electric constant flow injection thoughts?

[PSlem]

There is increased interest among hot rodders in the old multi-carb intakes that use from 4 to 8 Stromberg 97 carburetors. Since these are pre-war, prices are from $150 to $175 each rebuilt so any complete setup with manifold, linkage, and scoops or filters will run over $1000. EFI throttle bodies are offered that duplicate the look but the cost is $4000. I am wondering the feasibility of an RPM and TPS voltage control for an inline pump with a temperature and vacuum switch for choke and power valve functions, then installing the nozzle into a base plate topped by a resin cast replica carb.

 

[WGJ]

What about 'throttle body injection' EFI system, like the ones used by GM and Ford in the '80s?

 

[overmod]

Not sure I understand exactly what you're thinking here -- seems like several different kinds of fuel system being mixed up...

If I understand correctly, you want something that:

1. Provides appropriately metered fuel and air to the intake ports on a V-8 engine;

2. Has the *appearance* of a bunch of Strombergs on the manifold, but without internal working parts;

3. Costs considerably less than the real carbs and manifold.

I would start with an existing port-injection engine which has the injectors located fairly close to the heads. You will need to duplicate the mixing geometry from the port injectors down to the heads in any system that does this, which might involve making an 'adaptor manifold' or using a composite casting atop the bottom half of a 'stock' manifold.

IIRC, the 6-carb manifolds were similar to 3-2s in that the engine idled on the center carbs. That would require only that the center 'replacement' carbs possess air mass sensors, etc, as required for a port injection system, assuming that the manifold log has good *airflow* characteristics from the intake to the four ports on that side.

If the manifold uses all six or eight (shades of the Maserati engines with four Webers!) equally, you'll have to figure out how many of the 'holes' need to be 'open' for flow to the ports to be equal, and put air-mass sensors there.

Naturally, each "carburetor" that contains an active sensor must also be fitted with a working air cleaner and good seal.

All the TPS, etc. would be handled as for the base engine and would not involve what is now a glorified air intake. Fuel lines, etc. to the resin carbs would, of course, be dummies. Remember that for best "realism" you will also want dummy throttle linkages to the carbs that move along with the accelerator pedal (NOT in sync with the motor that's responding to the TPS connected to the accelerator pedal!)


I don't know if you can package the working bits of something like a Holley Pro-Jection setup inside a resin shell, although I see no reason why that couldn't be done. This simply replaces the carb jet, etc., with a metered throttle-body injection system. Same considerations as above for the number of carbs, proportioning to manifold, etc.

I think your original idea involved using something like a single 'throttling' booster fuel pump, supplying fuel to all the injectors, and then hand-adjusting each of the injectors (just like screwing with the jets of carburetors) so that the engine idles evenly. Then change the delivery rate of that pump to control engine speed. You might be able to get this to work with a single throttle-body-style control computer, BUT I somehow doubt that the injector characteristics will be proportionally correct across the full range of engine performance. Only a continuously-adjusting injector guarantees that -- and to provide that in six or eight units puts you right back up in the high-dollar range again.

 

[PSlem]

You are on track. I currently have 4 modern H/W carbs on a homemade intake, but their size does not present the visual appeal and they require 8" filters in lieu of cast scoops or stacks, and looks are paramount with these systems. With constant flow, mass air sensors would not be part of the system. I'm thinking a TPS would function like the barrel valve on a mechanical system and RPM voltage control supply fuel like an engine driven positive displacement pump. With four twos they would be simultaneously linked, but I would run smaller butterflies so I could stagger the shafts and separate them and run staged linkage for the second butterfly in each carb. For more than four carbs, I would run progressive linkage. I'm just playing with the idea of a garage business for my retirement.

 

[wildcatgeorge]

This is a cool idea, modern EFI and reliability, in
a classic style carb. I don't know how well resin
would stand the heat but a cast pot metal dummy
certainly would.

Why not place the injector into the "floatbowl" and
plumb the high pressure lines to the stock carbs
origional location, yet it feeds the injector, now
place a throttle plate in the dummy carb, no need
for a TBI or throttle body unit.

Now you can use a MegaSquirt EFI computer to
drive the injectors, a TPS sensor hidden in the side
of a dummy carb and a single vacuum line to the
EFI box, now you have a multi throttle EFI system
that looks and works like the stock carbs, but are
more efficient and reliable, and hopefully cost less
than 1000 of 4000 $$.

Development is likely to cost the most with this system.
Get a mold of the orig carbs and use for casting new
units for the EFI injectors.

Am I making this too complex ???