Exhaust Systems

[smsazzy]

What is the reason I would want a 4-2-1 exhaust system?

Why not a 4-1 collector?

What benefit am I getting? Am I giving something up? Has anyone done any back to back dyno testing of various systems? I have reason to believe my system may not be extracting maximum power from my motor and I want to understand why.

[brian]

4-2-1 or merge collectors are designed to enhance mid range power with a minimum effect on the high end. These systems are tuned to specific engines and sometimes do not work as well on other configurations. Years ago we found that the exhaust systems available in vees had this trait. Those that were developed with say a Mid-west engine didn't work as well on a West coast engine. I'm not saying one engine is better than an other; it's just that the exhaust systems are part of a system that includes the entire induction end as well. While our carbs and manifolds are very similar, the same cannot be said about heads. Sometimes, Joe Blow's heads don't like BIlly Bob's headers.

My dyno results show an improvement in the 4200 to 5000 rpm range with a merge but usually at some cost at 6000. Your results may vary.

[SR Racing]

As Brian points out the intent is to broaden the HP curve. This is typically done by tuning the primaries of 2 cylinders at 5200 RPM (e.g.) and the other 2 at 5400 RPM. (e.g.)
So the peak HP instead of being at 5300, would now ideally be from 5200-5400. Also, in theory, that peak MIGHT be a bit lower.

Since FV has no real good capabilty to match gear ratios to a particular track, a 4:2:1 system makes good sense in (in theory). Those cars with quick change close ratio transmissions, adjust the gear ratios to make sure they are in the best portion of the HP curve in their passing zones at each track. We can't do that.

[smsazzy]

Would it not be a better idea to have two or three sets of primaries to use on a given track?

[neilcox]

Oh Boy! Here we go, time for Spec Exhaust!

[smsazzy]

Not sure where that came from.....

I have also seen little primary extensions. Do those work? Can you really "tune" the HP to match the track by changing the length of the primaries?

[FV80]

Yes - you can use those short extensions to LENGTHEN your primaries .. but they aren't much good for making them SHORTER
SOME times, you can tell a difference ... SOME times you can't. I think there is a LOT more going on than anyone knows and it's pretty much impossible to measure more than 2 or 3 of the ... maybe 20 variables that might be playing here.
On track experimentation can tell you SOME things ... but it's VERY easy to be swayed by a 1 MPH head wind that changes to a 2 MPH tailwind or vice versa. It's EXTREMELY rare if not impossible to have NO WIND at all.
Steve

[brian]

I use extensions to lengthen the primaries on shorter tracks. The diffference between primary lengths and merge collectors is that the primary length shifts the curve rather than, as Jim says, broaden the curve. Those blessed with quick change trannies match their gear sets to the track in question. Most DA systems can generate reports on rpm that track the amount of time spent at specific rpm ranges to aid in matching track rpm to the most favorable range in the engine.

[smsazzy]

Yes - you can use those short extensions to LENGTHEN your primaries .. but they aren't much good for making them SHORTER
SOME times, you can tell a difference ... SOME times you can't. I think there is a LOT more going on than anyone knows and it's pretty much impossible to measure more than 2 or 3 of the ... maybe 20 variables that might be playing here.
On track experimentation can tell you SOME things ... but it's VERY easy to be swayed by a 1 MPH head wind that changes to a 2 MPH tailwind or vice versa. It's EXTREMELY rare if not impossible to have NO WIND at all.
Steve

Maybe the trick is to start with the shortest practical primaries and havwe a few different extention lengths.

[brian]

Steve makes a great point about controlling the variables; it's virtually impossible. If there's something I want to test I try to arrange a change in the pit lane and use my data acq. afterwards for comparison. Two examples of things I have tested this way are noses and collectors. I find myself amused when reading quotes regarding 1hp test results. With the posible exception of SR's Super flow, 1 hp is within the range for error on virtually all of the dynos in our class.

I know that generally speaking, longer primaries will help the low end power. Smsassy is right as well. I start out with quite short primaries and have extensions to match the shorter tracks. Frankly, i have never heard of different lengths used to spread the curve but it sounds interesting.

[jpetillo]

I'd never heard of using different length primaries to spread the curve, either. The whole point of the 4-2-1 is to do just that. Basically you have two resonances at different frequencies, so you will lose some peak power in exchange for a broader spread. If you use different length primaries on two of the four cylinders, you have two more resonances. That can broaden the curve still more, and lower the peak power a bit as well. You can then go to all primaries having different lengths, but I think you'd have hit diminishing returns by then. John

[fvkartguy]

Now my question would be, why have I never seen a drag bug with the 4-2-1 exhaust?
I've talked to a drag racer before and he just had the 4to1 collector because everyone else did and no one bought the whole broadening the curve thing...
Is it that they're working with a much smaller rpm range or something along those lines?

[SR Racing]

Yes. Different length primaries are how you broaden the peak HP and a 4:2:1 does exactly that. I have not seen it (different length primaries) done on a 4:1 but I assume it could be.
In the same vein (scavaging) we were doing some experimenting on the chassis dyno several months ago and found something interesting. Taking the engine at full dyno load to the peak HP point (5200+) we held it there and aimed a leaf blower at about a 45 deg angle across the tip of the tail pipe.We measured the velocity of the leaf blower at about 70mph in a 3 in diameter. INSTANTLY the HP jumped by more than 1+HP. We did this several times and it was like clock work. Now remember that on the chassis dyno there is no scavaging behind the car normally since you aren't moving. On the track there is already a low pressure area behind the car so MAYBE this effect can't be put to practical use. But it would seem that if your tail pipe extended out to the side with the boundary air off the side of the car hitting the pipe at an angle you could get some additional scavaging. A dyno in a wind tunnel would be the only way to really verify it.

