Found the backfire/misfire problem!

[rickyll7]

The engine misfires and backfires completely go away when I disable the Kenne Bell bypass valve (keeping it closed). Why is that?

My theory is this: The Kenne Bell bypass valve design is starving cylinder number 5 of air at low loads, and the engine is still giving that cylinder the fuel for the amount of air it thinks that cylinder is getting. So, it misfires nearly every cycle and backfires randomly as well. How can that be you ask?

When the bp valve is open and the engine is at low loads, lots of air is going through the blower and little is needed in the engine, so almost all of it ends up going back through the bp valve to re-circulate through the blower and that is exactly what is supposed to happen, right? Of course, but...

Here is the problem: The KB bp valve is located in such a way that it draws air straight out of the intake runner for cylinder number 5. It draws air from that intake runner and that intake runner only because that intake runner is where they put the routing to the bp valve. So what, why is that bad?

Here is why: As stated above, at low loads, very little air is going into the engine, and a lot is re-circulating around the blower through the bp valve. That means that air at a high velocity (relative to that of the air entering each cylinder) is entering the intake runner of cylinder number 5 and running straight past the intake port and into the bp valve. Anyone know what happens to pressure as velocities increase? That’s right... the pressure goes down - vacuum.

What happens then you put more of a vacuum on one cylinder than the others? It gets much less air than the others and the ecu doesn’t know about it, so it puts in the same fuel as all the other cylinders. That cylinder (#5) then misfires, the driver’s side o2 sensor sees excess o2 and adds even more fuel to bank 2 making a bad situation worse because now all 4 bank 2 cylinders are running rich.

Does this make sense or am I crazy? Is there something else I’m missing?



Info I gathered before discovering the problem goes away with the bp valve closed:

Only backfires through exhaust. I can hear it poping well from under the car, but not really from above the engine.

It only does it at low load such as when the engine is slowing down or if I hold it anywhere above 2000 rpm without load.

It won’t do it at idle, during acceleration, or while cruising.

It happens regardless of ECT and ACT.

It is random/sporadic - not rhythmic or synced to engine speed.

I've replaced spark plugs, wires, IAC valve, and the throttle position sensor.

Vacuum is at about 15 inches of mercury at idle.

I've checked for vacuum leaks by spraying engine starter fluid around the manifold and vacuum lines and didn’t hear any rpm fluctuations.

It also does it when I run my old injectors with my old tune, which never had problems before the blower install.

The AFR fluctuates between about 13:1 - 17:1 while it's happening and only when it's happening.

Timing advance is about 40 degrees when it's happening at 2000 rpm.

[RWTD]

I will say this, it is a known issue on some of the KB S197 systems that the bypass arrangement will cause 1 bank to ingest differing amounts of airflow, thus causing differing fuel trims (generally 5 to 10% different). I'm not sure if Whipple has this problem or not, but I believe I read some do.

[rickyll7]

So what can I do to fix this? Can I run the engine with the bp valve disabled? I noticed some of the earlier kb kits didnt have bp valves at all, like the 96-98 cobra kits.

[rickyll7]

I will say this, it is a known issue on some of the KB S197 systems that the bypass arrangement will cause 1 bank to ingest differing amounts of airflow, thus causing differing fuel trims (generally 5 to 10% different). I'm not sure if Whipple has this problem or not, but I believe I read some do.

If an entire bank of 4 cylinders is 5-10 percent richer due to one cylinder being different, wouldn't that mean that the one cylinder is 20-40 percent richer than the other 7?

AFR=14.6, Lambda=1

[1+1+1+(.8 to .6)]/4 = (.95 to .9)

lambda = .6 to .8 would wash that cylinder wall, right?

[rickyll7]

It seems to me that I've only got a few options here and they are all going to compromise something (thanks a lot KB).

Option 1:

Tune it with the bp valve closed to get the A/F where it is supposed to be. Then let the bp valve open as it should, but disable closed loop and learning or find some way of getting the engine to ignore bank 2 (as stated above).

