I am in the process of restoring one if not the only existing Rochester mechanical fuel injection units built by Cadillac engineering during the development of the Corvette/Pontiac production units. I will post pictures shortly prior to beginning the disassembly.
Can some wizard rotate the pic 90 degrees???
How's this?
If you click on the link it comes out correctly.
Nope
Is there more story to this? What cars was it going to go in? Why didn't they do it? Pontiac and Corvette already did it so Cad didn't want to be 3rd? Chrysler had tried it and failed so they really wanted to make sure it worked if they were going to do it?
Pontiac had an electronically controlled fuel injection in 1957 that used vacuum tubes in the circuitry. When it worked it was marvelous, but as you can imagine, vacuum tubes and engine compartments do not go well together
Greg Surfas
Chrysler was the one with the electric fuel injection in that 57-58-59 era . It was a limited production option on some models and only something like 30 built and supposedly all 30 recalled / replaced. I think it was designed by one of the big companies like Boeing based on ww2 research and technology. I think they set some speed records with one of the cars, a 300 maybe? at the time. If there are any survivors they are in the hands of people that don't show them off. I think one of Jay Leno's mechanics has one of the cars and has done a cosmetic replica of the system.
The Rochesters as far as I know were all more or less the same and overall concept was similar to what Bosch was doing at the time. It had a mechanically driven fuel distributor and some fairly large diaphragm units it used as map sensor to meter the fuel. I have seen them on cars and I saw one apart one time but the guy that took it apart wasn't there to answer questions.
TJ, you are correct. 33 cars were equipped as such and it was a Bendix multiport electronic fuel injection system only available in 1958. There is only one known that still has the Electrojector system.
Thats right, Electrojector. I knew it had a name. I was close with Boeing too, correct first letter and were known for a lot of aircraft related stuff. It would be neat to see some details on how the system worked. I would assume cost was a reason the Chrysler one didn't work. I would imagine it was based on something used in aircraft that likely worked well but it wasn't going to sell if it cost as much as the rest of the car so they had to value engineer it. I'm sure it was still expensive but nothing compared to what the aircraft version cost.
Hopefully Bruce will post some more photos and let us know any history he knows. Looking at the small sideways photo it looks like all the early systems. The plenums and intakes were at first sheet metal presumably because sheet metal was much easier to prototype and experiment with than the castings that were eventually used.
I have pictures of the only Chrysler Corp. "Electrojector" that I know of remaining installed in a fully optioned '58 Desoto convertible, it even had the Benrus self-winding clock (by motion of the steering wheel) in the steering wheel hub. I took several pictures of the car at an AACA show in Melbourne in about 2008 or 2009. In as-new condition, it won an award as I expected it would.
Sorry for the short post. I will try to give a few more pictures and some details later today. I do have pictures of the remainder of the components to operate injection, distributor, drive cable through the air cleaner.
Those pre 60s injection systems were interesting, but extremely limited
in operational features. Most significantly, they still used standard ignition
systems and no feedback fine tuning of operation.
The basic engine fuel problem, is to match the right flow of fuel to air,
BY WEIGHT. But carbs and mechanical injection generally work by
VOLUME. Fine for fuel, volume is pretty constant. But the air density
varies widely with temp and altitude, throwing the fuel system off.
Mechanical attempts to compensate can help, but are only approximate.
WWII fighter plane engines had a problem with certain maneuvers,
losing power when gravity failed to keep feeding gas from the carb.
Some fighter engines were built with forms of fuel injection which
avoided this, completely mechanical, not electronic.
I have never seen a vacuum tube controlled fuel system in a car, there
is a whole list of practicality problems. Transistors made the problem
easier, but the germanium transistors available in the 50s were rather
inferior. By the 70s far superior silicon transistors were available, and
the Integrated Circuit was beginning to be available. The IC at first
might contain many transistors (later millions, then trillions) in a pre
designed functional block, making a far more complex Engine Control
Module practical.
New sensors in the 70s actually measured Manifold Absolute Pressure
(MAP), Mass Air Flow (MAF), temperature, remaining oxygen present
in engine exhaust, and more. The MAF actually does measure the mass
of incoming air, the elusive goal, at a cost of a rather sophisticated
electronic control. Better fuel control decreased emissions and helped
power some.
In the 80s ECMs took over ignition advance, with manifold absolute
pressure and rpm integrated into a 3 dimensional control map, plus
other compensations such as temp. With digital microprocessors now
practical, more functions like shifting, cylinder deactivation, boost
control, fault detection, and much more could be done. Individual
engines were fine tuned while driving using sensors and learning ECMs.
I would not refer to any US built fuel injection system as ELECTRONIC
until the 70s. Chrysler did attempt a limited capability unit in the 50s,
but it failed in part because of issues mentioned above. One of these
does exist and functions with an upgrade to a modern ECM.
