Before I ask this quesiton I'd like to make a couple things clear...

- I understand perfectly well the concept behind equal length manifolds.
- The runner lengths proposed are mainly due to fitment constraints, not any mathmatical wizardry.
- I'd like answers that are based from either previous experience or manifold theory.

As I was fabbing up a manifold, I came accross a possible runner configuration for the #4 cyl that would save a ton of space, and a couple of els in the process (less material cost). However, it would drastically cut down on one of the runner's lengths compared to the other 3.

The lengths would be about:

4- 6"
3- 12"
2- 16"
1- 20"

It's obvious to see that this manifold is far from equal length, but then again I have yet to see any quantifiable evidence that equal length runners have any real measureable gain in spool or power. I firmly beleive that full-race's equal length VS log test results mainly came from the efficient collector and WG placement.

I'm curious what you guys think about this. I honestly have no idea if this is an issue or not, so if you have any evidence that equal length runners make a substaintial gain PLEASE post it!

TIA,
Bryson

 

I think in turbo applications, the advantages of an equal length manifold are blurred by the simple fact that so much power is inevitably created, therefore people overlook or discount it's effectiveness. When you strive to make power N/A you address thermal efficiency to great lengths, head work, cam science, etc. FI though, there are guys making 500+ on stock heads. Your not scratching and clawing for 7hp like the N/A guys will, and it's the small things that add up in that world.

In N/A applications though, it is undeniable fact that equal length is ideal for proper scavenging and peak output. The more evenly and equally that each individual cylinder can breathe, the more output as a unit they can create. Timing the period where pressure at the exhaust valve is below atmospheric is vital in getting a motor to breathe big at certain rpm points.

Runner length also has an undeniable effect on the powerband of a motor, so if your attempting to tune for peak power, you can only do if all cylinders are able to resonate at the appropriate rpm.

Forced Induction just adds alot of variables that I'm not sure have been truly experimented with. You obviously don't want intake reversion, which is very possible with an FI setup, which is why you see lower overlap cams being common for FI. The less overlap though, the less an equal length manifold will apply in terms of scavenging. In the end, the pressure differential between the intake vs exhaust when all valves are opened is key to purging the cylinder. The manifold clearly has a place in reducing pressure, at certain times, at the exhaust valve upon opening.

I personally think its an arguement that can't be won until someone is able to do an objective test. Noone is really making a turbo mani that I've seen that has runners all within a 1/4" of one another...it'd make for some runner lengths for certain. Compare that mani versus something grossly unequal, with all other variables as constant as possible...maybe it will become clear.

In the end, my opinion is, that regardless of how the motor breathes, all cylinders must breathe efficiently and equally to make optimum power as a collective unit.


Modified by RC000E at 2:49 PM 5/20/2007

 

I'm just really not sure about this setup though. I don't think i've seen a turbo manifold with this much runner length variation. Although, just because I havn't seen it, doesn't mean that it won't work well and make power.

IMO, the most important part of a good turbo manifold is the collector, and WG placement.

Also the whole point of equal length is so the pulses hit at equally spaced intervals on the turbine wheel. I'm curious how the pulses would look like dynamically with this setup?

I wanna just say screw it and try it out, but this manifold is for a buddy of mine. I guess I'm looking for enough support to just try it out.

 

If you are serious enough about it, you can compansate with a custom ground cam, using cam compensation, but like it was already stated to what point. Is equal length more effciant yes, is pulse fired equal length more efficiant yet, yes, but at what point does it matter?

I have built manifolds that have had 6" in runner length differnce, but pulse fired. You are obviously running alot more variation then that. When compared to the ramhorn I was running before or compared to the log manifold before that, this was the differnces I noted:

-Log was always quickest to spool, it pressurized the fastest, but always had the highest Drive Psi, highest oil temps, engine temps egt's ect... most innefciant

-Ram horn (Equal length but not pulse fired) spooled almost identically, to the point that I could show you on a dyno graph, but if I drove one on the street probably not. within ~75 rpm. With the advantage going to my non equal length. Ram horn was smoothest on the dyno while the boost was building, but was choppier and had some creap @ 20psi+ and 6500rpm+ once the gate opened and peaked out ~5hp less.

