SW20 jdm 3sge pinout. WHAT! No G- pin?? (D8 or D-? instead) huh? updated with pics.

[bajachris88]

Update:

Got out there this afternoon...

the good:

Loom and ECU have matching G-, G1, G2, Ne, ACIS, KNK points.

the bad:

ECU has pin for FPU 5th pin from the top-right of the 26P plug, loom does not have any pin there its left as a blank at the 5th pin. BUT on either side of the 5th pin, at the 6 th pin from right there is a cut brown and yellow wire, and at the 4th pin from right is L-Br (Blue brown in the SP2) pin location. It was an automatic engine. The fuel pressure regulator has one vacuum fitting thats plumbed straight into the intake manifold. I put it there after revising with new lines. i assumed it was correct.

I don't think this will have any devistating effects anyways. My original 3sge from the st162 didn't have any fpu provisions....
But does it sound unusual? no worries? it won't confuse the ecu?
I still got a handful of pins to check out... but its too dark now.

will update further soon.

Chris.

[bajachris88]

NEW UPDATE: pin E2 of ecu is not connected at all, it is not grounded, i will do rectify and report back.

would that have devistating effects?

[bajachris88]

I'm writing up an update, will be a few mins ish...

[bajachris88]

I did some extensive research using the BGB service tips with voltage checks as supplied for the st18 series for the USDM.

IE: wasn't for 3sge, but for 3sgte and 4afe, 5sfe etc...
Seeing i don't have a bgb for 3sge. but the running methods and outputs i assumed would be equal and relavent.

E2 (signal earth) was earthed on teh chassis, rather than going to the ecu to earth the signals. This was corrected.

the 12 volts in the black-yellow wire to supply current to the ACIS vsv, ISC and check connector etc was not connected. It is now soldered to the black and yellow lead from the efi main relay, and now has 12 volt supply.

ONLY WIERDNESS:

1) At terminal IDL from the ecu, the ecu is pumping out 11.4 volts approx, RATHER than 4.5 - 5.5 volts with the throttle open and ignition switch on, as described as it should from the 3sgte st185 ecu. IT IS the ecu pumping out this higher voltage, there is no wrong connections to the pink wire to supply such a voltage because i cut it at the ecu connector and used the multimeter to make sure.

AND!

4) I have never recieved a flashing check engine light, yet the engine light shines when i turn the key to on and try to start. despite all the issues with the sensor wiring that i had corrected.

EVEN when i have a sensor or component like the MAP sensor unplugged, the ecu still doesn't shine with a flashing check engine light.

3) THERE IS STILL NO FUEL SUPPLY.
Yet the injectors are getting current, and i can get it to start and run for a VERY short time if i spray fuel in the throttle body SUGGESTING ALL ignition and timing components are working fine.

Injectors still recieving 12.5ish volts, E01 and E02 are earthed.

COULD THE ECU BE STUFFED?
How does one test this out?
I don't believe it could be an injector not working, as it is highly unlikely that ALL injectors would malfunction at the VERY same time.



I dunno...

Are auto electricians relatively expensive?

[thechuckster]

3 - this suggests suggests no IGf (or what ever the Good Ignition Signal is called - am not at home) signal from igniter back to ECU - and possibly other signals from dizzi are not getting back to the ECU

4 - check engine light is weird. You have 12v supply to it and the ECU provides the earth? If you earth the check engine light at the ECU (disconnect plug from ECu first), does light come on?


electricians are expensive and will make you pay for (what they see as) butchering the loom

[jonathan yee]

is the fuel pump coming on? the B-Y wire that you connected for the ACIS should also be goin to the circuit opening relay for the fuel pump.

i could lend you my 93 ecu but then you'd have to reverse all the wiring you've done.....damn!

[bajachris88]

I checked the ecu inputs for determining crank shaft positioning. According to the manuals for the Turbo varient (3sgte rather than 3sge), crankshaft position is input as a control signal to terminal G1 and G2 of the ecu. RPM input terminal is Ne.

