Not too shabby...

ProJection 2D ECU
8 (x) 28pph MP injectors
Weiand MPFI conversion SP manifold, Vortec intake ports machined match square E-Tec 200 head ports (PITA)
E-Tec 200 Aluminum heads (200cc intake ports, matched to intake)
Large-cap, non computer controlled HEI
Hedman V8 1-5/8 med-length headers
ProJection TB for air control, injectors removed, but pod still in
ZZ4 short block, about 9.4:1 CR or so...
Comp Cams XR269HR (218/224, .495"/.503", 112 LS)

Pulls through 5,500 likes there's no tomorrow...

Would still like to use the 7747 GM ECU, BUT I could not get the thing to stop "hunting" under low throttle loads at about 1600rpm, kinda like RICH then LEAN the RICH and shows that on the AF Meter, BLMs around 128 when doing it as well. Stytmied. Tried changing fuel and spark in the bad areas but did nothing at all. If anyone can help me with that, I'd appreciate it. I'd like to get back to 7747 control but it is running VERY nicely...

 

Did the RPMs hunt as well? The afr gauge is supposed to operate like that when "all is well".

 

I guess I should be more clear...

The thing will be RICH for a second, then LEAN for a second (not stochiometric passes...), the whole time the engine is surging up and down at the same rate... Absolutely terrible, so bad you have to get out of that RPM range, its like you are hitting the gas then letting up only you are not actually doing it...

Only happens from about 1400-1800rpms under light loads and it SUCKS. Actually so bad it's embarrassing.

WTF? I gave up and used the ProJection ECU it was so bad. Getting ready to try again but need some help. I think Grumpy might be able to offer some assistance. He's the 7747 master with big cams although I don't think my cam is all that big... I went 112LS on purpose, as well.

 

Most likely the ECM is switching between sync and
async injection modes. Need to play with the
transition PW's to alleviate the problem.

Physical EPROM locations $2D6 through $2DE.

RBob.

 

Can you elaborate? PLEASE...I've got the same problem on my friends T/A. He has a 224/230 @.050 .500/.510 lift 112* LSA cam, with Sportsman II heads, 9.5:1 comp ratio, 7747 ECM with 90pph injectors.

 

When the sync PW gets too short the ECM will switch
to async injection mode. Sync mode is in time with
the spark plug firings, async mode is at 80Hz. The
80Hz rate is the internal ECM minor fuel loop rate.

However, the async injection will only take place
once enough PW time has been accumulated. In other words
the ECM keeps adding together all these short PW until
it has one large enough. This is done due to the injecter
not being able to meter a little-tiny-pulse of fuel.

Now, the problem is that the injectors are not linear
in the PW vs fuel metered graph. Here's the scenario:

The sync PW is currently short, and is getting shorter
as the engine slows down. The PW then reaches the
minimum allowed (by the cal). The ECM then switches
to async mode. Due to the non-linearity of the injector
the AFR goes a tad lean. The INT starts to climb.
As the INT increases the PW gets larger with the
result being that the ECM switches back to sync mode.
Now the AFR is a rich and the INT starts to drop.

As the PW decreases the ECM will then switch
to async mode. And the cycle repeats.

The change in AFR is magnified by the response of
the high-po engine. Regular old car/truck engines
do not respond as quickly or to the same degree.

The problem manifests itself as the sync mode is rich
while the async mode is lean. It is made worse by
the INTegrator being high from async mode then the
switch to sync mode being very rich.

The '747 also has an open loop decel mode that
can be enabled through one of the option bytes.
This may also alleviate the problem.

RBob.

 

wow, never would have guessed that :hail:

So would it help to turn up the MAX asynch pulse width so the computer would enter asynch at a higher pulse width?

 

That would expalin my problems with MPFI injectors and the dry manifold. Not enough fuel hangin around to mask it. I've also heard that this problem may be accounted for in the Edelbrock MPFI conversion BIN file. I have that BIN file, BUT here is what I did to make my MPFI chips:



The file I used was my latest TBI PROM updated to the Edelbrock MPFI conversion BIN by "comparing" and making things the same (ie, Fuel table), though, which would not carry over this operating fix, if it is indeed on the Edelbrock MPFI BIN. Maybe I should try the EDELBROCK MPFI BIN with just a BPW change to see if it still exists... Maybe that's why is does NOT do it with the ProJection ECU as it only uses one mode???

Does that make any sense??? I was not aware of this until it was brought up. that 7747 is more sophistocated that it appears. So, I could make the PW change to a level LOWER than 80hz as I have MPFI injectors capable of lower fuel spits at the low PW?????????????????? Would that help?

