1950s "computerized" fuel injection

I didn't know this was a contest. Sorry.

 

I read the same articles from 58 and I assumed it was a system that actually hit the street. Ive never seen one of these so I have to wonder was it ever actually sold to a customer or was it just a demo that we are seeing pictures of?

My memory is not that good on the modern stuff but Chrysler had some kind of fuel injection that was sold on the street and then recalled or at least they sold the replacement parts as a package to eliminate the system and make the car useable. Sort of a retrofit kit.

So... 2 questions is my memory reasonably correct and did they ever actually sell any of the 58 units to the public?

 

According to the Allpar article, they actually sold some to the public, but they were "all" retro-fitted with carburetors after a period of time.

Glad this thread returned from the popcorn fest it was becoming...

 

Geez squirrel, lighten up. Using that word was just making fun of our exchange.

 

The WWII battleship fire control analog COMPUTER.



The Norden bomb sight was also an analog computer.

 

The Rambler was the first car to use the Bendix Electrojector unit. AMC only built 3-4 Rebels with the Electrojector, and let the automotive press drive one. They didn't release it to the general public due to reliability issues. They didn't run them long enough for electronic failure, though the main problem was temperature related. The engines were hard to start under 50 degrees, and didn't work well above 85-90 degrees. The electronics of the time just weren't up to the task. I've got a couple extensive articles on it, including the AMC marketing brochure (printed before the decision to axe it went out).

The system wasn't throttle body -- there is an injector for each cylinder. The distributor was modified to add a second rotor. That second rotor fired the injector. One article I read stated that fuel and spark were simultaneous, but I'm not sure. Should work -- is usually sprayed on the back of the valve before it even opens in modern systems. Would be easy to alter the rotor or contact positions to fire the injector different from spark. There was also another set of points to send a signal back to the brain box. The box is just "fired" when the points open. It then sends a modulated signal back to the distributor where the second rotor sends it to the appropriate injector. The brain box received signals from an intake manifold pressure sensor, an acceleration sensor (vacuum operated points, would kick in extra fuel if a sudden drop in manifold pressure), a starting enrichment sensor (variable resistor controlled by a thermostat -- as engine warms resistance changes), deceleration sensor (vacuum -- cuts fuel off is a sudden high vacuum reading), and an altitude compensating sensor. Note that most of these sensors worked more or less independently. The brain box (Bendix called it an Electronic Modulator) used the varied resistance signals to modulate the injector pulse. The modulator controlled the pulse width only -- or how long the injector would stay opened. It was very similar to mechanical fuel injection systems of the time, but used electricity instead of fuel pressure to time the injectors. The Electrojector added 33 hp, but torque only increased by 5 ft/lbs. The system used only 20 psi of fuel line pressure and had a return line.

The first Bosch electronic injection system was a virtual copy of the Bendix with some modifications. The modulator box was transistorized, of course, which meant better reliability and faster action. Another break-through was adding another injector in or near the throttle body for cold weather starting -- enrichening the mixture like a choke. They also refined the sensors and may have added one or two, I don't recall right now.

I know that a 1972 Bosch system on a V-8 Mercedes looks remarkably like the Electrojector system on the 57 Rambler Rebel. Unlike the Chrysler version, AMC only used a single 2V throttle body. I've often thought about getting an old Mercedes system and adapting it to an AMC 327 and building a 57 Rebel clone using that engine. Would certainly turn some heads! AMC retrofitted the engineering examples they built with 4V carbs and sold them as demos.

 

Yes your memory is correct. And yes, they did sell them to the public. They were recalled almost immediately to have the EFI units replaced with dual-quads. According to a Bendix manual, two Plymouth Furys with 350ci wedges, 16 Chrysler 300s with 392 Hemis, 12 Dodge 500s with 361 wedges, and five DeSotos with 361 wedges were built with the dual-quad option and taken to a DeSoto plant in Detroit to have the Electrojector system installed along with a 40-amp generator, electric fuel pump and a pair of Fuel Injection emblems.

So to make this even weirder, here is a '58 EFI Plymouth... but it's a Belvedere, not a Fury.

"I received this photo from the car's original owner... you can see the fuel injection emblem on the front fender. I corresponded at length with the owner of this car, who drove it close to 100,000 miles. He also shared with me the manuals, etc., for the car so this was the "real thing". He claimed two cars "got out" - the other one was wrecked very early after delivery. His remained with the fuel injection unit intact. Where the car went when he sold it, nobody knows...."

 

Yes, it took thirty years for technology to catch up and allow the 1957 Bendix Electrojector to work to it's potential.

Bosch bought the Bendix EFI design and patents, worked on it for a dozen years and put it out on 1969 VW, SAAB and a few others in the late '60s-early '70s. Worked, but not well enough. I owned and fought with a few of them. Injectors clogged, so better filters were necessary. Electro-mechanical points for both the injector and the ignition were an anachronism which had to go.

