• 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!

B204FT/GT getting rid of the EGTC sensor

bobxyz said:
As an educated guess, the circuit consists of:
- a LTC chip to compensate for the thermocouple junction (EGTC-2)
- a LTC op-amp to condition/boost the thermocouple signal (EGTC-1)
- a Signetics AU2901N quad comparator to generate the output signal (code 8944 = manufactured in 44th week of 1989)

I'd guess that the temperature is sliced into 3 ranges by the quad comparator, with 2 different pulldown resistors being connected to the output depending on the temperature range.

For LH2.4+EZ116K, the EZK grounds the knock signal from EZK to ECU to increase fueling. I think it's just a simple open-collector driver that results in either 12v or ~0v on the knock wire. Does the B204FT use a different version of the LH2.4 box, maybe one that can do a 3-way sensing of the knock signal voltage?

[If you're interested in thermocouple stuff from the late 1980s, try a search for "Linear Technology thermocouple amplifier application note" (LTC is now part of Analog Devices, so it may point you to their web site if they still post the old app notes.)
Click to expand...

Advice from someone that actually knows how things work :cheers:

One thing though... I really have changed my opinion on the input the fuel controller expects on pin 28. I believe the signal is Pulse Amplitude Modulation.

If indeed the fuel controller is expecting a PAM input and it is simply grounded instead, do you think the controller will throw a code for a missing signal?
 
I got a complete 16V turbo car and pulled all the ECU's and sensors etc... I was thinking about trying to use it but I will most likely just use a standalone ECU when its time to use that motor. I dont want to have to deal with having 3 different ECU's (spark, fuel, turbo).
 
Really this, I followed one guy struggling with his LH EZK stuff and modified chips. Live without logging is hell when tuning a engine. Cost for Ostrich, WB-lambda and big working AMM is not that much cheaper than having entry level ECU.

LH world is somehow close Commodore64. Nice to have as a hobby, and could be nice entertainment.
 
dl242gt said:
it's LH2.4. There is probably a performance chip you could use that ignores the sensor. Check with sbabbs or thelostartof.
Click to expand...
Nope. And no way to cheat its output. Has been tried in a number of ways years ago...
tomasss said:
Any ideas? Wanted to measure mine but found out it is actually already shot...going to disassemble it and try to make it work at least for the measurement....
Click to expand...
I was about to tell you not to open that and than I saw you already did... Proprietary IC's, Pentatronics that made those sensors is long defunct, last time one was available, it was ~1300€...

OttoB said:
Sensor is /was expensive from Volvo, like 700€/$. Some say it is K-type thermocouple, so there might be cheaper solutions.

I would throw away all Lh stuff and get more modern ECU.
Click to expand...
Sensor should be a thermocouple, I still think, but what gets us is that integrated proprietary controller...

OttoB said:
LH world is somehow close Commodore64. Nice to have as a hobby, and could be nice entertainment.
Click to expand...
This sums it up.

Tomas, I've replied to your PM just before seeing this thread...
 
We'll I could take the 16v turbo fuel map out of the stock bin and put it in say a 937 turbo 8v bin and then burn a chip and you could put that in any 9xx 8v ECU like 937,954,967,984 and wouldn't that be the same mostly except for no EGTC? Might be a few other maps to change, but are those critical? Wouldn't that work? So I'm saying use modified 8v ECU with 16v maps instead of 16v ECU. And use manual boost controller also.

LH is 8051 I think the commodore64 was a few bits more. But I know nothing about commodore64's. I did 6502 apple II's. 8 bit assembly and Basic and dos.
 
nordmaschine said:
Tomas, I've replied to your PM just before seeing this thread...
Click to expand...

Thanks. It's quite clear to me that cheating the TCU is not a big deal...just remove the EBC solenoid and replace it with MBC and bypass the LH<->EZK signals that goes through TCU and done. No matter what TCU wants to do, it can't do anything anymore.

But cheating the LH in my opinion has to be done in the software. I am currently in contact with two chip modders and they are both stating they can disable the EGTC check if I want. I still need more investigation with them, once I know more I will updated it here.

I was originally considering making some PWM/PAM circuit that will feed the pin with some constant "normal" signal but as I am anyways going to bypass the TCU and chip the LH/EZK this solution above would be cleaner.
 
Last edited:
Hi everyone, any news on fighting the EGTC? I presume that mine can be actually dead, and any help would be of use...
 
You must log in or register to reply here.
Back
Top