Cadillac and Fuel Injection

[Bryan J Moran]

I was reading Perry Bs ad selling his 1957 EB and it caught my attention that the car has 2 x 4 carburation.  Then he goes on to mention what we all know - that the EB was a very expensive car. 

Why didn't the most expensive car in the world in 1957 have fuel injection? 

Chevy and Pontiac had fuel injection in 1957.   

[TJ Hopland]

Seems like we got fairly deep into the subject a few years back here.  Someone had a apparently prototype intake/setup and it seems like someone popped up that knew something about what was going on in that era. 

I don't have time to read em at the moment but maybe this thread?
https://forums.cadillaclasalle.club/index.php?topic=163139.msg484423#msg484423

Or this one from a few years earlier apparently started by the same person.
https://forums.cadillaclasalle.club/index.php?topic=158334.msg450306#msg450306

[jwwseville60]

GM would experiment brand new tech on lesser models until it was mature.
It's a smart move. Olds had the first Hydramatic in 1940.

Rolls Royce was never into innovation. They waited until the hydramatic/TH400 and the GM AC system was mature and reliable before putting them into their late 1950s and early 1960s cars. But customers could always ask for anything they wanted.

F.I. sorta-kinda worked for the high-performance Corvettes/Pontiacs, but small local shops had trouble fixing them back in the day. Only certain GM dealers could handle the job.

Chrysler tried its FI system in 1958. It was a total failure.

All through the 1960s, certain carmakers tried mechanical fuel injection with little success or benefit other than for pure racing. Only the Mercedes 6.3 V8 worked well because Mercedes had used FI in WW2 on their 601-605 series aero engines and their postwar 6-cylinder street cars.

I had a 1968 6.3 Mercedes. Very complex mechanical FI system but reliable. Yet that engine and all the other Mercs of that 1960s era were pretty loud and mechanically harsh. In contrast, my 1965 Caddy FW runs almost silent in comparison with more torque. Mercedes was/is known for over complication.

Only until late 1970s electronics came along, and made the FI tech work well, was it a success. Mostly for fuel economy and emissions, not power.

Rolls Royce didnt go to Bosch FI until 1980! Ditto Ferrari. Bosch had finally perfected it in Mercedes and Porsches. Pretty smart. It was a reliable and tested system.

I had a 1972 BMW with the early Bosch system. No BMW expert could get it to perform correctly. "Early days FI" they said. On went the dual Solex carbs. Ran perfect.

In my opinion, GM didnt perfect the EFI system until the 1985 Corvette.

Modern "one size fits all" cheap EFI kits for Caddies? The very capable mechanics I work with all hate them. Fiddly, underdeveloped and unreliable. Constant tuning, etc.
Carbs will get you home on old vintage pre-1980 Caddys.

[TJ Hopland]

Wasn't the 50's Chrysler system electronic?  Thought it was based on some of their tech that actually made it into late war production which is why they thought it would work especially since they had another 10 years to refine it.  I have never heard anyone get into detail what the issues actually were.

Was the 85 system you are speaking of the first year for what I think they called the 'tuned port' system?  When did that go sequential?  I don't think it was initially.  When did it get a MAF?  Maybe about the time it went sequential?

The 87 Allante was sequential wasn't it?  And it was a tuned port thing so at least Cadillac had one model with a fairly cutting edge system.  Rest of the Cads didn't get sequential multi port till 90 and when they did they didn't get the tuned port treatment. I would imagine since they were working on the Northstar they didn't think it was worth putting a lot of effort into the 'old' motor.

That 83-85 range seems like when Ford and Chrysler were doing the TBI systems too but then I think they pretty much switched to multiport vs GM that stuck with TBI on many models.  Oddly the trucks didn't get EFI at all till 87 and then stayed with TBI till 96.                 

[David King (kz78hy)]

The Eldorado Brougham was supposed to have fuel injection, but it must not had been ready to meet launch timing.


Al Haas provided this image of a test car being disassembled with the fuel injection on it.  Pretty cool stuff.

[jwwseville60]

Rochester GM:

https://www.youtube.com/watch?v=yqR0ebOcrxM

https://www.curbsideclassic.com/automotive-histories/automotive-history-1975-1979-cadillac-electronic-fuel-injection-gms-forgotten-fuelie/

Anyone know why Cadillac didnt use this well-tested FI system by 1963 with the new generation of 390/429 engine?
Was it just a complexity issue in an era of local service garages?
Cost? Both?

