• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

more info on volvo ecus (yay)

linuxman51

Railspeeder Enthusiast #1
300+ Club
Joined
Dec 25, 2002
Location
mont, AL
So having aquired a stash of ecus that represents an average of whats found here (lh 2.2/2.4/rex) i've got some stuff to share (this is a continuance of a post i made about two years ago on the subject at hand):

lh 2.2 NA:
fuel:
8.3mhz (crystal on the board)
no socket
Board layout is the same as the turbo ecu, most capacitors are the same values, however theres a large number of resistors that differ. at a glance both the turbo and na ecus appear the same, a closer look reveals several differences (of which I dont and dont care to track down)

lh 2.2 Turbo:
fuel
8.3 mhz
no socket
Ignition:
8.8 mhz, two connected boards. rather dull really.

lh 2.4 Turbo (no na ecu) last three digits are 563:
12 mhz (crystal)
approximately 2x the number of FET's as found on the lh 2.2 board
more diodes, and about 1/3rd the resistors.
socketed eeprom (thats chillin in my stack of chips). Larger harness connector than lh 2.2

Ignition:
8.0 mhz
one board (as compared with the earlier ezk versions that have two connected pcb's)
far fewer components, perhaps 1/4th the number of resistors, a few extra chips, one largely heatsinked fet. Part number on the siemens chip is: 2-227-355-302 (9019) (unresearched beyond what lh 2.4 info i've managed to glean. 16bit capable of running up to 16mhz iirc)


REX... Since info is largely sparse and from what i've been able to gather also largely incorrect, heres some thats correct with some things I noticed whilst playing with my rex car before ditching that and going with t3h ms:

FUEL:
there are multiple ecus, the 1990 ecus have one pcb, they run at 8mhz, and the processor is a motorola zc407531vfn (motorola's website doesnt pull up any info, still searching around)
more fets than the lh 2.4 board

later model ecu (p/n: 3531658)
has two pcb's
8mhz
several more ic's including what appears to be an external rom (soldered not socketed)

Ignition:
both appear to be the same (but then I could be looking at two 90 rex ICU's)
socket used is the same as the lh 2.2 ecu/icu and the 2.4 icu
Chip is a motorola ZC98909VP, running at 4mhz, aside from this its not any more or less exciting than the ezk 2.4 box.

as i research the various chips i'll post findings here. I'm more interested in the Rex stuff as its largely unknown, and well goddamnit i'm curious, even if its largely worthless from a performance point of view.

Rex notes:
It relies *heavily* on its o2 sensor (which is different from most other narrow bands that i've seen, but this could merely be a coincidence)
the slightest bend in the trigger wheel on the flex plate will yield wild results with the ecu, including but not limited to : it firing only on two cylinders (whoops) and/or it throwing a check engine code and reverting to fail safe ignition timing (bastard).
When coupled with a wideband hooked up to the ecu running narrow band sim, the ecu will hold the car almost perfectly at 14.7:1...... at all times. punching it will not yield a drastic change in air/fuels (it would occasionally go down to 13.9 for a second). without the narrowband connected the ecu pitches a huge fit.
with inaccurate narrow band readings it won't nesc. throw a CEL, however I noticed on mine when i'd connected the 5v wideband signal, the higher the voltage (hehehe whoops) even over 1v, the more fuel the ecu would pull, literally until the car died at 23:1. in an intersection (****er). and would not crank back up until the signal was disconnected, then it promptly threw a check engine light and I watched my fuel gauge plumet towards E.
How does Rex react to positive pressure? I dunno. One of yall try that **** out sometime and fill me in.
 
linuxman51 said:
Rex notes:
It relies *heavily* on its o2 sensor (which is different from most other narrow bands that i've seen, but this could merely be a coincidence)
the slightest bend in the trigger wheel on the flex plate will yield wild results with the ecu, including but not limited to : it firing only on two cylinders (whoops) and/or it throwing a check engine code and reverting to fail safe ignition timing (bastard).
When coupled with a wideband hooked up to the ecu running narrow band sim, the ecu will hold the car almost perfectly at 14.7:1...... at all times. punching it will not yield a drastic change in air/fuels (it would occasionally go down to 13.9 for a second).
I find the abovequoted snippet extremely interesting!
That would surely mean it's very happy coping with, for example, bigger injectors & more fuel & more air ... and even somewhat different fuel-delivery curves? I wonder what would happen if you stuck one on a technologically-similar engine, for example a SOHC turbocharged 2.0 Ford 'Pinto' out of a Mk2 Escort?
What are these things worth on the used market; and are they common?
 
