - Richmond Raceway.

The engine is at WOT for 6.5 seconds.
Beginning rpm (at throttle open) is 6500.
Ending rpm (at throttle closed) is 9500.
So we've established that the engine accelerates "the mass" (as this is data straight from the car's data acq - aero and all other friction-loads would be included in this E.T.) from 6500 to 9500 rpm in 6.5 seconds. [Call this "the run".]

Now, you set-up the gear ratio and inertia wheel arrangement of the dyno to exactly mimic the run data. (Engine accelerates the inertia mass from 6500-9500 in 6.5 seconds at WOT.) At this point, the dyno run is pretty close in simulation to the on-track run. [I agree that the aero and other load curves would be slightly different between the 2 runs, but not as much as one might think.] I then go to work on the engine and try to reduce the E.T. of the dyno run. If I am able to reduce the E.T. - wouldn't this always result in better on-track performance? The performance gain may not be linear from dyno to track, but it seems that it would be very close.

Again - please offer your thoghts.

Thanks.

 

[Lou Scannon]

FYI, there is a published method for determining vehicle road load (includes inertia, rolling resistance, and aerodynamic drag), for use in dynamometer testing. It is SAE standard J1263. My volume is way out of date, but this reference should put you on the right track.

 

[patprimmer]

I think gear changes might also impact the comparative loads and reaction rates on an accelerating engine.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[fishracer]

pat:

There is no gear change made during the acceleration run. It is more like the final drive ratio on a vehicle.

 

[patprimmer]

There is at the track, which is my point.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.