We did this as we were playing with seeing how much frontal boost to the carb helped HP. It does. However on the track aerodynamic boost costs frontal area loss. So again only a dyno in a wind tunnel could verify the trade-off advantage. (or a good long day of testing on the track with D/A)

We don't keep secrets here on our FV engines. (at least not many).

After our 6 axis CNC and digitizing machine comes for building heads and block from billet we will order a moving road wid tunnel with built in dyno. <G>

To stretch the post we just completed an engine for a Bug customer (1900cc and 2 2bbl webers and a very mild cam.) 160HP and it idles like your grandmas Buick. The first ACVW I would ever want in my street car. <g> (Other than the 960HP street gas Viper V10 we did a few months ago. <g>)

BTW, if one of you guys, builder or otherwise come up with some interesting ideas you would like to test, give us a call. If it looks interesting to us, we may be able to work out something to both our advantages.
If it is something only for your purposes, either dno can be rented with an operator for the day or by the hour.

[SR Racing]

Now my question would be, why have I never seen a drag bug with the 4-2-1 exhaust?
I've talked to a drag racer before and he just had the 4to1 collector because everyone else did and no one bought the whole broadening the curve thing...
Is it that they're working with a much smaller rpm range or something along those lines?

Exactly! They know exactly what RPM range they are in, they can use any clutch, gear set etc to keep the engine right at the peak HP point for most of the run. They are always running a 1/4 (1/8) mile with few unknowns. There is no reason to compromise with a 4:2:1. We could also do this if we always ran on the same road course and had close ratio gearing that we could change.

As far as "not buying the broadening curve thing". That is NOT an urban legend. It certainly does broaden the curve. It is just a matter if you need it or not. Remember you are sacrificing some PEAK HP to do this.

[smsazzy]

Measured my primaries today, the 1 and 2 cylinders are 42.5 inches, the 3 and 4 cylinders are 40 inches. Is that a good thing?

Am I the only one that thinks that is too long? Especially since they are stepped headers? Doesn't the step system effectively lengthen them even more? I would think this setup is losing power at the top end.

I also do not understand the different length primaries. Why wouldn't I want them equal length? I thought that was the point of headers; to have them equal length to maximize scavenging.

[smsazzy]

Also, after the 4 into two, how long should the runner be to the 2-1 combiner? What effect does that length have?

[DanRemmers]

My solo vee (1600) has primaries that are about 48" long, with a 4 in 1 collector. (I hope it helps with the low end.)

I assume you should measure the center line of the pipes. Although, if you want to be precise, I guess a volume measure would be more exact.

Is it legal to have cockpit adjustable exhaust? Just shorten up the primaries on the straights, and lengthen in the corners.

[smsazzy]

I measured the center lines.

Brian Harding made an adjustable exhaust system. I think it was hard to keep alive due to the heat.

[SR Racing]

Measured my primaries today, the 1 and 2 cylinders are 42.5 inches, the 3 and 4 cylinders are 40 inches. Is that a good thing?

Am I the only one that thinks that is too long? Especially since they are stepped headers? Doesn't the step system effectively lengthen them even more? I would think this setup is losing power at the top end.

I also do not understand the different length primaries. Why wouldn't I want them equal length? I thought that was the point of headers; to have them equal length to maximize scavenging.

If they are all equal length the MAXIMUM scavenging effect takes place at a specific RPM. The logic for 2 lengths is that you will have 2 peaks close together. These peaks will broaden the RPM range that maximum HP is made. That maximum MAY be less than the single length headers peak HP, however since it is broader it may be the optimum to use on some/most tracks. In regards to your headers I have seen many close to that length. Again it is a compromise that you have to test. On a track where you spend a lot of time at 5800 RPM you would want a different set from where you spend a lot of time at 6300 RPM. You could certainly build an ACVW to max HP at 4800 and pull everyone out of the corners and then watch them leave you at 5200 RPM. YOU have to figure all this out on the track. Same issue as with a long box vs. short box, 2 degree vs. 4 degree cam retards etc. If you had an infinetly variable drive ratio this would be a no brainer and you would tune everything for peak HP.

[DanRemmers]

The logic for 2 lengths is that you will have 2 peaks close together. These peaks will broaden the RPM range that maximum HP is made.

That's how a piano works. There are two or three strings that are tuned slightly different, and the result is a louder and longer sound.

[brp]

There are other reasons for having different length primaries. Several of the systems I have built had a different length for the front two pipes versus the rear two pipes.

The testing that I have done also indicates that the 4-2-1 collector does not necessarily hurt the top rpm power. Some have shown even more power at the top compared to a 4-1 collector. Of course there is a bunch of experimentation and work done on the 4-2-1 collectors to achieve this.

If you want to try some different things, be sure and have your pieces ready and try them on an engine dyno during the same half day session and go back to the first set of tests at the end to verify that the conditions didn't change enough to impact the test results.

My 2 cents,
Mike B.

[AJP]

That's how a piano works. There are two or three strings that are tuned slightly different, and the result is a louder and longer sound.

Hmm........I always knew my engine sounded like a Steinway!

-A