Intake Plenum Volume?

[Jay Fleming]

I know this is probably either a hush-hush topic or "use your head", but is there any equations or theories that you use to calculate the optimum plenum volume. I hate to ask, but our engine capstone was at first to optimize the intake, exhaust and restrictor designs. We've done the restrictor and now we're lookin at publishing a guideline for our future teams to follow that will help calculate plenum volume and runner (intake and exhaust) length. I've got the calculations for the runner length done already, but can't find anything for the plenum volume. I've seen in Winterbone and Pearson's book about the effects of different plenum volumes compared to manifold pressure as a function of crank angle, but that doesn't really help because of the restrictor. If anyone has any definitive equations or theories, I would be greatly appreciative. Thanks in advance.

[Jay Fleming]

I know this is probably either a hush-hush topic or "use your head", but is there any equations or theories that you use to calculate the optimum plenum volume. I hate to ask, but our engine capstone was at first to optimize the intake, exhaust and restrictor designs. We've done the restrictor and now we're lookin at publishing a guideline for our future teams to follow that will help calculate plenum volume and runner (intake and exhaust) length. I've got the calculations for the runner length done already, but can't find anything for the plenum volume. I've seen in Winterbone and Pearson's book about the effects of different plenum volumes compared to manifold pressure as a function of crank angle, but that doesn't really help because of the restrictor. If anyone has any definitive equations or theories, I would be greatly appreciative. Thanks in advance.

[Charlie]

I totally understand, I looked for equations too. With the restrictor, there just aren't any.

The best thing you can do it build a plenum that has variable volumes, and test it.

I found in some simulation I did that the volume of the plenum is a very small factor in power (at least at the size range tested, 2x+ engine volume), but the bigger the plenum the more unrestricted power made. This improvement was very small though.

[strong_auto_concepts]

I imagine one of the engine guys will chime in with this (I'm just a lowly EE, I dont' get to touch the car).

We had huge differences in HP with plenum volume. Sometime like 12-15Hp depending on runner distrobution and plenum shape.

[Charlie]

Certainly it's possible that there's more power in plenum volume than we've seen, we didn't do extensive testing.

However, especially if plenum shape with the same volume is what you are talking about, I suspect that the power you found was related to flow, rather than strictly plenum volume. From what I've seen, making your intake system flow with as little restriction as possible is 100x more important to power than plenum volume itself.

[Jay Fleming]

Let's see if my thoughts are correct, and may help everyone understand my frustration. I know that with too little volume, the restrictor will get choked very early in the RPM range because of the pulsation. Too large a plenum, and there will be very little throttle response. A large plenum would be great if the engine ran at a single RPM or a very narrow range. And a small plenum would be good for all RPMs if there was no restrictor. We don't have time to do any testing, so i was just trying to come up with something that future teams could follow and test their findings to help clarify in later years.

[BeaverGuy]

When we did our plenum volume testing we did se some variation in power. Though, it was nearly the 12-15HP quoted. The difference was primarily around the Resonant RPM which varied by about 500 RPM. The results were pretty much exactly what we were looking for.

The larger power increase came when we had to throw together a plenum and runners when the designed ones wouldn't fit in the car. The increase in power was due to the new runners being larger in diameter and shifting the torque band higher in the RPM range and better distribution to the runners.

For equations that will help in defining what volume is best for throttle response or negating the restrictor, I don't know that there are any. However, there are equations relating plenum volume to resonant RPM. I know that some of the Helmholtz Equations include plenum volum and the pipes upstream in there calculations.

[Jarrod]

we tested plenum volumes in 2002, from about 1L, to over 4L, and saw about 3-4 hp difference maximum (biggest volume made more), but 90% of the rev range the difference was less than 1hp. OUr dyno is only an old steady state water brake, so a ramp test may show a different story. The runner length tuning we did made a much bigger difference.

[John Bucknell]

Plenum volume is an interesting topic, something that I think I've touched on in other threads - but I can't be bothered to look them up http://fsae.com/groupee_common/emoticons/icon_smile.gif

Several good points have been made already in that performance variations found from varying plenum volume most often are due to changes in flow loss - that is to say momentum effects. The impact of plenum volume on damping restrictor pulsation is most likely minor on multicylinder engines, but I'd love to see data on that particular topic. Since runner geometry is the dominant influence, creating a plenum that adequately feeds the runner mouth is constrained to some minimum dimension such that the plenum does not act as a continuation of the runner (having cross-section change sufficient for wave reflection) and has sufficient 'quiescent' volume for induction events. Beyond that volume, any more reduces throttle response but may be offset by reducing flow loss in a multicylinder manifold. That is about as good a guideline as I use as an OE manifolding engineer.

[Z]

There is an article in the current issue (April 2005) of Race Engine Technology (sister mag to RaceTech) by Gordon Blair & Assoc., that covers some aspects of this subject. It concerns N/A MotoGP engines which are 990cc. Some earlier issues of RET also cover intake tuning, etc.

The bottom line, according to Blair, is:

1. If you don't buy his Virtual Engine Simulation Software, you're going to lose! (Although the software does look neat.)

2. Start with a plenum volume of 10 x capacity (ie. 6 litres for FSAE), then go bigger. This does assume throttles near the head.