Bruce Roe
So the story behind the Cadillac system was that there was a collaborative development done engineering staff where Cadillac decided that the improvement in performance vs. the cost was not warranted. There was a measurable improvement over the single 4 barrel carburetor, but unlike Chevrolet and Pontiac, it was decided to to proceed to production. The intention is to have both the meter and the pump restored by Jack Podell, while I will restore the remaining components. Given that there are multiple spare components, it should not be difficult to have a running unit from what is here.
That is just cool!! It does look like there are some external adjustments that make up for changes in temp/density, but I've never messed with one. The FI on our Seville is 'fun' enough.
Thank you for sharing.
Looking at the plumbing it looks like its a single batch fire system meaning all fire at the same time. I guess I knew it wasn't sequential but for some reason was thinking it was at least 2 batches. Or did later ones have at least 2 batches?
I like how nice that intake flange looks for what is likely a hand made prototype. I suppose it was cut on a bandsaw then cleaned up on a manually operated mill. I also wonder who made the plenum? I would imagine it was prototype body guy, maybe even the same guy that also just finished banging out a 59 tail fin on the same bench?
http://www.superchevy.com/how-to/0306vet-rochester-fuel-injection-system
https://www.youtube.com/watch?v=yqR0ebOcrxM
It seems that one prototype of the 1957 Eldorado Brougham was fitted with fuel injection. I'm wondering if the above system was the same...
Given the history of these parts, I would say they most definitely would be one in the same.
You guys who have contributed to these postings on this subject have caught my attention more than any other this year.
It is one of those things that is just so interesting to me and I can't wait for more posts and pictures to help fill a hole in my knowledge and experience.
You each deserve an A-DA-Boy.
Have fun,
Steve B.
I would like to note, that some advantage to many early FIs was in the
use of tuned intake runners. This works when the fuel is injected at each
intake valve. A single carb at the inlet to (wet) runners would drop a lot
of fuel along the length of the runners, causing response issues on the
street, though Chrysler did try it once. Bruce Roe
Its hard to tell in the photos but the mass produced cast version of the intake has the same tuned port design as the sheet metal one we are looking at here. Presumably they are tuned for the engines and cars they went in. The Vette's high revving 283 likely required a much different design than the low revving Cadillac 390. Who knows maybe that was a reason they decided it wasn't worth it. Maybe the advantages only really kicked in at higher R's which just wasn't Cadillac's style at the time.
Quote from: bcroe on December 29, 2019, 02:30:50 PM
I would like to note, that some advantage to many early FIs was in the
use of tuned intake runners. This works when the fuel is injected at each
intake valve. A single carb at the inlet to (wet) runners would drop a lot
of fuel along the length of the runners, causing response issues on the
street, though Chrysler did try it once. Bruce Roe
I have always wondered about the fuel in the runners on those Chrysler Cross Ram Intakes where each 4bbl carby would sit way outside the width of the engine, and feed the opposite cylinder head.
They do look neat though.
Bruce. >:D
Wasn't long runners one of the reason inline 6's and especially the 8's kinda hit a performance wall? Had to be really difficult to get the same mixture to a port 20 inches away as the one 4 inches away from the carb. It likely wasn't a big deal with 5:1 compression motors that were making 10 horse per liter. When port EFI came in they seemed to solve it by just having a massive long plenum with equal runners coming off it sorta like you can see in the photos of the Cadillac unit here.
Didn't someone also try a variable length intake in that late 50's time period? I know there were some in the 90's doing it and I presume some are doing it today. The one I saw apart was I think a Yamaha built motor in a Ford and there were essentially throttle blades in the runners to divert the flow to either a longer or shorter path depending on the current needs.
The variable displacement manifolds were intended to maximize the velocity across the engine operating speed. The higher velocity improves the VE, (volumetric efficiency).
Step 1 is complete and I will begin to work on other components. During the cleanup the identification showed that this was manifold #4 from 1958 stamped in the flange for the air meter.
Amazing what difference a little paint makes. Looks like a new piece just off the production line.
Slow and steady ;). The meters are soon to be sent out for rebuild.
Hope it works. If not you could bolt a modern MAF sensor and injectors
to it. Bruce Roe
Pretty sure that it work. This unit came off of a running engine. As far as a MAF and injectors, that can't happen with any of these components. This is a mechanical system that was never intended for that use.
Very familiar with the electronic fuel injection, I am a dealer of FAST and do calibration work for my customers.
I also race a C4 corvette that uses the FAST system.
What sort of performance gain do you expect from the tuned runners,
and what else needs to upgraded on the car (such as exhaust)?
Had a FAST analog system running in 1990, but now the car is gone
and the FAST is on a shelf. Bruce Roe