Don Terrill’s Speed-Talk

is there a formula to determine plenum size?for instance, to determine where you want your power at intake?

If my memory serves me correct its 1 1/2 times a runner volume. And thats the total runner not just the head. I will check my notes and repost if I'm wrong but I beleive this to be true.

sc2dave wrote:is there a formula to determine plenum size?for instance, to determine where you want your power at intake?

It depends on your RPM range, Engine size and intended use of the engine. For instance, An engine utilizing an automatic transmission will have a much smaller plenum than an engine using a five speed. An engine with a five speed behind it can use another 20% more plenum volume. You must be careful here! The dynamics of the race track vary slightly from the dyno and in this regard, they vary a LOT! On the dyno you can tune an engine to produce max power by adding a considerable amount of plenum volume. In some cases you can almost double the volume and see a considerable power gain. On the track the engine will lag against the converter or worse yet, just not accelerate at all. Smaller engines are much more susceptible to this than larger engines. We know from extensive track tuning that our P/S engines will not take any more plenum volume but on the dyno, I could add more volume and see more power. In this case, equations take a back seat to experience and practicality.


Tell us a little about the engine in question and we might be able to help.

RPM range
Engine size
Intended use

Keep the purpose of the plenum in mind. It's supposed to simulate atmospheric conditions. The larger the plenum, therefore, the better job it does in simulating the constant atmospheric conditions. The smaller the plenum, the more adjacent cylinders are going to affect the resonance you're seeking in a "tuned" manifold. Unfortunately, during acceleration, you have to wet the walls of the large plenum before the engine "sees" the richer mixture and reacts properly. During the development of the "ram" manifolds at Chrysler, the engineers always asked for a 4x4 cross section, but the production manifolds ended up smaller than that. (Weight and cost.)

i feel a little embarrassed by the engine that i'm working on. i am working on a saturn dohc . i plan on fabricating a header for alot more torque and/or power increase from off-idle to around 4,000 r.p.m. this is only a "just for fun" setup. this engine is a 1.9. that's why i also want to fabricate a intake manifold that works at my intended rpm range, not at the manufacturer's range.this is a daily driver, so i want to feel more power at the bottom, instead of having to wind it up. So, is there any hope?

Since this is fuel injection, you can make the plenum as large as is practical. The empirically determined equation used at Chrysler was NL=84000, where "N" is the engine rpm and "L" is the distance, along the tube centerline, between the entrance to the plenum and the head of the intake valve. This is for a 200 degree at 0.050 cam with 13 front traverses during valve closure, if you want to "fine tune" the results. If you figure 17 traverses, the tubes get a lot shorter, but the effect isn't as strong. Smooth the entrance (bell mouth) at the plenum/tube interface.

man, all these numbers sound confusing! i'm using stock cams,i just want more lower range power. what if i just lengthen the runners,and maybe make them smaller diameter?

man, all these numbers sound confusing! i'm using stock cams,i just want more lower range power. what if i just lengthen the runners,and maybe make them smaller diameter?

you don't want to lengthen the Intake Runner to any random longer Length, as its possible to lengthen the Runner where it will be out of tune/phase at Intake Closing Point and Intake Opening Point for your desired RPM range .

not as big a problem on the Intake side, as it is on the Exhaust side.
just randomly lengthening the total exhaust system length can cause a positive pressure wave to occur during a good portion of the overlap period hurting Torque/HP where you might have wanted gains instead .

[Since this is fuel injection, you can make the plenum as large as is practical] wouldn't this affect the power range, either raise it or drop it? so, how then can i know what length and diameter runners to use so that i can lower peak power and bring it down to where i want it at?[/quote]

-how are you thinking about building this manifold?
-If you intend to make it as a one piece unit with no adjustability as far as port lengths and as far as plenum volumes then you'll probably need to build about a hundred manifolds.
---------------------------------------
-I'd look at making the port lengths adjustable and the plenum volume adjustable and maybe even incorporate a way to change the port diameter and/or the taper of the intake ports' length.
------------------------------------------
-My question is what are you tring to get here?
-It sounds like you want to take a 1.9 engine that comes on the cam-comes on the pipe and comes on the manifold at something like 4500 rpm now and you want make it run better at what? 3400rpm or something?
-----------------------------------------------
-Then I wonder what you have in the way of equipment to do all of this?
-What about advancing the cam/s on the engine? have you tried that yet?
-------------------------------------------------
-What sort of manifold does the engine have now?
-Is it a two intake port per cylinder manifold with one port per cylinder shut off until some ECM controlled rpm?
-Why is this stock manifold such a problem?

max stock power is at 5600. all i want to do is try to make more power where i drive at, under 4000 rpm. i've heard these cams are not the adjustable type.this manifold is one runner per cylinder,then the head has 2 intake valves per cyl. there is nothing that the ecm controls on the manifold.

The plenum volume has nothing to do with the resonant speed. With fuel injection, you don't even need a plenum. That is, you could have a tube for each cylinder and no plenum. I'm sure you've seen this sort of arrangement on racing engines. For street use, however, you usually want to run an air cleaner, which means a plenum is required. And, there is usually a fuel injection sensor mounted in the plenum to control mixture during startup.

With carburetion, however, the interior area of the plenum affects the accelerator pump performance. Plenum size then becomes a compromise.

