Dual Runner Lengths

[Zero_12]

I was curious if any teams have recently run dual length runners?

A judge mentioned this idea at competition last May. The question I have about this idea is:
if we spend so much time creating models to match the intake lengths with the exhaust lengths, why would straying away from this designed length improve performance?

From what I think I understand from reading is that by changing these lengths, we may be decreasing torque at that designed RPM, but increasing torque at another RPM. Based on that concept, that's how the torque curve is smoothed out.

Does anyone agree with this logic, or is there some other aspect to intake and exhaust lengths that I am missing?

Thanks

[Zero_12]

I was curious if any teams have recently run dual length runners?

A judge mentioned this idea at competition last May. The question I have about this idea is:
if we spend so much time creating models to match the intake lengths with the exhaust lengths, why would straying away from this designed length improve performance?

From what I think I understand from reading is that by changing these lengths, we may be decreasing torque at that designed RPM, but increasing torque at another RPM. Based on that concept, that's how the torque curve is smoothed out.

Does anyone agree with this logic, or is there some other aspect to intake and exhaust lengths that I am missing?

Thanks

[Trevor]

Teams have done this in the past. Try the find button, there have been a few discussion of variable runners and plenums.

http://fsae.com/eve/forums/a/t...10120821#23410120821 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/48010410821?r=23410120821#23410120821)

http://fsae.com/eve/forums/a/t...10322711#44310322711 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/46310791711?r=44310322711#44310322711)

http://fsae.com/eve/forums/a/t...10325141#98910325141 (http://fsae.com/eve/forums/a/tpc/f/125607348/m/52710754141?r=98910325141#98910325141)

etc.

[Chris_S]

At around 3/4 of the way through this video, there is a short clip of Graz's system on its 2007 car: http://racing.tugraz.at/media/videos/t07-making-of/

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Zero_12:
The question I have about this idea is:
if we spend so much time creating models to match the intake lengths with the exhaust lengths, why would straying away from this designed length improve performance? </div></BLOCKQUOTE>

I think that if you spend the time working with the effects of different lengths, the possible benefits of variable or multiple lengths will become quite obvious.

However, the concept is much simpler than the execution. I'm not sure that this is a project worth taking on for a FSAE team.

[kapps]

Agreed. Depending on how your current dyno setup is, it can be a lot of work just getting ready to tune intake lengths. Myself along with three other guys just finished a Sr. Design project on our intake/exhaust. We spent most of our time adapting the engine to and troubleshooting the dyno. If you think your going to do it all through software, have you already tested to determine how close your software is to actual practice. Using Helmholtz, you'll be within 4 or 5 inches of actual...

[Zero_12]

Thanks for the responses and the links.

To clarify my intial post, I was wondering if a team has run two stationary runner lengths, like say 8" for the middle two cylinder and 10" for the outer two cylinders. I've seen the posts about variable runners. There's no way that our team would be able to pull off that at this point in the year. We also do not have the time to find the optimal runner length through dyno tuning. Most of our time on the dyno will be devoted to fuel and ignition tuning.

kapps,
We are still working on getting our dyno fully functional. Right now, a local shop helps us out by allowing us to use their dyno.
We have used the dyno at competition as a benchmark and used those results to compare and calibrate our models created in the engine simulation software. So far, the location and amount of torque at its peak has been pretty much dead accurate, but the pattern of the torque curve has only been relatively accurate. So we know to take the results produced by the simulations with a "grain of salt". It is this reason that I am wondering if anyone has run dual stationary runner lengths. As of now, our software is saying we should run two different stationary lengths, but I was just wondering if anyone had one this in the past.

Thanks again.

[kapps]

Ah, gotcha. We haven't tested anything like it but it would definitely be interesting to see how it does. I wonder if you would have to worry more about crosstalk between runners with a setup like that.

[Charlie]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Zero_12:
We also do not have the time to find the optimal runner length through dyno tuning. Most of our time on the dyno will be devoted to fuel and ignition tuning.
</div></BLOCKQUOTE>

Then my advice is don't try anything fancy. Different length runners across cylinders would require cylinder specific maps to get the most out of the setup.

