Chris.Palmer

Help Please!

Just completed building John's AFR Meter & want to connect it up.

However, I cannot find anything that matches the description of B136 on MY96 ECU.

Connected to the left side of MY96 ECU are 4 connectors - all yellow (not Blue Black & Grey as described in the PossumLink Series 5 WRX manual), and each connector has only two rows of pins not 4.

Am I looking in the wrong place or is the ECU & connectors different Pre 97.

Any help gratefully appreciated from anyone who has fitted John's AFR or a LamdaLink Meter to a UK MY96.

Cheers,

Chris

[Edited by Chris.Palmer - 1/26/2002 6:18:59 PM]

john banks

Hope you find the answer. There were similar issues with MY94 IIRC.

Scott.T

iTrader: (2)

On Coil packed Scoobs i.e B4 97' the ECU is different and has different connectors.

The wire you want is the 5th one in from the right (White with silver strips).
Or an easier way to remember, if the ECU is still mounted on the floor, it's the 5th wire up from the bottom.

That's were my MY94 is connected, just connected to it using a scotch lock connector.

Double check the colour tho's as I may be wrong about it being the 5th, as it may be the 4th.

mail me if your not sure and I'll check it for u and get back to u.


[Edited by Scott.T - 1/26/2002 8:09:39 PM]

Muncher

I used to be in the car stereo/car phone fitting trade a few years ago and I would not trust a scotch lock for tapping into an existing wire.

It might be fine for a while but I've seen cases (very few though) where they have failed. This might be ok when your car stereo ignition supply is connected using a scotch lock but a sensor input to a circuit controlling the solenoid duty cycle is probably a different matter.

Personally I would do a soldering job myself.

Giles.

john banks

Giles I have been wondering a bit about this. I am making an electronic boost controller and was using snap locks to tap into the loom to avoid cutting it - the plugs are very hard to source for the ECU to make an adaptor.

Do you just cut them, pull back the insulation and then solder together? How do you insulate? I have some of these connectors which you can put wire in and then crimp and heat them, but I never thought they were trustworthy.

Any advice welcome.

Chris.Palmer

Scott,

I assume you mean the first connector on the right, &
the fourth wire in from the right on the top row (although it should be the 5th but there is no 4th, if you see what I mean from the image below)

http://www.saxonfields.freeserve.co.uk/my96ecu.jpg

Please confirm.

Thanks in advance,

Chris.

PS. John in a vain attempt to try to contribute something to your herculian efforts. If your loom is as generous as mine, I would cut the wire solder to connect & use heatshrink to insulate.

If I ever get my damn AFR working I'll post some picks.

Thanks again to John, Scott & every one that's freely offered advice RE: DAWES & AFR.

PPS - I dread to think how many dumb questions you'll get when your EBC is working John.




[Edited by Chris.Palmer - 1/26/2002 11:12:15 PM]

Muncher

Hi John,

Yes I've been following you projects with interest over the past few months

You could solder in two ways: -

1. Cut the existing wire, place heat shrink sleeving back along the wire, remove insultation from the ends of the wires, solder 3 ends together. Pull heat shrink sleeving over the soldered part and apply some heat to make it shrink over the solder joint.

or

2. Use a pair of wire strippers (these are probably the best)



to strip the existing cable without cutting it (Try it on an old piece of wire if you haven't done this before). Solder your new wire in and tape up well with electrical insulating tape.

Option 1 gives added security of the heat shrink but means you must cut the existing cable.

Option 2 is easier and doesn't mean cutting the existing wire but you have the nagging doubt of using insulating tape that may unravel itself over time although unlikely as the ECU is in a relatively cold place away from heat and moisture.


Giles.

[Edited by Muncher - 1/26/2002 11:28:26 PM]

john banks

Thanks very much Giles.

Scott.T

iTrader: (2)

A Scotch lock is fine for the purpose of tapping in the AFR, as you are only monitoring whats going on, on that particular wire.

Plus it's a damn site quicker and also alot easier to remove when you want to take the AFR out.

I agree with the genral idea about scotch locks for stereo's and alarm fitting as this should alway be done using solder joints and heatshrink, as you don't want to run the risk of the stereo buggering about or the alarm generating false alarms due to a faulty scotch lock.

In Johns case I would use solder joints as the EBC that he has built relies heavily on getting a good signal from the ECU.

In tha case of fitting the AFR, you are only visually monitoring the signal. If the joint fails you will have no LED's.

I'm just gonna go and check the connection.....will be back in 5.

[Edited by Scott.T - 1/27/2002 10:19:43 AM]

Scott.T

iTrader: (2)

Connection is :

Top Row, 4th wire in from Right.
Which is position six as the 1st and 4th positions are un-used.

[Edited by Scott.T - 1/27/2002 10:28:41 AM]

Chris.Palmer

Cheers Scott, You're spot on - I missed the first one. Its a bit annoying that there are several white wires with silver stripes on a pre MY97!

Anyway my "John Bank's AFR meter" is in MY96 & working (I.E. LEDS come on & I can see stoich on cruise, & at WOT see approx. 8% Co2).

Would be interested to hear if other people get the same sort of results as a couple of things worry me :

a) With No Signal Wire connected to PIN 5 I get all LEDS on! A multimeter shows approx 2volts. This does not seem correct. Once Signal Source is connected multimeter shows 200 - 900 mv and operates as expected. Could there be a problem with my chip?

b) When starting from cold it takes a minute or so to get any readings on the AFR at all. Is this just the Lamda Sensor taking its time to start giving readings?

c) Before the car gets up to temperature it seems to run very rich - Eg WOT at 1/4 temp is 900mv - 10% Co2

d) Once up to temp WOT is approx 870 mv - 8% Co2 - as expected.

e) Having driven for over 1/2 hour - some cruise some WOT - the Car leans out & max Co2 at WOT drops to around 840mv 5% Co2.

Am I being paranoid? Are you seeing the same sort of results & is there anything to be concerned about.

Any Comments appeciated,

Thanks all,

Chris

MY96 - Running Optimax - Peak boost 13.5 PSI

john banks

a) Normal - floating input like a lot of chips

b) Yes

c) Expected and also magnified due to cold sensor

d) Nice

e) After a few minutes off boost then it should be 7% CO or higher. Part throttle boost or full boost driving can give higher EGTs and this will result in the sensor reading over-lean.

Scott.T

iTrader: (2)

Chris,

In answer to your questions, here are my thoughts

a) With no signal you will get strange readings. Before I fitted mine in the car the Min Led was always on very slightly and the Max LED was quite bright. Don't worry it's just a characteristic of the chip. It gets confused with no input source.

b) Yes, the Lambda sensor takes a while to heat up. If you just sit with the ignition on position 2, after a few minutes the LEDs will start to light. Mine went up to about 720mv. It's the heater element in the Lamda warming it up.

c) & d) The warm up cycle will run richer hence the higher readings from cold.

e) If you thrash it for several runs in a row the Lambda Overheats and starts to read low. This is a design fault due to the loaction of the Lamda on a Subaru.

I hope this puts your mind at rest.

At WOT mine lights the 840mV, 860mv & 880mV LED's

Chris.Palmer

Thanks for putting my mind at rest,

I'll keep my eye on it & see how it performs under "normal" driving conditions over the next few weeks.

Would also be interested to see how accurate the meter is after a few weeks and whether or not it is affected by operating temperatures & extensive usage.

Next Step Dawes!

Cheers,

Chris

john banks

After several weeks on the car my references are still within 1mV. Your errors are the limits of the sensor and its location.