Need electrical engineer input

[Phil1934]

I've touched on the subject of doing something for multi carbs before and it seems everyone wants the 97 look. What i would like to do is make a 97 throttle body with constant flow injection. It would require an inline elctric pump for fuel injection with controls for RPM and throttle. It would be simple to add choke and power valve with a couple switches. A knob to allow 10% rich or lean would be added for fine tuning without the possibility of engine destruction. It needs an accelerator pump, but that seems tougher. Nozzles would come from a source like
http://www.mistcooling.com/nozzle_cost.htm There would be only one nozzle per carb, in the primary bore. There should be no need to aerate the stream as is done with carbs and mech. FI. I picked up a Spearco water inj. unit for the RPM to voltage capabilities, but don't know if it would handle increased amps. Also could get a Mallory voltage filter so unit would not run rich every time alternator kicked in. I already have WB O2 sensor for air/fuel ratios. I envision the carb bases as being in two parts with smaller butterflies. This would enable the two shafts to be offset and apart so linkage could stage the secondary opening. Linkage would be in the middle which is why two parts. This allows cheap synchronized external linkage. Carb bodies could be some sort of resin cast with an aluminum tube to handle backfires and carb fires. They could be cast aluminum if the demand for polished carbs is strong. should be less than $1000 with manifold and six carbs. Is this doable?

[Fat Hack]

Moon is already selling EFI units hidden in new 97 style castings. Pricey, but they're out there!

You would be far better off to use your 97 castings as throttle bodies like an older GM OEM set-up before they went to port injection. It would be a little easier to "disguise" your carb castings with the injector inside of them, which is how the stuff that Moon sells is configured.

If you HAVE to run EFI, I'd elect to use the "phony" carbs as nothing more than throttle valves with injectors at each port, the way most modern systems operate. Ford used staged throttle plates in a trick looking casting on their newer Taurus engines...the plates meter air flow only while fuel is injected near the ports and controlled from a CPU through sensor input.

I had considered doing something like you're talking about, but working it out on paper and in my head lead me to believe that it would be very prone to flooding and would only deliver decent performance at wide open throttle...not very "street friendly" at all.

That said, I'd like to see the results if you decide to try it! Clever rodders have always conceived and built their own speed equipment and induction systems, and today is no different! If not for the sake of better performance, do it just to satisfy your own curiousity!

[Phil1934]

Electronics are cheap. If you can solder Meagsquirt will sell you a kit for $150. The problem with EFI is the solenoid operated injectors will set you back over $200, then the MAP and inlet air sensors, etc, are also pricey. Carbs get by with a single trip point power valve instead of variable, so a simple Hobbs switch is only $15. We are not trying for maximum fuel economy and minimum emissions. And while you could duplicate Moon's offering for half th emoney, with a little creative thinking and a few concessions, I don't see why it could not be done much cheaper.

[Fat Hack]

Yes, you're talking about 'electronic carburetion" almost, for lack of a better term at the moment! I think it could work, if you're willing to put lots of time into developing and testing your system. I'd love to see how it works!

I would rob injectors, and other related hardware off of junkyard motors to build me own EFI system for cheap as possible, but I'm not ready to go down that road just yet! A simple, well tuned carb (or carbs) is still tough to beat, but your idea has merit and shows creative insight...I like it!

[four-thirteen]

I'm not quite a EE yet, but am working on it. The only thing that a carb/injector setup has to do mix fuel at the proper stoicometeric ratio. Somewhere around like 13 or 15 to 1 by mass. Carbs do it by metering the volume of air over time, and work quite well at it, only when the air flow isn't constant do they have trouble doing there job. So they use an accelerator pump to fix that.

Stu hilborns lawn sprinkler does it but metering engine rpm and throttle posisiton. This also works good, and doesn't need an accelerator pump or power valve.

I guess what I see here is a system that uses a variable voltage or current to inject fuel based on RPM, manifold pressure and throttle posistion. Interesting idea. I think that to make it work right you are going to need someone who knows analog circuit design. Thing is, most EEs are people how can't do analog. If it was digital, you just have yourself another EFI setup with a different name, but with no feedback, which would be hard to calibrate. Dave

[leadsled]

I'm not an electrical eng but I work on some components we call linear variable transformer (LVT) they are usually 4-20ma and often used for engine speed control. This would be placed some where as a TPS. I set them up as 6ma 0 fuel and 18ma max. Anyway it's do able but pricey.

[CharlieLed]

I am an electrical engineer licensed in the state of California, but what you're asking goes well beyond the scope of electrical engineering. What it sounds like you are trying to do is add a bunch of complexity to a very simple system that has been proven to be effective for over a half a century. I agree with Fat Hack, modify the 97's to be throttle bodies and get the injection (I like direct port injection) from existing technology (read "junkyard"). There was an excellent article in a recent automotive magazine where a guy did just that, converted the 97's to be throttle bodies and placed injectors in the underside of an Edelbrock manifold where they were out of sight. When installed on an engine in an open-fendered rod it was hard to tell that the 97's weren't completely stock/functional.
IMHO...if someone has already invented an elegant solution, I'm with it unless I can find a better way. I think that you have a good idea but someone has already done the hard work for you. Here's a book that I like that covers may variations of FI...

[Phil1934]

Junk yard would be OK if I just wanted one, but it would be tough to sell junk yard parts. Forget EFI and start thinking Hilborn injection with an electric pump in lieu of a belt driven one and a TPS for a barrel valve. The only hitch I see would be integrating RPM and TPS. You could not weigh them equally, i.e. run them parallel. You would have to run the RPM control through the TPS, in series, so the TPS controlled the amount of fuel and the RPM added as engine speed dictated. I'm talking a few resistors in lieu of a computer.