When I judge I am the most critical of teams that use unverified software to make decisions. This is one of those cases; you can't verif a slope with a single point and you can't use a single dyno pull to verify a simulation. You need at least one change to verify that the shift in real life and the shift in simulation is the same.

What is keeping you from putting the old car on the dyno at your local shop, and trying a couple more runner lengths?

There are a lot of ways to test empirically, some of them don't even require a dyno. Be creative.

[Zero_12]

Charlie,

I see your first point with the cylinder specific maps. Maybe this is why I don't know of any teams that have run the dual runner length set up. Good point, I hadn't thought about that, and as of right now, I'm not sure if our ECU would able to do that or if we would have the time to properly tune for that.

For the software verification, we have used the results from the 2006 and 2007 competition, so we have used more than one graph to verify the accuracy. In both cases, the peak torque was accurate as well as the the trend of the torque curve. We do need continue to improve the model, but that is an on going task.

I see your point in trying various runner lengths on the dyno. That would not be bad, just simply making a couple of tubes, and maybe an interchangeable intake mount to keep the intake steady on the dyno. I think this may be the best route to go. We have a couple of lengths that we would like to try, and this sounds like something we could get done with a solid weekend of work.

Thanks for your time and helpful responses.

[kapps]

That's basically what we did. Our aluminum runner tubes were left long then cut down an inch at a time after running three or four dyno pulls at each length. The very long one's didn't really show much of a trend but as they got shorter, the real trends could start being seen. Remember, depending on what engine you guys are using, the cam is going to have a pretty big influence on what the curve looks like.

[Zero_12]

Kapps,

I am guessing that you modified the cams on one of your past cars. That was also my understanding that modifying the cams should change the torque curve. We are using a F4i engine. If you don't mind, what engine is your team using?

My question based on your previous post is, did you design your intake and exhaust around the stock cams, or did you design the intake and exhaust, then modify the cams? Depending on which way you did it, is there any specific reason that you went that route?

[kapps]

We also run an f4i. I can't really help with cams though. We run stock cams and, to my knowledge, have always done so. The stock cams come with some serious overlap which doesn't exactly help our restricted engines. Our design took into account the fact that the stock cams are designed for high rpms. This is why I believe our longer runner lengths didn't really show much trends. Actually, the peak torque between our longest and shortest runner lengths were within a lb ft or two. The rpm of these peaks were also within a fairly close range. When we tuned the intake lengths, we were looking maintain a smooth curve for a long an rpm range as possible. The long runners hit the peak and then dropped pretty drastically. I'm not 100% sure if our findings were because of the cam but it's my best guess.

[Wesley]

In my opinion, cam geometry has way more effect on power output than any runner length changes. However, it's also harder and more expensive. You can't buy racing cams, because they usually make the problem (overlap) worse. It seems that the "power wall" most teams run into without cam work is about 80HP.

We did some studies on advancing the intake cam to reduce the valve overlap, but the gains were minimal, 4HP or less predicted at only some of the RPM range. What I felt would be more appropriate would be a late opening, higher duration intake cam with a decreased duration exhaust cam. Since the bottleneck in these systems is the intake system, cam geometry should be all about maximizing cylinder filling. Evacuation is not really the problem. This would definitely help with the low end, but it's somewhat of a sacrifice with the high end, since you lose that scavenging effect due to the overlap.

I wish we had a grinding post and the time and know-how to develop proper ramp angles so we could just grind and test cams all day.

[kapps]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Wesley:
It seems that the "power wall" most teams run into without cam work is about 80HP. </div></BLOCKQUOTE>

Yep, thats exactly what we found as well. We might try some adjustable cam gears but the real change would be with a different cam. In these cars, I feel it's more about having the broadest torque curve rather than a big horsepower peak. Unless you have some darn good drivers and a good shifting mechanism, it's going to be hard to keep the engine where it's making power.