Hello All,

I thought I'd start a new thread to get some feedback on my ZFuel project. To recap, I began a tuneable LJet ECM replacement long long ago in the early 90s, stalled out due to life/growing business, and then rekindled it last year when my son was driving a 77 280. He went away (along with car) to college, so again I stalled on development. I'm glad to say that I have the car back now, and I'm excited about getting this unit off the drawing board and into the garage. I think at the rate that AFM's and ECMs are aging, it's long overdue. Here is a quick overview:

Standard Features

-100% drop in replacement for stock (should be a 10min install - pull kickplate, ecm, unplug old, plug new, hit the starter)

-Also can run in Speed-Denisty (MAP) if you wish. This would allow you to nuke the stock AFM. I think this will be handy for anyone who thinks their AFM is suspect with aging wipers)

- Realtime data monitoring via USB/Laptop (now you can check the status of those pesky sensors easily)

- Internal diagnostics/logging of sensors.

- You can of course tune it and adjust the map with the laptop interface.

- Run any size injectors you want to plumb in (within reason of course)

- Tune for aftermarket performance parts (cam, headers, porting, ect)

- Some general purpose I/O for shift lights, ect.

- Will have CAN bus output on an optional connector (thinking aftermarket gauges).

- Idle air control/quick warmup mode

- Support for O2 sensor and closed loop mode.

- Run any throttle body and TB sensor you wish. (analog position, or just off-idle/WOT)

- copy/save/email tunes back and forth between users.

The primary concept is to have it plug-n-play with a bone stock unit, yet still have some flexibility to accommodate other configurations. While one could run it alongside ALL the existing stock equipment, to me the following would be an ideal setup. I say ideal mostly because all the additional plumbing/wiring complexity of the L-Jet under the hood annoys me.

- Replace the AFM with a straight pipe and run with the Map sensor. Now, no worries about the aging AFM, no restriction for incoming airflow, no but ugly AFM in the engine compartment with it's 40year old connectors.

- Leaky cold start valve? Remove all of this nonsense & the accompanying plumbing/thermo-time switch/wiring.

- AAR would be replaced with more modern idle air control. (or just remove it completely if you are in a warmer climate).

Well, that's the concept/idea. I'm actively soliciting feedback as to likes/dislikes or things I might have not thought of. Ignition timing is also possible, although I'm going to ignore that aspect on the first pass.

Superlen

Excellent feedback everyone.

Captain, Yes please email me the xls you have if you don't mind. I remember you saying the points shown in the FSM were not quite spot on. My readings confirm that.

However, assuming this calibration is correct, it could become the closest thing we have to a gold standard, unless someone somewhere has a better AFM, perhaps a NOS unit, from which we could take measurements. I'm not confident that a refurb AFM is really calibrated any better, using any more "correct" standards, than what could be done with the AltlanticZ method.

Sarah, I agree with this.. I have looked at the Atlantic Z site before, but I didn't realize it was Blue's creation. I also agree with your thoughts about a loosened clock wheel is going to peg out at higher RPMS. With the stock ECU we just had to live with it. For HellFire the simple solution is to tighten the clock spring until the 4500-5000rpm range does NOT peg it. Then we can calibrate the lower portion no problem. Poof! Now we have a nice full scale to use.

As for the inputs being spot on in order to have it work perfect out of the box, I totally agree. However, If you have a good running Z now & you plug & play a HellFire, you will still have a good running Z(and you can tweak it to a great running Z). The same goes for a crappy running Z. Assuming it wasn't the ECU just being thrashed, plug & play HellFire will still result in a crappy running Z. However, this is exciting to me as now you can get all kinds of feedback via the GUI to troubleshoot your engine. we can validate your temp sensor(s) are close or waaaay off. We can validate your AFM doesn't have dead spots & we can validate the clock spring adjustment is reasonable. We can validate your idle switch & WOT switch operate properly. HellFire can also read shorted and open injector circuits. We can't *easily* tell if your injectors are spraying correctly, but if one want's to pull the rail & stick a graduated cylinder under each injector (one at a time of course) you can run an injector test and it will tell you the exact flow of each injector using *your* fuel pressure. This would even calibrate out an adjustable regulator if someone was running 40psi or something silly. I need to get busy. I have lots to do.

Blue,

I have been thinking about building a flowbench as well, but too much to do. My redneck solution was to use the Infinity Q45 MAF plumbed in front of my stock AFM. The Infinity MAF has a know transfer function that I can read on the fly & then do the math using that output plus IAT to derive the CFM of the stock AFM. My software is written to support MAF anyway so it's not much extra work. The MAF sensor reading comes in on the wire that used to be the AFM Reference voltage so the wire is already right there at the correct location in the engine compartment. ( I don't need the reference as I'm sending regulated +5V to the AFM, not bat voltage) It would be awesome to compare the graph I get with your flowbench measurements. We could send some AFMs back and forth to compare apples to apples too. As you mentioned VE would work too, but I don't have a starting map for the VE of a stock LJet. I'm sure people have them from MS installations that would be close. My VE tables are 16x16. (smaller works too, it's configurable, just max is 16x16)

Zed,

I agree with you and I'm sure I'm going to be doing some of that, but I would still like to know as close as possible the real numbers. That way I won't be tuning the AFM thinking it's off when really it's some other issue I have that's causing the AFR to be off. I do think there is still going to be some "seat of the pants" tuning. Or is it playing? I'm not sure. haha.