Here is the problem with that (Option 1):

If I tune it to within 2% commanded AFR with the bp valve closed, everything works great until I let the valve open. When that happens, the misfire and backfire return and the wideband shows anywhere from 5-10% leaner (the WB is in the driver’s side, bank 2). Now I leave closed loop and learning disabled for now to keep things from getting any worse. 5-10% doesn’t seem to bad right? After all, it's only at low loads, right. Wrong, it's not ok. Here is why (correct me if I'm wrong):

The WB is reading 5-10% leaner on bank 2 (4 cylinders total) due entirely to cylinder number 5 misfiring. That 5-10% extra o2 is coming from cylinder 5 only. That means cylinder # 5 is running off by as much as 20-40%. If it is getting so little air while still getting all the fuel that it misfires 20-40% of the time, it must be running terribly rich and that will eventually wash that cylinder wall, I think, but correct me if I'm wrong.

[1+1+1+(.6 to .8)]/4 = .9 to .95

Maybe I'm wrong, but that’s how I see it.

Option 2: Completely disable the BP valve. KB has a kit for the 96-98 cobras with a 2.1L and no BP valve and no intercooler. They allow it to run up to 11psi at 13000 rpm (blower speed, engine speed is 6500). If it works there, it should work for me. With a stock crank pulley and a redline of 6000, I can run down to a 3 inch pulley before I hit 13000 rpm.

Option 3: Use some sort of mechanical stop to prevent the bp valve from opening beyond some point at which the engine still runs without issue. This still allows some air to bypass, which I'm sure we can all agree is better than none. Also, some OEM blowers bypass air before it can run through the IC, so they are only re-circulating the same hot air. The KB kit re-circulates air after the IC, so If I can get at least some air to bypass, that could be better than the OEMs that are bypassing more air, but hot air.


I'm obviously leaning towards option 3, but I'm open for suggestions as I'm sure I'm not the first one to have this issue as it seems that KB designed this flaw into their 99-04 GT kits. Perhaps they have some sort of way to “tune out” the issue. Although, I don’t see how they could unless they are able to shut the fuel off for that cylinder. Even that isn’t good for the engine.

[rickyll7]

I just got off the phone with KB and they say the bypass valve design only causes problems on non-stock cars (mostly a cam overlap issue). They say that their kit was designed for stock cars and works well with stock cars. I guess that’s reasonable and is only fair. They suggested I call Tim at MPH and ask him about it. They say he has developed some cams that supposedly work well with the kit.

[Billf6531]

I haven't seen this s/c, but is it possible to relocate the air intake to the bypass valve so that it doesn't draw from the #5 intake runner, and instead draws from the upper part of the intake?

[rickyll7]

I haven't seen this s/c, but is it possible to relocate the air intake to the bypass valve so that it doesn't draw from the #5 intake runner, and instead draws from the upper part of the intake?

I guess anything is possible, but it wouldn't be easy. It would take some real in-depth fabricating with the manifold off the car. It seems like a good place would be behind the intake plenum and inbetween the intercooler hoses.

This would be a last resort for me.

[BlueStreak]

i like the sound of blocking it off and relocating it much more than pulling the engine and having to rebuild it in a few months =/

[Mep_q8]

any update to this? im running into the same problems

[Skud]

I've had this issue as well and unfortunately KB is content to not resolve the issue because it works on "stock" cars and Tim @ MPH can tune the issue out. Doesn't make much sense to me that they design a supercharger that only works on a stock car. That's somewhat of an oxymoron if you ask me. I think it boils down to cost. They would have to re-design the bypass and as long as they can pass the buck on "non-stock" camshafts they will continue to do so.

As said above, Tim also has special cams for the 2v Kenne Bell cars. I have a set and they do make the car quite drivable. I have very little popping and a little misfiring but the bank to bank trim differences are still there. Also, the car loads up at idle. Apparently, Tim is able to tune that out as well.

I got in contact with Tim and I guess at one time he did tune these cars over the phone, but due to the time it takes he will only tune if you bring him the car. That's definitely his choice and I respect that. I would have my car there in a heartbeat if I didn't live 10+ hours away. I'm pretty angry with KB that their solution is for everyone with this issue to take their car to Tim or use "stock" camshafts. These little morsels of information are curiously left out when you purchase the kit. I wonder why.

Apparently, there used to be a good thread over at modular fords in the KB section dealing with all of these issues. It somehow got deleted though.

