Helmholtz Tuning

[Zero_12]

I've been reading some papers and looking over calculations for helmholtz tuning. One question I have relates to the 1989 Cornell paper and the 2001 Lawrence Tech paper. In their calculations to find the RPMs that this tuning will help at, it seems like they didn't accounted for the ram pipe. Is this accounted for somewhere else or is there something I am missing?

[Drew Price]

I thought the most pertinent pressure waves were caused by reflections at large changes in the cross section of the manifold, either the distance back up the intake runners to the plenum, or in the exhaust the distance down the primaries to the collector. Could this account for the omission of the length of the intake plumbing upstream of the plenum?

Best,
Drew

[Jersey Tom]

From our own experience, and from having talked with John Caldwell face to face.. you get a LOT more benefit from changing exhaust header layout and lengths than intake. I have my own theory as to why.

All straight wave speed and reflection calculations. Real easy if you know exhaust gas temp (roughly), exhaust duration, and where you want the power.

[Drew Price]

Tom,

Does that theory have anything to do with the stored energies in the gasses in the intake and exhaust case (thermal or otherwise)? Because my own theory is that it seems like it would be easier to evacuate the combustion chamber, (and help pull intake charge in, depending on valve overlap) with a pressure wave than it would be to push more intake charge into the CC, since the flow during valve overlap already has the inertia of the intake and exhaust charges going that direction.

Best,
Drew

[Jersey Tom]

Way I reckon..

At EVO you've still got considerable pressure in the cylinder, and when the cylinder "blows off" with that kind of pressure ratio accross the port you get a strong pressure wave to propegate out the header, reflect strongly and come back around.

At IVO.. I'd think there would be much less pressure ratio and hardly any wave propegation. Hell. With the intake valve opening BTDC you'd kinda think the pressure ratio would the opposite of what you'd want for ramcharging.

That's only a theory though.

[absolutepressure]

Get the book "Advanced Engine Technology" by Heisler. I goes into detail on helmholtz, and why we shouldn't use it. It shows that it's quite effective...up to 3600rpm, then it's very detrimental because it disrupts the intake pressure waves (tom) and produces an insane amount of destructive interference. So what you could do is create a dual plenum, with both sides helmholtz tuned, then have a butterfly valve connecting them open up to destroy the resonance (since the volume will be changed).....if your engine's mid range is around that area. Depending on your engine, you may not drop below 8k rpm while racing. But I guess helmholtz is used widely in diesel engines seeing as they never go much higher than 3600rpm.

And as I've said before, who cares if it's not as effective as exhaust tuning (if that's true), do it anyway (wave-ram, that is).

[Erich Ohlde]

we have used the equations given in the lawrence tech paper to torque peaks given by resonance tuning. this method is exact and is verified by engine dyno testing. don't even bother with exhaust tuning using hand equations, just use software and verify it.

[Matt N]

... after what absolutepressure said about tuned runner interference...

go look at the stock trumpets/airbox for your CBR600F4i (if its still around). Look at the runner lengths. It oughta get you thinking...

[absolutepressure]

don't even bother with exhaust tuning using hand equations, just use software and verify it.

What do you think the software uses? Seeing as we can't afford expensive software like that, we have to go old school and think about things. We can't just let a box do all the thinking for us while we sit back and say "Hey, they were right, it does get more power." No shit, they don't sell you crap. I'm a fan of doing it old school so that it gives me a better understanding of what's going on in the equations. Equations are representations of real life, and the more we understand both, the better engineers we'll become. One day there'll come a time when we're confronted with a truly novel situation, and then you can't just walk over to the keyboard and hit the easy button. That's why we need to get to that point where everything clicks, and we can start relating things that we know to a difficult situation, until we find the solution.

Matt, runner lengths get very short when tuning for very high rpms, but also realize that the port length is part of the dimension too.

If anyone want's to know what the formula for both intake and exhaust length is, it's here (http://fsae.com/eve/forums/a/tpc/f/125607348/m/57610127831?r=85410348831#85410348831). The C for the exhaust eqn will be much higher due to the high exhaust gas temps. Guess what, I used that equation to reverse engineer the stock headers, and it gave me an rpm right around where I thought they would've tuned it for. No compy program can do that for you.