Runner cross sectional area also does not affect the resonant speed, but it can tend to mask the effects of runner length changes.

With that NL = 84000 equation I posted earlier, a tuned engine speed of 3500 would require a runner length of 24 inches. This will shift slightly with intake duration. The tuned engine speed is associated with the number 13. There will be other "bumps" in the torque curve below and above the calculated speed. These bumps will be associated, inversely, with the numbers 9, 13, 17, 21, etc. The "13" length seems to offer the best compromise between runner volume and flow losses.

sc2dave wrote:max stock power is at 5600. all i want to do is try to make more power where i drive at, under 4000 rpm. i've heard these cams are not the adjustable type.this manifold is one runner per cylinder,then the head has 2 intake valves per cyl. there is nothing that the ecm controls on the manifold.

Are you talking about part-throttle driving? I think most of the posters are assuming you mean wide open throttle (WOT).

Why is it you want to keep the revs under 4000 when you are at WOT? Even the automatic shifts north of 6000 at WOT if I recall. The 1.9L SC2 ("2" for two cams) has a fairly flat torque curve as designed with a torque peak at 4800, but with good torque from the low 2000s on up, unlike many import engines that sacrifice low-mid torque.

If you are short shifting at WOT to save the engine from wear, don't bother. It's one strong sonofagun. Our family went a few hundred thousand miles on 3 of them. One in particular had the stuff beat out of it for almost 70,000 miles with nearly every shift above 6000 rpm. It was SStrokerAce's (an occasional poster here) college car.

There are some aftermarket parts available. (We had to build our own cat-back in '98.) Most of the parts show better increases at WOT and higher rpm, of course. While cold air inlets, headers and cat-back exhaust will help under 4000, the torque/$ ratio is very low. If you want to help some, increase the compression 0.5 or so.

Here's a link to some mostly higher rpm improvements:

http://www.teamscr.com/power.htm

If you are truly trying to get significant increases below 4000, I think you are tilting at windmills. The SC2's 64.2 lb-ft/L is fairly good. It's just a llittle better than my 345 hp/350 lb-ft, (61.7 lb-ft/L) LS1 C5 Vette.

sc2dave wrote:..so, how then can i know what length and diameter runners to use so that i can lower peak power and bring it down to where i want it at?

Intake manifold alone isn't going to do that. You'll need to change the valve timing significantly which isn't practical. The aftermarket cams available are made for more higher rpm power, not less. Getting custom cams made for a 4000 rpm power peak would be very costly.

If your goal is more grunt below 4000, either increase the displacement significantly, which isn't practical, or put a positive displacement supercharger on it and make the engine think it's bigger than it is.

My $.02

I would agree that it's really silly to manifold tune for an rpm which you won't even "see" after the WOT shift out of first gear. I say this even though I was at Chrysler when they did exactly that with their long branch intake manifolds. My boss in the Vehicle Performance Department tried to point this out at the time, but he didn't have enough clout. The bean counters were given rides in the early project cars and they were impressed and that's all it took. After it was too late, a "drag race" was surreptitiously arranged, at the proving grounds, between a Chrysler with the "ram" manifold and another identical car with the conventional dual quad setup. The conventional "ran off and hid" from the ram.

But, with the throttle blades upstream of the plenum, the ram did give a very good part throttle "feel," which is why the bean counters were impressed. So, I'm sure there are those who would opt for the ram, even though WOT performance was actually poorer.

As far as WOT performance is concerned, you'd want to tune for an rpm midway through the range seen in an intermediate gear. (Unless, of course, you're building a Bonneville car, in which case you'd want to tune for the horsepower peak.)

BillyShope wrote:I would agree that it's really silly to manifold tune for an rpm which you won't even "see" after the WOT shift out of first gear. I say this even though I was at Chrysler when they did exactly that with their long branch intake manifolds. My boss in the Vehicle Performance Department tried to point this out at the time, but he didn't have enough clout. The bean counters were given rides in the early project cars and they were impressed and that's all it took. After it was too late, a "drag race" was surreptitiously arranged, at the proving grounds, between a Chrysler with the "ram" manifold and another identical car with the conventional dual quad setup. The conventional "ran off and hid" from the ram.

But, with the throttle blades upstream of the plenum, the ram did give a very good part throttle "feel," which is why the bean counters were impressed. So, I'm sure there are those who would opt for the ram, even though WOT performance was actually poorer.

As far as WOT performance is concerned, you'd want to tune for an rpm midway through the range seen in an intermediate gear. (Unless, of course, you're building a Bonneville car, in which case you'd want to tune for the horsepower peak.)

Great story! I recall hearing that a long time ago.

Yep, "feel" is very important when selling a design to Management (or bean counters).

At the OEM where I worked, we played with accelerator pedal-to-carb linkage to get the "feel" of more off the line grunt. The first 25% of pedal travel was more like 50% throttle blade opening. WOT was still WOT, so it only mattered at part throttle driving.

Interestingly, a few years ago when VW-Audi went to throttle-by-wire, some folks were selling "chips" that changed the pedal-TB blade ratios for part throttle "feel". Of course, WOT was no different and max power wasn't either. Deja vu all over again, but electronically!

I suspect the overwhelming look of the ram manifold was a great marketing tool. It certainly wasn't cheap or easy to work with.