-Pulse fired unequal length, spooled slightly quicker then the ram horn, ran the lowest egt's and made the most power, though it wasn't enough to actually qualify in my experiance ~5hp peak. This also had the lowest drive psi, and oil temps. I think alot of it had to do with the fact that the waste gate instead of coming 90 degrees off the collector as on the ram horn, split off at about 35 degrees. The other added bonus is that boost seemed much easier to control precicly @ 20psi+ and 6500rpm+

Now to add a cavet, this is all still very subjective, as the dyno runs were not done back to back, but the setup was the same, the dyno was the same, it is indoors, and the temp, humidity ect.. were all the same. I didnt have the time or the patience to tap all the indivdual runners for 02 bungs and run a bunch of differnt runs with the wideband in each cylinder. So take it for what is it worth ~.02c

You need to decide how much effort it is worth, and what you are trying to accomplish. As if you want to get really serious the exahust manifold is just a small part in the system of the process and the intake and cam and combustion chamber, tuning ect... all have a roll to play in how well your exhaust manifold will work, or not work. There will always be something better as that is human innovation. If you have the time and money to push that envelope I apluade you, and will be copying you designs in the future. If not decide what is enough for you, try your design but document the Drive psi, egt's, oil temps, ect... vs this new design, and make your own conclusion, I always learn more by screwing it up, then by getting it right.

Ok I'm done now.

 

What do you mean by pulse fired? I'm assuming paired complimentary runners, with a twin scrowl housing?

Do you have any pics of that pulse fired unequal length manifold?

This manifold I'm building actually has a WG prioritized collector design (the collector points straight at the WG port). I just want the manifold to make good power, be compact, but not cause any unbalanced reversion or knock issues due to the uneven runner lengths.

 

On another note, how come so many poeple think a ram is equal length.

Good topic none the less Bryson

 

Get a custom cam ground? lol right...

Build it, it'll work just fine. We're talking about very small hp changes. Lets not forget that people get by with log manifolds.

If you're concerned with 5-10hp difference between that and an equal length, then don't do it, but i doubt you're in that situation. I doubt there will be any more power loss than that...

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Boostwerks.com &raquo;</TD></TR><TR><TD CLASS="quote">IMO, the most important part of a good turbo manifold is the collector, and WG placement.

Also the whole point of equal length is so the pulses hit at equally spaced intervals on the turbine wheel. I'm curious how the pulses would look like dynamically with this setup? </TD></TR></TABLE>

Well sure, wg placement and collector length/angle/design has alot of impact. These are very obvious design aspects which can make or break a manifold, and these are obvious. There are other not so obvious aspects though, and that's the tuning impact that manifold design can have.

Equal length is about far more than "pulses" hitting the turbine wheel at equal intervals and impacting spool time, etc. This is the area where I think there is some unknown. How does the turbine wheel effect pressure in the manifold, and effect pressure waves inside the runners.

In an N/A setup, when the exhaust valve opens, the pulses that leave the chamber will travel down one runner, then up the others, rarefract and so forth. The theory behind that goes very indepth, tons of which I have read, but cannot and will not lay out here.

The simple matter is, the runner lengths, I feel, will have an effect on cylinder fill. This will have an impact on tunability and where certain lines are drawn. Any variation in how each cylinder breathes, burns, and produces power is going to have an impact on tuning. If you have greatly varying conditions, then you impact how stable the setup will be on pump gas, you impact how much overall power the motor can make.

Would you run a motor with 8:1 compression in one cylinder, 9:1 in another, and 10:1 in the remaining two? Would it run...sure, but can you imagine the limitations that would surface?

Like I said before, treat each cylinder as 4 individual motors. You want to make it easy to manage, so you make all four as closely related as possible. Together, as a unit, they will make more power, and they will require more similar fuel and ignition needs.

This is where I think alot of guys miss out when they just jump into FI. Learning to make power N/A allows you to realize every bit of what it takes to build an efficient motor that makes maximum power.

So, will your manifold work....absolutely. Do I feel that an equal length manifold would outperform yours...absolutely. I think if an objective comparison could be done, the results may surprise people. Just think of the impact cam changes have made for those pushing say 20 psi, 25 psi. I think some substantial gains could be seen, and specifically the limitation on pump being pushed higher as a result of more equal breathing.