Riley

[03StangWV]

Did you ever get this problem resolved? I have a 2.1L Kenne Bell on my '03 GT and have the exact same problem. I'd like to be able to fix it without having to relocate the bypass valve. Possibly through tuning. It's ridiculous that Kenne Bell hasn't changed this design, knowing its a problem.

[Eric@HPTuners]

Did you ever get this problem resolved? I have a 2.1L Kenne Bell on my '03 GT and have the exact same problem. I'd like to be able to fix it without having to relocate the bypass valve. Possibly through tuning. It's ridiculous that Kenne Bell hasn't changed this design, knowing its a problem.

Yes. There are two options. You can just put the car in open loop permanently, or what I like to do is make the car run off one 02 sensor, on the bank opposite of the bypass valve.

[brightblue00gt]

Yes. There are two options. You can just put the car in open loop permanently, or what I like to do is make the car run off one 02 sensor, on the bank opposite of the bypass valve.

Are these tuning option or can open loop be the same as disconnection the vacuum line from the bypass valve? I assume the 02 sensor option is tuning only? A MIL Elim on that banks sensor wouldn't work right?

[Eric@HPTuners]

Both options I mentioned are things to be done in the tune.

[brightblue00gt]

Both options I mentioned are things to be done in the tune.

I think my 'tuning' options are limited, because KB sent me the chip with the blower when they shipped to me. I'm stationed in Korea with my car. KB burned a chip and shipped it and the blower to me over here and I installed it all myself, as I do all my mods/parts. I've asked other tuners and they all say the same thing... can't do anything with the KB chip and it's best to keep using it only?

On another note concerning the bypass valve, it's either Open or Closed right? Open at idle and partial throttle, and Closed in boost? Can it be partially Open vs Full Open and still perform it's function while not hurting the blower? What happens if the valve ran at only half open?

I've had the same problem with the "unplug the vacuum line" (closed) and all the popping, backfires and hesitation goes away, but with the bypass valve connected up all those problems exist. I just live with it. But, I've noticed I can unplug the vacuum line and operate the valve by hand and at about half open those symptoms are very unnoticeable. I just don't know if its safe on the blower to operate that way.

Thoughts?

[Eric@HPTuners]

I think my 'tuning' options are limited, because KB sent me the chip with the blower when they shipped to me. I'm stationed in Korea with my car. KB burned a chip and shipped it and the blower to me over here and I installed it all myself, as I do all my mods/parts. I've asked other tuners and they all say the same thing... can't do anything with the KB chip and it's best to keep using it only?

On another note concerning the bypass valve, it's either Open or Closed right? Open at idle and partial throttle, and Closed in boost? Can it be partially Open vs Full Open and still perform it's function while not hurting the blower? What happens if the valve ran at only half open?

I've had the same problem with the "unplug the vacuum line" (closed) and all the popping, backfires and hesitation goes away, but with the bypass valve connected up all those problems exist. I just live with it. But, I've noticed I can unplug the vacuum line and operate the valve by hand and at about half open those symptoms are very unnoticeable. I just don't know if its safe on the blower to operate that way.

Thoughts?

I have only experienced the bypass valve issue on cars with aftermarket cams. Odd that you are having it on an stock engine, I am assuming?

Correct on the opening and closing of the valve. With the bypass valve barely open it will not make full boost, and will increase air temps during part throttle driving.

The absolute easiest thing to do is put the car into open loop, so the skewed 02 sensor readings do not affect how it runs.

[brightblue00gt]

The absolute easiest thing to do is put the car into open loop, so the skewed 02 sensor readings do not affect how it runs.

And to do this is only through tuning? I don't suppose a MIL Elim works on a front 02 sensor right?

And correct, no cams. Lots of mods, but no cams or internal changes.

[Eric@HPTuners]

You could unplug the front O2 sensors, which will force it into open loop, but you will have a check engine light.

[caniggia]

You can try to install lambda sensor emulator, so it will emulate always stechiomtric, so most likely nearly like going OL, but no mil light, also no mixture control.
Second option is to install innovate lc1 in good working correct bank and direct 0-1v output to wrong 02 sensor cable.
Second option is better, due it will still control mixture and it's not expensive.