Intake manifold runners flow equaling?

[SilverXJ]

I have been thinking about equaling the flow on the intake runners on my 99+ intake manifold. However I tested a stock 99+ 4.0L intake using a manometer and a shop vac. While it isn’t 100% scientific it still shows difference in flow. Previously I thought all the intake runners flowed the same. However what I saw was that the intake runners flowed different in a mirrored fashion. Runners #1 and #6 had the lowest reading (more flow), followed by #2 and #5, and #3 and #4 had the highest reading (less flow). This also correlates to the bosses under the injectors on the intake manifold with #1 and #6 having the smallest protrusions into the intake runner and #3 and #4 with the largest protrusions in to the intake runner. I tested it with all runners closed off but the one I was testing (pulling air through the throttle body opening only) and with them all open. Both tests show the same pattern. The test was setup like this: intake manifold --> intake runner --> manometer --> show vac.

Is this designed with that flow pattern for a reason? Especially since the #3 and 4 runners have the largest fuel injector boss protrusion in to the runner, and #1 and 6 have the smallest.

I would just bee removing material from #3,4,2,5 runners to equal the #1 and #6 runners, but with what I found perhaps it shouldn't be modified in this manner.. especially if that flow was engineered that way.

-Chris

[Mgardiner1]

The ECU reads "global" sensors, and does not monitor A/F ratio per cylinder. If the average for all 6 cylinder is showing a good A/F ratio in the exhaust pipe, it will inject the same amount of fuel to all 6 cylinders, but the computer is unaware how much air flows into EACH cylinder.

My vote is for as close to equal flow accross the board.

[SilverXJ]

Yes, I know that the PCM can't read each cylinder. However, what I don't understand is the pattern that I see. Especially since from the way the fuel injector bosses cut into the runner it is almost designed to flow that way for a reason... one which I can't under stand unless I am missing someing like air pulses or soemthing.

[1bolt]

Interesting I'm thinking that those protrusions and added "pinch" in the intake runners have more to do with air velocity and less to do with total CFM (volume and velocity).

The speculation in past threads about these differences between the runners has been that they increased the injector boss size on those runners to equalize CFM. But those inner runners (3 and 4) have more "right angle". Because of the need for equal length but short path to do it in; they put an S-curve in these runners and made them meet the plenum in a right angle to get the length right. That also should make them the slowest runners. So they probably added the "pinch" (look at the pic below) and larger injector boss restriction to increase air velocity (venturi effect) right before the injector.

These are pics from an older thread we had about this:



This makes a lot more sense now. Last year when we were talking about this, something was bugging me about the idea of some Chrysler engineer optimizing that manifold so much for max flow CFM... Not something OEM's tend to do being that they are more concerned with lower lift flow and almost never optimize for WOT performance! Especially not Jeeps.

Our strokers shouldn't get anywhere close to maxing out the CFM of the 99+ manifold. So balancing absolute peak CFM numbers is pretty unimportant, however having balanced velocity from port to port will be vital to making the best torque throughout the RPM range through low lift inertial ramming.

There are definitely improvements possible. For starters I'm sure that the OE engineers just got the velocities matched "in the ball park" and not fine tuned. They probably optimized for 1000 to 3000 RPM's, so opening up things mildly but keeping the same proportions could bump the power band up higher in the mid and high range. If high RPM is the goal reworking it agressively for best balanced MAX CFM is certainly doable However prepare to sacrifice torque through the band (less inertial ramming from loss of velocity) for added HP at peak RPM's.

Either way this is the kind of thing you would want to do with a flow bench and velocity probe.

[SilverXJ]

The CFM on the shop vac I was using I don't know for certain, but from their current models it was pulling 130- 210 cfm ( I don't know the exact HP of the one I used.. its across town and I don't feel like driving there now). Maybe I noticed so much change because I was pulling more CFM than the engine actually needs per cylinder, i.e. over working the intake runner. As I said, this wasn't a very scientific approach.

BTW, this whole thing started when I pulled my exhaust manifold off of the engine and one cylinder looked to be running lean.. viewtopic.php?f=15&t=458.. That is on a manifold I possibly modified wrong and created too much flow on that one cylinder... That isn't the stock intake manifold I was talking about earlier. It also could be a fuel injector issue, but looking at the test results it is the manifold... the injectors have been sent out for testing though.

So now I have the option of using the completely stock manifold or modifying the already modified one in such a way that it shows equal figure over all the runners.

I also posted a similar question on the Hesco website and this was Leee's reply over there:

When the new style intake was developed the flow was done on a flow bench that pulls air thru all ports at the same time and mano readings were taken from holes drilled along the runner. The flow at this time was designed more for low end and mid range flow.

When the intake was run on the dyno the EGT readings are close enough not to warranty a injector difference.

With that said the ACELL DFI kit has a map to correct injector flow for each cly. The only problem is when the injector is dialed in for WOT it is different at idle and cruise.

The answer to this problem is a TRUE IR intake that has a butterfly for each port. Then the MAP is a problem as there is one one MAP sensor and now we need 6 readings. WOW!!!

Bottom line is you can make some of the cly happy some of the time but never make all the cly happy all the time!

[John]

The new intake was another effort to address the low end torque loss with the changes in the 4.0 vs the 258. The changes you are talking about, could be done and the runners balanced, as 1bolt said on a flow bench,
You could cut for WOT, or to shift upward, the power band. I only clean the castings of imperfections from manufacturing. Same technique as basic head intake porting. This cuts really fast................
John

[Comanche91]

Here's an older post showing how one guy tried to equalize runner flows:

CLICK Here

[SilverXJ]

I saw that. I did my testing similar to his, but I didn't place it in the injector boss.

[SilverXJ]

Instead of doing anything useful today I decided to do some grinding. BTW, its a Rigid vac and only pulls 187 CFM new. Here are the results:


Again I did the test with all runners but the one I was testing blocked off (only pulling through the throttle body port).. and one with all the runners open. The Injectors holes were plugged. I'm fairly close.. looks like # 1 and #4 could use some extra work. However, with all the work I put in the other runners, the #5 one wasn't modified that much and still retains most of its injector boss, but yet it has less resistance than the others... and that doesn't make sense.

[SilverXJ]

I think I finished fooling with it today.

Final numbers:
1 - 9.25 in/h20
2 - 9.25 in/h20
3 - 9 in/h20
4 - 9.25 in/h20
5 - 9.25 in/h20
6 - 9.25 in/h20

Those set of numbers are higher than the previous set because someone cleaned the vacuum at work, which included blowing out the filter on it. Those numbers are from the test with all but the runner I was testing sealed off, only pulling air through the throttle body opening. I saw no point in testing it with all the runners open

Also, for reference these are the numbers from a stock manifold.
1 - 10.75 in/h20
2 - 10.5 in/h20
3 - 11.875 in/h20
4 - 11.875 in/h20
5 - 10.5 in/h20
6 - 10.75 in/h20