AFM Pull Through Stutter / Hesitation

[ianwood]

Following up from a thread a while back on our shift hesitation ( http://www.mye28.com/viewtopic.php?p=787770 ). We switched to a pull through setup to eliminate boost leaks. Good news is the boost leaks are gone. Bad news is we get stuttering / hesitation on shifts. I suspect part of the flow of air from the BPV is going against the intake flow and pushing against the AFM door. Currently the BPV rejoins the intake flow at 90 degrees. I think it needs to be angled such that the BPV flow is aimed in the right direction minimizing any reverse flow against the AFM door. I drew up the two diagrams below to illustrate what I mean. Do you think that will help? I don't know if it stutters from running rich or lean yet. I ordered a new wideband with a display for the dash so I can video it and the shifts at the same time. Thoughts?

Oh, and this is all in preparation for Infineon in three weeks.

[Jeremy]

Do you have the bypass valve installed correctly (they're designed to allow air to bypass in one direction), and are you sure it's functioning?

[ianwood]

Do you have the bypass valve installed correctly (they're designed to allow air to bypass in one direction), and are you sure it's functioning?

It worked fine as a BOV when the car had a push through set up. We pressure tested it a while back and it seemed to be holding up to about 15psi if I remember correctly. It and the flange are from a Mitsubishi Eclipse.

[Eta power]

What if the BPV dump pipe (or whatever the hell you want to call it) was in "front" of the AFM? It seems to me that with your current setup, the BPV could be causing the AFM door to slam shut, causing a lean condition (read: hesitation) for an instant. But if the BPV dump pipe was in front of and pointed at the AFM, your problem could be eliminated. Just a thought.

[M. Holtmeier]

The problem there is the air would then be metered twice.

[Eta power]

The problem there is the air would then be metered twice.

What if the BPV vented right at the compressor intake, so as to be as far away from the AFM as possible?

[marc79euro645]

the y pipe design was bmw's approach with the 81-83 m102 Ljet setups in the 745. Thats what I have,& it works great.
At high boost with the hood off I was surprised to see dust blow out of my cone air cleaner as I shifted,and the bypass whistles.
Obviously, there's still more air than can be recycled.
Sounds like you might be better off venting to atmosphere,than recycling.
good luck
marc

[turbodan]

Vent to atmosphere isn't possible without standalone or a blow through setup.

I don't think the angle of the diverter is going to have a significant effect on driveability. It may cause excessive turbulence ahead of the compressor when the valve opens up but thats not going to really do anything. You want the AFM to slam shut when you shift because you're going from full load to zero throttle in an instant. The AFM slamming shut accurately describes the amount of air flowing through the engine while you shift.

What are the AFR's like during this hesitation? I'd imagine they're excessively rich. Any black smoke?

For all the trouble you've had Megasquirt must look pretty good. Might be time to take that step.

[ianwood]

The current angle of flow from the BPV must create a good deal of turbulence and I am hoping that aiming it will help to reduce turbulence into the compressor. I agree that it might not help that much in terms of the hesitation but am hoping all the same. There's no black smoke but I am guessing it's a rich condition. I won't know until I get the AFR display installed and take a video. We haven't been very good at tuning the BEGI and probably have it running a little rich to begin with. Again the AFRs will hopefully reveal more.

Heading to Buttonwillow in a week for some track test and tune. Megasquirt was (is still) the right way to go but I was overruled by my teammates so for now it's still the antique RRFPR and AFM.

[turbodan]

Do you have any preload on the diaphragm? That could cause problems. I always ran on just the bleeder valve with no preload. You could be going too rich as it crosses from manifold vacuum to atmospheric. The eta has rather large injectors, so it wouldn't take much extra fuel pressure to flood the engine.

[ianwood]

Do you have any preload on the diaphragm? That could cause problems. I always ran on just the bleeder valve with no preload. You could be going too rich as it crosses from manifold vacuum to atmospheric. The eta has rather large injectors, so it wouldn't take much extra fuel pressure to flood the engine.

The diaphragm on the BPV? Is there a way to adjust it? It's right off of a Mitsubishi Eclipse with no modification. However, I am not sure I understand what you mean. You've exceeded my limited knowledge of forced induction!

Another note on the BPV: When we redid the intake (from push to pull) and weeded out the myriad of boost leaks, we tested the BPV. It opens gradually starting at around 11psi and then is pretty much open at 15psi (unsure of the exact numbers). Not knowing if this was right, I got another and it pretty much did the same thing.

[turbodan]

Referring to the diaphragm for the RRFPR, for adjustment of fuel pressure in the transition between manifold vacuum and boost. That big center screw on the RRFPR. I ran no preload on mine.

Your diverter valve should have no problem staying closed under boost. I use standard Bosch diverter valves and have had no problems up to 1.4 bar. You need one of these:

http://shop.achtuning.com/bosch-motorsp ... valve.aspx

Thats for the top of the line, hot rod -114 valve. I've also used the -103 valves without any problems. I think I've got one -103 and a -109 in there now. Under boost they pressurize through the signal line and hold shut. I've had them blow open on one occasion when I had a section of vacuum line come apart. The spring pressure isn't enough by itself to hold shut.

[ianwood]

Referring to the diaphragm for the RRFPR, for adjustment of fuel pressure in the transition between manifold vacuum and boost. That big center screw on the RRFPR. I ran no preload on mine.

Ahhh, of course. OK. This is something to look at. As I mentioned before, our BEGI tune is not exactly pristine and could use some tweaking. Installing the new AFR gauge today. Might see if I can pull a couple of runs locally to get some detailed AFRs.

Under boost they pressurize through the signal line and hold shut. I've had them blow open on one occasion when I had a section of vacuum line come apart. The spring pressure isn't enough by itself to hold shut.

This makes perfect sense and, of course, we did not factor it into our testing of the BPV and instead left the signal line open to atmosphere when we tested its boost holding capability.

[turbodan]

Sounds like your tune is close. If it runs well under boost you may not need to adjust the bleed valve at all. Just take all of the preload off of the large center screw and see how it likes it. That could be the cause of the problem.