While they worked out those EFI problems, Bosch went back to an updated version of the '50s constant flow injection. Thus, in a way, Bosch also perfected the Rochester FI, which GM never bothered to do. By the mid-80s, Bosch reintroduced pulsed EFI, along with crank-triggered ignition, became the standard of the industry for the next twenty-five years.

Today, high-pressure direct injection is replacing EFI.

jack vines

 

Jack, I can't tell which system you said was no good, but I have driven my 70 VW T3 for about 5 or 6 years now as my only daily driver car. It starts well in any weather and gives better mpg than the much smaller/lighter beetle. Car has never been garaged, either.

And...it is a generator, not an alternator; these cars never had an alternator, which you would think was necessary for EFI. Maybe this is something carried over from the Mopar setup.

I went to school for those in NY at the factory owned school. Like is still said by diehard VW mechanics, "99% of Bosch EFI problems are actually ingnition or something else".






And thanks to all for putting up the thread.pics, and additional info. I never ever heard about the Ply, Dodge, DeSotos having this setup. I only knew of the Chrysler 300's. Interesting history IMO

 

The early '70s was when serious emissions regulations started taking effect. To one degree or another all those engines were down on power and had some level of compromised driveability. In the early '70s I worked with a guy who bought a new first generation 914 Porsche. He was so unhappy with the non-sequential Bosch fuel injection that he put carbs on the engine. No doubt it no longer met emissions requirements, but it was nicer to drive.

 

Cloning an AMC "Bendix Elctrojector
unit using Bosch Mercedes components
would probably be a big project, but
so, so cool! Besides Mercedes components,
you might be able to scavenge some bits
and pieces from the Bosch EFI units used
on the first-generation '75 to '79 Cadillac
Sevilles too.

Mart3406
==========

 

Yes they do, especially in guitar amplifiers.


Sorry to cut up your post but I had to make this point.

Good thread.

 

very Forward Look

 

Very... but I'm still going to be running Hilborn on my '57 Chrysler.

 

Yes, I am familiar with the 914 4, as it was also used in the short lived VW type 4, which also was known to hesitate, and also the rpm's would "hunt" at idle, much like blower lope/detroit roll.

Switching to carbs always lost 3 to 4 mpg back in the early-mid 70s.

There always was at least one guy in each dealership that would figure out how to modify what was needed, just to say they could fix a factory problem.

In my area, the "fix" was to adjust the factory sealed manifold pressure sensor. We never knew there was a second adjuster hidden inside. I found out when I got mine from Nevada. It had a "new" hole drilled through the bigger main (factory sealed up)adjuster screw that most mechanics thought was the only ajustment. I used a jewelers screwdriver through that hole and barely moved the inner screw, and it really woke the motor up. Mine previously ran lean, under accel, because I put in larger pistons, which also boosted the compression. It had run so lean/hot, that it melted some rubber pieces near the engine! That went away with tweaking that sensor. (interesting that some bosch guru websites say to never touch that adjuster, because the world will come to an end )


Another mod is to make a quick adjustment to the fuel pressure regulator, bumping from 28 factory, to 30-31 psi.

My car never hesitates, never needs warm up, stuff the gas pedal when 0 degrees temps out, and it takes it.


anyways, this system seems so hot rod friendly. I always wanted to try to use the same family Bosch D-jet that came on the Mecedes 8 cyls. A friend said it's the same system?


As the gas we buy keeps changing, and our old carbs with open bowl vents that promote evaporation, make gas fumes when parked, not to mention possible fuel boiling and vapor locking. I'd like to see what some talented hambers could do with that setup and use some sort of vintage speed manifold and modify it?

 

just today I was thinking about those problems....and also thinking about what you could do with modern sensors, and some analog circuit design expertise (that I don't have). Interesting stuff.

 

Aside from the electronics inside the ECM, the rest of the Bosch system is kinda mechanical.

When these were new, I was sort of intimidated by them as I was the young guy in the shop. Now that I have to keep one running, or buy a new car, I got into them a bit deeper.


Mine is 42 years old, and I do have twin carbs to put on, but I refuse to do so....so far. This runs better at all temps, compared the 66/67 twin carb version. More power also!

 

Some thoughts I've had thru the years that I find interesting on fuel injection.

* carbs work well and are based on simple principals.
*a - air flow / jet size & vacuume/jet size


* distributors work well and are based on simple principals
*a rpm, advance, vacuum

In theory a sophisticated distributor should be able to fire electric injectors.
It would need throttle input and vacuume input and variable fuel pressure based on vacuum and temperature.


Maybe one day ill mess around with the idea.

 

A little off subject but why don't they computor control the spill valve on a hilborn for a street unit.