[Matt N]

absolutepressure,

I was not referring to runner length. I was referring to the fact that there are two different tuned lengths on a stock F4i - cylinders 1 and 4, and 2 and 3.

[Jersey Tom]

Interesting. All I know is between '03 and '05 we tried medium length, fairly long, and real short intake runners while keeping the exhaust primary length and configuration (4-1) pretty much the same. Power band never changed appreciably. Was nothin there till about 8000rpm and then itd turn on like a lightswitch and off you'd go.

In 06 we kept intake length pretty similar to what it was in 03 (7-8" from valve to reflection point IIRC), and lengthened the exhaust headers and ran 4-2-1.

Night and day difference. Powerband was much broader and driveable and picked up much lower.

[Chris Allbee]

absolutepressure,

Ricardo sponsors FSAE teams with a free license of WAVE, if i'm not mistaken. 1-D software that is relatively easy to begin using and even with a crude model you can iterate through different intake/exhaust tuning combinations fairly quickly...

[Mike Claffey]

As people previously mentioned Helmholtz tuning theory breaks down at higher rpm, but it can be a useful tool to begin with, if only to get you thinking about system design/interactions .. and the equations are straightforward and well documented, so I don't think its completely useless to the fsae designer.

On the other hand, Ricardo is a great tool, but back it up with Heywood and Blair. Jersey Tom, am I right in assuming you would still see the movement of a torque peak in the curve with those Intake runner length changes?.. cause they are there, and they do affect things below 8000!

I did a similar thing to you with lengthening our pipes on our exhaust(4-2-1 setup) from 05 to 06 and it seemed to help out on the dyno (quite alot), especially in some of the transient tuning. If anything, making sure that you have a combination of intake/exhaust lengths that do not create a torque dip/anti resonant effect just before 7-8000 can be sort of tricky, because the Honda doesn't seem to like it around there with alot of designs that steady state seem to look really good.

By the way wave's coupling facilities are excellent for finding out whats really happening in your manifolds - doing a few runs can teach alot, if you have the time & patience.

[Erich Ohlde]

if you want to waste your time doing what claude rouelle calls intellectual masturbation go ahead and make horrible assumptions that try and get hand equations for exhaust tuning correct. The reason we use software is because the math would take so long to do correctly that it would be worthless.

[Zero_12]

My concern from the Lawrence Tech and Cornell papers was that their calculations include the throttle body and the restictor, but not the pipe connecting the restrictor to the plenum. Was their some kind of assumption the restrictor went all the way back to the plenum?

Doing these calculations is a starting point for our design. I plan on using Virtual 4 Stroke to verify that the set up should work fine.

[Chris Allbee]

Erich is kindof on the right path here, I believe. Do just enough hand calcs to that you understand what the software is trying to do, this will help you make sense of the results and any anomalies that might pop up, but beyond that you are going to waste hours and hours of your life trying to use hand calcs for a tuning tool.

Do enough to know whats going on then use the programs to get the grunt work done. I mean, how many of us still do long division without a calculator? You did enough to know how it was done, but you know that a calculator will be much faster in giving you the result of 347.6^2/(pi^3)

Smarter, not harder....

[drivetrainUW-Platt]

Originally posted by Chris Allbee:
absolutepressure,

Ricardo sponsors FSAE teams with a free license of WAVE, if i'm not mistaken. 1-D software that is relatively easy to begin using and even with a crude model you can iterate through different intake/exhaust tuning combinations fairly quickly...

Denny get on that free copy!
Also reply to my email about the imagionary intake that we need to actually build to make the car run!

[Jersey Tom]

Hand cals have some merit. I got to talk to race engine guru John Caldwell a while back, and he gave me a 'worksheet' that he uses for exhaust length design.. pretty straightforward. Not an awful lot of math. Made it into a spreadsheet. He trusts it to "within a couple hundred rpm".