I checked out the G-, G1, G2 and Ne in the loom. thats all connected right. But i'm assuming 3sgte is same as 3sge with the dissy.

But the pinout on the ecu is still in question.. particularly when the printed circuit board does not have ANY pin out of all of them labelled G-.

I used the resistance tests via the "Service Tips" to check on G1, G2 and Ne and they all checked out fine.

NOW! take a look at this! this is my pinout, and the D- or D8... wateva way you read it pin, is what i have been 'taking' as G-... as there is no G- printed on teh circuit board and i assumed it was so. If this D- is not G-, then that will explain why the ecu has no idea of the position of the crankshaft, and hence no injector pulse.




What do you guys rekon about this?

[bajachris88]

Charles, i will check the IGf signal and report back.



Its all good Jonathan, i was reading quite a few articles and they were saying ecu failure is quite rare, although possible. I got a spare one of the same part number lined up on ebay that i'm watching just in case, but don't wanna have to wait the 2 weeks to get it from teh uk, and pay the 40 or so pounds... I'll go there if i need to .

And chances are that its not the ecu, although it makes it an easier fix i guess. The componentry doesn't look or smell burnt... we'll see what happens.

[thechuckster]

D8 is not G-

[bajachris88]

Yea ur right... But if u squint hard enough it looks like D(-).... lol.

I figured it made some 'small' sense cause in the pinouts G- is located there... depending on the yr of manufacture of course. Otherwise... well, there is no pin labelled G- at all. And where the 'alternate' location of G- is... as you can see its not labelled and not soldered to any printed circuit on the circuit board.

i checked IGt, its got a voltage that bounces around near the range... so it looks good.

BUT DO NOTE:

I cranked it over... and checked the check connector (for auto diagnosis)... i got the following error code and nuffin more. this confirms the ecu is somewhat operating so far, + the warning light setup is correct.

code: | | - |

According to the st184/st185 manuals, this is just a warning for the oxygen sensor i haven't got plugged in as i'm in the process of ordering (4 prong one). But i have a single prong one blocking it off and sealing the exhaust. I have always assumed it would not stop the engine running, as it would just decide to operate in Open Loop mode as opposed to closed loop, hence why it has never been a massive priority on my list of 'to do's', as opposed to getting it to run.

If you think this ideology is flawed please tell me. It would be good news to know that this is whats stopping me

[jonathan yee]

i'll list the differences between the post/pre dec91 ecus.

91/93
RSO:B22/b10
STJ/EVP:B7/B10
FPU:B5/B6
G2:B2/B5
NE:B1/B4
G-:C10/B17
ACIS:B18/B22
KNK:C5/C13

also were the KNK is on the 93 ecu the 91 has VAF which goes to some variable resistor which is also linked to the VC which is a signal that goes to the vac sensor and the throttle position sensor.

and where AC2 is on the 93 ecu which isnt in the bgb its PS in the 91 bgb.

and finally TE2 in the 93 bgb/ecu is OX2 in the 91 bgb

just quoting myself.....G- should be at pin C10 or B17...what do you have at those pins?

[jonathan yee]

actually G- is B17 post 91 and C16 pre dec 91 dunno where i pulled C10 from.

so C16 is where you have that funny D6/B?

[jonathan yee]

delete....dumbarse me just repeated what you already said.

another thought that may be possible is if you have removed the dizzy at any point? it may have been installed wrong by 180deg.

is there an NSW on the ecu at all?

[thechuckster]

dizzi should only go one way? the dog teeth are slightly offset (or at least are in a 3sgte)

[bajachris88]

Yea thats correct + the 3sgte's use the same distributor. the teeth going into the camshaft is slightly to the side for a one way installation.

The oxygen sensor was posted from sydney on tues... be nice to get it by friday. will keep yas updated.

Its a second hand one for $30, i can't believe the brand new price! GOSH!