 

Pulse fired as in 1&3 2&4 vs 1234 around in a circle, no it doesn't need a twin scroll housing to make a difference. Picture golf ***** going a manifold vs water, as this is a more accurate idea of what is happening. When you have it pulse fired the fire very close to the same time adding scavenging in those cylinders and avoiding revsions. Idealy you have equal length between 1&3 and 2&4 with the differnces in runner length coming between the pairs. If there is too much differnce between them this can actually be a hindernce as they will fire and run into each other. I am no engineer nor do I claim to be, but this concept has been explained to me time and time again by an engineer that worked with the porsche racing team in the 80's, doing headwork, manifold design, turbo development ect... I just grasp the basics.

I had pics but an office fire last year wiped those out. The car has been sold at least 3 times that I am aware of now, but I will write some emails and see if I can track one down.

I took a look at the pics of your manifold, I really like the wastegate location, I might move it up the bend a few inch's and between runners 3&4 a bit more for the most fluid outlet. Good looking manifold. Reminds me of a turbo mustang one I saw a few weekends ago at the track. The collector pipe looks a bit big for a T3 or T4 flange, but pics can be deceving, but as a prototype.

As far as if you will get any bad revsions, or knock, from that design, I couldn't tell you as the math involved is well over my head.

RC - The ramhorn I was running was within 5/8" from equal from the valves... I know truly equal length is conciderd 1/4" or less from valves, but I never have had a race car, so imho it was equal.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by tony1 &raquo;</TD></TR><TR><TD CLASS="quote">Get a custom cam ground? lol right...

Build it, it'll work just fine. We're talking about very small hp changes. Lets not forget that people get by with log manifolds.

If you're concerned with 5-10hp difference between that and an equal length, then don't do it, but i doubt you're in that situation. I doubt there will be any more power loss than that...</TD></TR></TABLE>

Thats exactly what I was thinking. However, i am concerned about the tuning issue that RC000E mentioned. Do you think the length variations might cause a lean or rich cylinder situation?

 

Are we ignoring the fact that one runner significantly shorter than the other will probably expletive the knock limit on the short runner cylinder? If the tuner isn't really on the ball I'd expect a number of "Low compression in cylinder 4" threads.

 

Runner length has much less effect on runner efficiency than actual flow. These manifolds with jacked up collectors, and/or no real collector will effect flow much more than a shorter runner. I wouldn't expect to see much difference in the knock limit on any of the cylinders.

Make your runners smooth, with no cheated bends, and a decent collector and the manifold will make decent power with no issues.

 

I would really like to see a comparison of plugs with dramatically different runner lengths. I could see it easily being a problem and individual cylinder tuning being absolutely necessary rather than strongly suggested.

 

I believe the exhaust manifolds on the GM Pro FWD cars that are going over 200mph are pretty far from equal length.

 

They are pretty close to equal length (3 to 4") longest to shortest.But they are really short.Like 12"to 14" at the longest.But thats only on the FWD. The RWDs and longer and more equal length.But i could be wrong.Hold on I'll walk across the street and see.There is a blue one and a dark red one (cobalt)over there.he he

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by tony1 &raquo;</TD></TR><TR><TD CLASS="quote">I believe the exhaust manifolds on the GM Pro FWD cars that are going over 200mph are pretty far from equal length.</TD></TR></TABLE>

That doesn't mean they aren't needing to pay attention more closely to the individual cylinders. I wouldn't be concerned about power. The power difference will be insignificant. I would be worried about one cylinder acting completely different than another. This is a fixable problem via tuning individual ignition and fuel, but again, less experienced tuners may have troubles doing this and it could result in problems.

 

It would have been very easy for GM to make them equal length on those cars. They have a ton of R&D time into them, and a ton of time on the engine dyno.

I know where you're coming from, i just don't think the difference in runner length does much to the backpressure per cylinder.

Now if you had equal length runners and 3 of them entering the collector normally and one coming into the collector perpendicular to the rest, i think you would definitely see the problem you are describing.

 

Tony the new manifolds are really close to being equal length.The old full race ones were not.

 

Like I stated before, will it work...yes. I am sure there are plenty of examples out there with varying runner lengths due to logistics issues, but the fact is that it isn't ideal. There are combustion studies over 50 years old to support the fact that equal length runners produce power gains, and that resonance tuning is 100% valid.

Highly variable runner lengths will inevitably cause cylinder to cylinder variations in terms of residual gases that remain in the chamber. To say this won't impact multiple tuning factors just isn't true.

A badly designed collector will certainly have effects such as increased backpressure and high turbulence at the outlet, but that's going to impact all four cylinders in similar fashion unlike highly variable runner lengths would.

