Hi Everybody, Right I've put an AGU 1.8T into my 1997 Mk3 Golf using the OE Ecu. The problem is that when I try and use Vagcom(genuine Ross Tech) to talk to the ecu, it only works with the engine running. If I try it with just the ignition on I get either "no response from controller" or "too many communication errors". Also I'm getting a fault code 18010 - terminal 30 voltage too low, but the battery is new, and the alternator is charging fine. I haven't checked anything yet as I haven't had a chance, but I'm thinking it might be a wiring issue, or could it be the terminal 30 thing thats screwing up Vagcom, surely not though as it works fine with engine running. Anybody got any suggestions of were/what I should be checking. Thanks from Paul.
thats usually an indication of ecu relay failure, i think on your car it'll be inside the big plastic box in the engine bay somewhere
If the ecu relay had failed though, surely the car wouldn't start. "big plastic box in the engine bay" are you sure you mean a Mk3, or are you talking about the engine? Seriously though I can't think of any big box in the engine bay of a mk3, there is an ecu relay in the fuse box though, number 30 I think.
no it can still run with a failed relay, but car wont be quite right. missed it was fitted to a mk3, doh! on the 1.8t loom in the engine bay there would have been a big plastic box which had 2 relays inside. what did you do with it when you converted the loom to fit the mk3? the 30 relay is indeed the ecu relay, the number matching the terminal 30 code is just a co-incidence though terminal 30 refers to the permanent live feed, this is what the permanent live circuits are referred to in German wiring diagrams you see.
Ah right, so it could be the relay then. I didn't do the wiring, but I think the 1.8t AGU loom was just spliced into the AGG one, so the AGU is running off the ecu relay in the fuse box. So the two relays you refer to are gone. From the wiring diagrams I've got it looks to me that an AGU gets a permanent 30 supply all the time with no relay involved, but the AGG engine works through the ecu relay ie getting 30 power only when the ignition is switched on. So, if I'm right, that means in my installation the AGU ecu will be getting no 30 power everytime I switch off, so its throwing that fault code. Does that sound right to you? PS : After 8 months of messing with this car I'm slowly getting my head round this wiring lark - 30 is permanent battery power, 15 is switched live, 31 is earth. See.
If the engine can still run with a failed relay does that mean the ecu is pulling all its power from the 15 supply
hmm actually yes you're right, no relay on the AGU! the loom I worked on was an aum/auq which has a relay. Sorry about that! yep you got it re: the circuits X is x-over relay output (switches off when cranking things like lights, wipers etc) 50 is starter live I would check ecu pin 3 is getting a good live feed, while you're there might as well check pin1 is ignition live too, and is a proper ign 15 feed not an X feed, oh and check pin 67 has good ground connection too If I had made this loom I would have run pin 1 from the old ECU relay and pin 3 to a fused feed from one of the Y pins on the back of the fusebox. Some MK3s have spare spades on the jumper wire from 30 to 30b, so that would be another possible place. could get a live direct from the battery too, depends how the loom was put together.
I'm sure the relay is clicking on when I turn the ignition, I can certainly hear something click, but I'll check it properly of course. RJ, I'll get my multimeter out this weekend and go over all the pins you said. In my mind it's got to be something simple otherwise vagcom wouldn't work at all, and neither would the car for that matter. Thanks for your input guys, at least I know were I should be looking now, I'll let you know what happens. BTW RJ your fusebox and reading wiring diagrams guides have been invaluable, thanks.
Update on this. I've checked pins 1 and 3 and they are getting power. Pin 3 is getting 12.5V, however pin 1 is only getting 11.7V when the ignition is on. I didn't check the voltage with the engine running as I couldn't get into the back of the ecu plug. I will do that though. I think it might be possible that the ignition switch might be at fault, which wouldn't be a surprise considering it is nearly 14 years old and done 171,000 miles. I've gone over all the earth terminals, including the major ones from the gearbox etc, and cleaned them up. I'm going to run extra earth cables just to be sure, including one from the ecu straight to the battery. The original ecu relay is not an issue, as it turns out the ecu is not being powered from that. I haven't traced where it is being powered from yet. More to come.
another check, make sure pin 1 still gets voltage when the car is cranking over. could be they used an X circuit not a 15 circuit like you're supposed to
yup I've still to do that. I didn't have anybody to help at the weekend and the leads on my multimeter don't stretch far enough. I've bought some extra cable now, so am going to attack it properly this weekend eg probe the back of the ecu plug to get voltages with the engine running.
if it was me I'd wire it up so pin 1 gets power from the ecu relay, just becasue thats how mk3 is wired from factory. trouble is you gotta find where pin 1 ends up in your loom!
This is good practice. I have my MS and wideband lambda wired to the fusebox as the original was. Also makes fault finding easier.
I agree with both of you. I thought this is how it would have been done since it would have been easy to just splice the appropriate wires from each loom together. Fault finding is obviously the reason why I want to get Vagcom to communicate with the engine ecu with ignition only, eg how could I find out the fault codes if the engine won't start? This problem is also not made easy to solve due to my inexperience, but I'm learning fast and am now seriously considering pulling the engine out and redoing the wiring myself, spurred on as well because of the fact that this is not the first problem I've had to chase with this wiring job. I getting ever so slightly p****d off with it. Don't worry though, this thing is not going to win, I will fix it. One way or another.
Don't think you need to remove an engine to do the wiring. You need to study the cars diagrams and rubjonnys fusebox FAQ. You can then cross reference it all. May be worth getting a meter if you don't have one.
Removing the engine would make taking all the wiring out alot easier as it is really tight around the brake servo as the wiring goes through the firewall. If I did do it then I'd want the whole loom out so I could rewire using complete lengths of new wire rather than splicing in etc. I do have a multimeter and am slowly getting to grips with the wiring, I've got all the wiring diagrams I need as well as I have Elsawin. Did a bit more tonight and have identified where the terminal 30 power is coming from - G2/9 on the fusebox - same as a VR6 ecu AFAIK.
No need to pull loom either. You're just gonna make work for yourself. A meter can be used to trace all connections. I managed all my wiring with dash and engine in place. Just removed underdash panels and clocks. Splicing to existing loom is probably best. Solder and heatshrink connections. Not twisting an taping.
you can pull the loom with the engine still in, just need to pull servo forwards a little then you can get it out. ive even managed to remove the servo from the car with the engine and inlet in place, that was a bit of a mission I can tell you i would definitely remove the loom if i was in the same situation, then i would strip all the wire wrapping off and inspect the whole thing end to end to see whats been done and to what standard. G2/9 is fine to use aye, same as later VR. ecu relay output is G1/10
I can imagine what you mean RJ about getting the servo out with the engine still in, its tight. Don't even mention getting the brake pedal off the servo rod. I got the special tool from VW and even that didn't work properly. Anyway, I've done more digging about in the car(its in pieces now, battery out etc) and I've now found where the ignition switched live(15) is coming from - G1/3. That goes to T80/1 on the AGU ecu. That's the 85 pin from the fuel pump relay. That doesn't seem right to me, because on the AGU wiring diagram T80/1 just goes to a straight 15 supply, through a fuse of course, and T80/4 takes the pin 85 from the fuel pump relay. But T80/4 is going to to G1/4 on my setup. In which case my setup is kind of the wrong way around. I mean going by the AGU wiring diagrams T80/4 should be pulling the 15 supply down from the fuel pump relay, and T80/1 should be getting a clean 15 supply. What do you think, am I going mad, or is this maybe the answer.