Forg said:
I find the abovequoted snippet extremely interesting!
That would surely mean it's very happy coping with, for example, bigger injectors & more fuel & more air ... and even somewhat different fuel-delivery curves? I wonder what would happen if you stuck one on a technologically-similar engine, for example a SOHC turbocharged 2.0 Ford 'Pinto' out of a Mk2 Escort?
What are these things worth on the used market; and are they common?

well maybe. Rex is a speed denisty setup, volvos second and probably last pure speed density (fi'd 140's and 160's anyone?) setup, so you might run into problems with significantly higher flow, however by all indications it would figure itself out quickly. it also has a bit of retention for things like that, as evidenced when the car wouldn't crank back up.
 
The Rex system isn't really anything special. Compared to the Bosch systems, it advances the timing in larger steps (1.0deg v 0.375deg w/ the Bosch systems).

The green manual plays up the Rex system's adaptive abilities, EZ-116K has adaptive abilities as well. The Rex-1 system doesn't make use of an oxygen sensor input (neither do the EZ-K systems). The adaptive bits are interesting, but if I'm reading things correctly it'll take preemptive action, and retard the timing further (by at least 1degree) if it detects a prolonged condition of high load (EZ-116K does this as well).

One other thing to keep in mind, is that the Rex system is dependent upon the Regina brain for load information (it translates the MAP signal into a square waveform for the Regina brain). This is in contrast to the EZ-K systems which take the signal straight from the AMM.

The titania oxygen sensor that the Regina brain uses is supposedly able to respond faster than the more common zirconia sensors.

--
alex
 
Last edited:
blarf said:
The Rex system isn't really anything special. Compared to the Bosch systems, it advances the timing in larger steps (1.0deg v 0.375deg w/ the Bosch systems).

The green manual plays up the Rex system's adaptive abilities, but I believe that EZ-115/EZ-116K have adaptive abilities as well. The Rex-1 system doesn't make use of an oxygen sensor input (neither do the EZ-K systems). The adaptive bits are interesting, but if I'm reading things correctly it'll take preemptive action, and retard the timing further (by at least 1degree) if it detects a prolonged condition of high load.

One other thing to keep in mind, is that the Rex system is dependent upon the Regina brain for load information (it translates the MAP signal into a square waveform for the Regina brain). This is in contrast to the EZ-K systems which take the signal straight from the AMM.

The titania oxygen sensor that the Regina brain uses is supposedly able to respond faster than the more common zirconia sensors.

--
alex

lh 2.4 works the same way, just with that silly MAF. the fuel ecu relies on the ignition ecu to interpret the trigger wheel, and the ignition ecu relies on the fuel ecu for load.
lh 2.2 works this way as well.
 
linuxman51 said:
lh 2.4 works the same way, just with that silly MAF. the fuel ecu relies on the ignition ecu to interpret the trigger wheel, and the ignition ecu relies on the fuel ecu for load.
lh 2.2 works this way as well.

Well, no. The impression I got is that the EZ-K ICUs get the same AMM signals that the LH brains do. The Regina brain translates the voltage into a square wave. The LH brains at least don't have any fancy inputs from the EZ-K computers. The EZ-K brains will simply ground an appropriate terminal for stuff like knock enrichment. The LH brains, if I'm reading this properly, don't have any logic for determining when stuff like knock based enrichment is appropriate.

OTOH, a Motronic brain would, as Motronic combines the two brains.

--
alex
 
blarf said:
Well, no. The impression I got is that the EZ-K ICUs get the same AMM signals that the LH brains do. The Regina brain translates the voltage into a square wave. The LH brains at least don't have any fancy inputs from the EZ-K computers. The EZ-K brains will simply ground an appropriate terminal for stuff like knock enrichment. The LH brains, if I'm reading this properly, don't have any logic for determining when stuff like knock based enrichment is appropriate.