Lenny

Lenny, your idea of tightening the clock spring a tad for expanding the upper end of airflow metering might work, except you have to be mindful that the earlier 280 models have the fuel pump cutoff switch in their AFM. If the AFM vane does not hover open far enough, the fuel pump can cut off. Apparently this is already a tedious little dance with some Z owners.

I love that the Hellfire will be such a powerful diagnostic tool. I hadn't even considered that. On that subject, you say you can "easily" tell whether injectors are delivering fuel (and/or whether there is problem with a cylinder?). I presume you would do this by selectively disabling individual injectors and measuring the effects on RPM. Yes?

Sarah,

I said we *can't* easily tell if the injectors are delivering the proper amount. At least that's what I was trying to imply.

I have thought about selectively disabling injections to check if a cylinder is misfiring/injector clogged, ect, but I haven't spent any time on it. What's more interesting to me is rotating through a squirting sequence in an attempt to do sequential injection. Without a sync pulse of some sort, I don't know which cylinder is where so sequential isn't really possible. However, if I rotated through all 6 possibilities and monitored rpm and MAP I *might* be able to detect which of those sequences resulted in the most stable idle & then use it. From my research sequential doesn't buy a whole lot, but it does help make a nice stable idle.

BTW, I was unfamiliar with the fuel pump switch touchiness on some models. The user would just have to make sure that their switch was adjusted so that it operated at idle or just turn over fuel pump control to HellFire and forget about the AFM switch.

Note: There is a slight caveat with the HF taking over fuel pump control. The Z's wiring harness is wired for the ECU to control the relay if it likes (via pin 20), however Bosch didn't load pin 20 on all ECU's. Even ECUs with the same exact part number don't necessarily have that pin loaded. So depending on what ECU someone sends in as a core they may/may not have that functionality out of the box. Because of this, I routed the signal to both pin 20 & pin 11. Pin 11 is unused in the harness, but is always loaded on the connector.

So...if one desires fuel pump control, but has pin 20 missing they need to move the wire in their harness connector from pin20 to pin 11. There's no soldering or cutting & the harness will appear perfectly stock afterward. Taking the connector apart & moving the wire/pin from one location to the other & putting back together is all that is required. A small sharp pick/tool will be needed to let the pin release from the body.

Lenny

Captain,

FYI, I keyed in the equation so I'm good to go. No need to email something. Thanks,

Lenny

FYI, I keyed in the equation so I'm good to go.

Lenny, Cool. So do the calculated numbers more closely match your measured?

( I don't need the reference as I'm sending regulated +5V to the AFM, not bat voltage)

Also, I don't remember where I read it, but I thought the reference was also intended to account for temp drift of the AFM resistance. Let me see if I can dig that up or if I'm imagining things. :bulb:

I created my table, but I haven't downloaded yet. I need to change out a pullup to adjust my range some & I built the table around the new pullup. I'll test tomorrow.

Also, it seems plausible the reference is also used for drift, but I don't recall coming across that anywhere. I do have it wired up in the stock configuration so I could use it if necessary, I just chose to ignore it.

Len

I love the thought of being able to drive my 77 280 without having to make adjustments almost every time I drive her! Now fine tuning is another story. That would be enjoyable. I swear my ECU floats from lean to rich and back.

Rail voltage can drift (bat 12.0V to bat 12.8V to alt 12.8v to alt 14.8 v or more) so cleverly real-time-scaling by measuring a reference (the value of a voltage divider circuit) can help keep thing under control. Especially during transients to the bus (such as turning on lights)

Blue,

I do measure Battery + on board the ECU. I need it for modeling the injector dynamics for a low/high battery. The stock ECU does this too, but I'm assuming because of the lack of technology at the time (or maybe "complexity needed" is a better term) they couldn't use the same reference on the AFM input circuit so they had a sensor in there as well. I suppose possibly the ref sensor in the AFM is what they use to determine their BAT + for injector dynamics, but I wouldn't understand why they stuck it there if thats the case.

Len

Blue,

I forgot to add. Have you seen this forum: Home There is lots of good info there and seem to be a friendly bunch.

Len

I think because the old ECU was 100% an analogue computer, they had to do everything relative to the supply voltage and use ways to compensate where required...like the AFM. If your system monitors the DC bus then you can compensate in code.

And I'll give you a different perspective...

I would derive all of my analog stuff using precision reference voltages or current sources. I wouldn't want to have to know what the battery voltage was unless I really needed to. You might be able to figure out a way such that you do not care what VBatt was at all. :bulb:

I'm assuming there's a good stable (maybe even band gap) reference source used by the on-board A/D's and D/A's? There's no way they're using the digital supply for conversions, right? Use the same or another good reference source to drive the sensors and the battery voltage no longer matters.

You might still need to know when it comes to the injector pulse because the opening speed is so influenced by system voltage, but other than that, I'm thinking you shouldn't care.