uneven air flow and detonation

[rossman]

I've had this theory churning in my head for a while now and I thought I'd share it to get some feed back. I am sure someone else has discussed it before but I couldn't find anything specific to this subject as it relates to detonation.

 

The theory is that uneven air distribution caused by stock or stock like intake manifolds setups with the TB @ 90 degrees relative to the intake runners, could be a significant contributor to the detonation problems with the L6, especially boosted L6's. I was looking at the CFD work that MONZTER did on the stock style intake manifold (results here). It appeares to me that while under boost, cylinders 4 - 6 received more air than 1 - 3. Assuming that all of the fuel injectors are all dumping the same amount of fuel then that would lead me to believe that cylinders 4 - 6, especially 5 & 6 are running quite leaner than 1 & 2. Combine this with poor coolant flow in cylinders 4, 5 & 6 (as discussed here) then you get a serious recipe for detonation.

 

Thoughts?

[WizardBlack]

I've had this theory churning in my head for a while now and I thought I'd share it to get some feed back. I am sure someone else has discussed it before but I couldn't find anything specific to this subject as it relates to detonation.

 

The theory is that uneven air distribution caused by stock or stock like intake manifolds setups with the TB @ 90 degrees relative to the intake runners, could be a significant contributor to the detonation problems with the L6, especially boosted L6's. I was looking at the CFD work that MONZTER did on the stock style intake manifold (results here). It appeares to me that while under boost, cylinders 4 - 6 received more air than 1 - 3. Assuming that all of the fuel injectors are all dumping the same amount of fuel then that would lead me to believe that cylinders 4 - 6, especially 5 & 6 are running quite leaner than 1 & 2. Combine this with poor coolant flow in cylinders 4, 5 & 6 (as discussed here) then you get a serious recipe for detonation.

 

Thoughts?

 

Absolutely. Per cylinder air flow variations can be noticeable; even on modern engines. I've no idea what a stock L intake varies by, but I am sure it's a contributor to some degree.

[ktm]

I agree, but in the case of the L-series it has been widely shown that cylinders 5 and 6 detonate before the others, all else equal. This is due largely to the cooling issue has discussed in the sticky at the top of this forum. The unequal distribution of airflow may be a contributor as well, since we are tuning to an average AFR (if running one wideband).

[kiwi303]

Has anyone made a 1-6 intake header like a 6-1 exhaust header and mated that to a TBI with the injectors in the throttle body? Long runners compared to a manifold, but equal air distribution?

[Robert]

If this is true, then a very rich mixture could help on knock.. I have the problem myselfe.. Interesting. But was this test performed on a running engine? Im just thinking, If its just the intake he tested, then this has little interest, cause if nothing stops the flow, theres no pressure build up in the manifold, and ofcourse #1 gets less then the rest......Is there a link to the rest of the testing? Or maybee a solution? First thing I could think of was cutting the fuel rail in half, and increasing fuel pressure on #4-6 compared to 1-3...

[rossman]

If this is true, then a very rich mixture could help on knock.. I have the problem myselfe.. Interesting. But was this test performed on a running engine?

 

It was just computer simulation.

 

Im just thinking, If its just the intake he tested, then this has little interest, cause if nothing stops the flow, theres no pressure build up in the manifold, and ofcourse #1 gets less then the rest......

 

I'm not sure what his boundary conditions were but pressure will build up due to flow restriction.

 

Is there a link to the rest of the testing? Or maybee a solution?

 

I pulled that link from MONZTERs Turbo intake plenum thread. I believe he just ran the Cartech one for comparison to his and the HKS intake.

 

First thing I could think of was cutting the fuel rail in half, and increasing fuel pressure on #4-6 compared to 1-3...

 

Ideally you would want EGT sensors on every cylinder, otherwise you are kinda shooting in the dark.

[WizardBlack]

As ktm has said, the biggest issue on the L series specifically is no doubt the cooling issue. I would address that first if you want to correct issues with the L. Most cars don't have that severe of a preference as to which cylinder suffers damage first.

I have also implemented the cooling bypasses (again, as found in the sticky) and found them to be very helpful. I have yet to change the plenum on my L-turbo; although it's on the list.

[WizardBlack]

If this is true, then a very rich mixture could help on knock.. I have the problem myselfe.. Interesting. But was this test performed on a running engine? Im just thinking, If its just the intake he tested, then this has little interest, cause if nothing stops the flow, theres no pressure build up in the manifold, and ofcourse #1 gets less then the rest......Is there a link to the rest of the testing? Or maybee a solution? First thing I could think of was cutting the fuel rail in half, and increasing fuel pressure on #4-6 compared to 1-3...

 

Fuel is not the proper solution to make such gross changes in cylinder cooling. Varying fuel has actually been done on a few cars that I am aware (at the commercial level), but that is to correct intake manifold imbalances; not coolant flow imbalances.