Cost effectiveness, logistics, and design simplicity have placed some highly unequal runner lengths on vehicles direct from the manufacturer, and it supports the fact that it obviously works. In the end though, if the challenge is to design a manifold for optimum power, I feel its clear that the goal would be to obtain as equal runner lengths as possible.

You (boostwerks) stated this is for a friend, maybe your hookin him up on price so of course you want to soften your costs, etc. It's fine, but as far as that being the final design for sale in general, then I'd personally say no.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by RC000E &raquo;</TD></TR><TR><TD CLASS="quote">
You (boostwerks) stated this is for a friend, maybe your hookin him up on price so of course you want to soften your costs, etc. It's fine, but as far as that being the final design for sale in general, then I'd personally say no.</TD></TR></TABLE>

well the runner would be about 12"s if I made it the other way (exactly like the first manifold I made of this style). Theres NO way that I can change the other runner lengths to get it closer to equal length. I highly doubt the extra 6"s is going to make any affect at all.

What Tony has said really reflects what I've seen as well. If people can make 600whp with a log manifold, that says to me that runner length isn't a huge issue on power OR tuning.

Dynamically, each cylinder is contributing equally to the pressure in the exhaust manifold, as long as each runner enters the collector at the same angle. I highly doubt the length of the runners changes the certain backpressure in any one runner.

I would love to hear what Ben Strader would have to say about this. I'll see if I can email him.

 

I think you'd be best served to do some research on exhaust design and theory. There is alot of information out there that I think would benefit you. If you firmly believe all runners contribute to a given "pressure" in the manifold, and that given various runner lengths have no effect on pressure levels in various runners, that means that you've just discounted plenty of documented research to say otherwise.

All I have to say is, a manifold in my opinion should be judged by more than just what peak number it's capable of making. You think that same log manifold that made "600hp" is making any substantial out of boost torque gains?

Alot of you guys are more concerned with a peak output number than the out of boost responsiveness, drivability factor, transient response, and true usable power curves. In alot of cases these motors are all motor for half of their usable rpm range, having an efficient exhaust design will benefit spool times and usable torque.

There's simply more to this than just what peak number it can make. In that case why not just run logs if they make 500 just as easily?

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by RC000E &raquo;</TD></TR><TR><TD CLASS="quote">
There's simply more to this than just what peak number it can make. In that case why not just run logs if they make 500 just as easily?
</TD></TR></TABLE>

For the bling factor and to be on the bandwagon

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by RC000E &raquo;</TD></TR><TR><TD CLASS="quote">There's simply more to this than just what peak number it can make. In that case why not just run logs if they make 500 just as easily?


</TD></TR></TABLE>

Although I can't say for certain, but like I said...I firmly beleive most good tubular gains can be attributed to the efficent collector and WG placement. Its hard to say how much impact the consistant runner length has on those power gains...unless someone does an unbiased test.

At any rate, I shot an email to Ben. Hopefully he'll be able to shed some light on this.

 

<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by RC000E &raquo;</TD></TR><TR><TD CLASS="quote">I think you'd be best served to do some research on exhaust design and theory. There is alot of information out there that I think would benefit you. If you firmly believe all runners contribute to a given "pressure" in the manifold, and that given various runner lengths have no effect on pressure levels in various runners, that means that you've just discounted plenty of documented research to say otherwise.

All I have to say is, a manifold in my opinion should be judged by more than just what peak number it's capable of making. You think that same log manifold that made "600hp" is making any substantial out of boost torque gains?

Alot of you guys are more concerned with a peak output number than the out of boost responsiveness, drivability factor, transient response, and true usable power curves. In alot of cases these motors are all motor for half of their usable rpm range, having an efficient exhaust design will benefit spool times and usable torque.

There's simply more to this than just what peak number it can make. In that case why not just run logs if they make 500 just as easily?


</TD></TR></TABLE>

By reading the things you have to say i'm going to assume you have read alot, but don't have much real world experience with this stuff....

 

my brother is finishing up his engineering program and working at GE right now...i've built a few things for his car that he is testing. Although this application is a vr6 where the distance from valve to port opening isn't equal amongst all the cylinders due to the offset.

The 1 /3 /5 runners are shorter to comp. for the distance in the head.


Early pics of intake manifold attempting to compensate for the unequal length..


For whats its worth, i told him to just get go 24v where vw incorporated cam compensation much easier for the intake