Is The 280Z Fuel Injection a "Good" System?

[TomoHawk]

I think it would be a much easier job to build a new system using a digital core that "reacts identically to input stimuli"

that would work too. Just make it fit into the L-jet box.

[Captain Obvious]

that would work too. Just make it fit into the L-jet box.

Come to think of it, I just remembered that there is a user superlen who I believe had been working on such a device. He said he started it in 1993...

Lenny doesn't post much, but here's some info in this thread:

http://www.classiczcars.com/forums/fuel-injection-systems-s30/45620-long-time-member-new-job-diyautotune-com.html

I'm surprised he hasn't poked into this thread and mentioned it. He should be long done by now. :classic:

[TomoHawk]

Thanks for mentioning that. It seems like just the thing. I wonder how it got by everyone- probably because March is the busy month for Z owners; cleaning the garage, putting the car back together, tuneup, etc.

[TomoHawk]

It looks like the Micro Squirt system is what you'd use for something like an L28 engine.

[FastWoman]

Fastwoman, I think that's what Tomohawk suggested earlier, and I agree. I think that would be cool.

I've got the output stage under control, and I'll take care of the power supply. What section do you want to do?

LOL Oh, WOW! Uh, gee... So you want someone to tackle the black box part?! LOL

But seriously, I suspect some custom implementation of a MS system, cobbled into an L28 box, would be the way to go. Perhaps there are even a few unused pins for connecting an O2 sensor.

I do admit, though, it would be fun to have an analog replacement ECU, just for grins. Perhaps this appeals to me only because I used to design analog circuits. Op amps are commonplace, and it would be easy to design a socketed board based on a commonplace pinout, as OP amps are fairly substitutable for a low frequency, low voltage application such as this. The magical hurdle to jump would be to introduce multiplication into the mix for compensation of fuel mixture for air temp, engine temp, throttle sensor status. That takes special chips that never seem to stay around very long. Analog Devices used to make a bunch of them, typically with product cycles of a couple of years (?). I don't know if these things are even made anymore, now that digital devices are so commonplace. Even after getting that far, I don't know what happens after the AFM pegs. Is there further enrichment based on RPM?

[Captain Obvious]

LOL Oh, WOW! Uh, gee... So you want someone to tackle the black box part?! LOL

Haha! No, I wasn't offering you the black box part. That's the big selling point of a design which is digital at the core and "reacts identically to input stimuli" when compared to the original analog design. The trick is that nobody needs to figure out HOW the black box parts work. You only need to experimentally determine WHAT it does, and that can be done on the test bench.

You need a power supply, some input conditioning circuitry, a microcontroller with enough A/D's to handle the inputs, and an output stage. The rest is software. As Bones says, "A child could do it"...

[FastWoman]

Captain, I remember giving the answer similar to the one your cartoon on an organic chem midterm back in the Jurassic era. I got partial credit for the chutzpah. I drew a magician's hat with the reactants, a reaction arrow with magic words underneath and a magic wand overhead, a "poof" and then the same hat full of product.

But back to the ECU, I understand that a MS board can already process input from antiquated hardware like air flow meters, so I don't think any additional ADC work is needed -- but I could be wrong. I doubt there would be such a flood of people wanting replacement 280 ECUs that it would be worth reinventing MS's wheel. Instead, perhaps someone could just figure out the easiest turn-key retrofit, possibly programming in the necessary parameters before shipping a MS system wired into a 280 ECU's box. I'm thinking of some sort of program whereby somebody could send in an OEM ECU, working or not, and have it gutted and retrofitted with a pre-programmed MS board for reinstallation. Or maybe it's not that easy.

[Captain Obvious]

Captain, I remember giving the answer similar to the one your cartoon on an organic chem midterm back in the Jurassic era. I got partial credit for the chutzpah. I drew a magician's hat with the reactants, a reaction arrow with magic words underneath and a magic wand overhead, a "poof" and then the same hat full of product.

Haha. Partial credit for the chutzpah. That's awesome!

The prepackaged MS option is certainly viable. I wonder if this is the kind of thing that the Diyautotune.com company in that other thread is planing to do.

I was thinking KISS. Less programmability than MS, but more target specific. In other words, it wouldn't need to be as complicated as MS, but it wouldn't be as adaptable either. Just does what the original system does and maybe a little more?

[TomoHawk]

I drew that cartoon in P-Chem class one day, and we spent the next hour debating the 'miracle' part. We actually came up with a mathematical algorithm (some quantum physics stuff) for the miracle, but it included a couple suppositions that were pretty much impossibilities themselves.

[Captain Obvious]

We actually came up with a mathematical algorithm (some quantum physics stuff) for the miracle

Haha! Nice. So you could either know how fast your miracle was going, or where it was located, but not both at the same time?

Am I the only one here who completely wiped chem from their memory? My transcripts say I took some chem, but other than that, I can't prove it!

[FastWoman]

Am I the only one here who completely wiped chem from their memory?

Hardly! I'm afraid most of my Francois is gone too.

[TomoHawk]

I'm into drugs... Er, I work at a pharmaceutical lab part time, so I get it more often than most.