Z

[Sam Zimmerman]

Not to badmouth virtual four-stroke, but it is not a magic ticket to a good design. If you do not understand the physics involved I would not recommend any simulation software (engine, suspension, or otherwise). The software becomes a video game with far too many variables and it can become a time drain, I've been there.

I would not try to design a suspension using any suspension software until I understood the kinematics involved. Likewise, do not use engine simulation software if you do not understand acoustics. (I would argue that acoustics have a greater effect on volumetric efficiency vs. engine speed than any other parameter.) If you don't understand acoustics, I think your time is better spent with some PVC intakes and a dyno than on simulation software. You would be surprised how well a crappy flowing PVC intake will show you the general shape of your torque curve before you spend much time building a real intake.

I am sure John or Charlie will tell you if I am way off on any of this. They are the pros.
http://fsae.com/groupee_common/emoticons/icon_wink.gif

[Sam Zimmerman]

I forgot to comment on the plenum volume. One aspect of my research seems to indicate that through proper intake geometry, the volumetric efficiency at certain engine RPMs can be increased by manipulating the plenum geometry and thereby decreasing the impedance at these engine speeds. If that is the case, the volume of the actual plenum becomes less important than the relative sizes of the plenum and inlet. I will be looking into this a bit more, I don't know if I will find anything conclusive or not. Thoughts?

Finally, the extra power that may be gained by increasing plenum volume seems to me to be insignificant compared to the loss of throttle response due to the location of our throttle bodies. What will do better in an autocross, a car with a couple extra horsepower of a car with a predictable, responsive throttle?

[osubeaver]

I agree that testing on the dyno is the best, but the virtual four stroke can give you a good idea of where to start - given you set it up properly and understand it is just a simulation and doesn't take care of all the physics going on in there. As for plenum volume, we have tested several volumes and we really havent seen huge changes, just some slight movements in resonant points. We of course, haven't tried anything huge like 6 liters though, so maybe that would work, but I have always thought that something that big would give poor throttle response - not to mention the headaches of packaging it in the chassis.

[MikeWaggoner at UW]

Could you put a secondary plenum before the throttle? Basically route the intake into a sidepod of the chassis, and then into the throttle, then into the intake? I realize there'd be problems with the buttefly, and weird resonance issues, but is this way off base?

[Patrick W. Crane]

I think what you are getting at is a Ram air type system with a proper diffuse to increase the pressure before the restrictor... i think that is a great idea (maybe not a new one) but still good.

[BeaverGuy]

The problem is that for the best effects the plenum needs to be placed after the Restrictor. In our case the throttle then has to come before the restrictor. So the order goes throttle, restrictor, plenum, runners, engine.

In Blair's Articles on restricted motors the throttle was located at the engine. This made the order of items restrictor, plenum, runners, throttle, engine. With that layout a large plenum volume helps to offset the restriction without as much affect on throttle response.

Adding a second plenum before the throttle would certainly be interesting. The accoustical effects could cause lots of interesting problems with the restrictor. If the plenum before the throttle resonated to create a higher than atmospheric pressure then the choked flow rate would be higher than normal. However, there would be anti-ressonant points too, causing a lower than atmospheric pressure. This would cause the flow rate of a choked restrictor to decrease.

[Etalon]

Plenum-throttle-restrictor-plenum is an interesting idea, but I think you're all reading into it too much. Think of packing issues, added weight and potential gain. Generally for 4cyl engines (half a V8), I've found 50-60% of engine capacity gives good dyno results.

As mentioned before, you're much better off spending your time tuning your runner lengths and reducing flow losses than worrying about plenum design. Having said that, let me know what you dig up...

[Sam Zimmerman]

BeaverGuy has hit on something I have wondered for a while. With the RPM range of the engine, there would be resonances and anti-resonances and the trick would be getting the pre and post restrictor volumes to work together at the frequencies you want. If you look at Engelman's work, that is how he conducted his testing, with a large volume followed by an inlet to the "log" (aka plenum). Unfortunately he seemed to ignore these 2 volumes since he was using one to slow pulsations and one strictly as a manifold.

My question is whether there would be enough benefit to justify this. It would be interesting, unfortunately I have to stop looking at stuff and write my thesis soon. http://fsae.com/groupee_common/emoticons/icon_wink.gif

I would like to see another student at the U of I continue my research by recording actual total pressures in the intake vs. crank angles. The possibility of changing valve timing to work with the phase of the waves would be cool, if not overkill.

[John Bucknell]

For throttle response purposes, we like to see no more than 5 times (preferably down around 2 times which is functional mininum for a low-speed production car) displacement volume for throttled volume. Typically less than half (more like a third) of this value is the plenum.

So GP Blair (who is an old curmudgeon, but I like him - he got me started on two strokes) is fielding a recommendation for a port-throttled geometry - I can't imagine trying to drive a 10 times displacement volume system in competition with a plenum throttle. We did about 20 times displacement volume once with an upstream throttled supercharger and a body-mount intercooler, you'd tip out of the throttle and the damn thing didn't slow down for like two heartbeats. Imagine a blower car with lag, only the inverse (there was some on tip in, but surprisingly small).

[Jay Fleming]

Thanks for all of the replies fellas. I appreciate all the knowledge. I will try and paraphrase this whole post when I put the info into our suggested guidelines (sort of). It's good to have some suggestions from a design judge and other veterans of FSAE.