I would like to add that going back over this story several times this hint is very powerful.
QUOTE from OP:
"but if you try to turn the throttle, it instantly dies."
That is almost always too lean problem if fully warmed up.
This configuration is going to take a #42 or #40 idle jet.
The stock #35 or even #38 will not work, with this configuration, even if the mixture screw is turned out.
It has been tried by at least two other forum members and would not work on these 1978 carbs. -LUCKY
You seem to be arguing that the original bike would not run well as driven off the showroom floor. I'm sorry, but I cannot believe that.
The original jets and adjustments were just fine when new. If that same configuration today is not working, then something is broken, or not meeting "as new" spec.
The Shop manual for the PD carbs, describes tuning each IMS for maximum rpm at idle speed. This corresponds to the leanest setting at idle. The accelerator pump was provided to compensate for this lean mixture, as when the throttle is twisted, it causes the metering jets to flow less, while more air is available to flow through the carbs.
The accelerator pump shooting raw fuel into the inlets should pretty much compensate for any lean mixture the pilot jets and IMS at set for, including if the pilot jets are plugged solid. (No, it won't hold idle with the pilots plugged. But, it will dang sure accelerate with that strong shot of gas entering the carb throats and the slides open.)
For the accel pump test, it is not enough to simply observe fuel coming from the squirters, it matters a great deal HOW MUCH is coming from the squirters.
I had them "working" on my 77 CB750F. And fuel was getting pumped. However, I had a check valve that was leaking which reduces pressure, and the diaphragm hole in the Mickey mouse ears wear swelled/smashed nearly shut. Fixing the valve and passage holes, changed (increased) the volume they shot into the carb throats. Those squirter posts shot a steady stream at least six inches after proper repair, and totally cured throttle twist acceleration. Even later when the pilot jets plugged again and it would not idle, if you could catch it before the RPM dropped below 800, a throttle twist would still rev the engine.
QUOTE from OP:
"but if you try to turn the throttle, it instantly dies."
That is almost always too lean problem if fully warmed up.
This configuration is going to take a #42 or #40 idle jet.
While true that twisting the throttle causes a lean A/F mixture, that is exactly what the accelerator pumps are supposed to cure on the PD style carbs.
Ideally, you would want idle speed to be >AT< stoichiometric. But then, any slide opening would go over lean and produce very weak to no power. The accel pump overcomes this if operating correctly.
As to using a larger pilot jet, this actually
IS true to obtain idle pickup with
non-accelerator pump carbs. Mechanical slide carbs must be tuned over rich at idle, so that when the slides are opened, the mixture remains in the combustion range while more air and less vacuum are present in the carb throats.
There are other things to consider within the carb design. But, the details must be framed about the specific carbs in question and not applied to all carbs in general.
Factors to consider:
How far away is the fuel from the carb throat fuel entry point? (Actual float bowl fuel level?. This effects how easy it is to get fuel into the carb throats.)
What is the fuel volume/weight of the emulsion tube chamber contents? (I don't expect many readers will understand the significance of this. But, a carb designer must, and a carb alterer, should.)
Cheers,