OTOH, a Motronic brain would, as Motronic combines the two brains.

--
alex


the lh/ ezk both have one input/output. lh sends ezk a signal that directly affects ignition timing, and ezk sends lh a signal thats presumeably knock enrichment.

lh 2.4 (and their respective ezk boxes) share the following info with the ignition ecu:
one wire from the maf goes to the ignition box, two wires go out from lh to ezk, and one comes from ezk to lh. (NA lh 2.4, assume turbo is wired the same as ecus are interchangeable)

lh 2.2 has no maf signal going directly to the ezk box, and has one line out and one line in (turbo system)

Rex has one signal line going out to the ignition box, and one signal line coming back into the ecu from the ignition box

the interesting thing is that in the wiring schematics for the various ezk ecus theres no indication that they get anything directly from the MAF, but straight from the lh ecus.
 
linuxman51 said:
the lh/ ezk both have one input/output. lh sends ezk a signal that directly affects ignition timing, and ezk sends lh a signal thats presumeably knock enrichment.

Right. Knock enrichment is achieved by grounding the appropriate terminal.

"On receiving a signal from the knock sensor (2) and having established that knock is present in all cylinders, the control unit (1) connects a terminal on the fuel injection system control unit (3) to ground, causing the latter to transmit a signal to the injectors (4) to extend the opening period."

All EZ-115/116K cars got this, and the B230FT w/ EZ-117K as well. #1 is the EZ-K brain.

linuxman51 said:
the interesting thing is that in the wiring schematics for the various ezk ecus theres no indication that they get anything directly from the MAF, but straight from the lh ecus.

Yeah, I realize that the MAF isn't connected directly to the ICU. My mistake, I thought that the MAF signal was passed via the ECU directly to the ICU. On both EZ-K and Rex systems, the fuel injection brain (LH or Regina) generate a "series of current pulses of a duration proportional to the load."

--
alex
 
interesting, which means that it could be simulated. I wonder what max advance and retard is on the various ecus?

Also of interesting note (posted before, but its been a while) theres a line going between the two that one can disconnect (i'm assuming its the load signal), put a potentiometer on, and dial in as much or as little timing as one could possibly want, deathwagon figured this one out messing around with lh back in the dark ages before he saw the light :-D
 
linuxman51 said:
interesting, which means that it could be simulated. I wonder what max advance and retard is on the various ecus?

Depends on the engine. Rex can retard the timing up to 15 degrees, EZ-K on most of the B230FTs up to 9.8 degrees, EZ-K on some of the other motors around 12 degrees. There's a lot of mention about retarded timing heating up the exhaust temperatures.

http://blarf.homeip.net/gallery/TP31397_1/p033

linuxman51 said:
Also of interesting note (posted before, but its been a while) theres a line going between the two that one can disconnect (i'm assuming its the load signal), put a potentiometer on, and dial in as much or as little timing as one could possibly want, deathwagon figured this one out messing around with lh back in the dark ages before he saw the light :-D

Yeah, that seems like an odd way to do things... but if it works. Honestly, I think there's more useful knowledge to be gained by reading up on EZ-K and Rex than by hacking it to work with a modded motor. Unless you start re-programming the EZ-K or Rex brains, you're ultimately still stuck with a base timing map designed to work in as many conditions as possible, with a stock motor. IOW even with a stock motor, it works okay most places, but not necessarily great.

OTOH, the design and function bits give some good ideas for stuff to implement with MSnS (for instance, the adaptation and knock prevention algorithms).

--
alex
 
blarf said:
What about it? The rev limiter is probably a function of the ignition control unit. Nothing special to see there.

--
alex


i think its actually in the fuel ecu somewhere, the guys hacking lh have been able to adjust the rev limiter somewhere in the lh boxes.
 
linuxman51 said:
i think its actually in the fuel ecu somewhere, the guys hacking lh have been able to adjust the rev limiter somewhere in the lh boxes.