 

It is difficult for me to say what will and will not
work for a particular setup. One thing that I did
was to increase the bias value and set the trip
points to async mode to 0 PW. This had the effect
of eliminating the switchover to async mode. It did
act like a fuel cutoff too.

I have also seen bins that did the opposite (from GM).
They set the trip points to 13msec so that sync mode
would never be used. I do not recommend this for high
RPM engines though. Duty cycle could get tight.

These are the parameters that I have discussed. The
first two are the terms that switch the mode between sync
and async.

The inj bias is added to each and every pulse of
the injector. It is designed to help linearize
the injector flow rate.

The last two is the minimum and maximum allowable
async mode PWs.

The bin value is: N = msec / .0156

Also note that each value is 2 bytes.

The switchover works as such: Once the sync PW is
smaller then the value at LD2D8 the ECM switches to
async mode. Once the sync PW is greater then the
value at LD2D6 the ECM switches back to sync mode.
(The ECM always calculates the sync PW. The async
PW is derived from that value).

RBob.

 

So, if I cahnge my 1.4ms and 1.1 ms values to 0.1 and 0.0 respectfully, I'll essentially remove the swithching and keep SYNC mode on all the time? and would this be bad with 8 (x) 28pph MPFI injectors?

Has anyone checked the 350cid MPFI conversion BIN for SYNC/ASYNC mode modifications, yet? My laptop has crapped out. Like met its maker...

 

Yes, changing the 1.1 & 1.4 values will keep sync mode
all of the time. I do not know if it would be bad
for the injectors.

The Edelbrock bin I looked at made no apparent
changes to these parameters. I was surprised. They
did remove a lot of accel enrichment though.

RBob.

 

Yep. You need VERY little enrichment vs TBI wet manifold...

How can I be sure that the SYNC/ASYNC switching is causing my problem??? Are my symptoms typical of this switching, or something, that is why you brought it up? Lessons learned???

Is/are there any other PROM/ECU related issue that could be causing it?

The reason I ask so many questions is that if we are relatively sure that the SYNC/ASYNC issues is the problem, I will go through the tedius task of swapping the ECU's, wiring harnesses and re-splicing the MPFI harness back into the GM TBI harness. Not easy chores.

 

Unfortunately, the only way to tell is to try it.

I've attached an Excel graph of the sync/async switchover
in action. This is a snap shot during the bucking-bronco
scenario. I hope this helps out.

Scaling:

RPM: 200 is 2000 RPM
O2: 100 is 1000 mV or 1 V, 50 is 500 mV (almost stoich)
sbpw: 100 is 1 msec or 1000 usec (sync BPW)
abpw: 100 is 1 msec or 1000 usec (async BPW)
INT: 100 is 100 (1:1), INTegrator.
Bottom axis: hh:mm:ss

Notice how the O2 drops and climbs as the injector
mode switches between sync and async. It can even
be noted tha the O2 response time is not all that fast.

This is a heated O2 in the reducer, past the header collector.

This also shows how the INT climbs and causes the
mdoe to revert back to sync. Then falls with the result
being back to async mode.

RBob.

 

OK, lets try some better glue. . .

RBob.

 

Thanks. That makes sense.

I had to laugh when you said "bucking bronco"! That's exactly what is feels like!!! I guess you HAVE felt it before.

How did you get that graph? During an actual bucking bronco episode???

I suppose you are right, I'll just have to try the changes in the PROM and see what comes of it.

 

Oh yea, BTDT, and a bucking it does! The graph is from an episode
of the bucking. Took me quite a while to figure out what was
causing the problem. The data capture and graph of the data
helped to pin-point the problem. I disabled the sync to async
switchover to prove it out (set terms to 0).

I've since switched to different injectors and fuel pressure. Along
with that I also changed the sync/async terms to those shown above.

RBob.

 

I am using MPFI injectors, 8 (x) 28pph.

What values would you recommend I start with? "0"'s???

EDIT:

That really is a lot of help you gave me. Thank you VERY MUCH!!!

Nice graph and explanation.

 

I'd say either set them to 0 msec, or set them to
13 msec. Either way will allow you to prove out
whether this is the problem or not. I am referring
to the values at $D2D6 & $D2D8 (2 bytes each).

If/when that clears up the bucking-bronco ride then
it is a matter of coming up with better switchover PW's
and min async PW terms. With port injection these may
be a bit higher then any TB injected systems.

RBob.

 

Looks like I'll be playing with the Tunercat editor for the 1227747 ECU this weekend, if I get my new laptop. I'll see if I can get the BIN changes done then swap back to 7747 control next weekend.

Thanks again.