I'm hearing of issues with fuel being evacuated from the fuel galleries connecting the accelerator fuel pump to the accelerator squirter nozzles. It seems that heat soaks the carbs and the fuel is pushed out of the galleries from the pump to the squirter nozzles. So, when first running after parking when hot, these galleries must be replenished with fuel before accelerator squirt is reestablished. It isn't easy to do while stationary since you would be flooding the engine; unless you want to sound like a boy racer in the parking lot after starting up.

The above is my best guess until I get a proper education as to the source of the issue.

Mount pump to front cross member if possible and while you are there pull the fuel screen from the rear of the fuel tank to check for blockage; mine was and it gave me trouble before I found it out.

Don't want to hijack at all, but since Paul's mentioned some related issues, I gotta ask-Paul what is your opinion on doing a recirculating system for the webers-do you have it on yours?

Paul,
Interesting diag on the warm start issue. Looks plausible, and unrelated
to the real problem.

I know it's taken a while, but update:
Got Carter fuel pump mounted with PMO regulator. With this I also basically
replaced all the non plastic hoses, fuel filter, and in-tank screen.

The in-tank screen was replaced by another generic filter before the pump.
One thing to love about carbureted cars is filters and hoses are cheap
and available. The screen, however, looked very good and should not have
been causing any issues.

So, with that finally done, I proceeded with a few of the same full-throttle
tests with the same type of results. From just below 6K RPM and higher,
AFR leans out by about 2 in that last 1200 RPM. I experimented with setting
the pressure at 4 psi and down about as low as it would go, ~2.8psi. There
was no noticeable difference. Also I did the the experiment of going to 100
air correctors and still at full throttle they operate over the whole 3K-7K
range, not making more of a difference at 'high' RPM.

New graph sample, looks like the old ones.
http://forums.pelicanparts.com/uploa...1342388914.jpg

Your e-tube is a F3, right? The latest graph says "F2".

Your AFR plot is flatter and looks improved over the plot from 19June; your mixture response is flatter as a function of RPM increase. You do have a rich transition but this would be a function of emulsion tube selection. The F3s provide a richer transition than F26s. If you need a set of F26s (yours are modified I believe) I can send a set to you to try.

Very weird that the 19June plot has larger mains (130 vs. 125) and smaller air correctors (100 vs. 170) and it produced leaner plots than those posted 15July using leaner jet selections! Is there any doubt regarding the wiring to your sensor; wires connected correctly?

Sorry for confusion that should read F26. I think it should be comparable to
earlier plots but I will check again when I get home. The other thing is that
the Innovate is not taking a huge number of samples per second and I'm
in second gear. So I am treating this data as approximate. After getting
things mostly sorted out I could get better data on a dyno.

Since the emulsion tube/main jet/air corrector system is so
simple, I'm starting to think that the issue is outside of carb
control.

The DC30 cam is not extreme but has more overlap than
stock. Could reversion issues cause problems over 6K? The
only thing I could think to do is try taller stacks.

Hi Mark,

I think your issue isn't getting much useful input now. I know I don't know what the issue is. I doubt the DC30 cams are a problem, they are a Solex type grind which is less aggressive than a "S" cam and reversion isn't an issue there.

I assume your headers are early style (pre-74) or are SSIs. If they have an OD larger than 1 1/2 inch at the exhaust port then they are way too big and could be causing troubles there.

Your F26 E-tubes were modified; I assume you have unmodified F26 tubes now.

Again, are you sure your connections with the oxygen sensor are correct and responding correctly? There is a bench test for checking sensor operation. Your AFR readings are completely disconnected with typical responses. Also, there is a huge change in mixture at 3K RPM which I don't understand. It appears to be more of a "step function" as opposed to a "transition" from idle/progression onto the mains.

Mskala, have you checked thoroughly for exhaust leaks??

I have a lift and have looked for signs of leaks, haven't seen any.
Headers are new this year MSDS and are the smallest size, though
I didn't measure them myself.

Think I'll postpone any more testing until after the season, nothing
is making a difference and it may take disassembly to see further.
I would think that there is a problem with the O2 sensor itself except
that when it is reading lean I can feel it for real in the car. The sensor
calibrated to fresh air and later re-calibrated to the same. It is
removed when not testing so it doesn't have huge miles on it.

I haven't been paying close attention to the values on the graphs
below 4K as I haven't disconnected the accel pump and it's hasn't
been exact what rpm I floor it. It's below 3K but the time scale is
pretty short in 2nd gear and those values are going to look weird.

Thanks all