[Eric DeVirgilis CLC# 8621]

The following excerpts from the article sum it up well.

Many of the sensors and electronic components used on this EFI system became common place in the 1980s, but during this time in the 1970s they were uncharted territory for the American manufacturers.  Like other technologies in their infancy, this made for some growing pains...

There were many reports of EFI problems such as no-starts, hard-starts, high fuel consumption and other driveability issues.  This combined with many technicians being completely unfamiliar with EFI meant not only did the systems fail, but more often than not they were not able to be repaired correctly.  And to top it off, parts for these systems were extremely expensive and often difficult to obtain...

Given the reliability of FI systems up to that point, the time wasn't right in the 1960s and not a whole lot better in the '70s. Between the higher costs, complexity, finding competent support and lower reliability, the benefit of early FI had a lot to overcome.

[TJ Hopland]

I believe where most of the fuel injection development was originally for was aircraft.  Aircraft had a lot more issues with G forces and orientation as well as great variations in air density.  I'm not that big of a historic aircraft guy so I really don't know how much it actually got used or why it didn't other than turbine/jets coming in at about the same time.

The mechanical systems as well as the earlier electric ones really had minimal advantages over carbs for general automotive applications.  Their 'sensors' and 'controls' were not really sensing or able to control anything different than what a carb could do.  It really took more sensors and the ability to process that data and make corrections to really make a difference.  O2 sensor was a big one in the early days, say around 1980(ish). 

The o2 sensor could monitor things at the end of the process and then make adjustments.  This wasn't a perfect system tho since it was after the fact so there was always some lag in having to wait for the results to see if the changes worked.  By the time you saw if the changes worked the conditions could have changed requiring more adjustments.  To make this a little better especially in the digital age they added a learn function so at various run states it learns what it had to do to make it run correct so when the engine gets in that state like say steady cruise rather than wait to see the results it starts out by using the settings from last time which in theory should be pretty close .

The next big leap was mass air flow MAF sensor.  Those didn't come up till the later 80's and didn't become standard till the mid 90's.  Before those they were just kinda guessing based on testing what the air density was going into the engine.  They/it could take a barometric reading usually only before the engine was started and then real time incoming air temp readings and estimate what the air density probably was.  It was better than a carb that for the most part had no idea or ability to do anything about it but still there were assumptions and delays involved.  You could also then throw in a dirty or different brand air filter and throw all those estimates off.  There were a few carbs that did have some ability to compensate for altitude and also in the 80's carbs that had computer controls to trim the mixtures. 

The MAF sensors along with the IAT (incoming air temp) were able to measure fairly accurately the incoming air density and they already knew and had full control over the fuel so finally they were able to be way more accurate on what was going into the engine in the first place rather than having to guess and make corrections based on the results.  They still use o2 sensors to check the results and tweak those learning tables but its a much faster responding more proactive system than what they started out with.

So you can see without several sensors and the ability to process all that data quickly and take notes so to speak it was hard to really gain a lot.  I built a EFI system 15-20 years ago and it was a pretty universal kit so I had to really learn and understand what was going on to calibrate and configure things to work on my car.  Spent a lot of time data logging and looking at graphs and it was amazing how much of the time things were kinda a linear relationship which a carb had handle just fine.  Its the times usually transitions between states where carbs and even the slower EFI with fewer sensors had issue.  Like 75% of the time all you need is a straw in a bowl of gas and you are fine.  All the rest of the crap on a carb is for that 25% of the time.           

[jwwseville60]

TJ

Can you come over and help me with my 1955 FW sedan?
I found a 1960s Hillborn 8 stack FI unit.
I built the lower end with 13 to 1 compression and a solid lifter 3/4 cam.
Heads are ported.
I think we can get by without any sensors...
 

[TJ Hopland]

I know of Hillborn and have seen them but never running that I know if and I really don't know much about how they worked or what made them unique.  I do know that those were for sure one of those things most people didn't have a clue how to work on so either would not touch em or really messed em up.  Your not the first to ask me to work on one, either joking or serious or both.  I'm not against the concept but the others were people I didn't know that well and on show cars which isn't my style or what I usually work on.  Whole different set of rules and expectations with those.

[jwwseville60]

Just joking!
The Hillborn mechanical FI was really for drag racing only, not for the street.
It's an ON/OFF throttle system. Brutal.
Only the most diehard hot rudders used it on the road.
Today they make a street-friendly EFI stack system that looks like the original from the 60s.

[CadillacFanBob]

Cool stuff


Bob