 

That would be fun....you'd need some sensors to do it though.

 

31vicky... you're reinventing the Electrojector system!! ;>

I had a friend with a T4 VW wagon that he put a single Holley 2300 on. I remember him saying something about the computer being too sensitive -- it made too many readings and adjustments too often, and after a few years the adjusting components would wear out. Sounds a little BS to me now, but maybe he had a point. He said the later systems had a couple fewer sensors and were less sensitive, but I'm not convinced that was the problem. Just new cutting edge technology that took service personnel some time to figure out, and had some teething issues is the typical reasons for such problems.

As far as a generator/alternator, the early Electrojector modulator (as used by Rambler) drew only 3.5A @ 12V. It wasn't a power hog.

Interesting the mention of vacuum tubes over transistors. I've often heard that tubes were a lot more interference resistant. That's one reason they work better for amps -- less (no?) feedback, so a "cleaner" sound. I know they are EMP resistant too. A Russian pilot defected with a MiG Foxbat in the early 70s. One tech magazine reporter who saw the partially disassembled aircraft commented that the Russians were so far behind in electronics that they were still using vacuum tubes. What he didn't realize is that they had developed tubes way beyond what we did, so they were much better than the old tubes he knew, and they were used specifically to make the electronics EMP resistant. At the time we still used the "Genie" nuclear air to air missile to knock out a small fleet of bombers with one shot...

 

Don't forget that vacuum tubes were used in all manner of military equipment. You think a B-29 might have some vibration issues? The Soviet and Chinese war machines produced vacuum tubes up into recent history, and even flew them in fighter jets (those factories now produce for the musician and audiophile market btw).

There are even milspec variations of common tubes, rated for higher reliability and tighter tolerances.

If this amazingly ahead of its time system failed because it wasn't reliable, it's because it was bean-countered to death and parts quality was skipped on.


Sent from my Rezound using Tapatalk 2

 

Farna, we're on the same page lol

Sent from my Rezound using Tapatalk 2

 

Tubes are neat, I have them in my wristwatch...and they sound good, but not because they're "better". We just happen to like the distortion they produce.

 

- Tubes produce more and less of different types of distortion than transistors. It's not a matter of finding tube distortion pleasing. It happens that the distortions produced by tubes are less objectionable to the human ear than what transistors produce. As a result, a transistor amp has to be very good to equal or exceed the listenability of a just a decent tube amp.

- Tubes are more linear than solid state. The ways that is typically compensated for tends to cause distortion.

- When transistors are overloaded they "clip". That can happen subtly even in normal listening. When transistors clip they produce a very harsh distortion. When tubes are overloaded in the same way they just lazily roll off and refuse to try.

- Tubes don't come close to the rise time(speed) of transistors, but when transistors are pushed to their limits of speed it can result in transient intermodulation distortion(TIM).

- Because tubes are slower and "lazier" than transistors, in some cases that will mask shortcomings or harshness present in the system.

- Tubes also have qualities that gives them some advantages over solid state analog in radios. FETs largely bridge that gap, but not entirely.


There is a lunatic fringe that clings to tubes based on misconception, nostalgia, or whatever. Those people generally don't have an objective point of view. At the same time, it is a fact that tubes are not in every way inferior to solid state.

 

So should... efi fuel injection allowed be in nostalgia racing now!!!

 

that "fuel injection" badge is awesome!

 

Then you would have a field day reading the trials and tribulations of the military getting their pre-solid state onboard radar and fire control systems operable in the B-36 and B-47 SAC fleet during the same time period. Even with unlimited budgets and dedicated R&D teams, any sophisticated electronic system of the 1950s was a mess.

Here's an example from the Convair B-36 wikipedia page (quick reference, take it with a grain of salt):

http://en.wikipedia.org/wiki/Convair_B-36

 

Well, I can tell you for sure that the vibrations from the 20mm cannon are a lot more than road vibrations! Vulcan "gatling gun" 20mm cannon were test fired at Warner-Robins AFB in an open shed covering the gun. I was in Civil Engineering and had to go tighten screws on the metal building at least once a year from the reverbrations of the gun firing. The gun wasn't attached to the building either! I can see vac tubes working out of sockets easily... ICs might even work out of sockets!!

Should EFI be allowed in Nostalgic Racing? Only if it's Nostalgic EFI... Some of the Mercedes/Bosch systems might qualify, depends on the cut-off date. Nostalgic racing isn't nostalgic with modern EFI IMHO. Maybe a separate class for old cars with EFI?? In the end we all know that a carb and EFI engine can put out the same power, the carb just has to be manually tuned for current conditions while the EFI self-tunes. Better driveability is the main hallmark of EFI, and better emissions control, and SOMETIMES a little more power due to better fuel atomization and/or disctribution (especially with multi-port injection).