On the EZ-117K ICU:

Pin 17 to LH 2.2 #1: "Transmits speed signal to fuel system control unit"
Pin 15 to LH 2.2 #12: "Transmits knock-controled fuel enrichment signal injection system control unit (B230FT only)"
Pin 8 to LH 2.2 #24: "Receives load signal from fuel system control unit"

Dollars to donuts it's just a normal tach signal (that's easy enough to check). Unfortunately I don't have an LH 2.2 book and the LH 2.4/EZ-116K book I do have is at my place (where I am not).

Of course I question the wisdom of running either system way off the maps.

--
alex
 
Kenny, this is a real long shot, but do you have any information at all regarding:

Electrical specs of the Rexx/Regina boxes, such as input & output acceptable voltages and loads? Any detailed info in this area would be nice.

The language/format used for the PROM on the ignition ECU?

I'm interested to see the code to see if anything can be changed...I've got access to a lot of nice equipment including surface mount soldering/desoldering stuff and read/writers for just about any type of chip you can throw at me. (Thank you WPI tuition...)


Also, am I understanding that with a wideband O2 sensor the Rexx/Regina works perfectly yet throws codes??

(Mike Baxter - Metalgod_Z posting from Mike volvo944ti's comp.....)
 
no doubt, you wont catch me out there fiddling with em, but info is info regardless. not really sure how it manipulates the rev limiter, cause when I unhooked the fuel ecu and was running ezk "open loop" on my car it did alright, i'm sure it was some sorta basic map, but there didnt seem to be any limiter anywhere
 
volvo944ti said:
Kenny, this is a real long shot, but do you have any information at all regarding:

Electrical specs of the Rexx/Regina boxes, such as input & output acceptable voltages and loads? Any detailed info in this area would be nice.

The language/format used for the PROM on the ignition ECU?

I'm interested to see the code to see if anything can be changed...I've got access to a lot of nice equipment including surface mount soldering/desoldering stuff and read/writers for just about any type of chip you can throw at me. (Thank you WPI tuition...)


Also, am I understanding that with a wideband O2 sensor the Rexx/Regina works perfectly yet throws codes??

(Mike Baxter - Metalgod_Z posting from Mike volvo944ti's comp.....)

I do not have access to such things, so I couldnt tell you whats on a rex ecu. Hey tell mike thanks again for the ecus and to lemme know what shipping ran (and whats up with these wheel chair feet props or whatever?). With the wideband hooked up on mine it threw a code, but I dont believe it was an o2 code (every time mine threw a code i got all my ignition timing taken away, i think there was something up with the crank pickup sensor). with the wideband hooked up to the ecu the car ran stoich pretty much 24/7, under acceleration the a/fs were closer to 14.2ish, but cruising it would rarely go over 14.8. It was pretty neat to watch, my fuel economy went up about 10 miles per tank (but my lackluster mileage had more to do with the codes the car was throwing elsewhere, and the code reader broke, never bothered to figure out what the problem really was, i just lopped rex off at the knees and installed megasquirt).
It would be interesting to watch lh's fuel curves with the stock narrow band hooked up on an otherwise well tuned car. I suspect on turbo cars it gets nasty rich, esp in higher than stock load levels.
 
linuxman51 said:
I do not have access to such things, so I couldnt tell you whats on a rex ecu.

I'll dig up the Rex/Regina book when I go back home (probably Monday or Tuesday). It's not as nice as a spare ECU, but it's a start.

linuxman51 said:
It would be interesting to watch lh's fuel curves with the stock narrow band hooked up on an otherwise well tuned car. I suspect on turbo cars it gets nasty rich, esp in higher than stock load levels.

http://www.carlton24v.co.uk/m15.htm

It's not a Jetronic ECU, but it's a start. I think some of the Saab guys have dumped the Jetronic ECUs.

--
alex
 
blarf said:
I'll dig up the Rex/Regina book when I go back home (probably Monday or Tuesday). It's not as nice as a spare ECU, but it's a start.



http://www.carlton24v.co.uk/m15.htm

It's not a Jetronic ECU, but it's a start. I think some of the Saab guys have dumped the Jetronic ECUs.

--
alex

I used mine as a soldering heatsink for a little while, then passed it on to jpbturbo, who then turned around and gave it back to me last weekend. I fulfilled the circle by taking it appart, ripping the connector off the board, and then tossed the board.
 
Back
Top