I moved up from a GL to a GTI with a 2e engine conversion. There are a few issues. Since i've had the car, the revs have stuck at 1100 or so for about 30 seconds when returning to idle. It's then been idling around 800 and the ECU has been trying to up it, as it would. The idle has been lumpy. I'm new to digis so i read up a bit and suspected the AFM and ISV. The AFM has been wound on at some point, sadly i don't know how much. I'd like to restore it to the factory position but need advice on doing that. The ISV seems to be working, when i disconnect it, the car invariably wants to stall. Today i discovered that the Lambda sensor wasn't connected and the connectors were just hanging around in the dirty bits. I've connected it up and made a temporary earth to the coil bolt. Then i did the procedure that i believe resets the ECU. Battery off for half hour, reconnect, ignition on for 15 seconds, key out, key in, fire up. I went for a few miles drive in an effort to get the ecu to learn to use the lambda. The car now seems to be running smoother but now the idle is hanging up at 1400 and eventually goes down to about 1100! The ecu then repeatedly raises it again Strangely, if i reverse before going to idle, it drops down straight away to idle just below 900. I've cleaned the breathers, bathed the isv in petrol overnight, checked the one vac pipe, checked the tps wiring, cleaned the throttle body. Could the PCV or it's bung be involved? I have been getting some mayo in the PCV. Eveything seems clear on the dipstick/ expansion tank front, could the oil cooler be leaking coolant into the oil and causing the bit of mayo? The ISV doesn't seem to sit right. It's literally resting on the cam cover so the pipe to the manifold is at a slight angle. The cam cover stops it sitting in line. The rubber connector between the isv and manifold doesn't look right to me. It pushes over the pipe on the isv and into the stub on the manifold. The manifold stub has a lip (like the ones on coolant hose stubs that jubilee clips sit behind). To me that implies the connector should go outside the manifold stub and be clipped. Is it possible that this arrangement is causing a slight air leak and that may be involved in the high idle? Does anyone have any close ups of a stock 2e isv connection? I know i've asked a few q's but any help/input would be greatly appreciated. Cheers
I'm assuming its a complete 2e engine + managment conversion? step 1 on these is to clean the throttle body. there is no idle or co adjustment, and if someone has fiddled with the afm I'd just bin it and get another from the scrappy from a 2e code engine, early passat, vento and mk3 gti all came with the 2e. Get the isv from the same car, always good to have spares. The isv pipe should indeed push over trhe lip not inside, so grab the isv pipework off the donor car as well. the pcv will get a gfair bit of mayo in if you do lots of short journeys, if the oil cooler had packed in you would be seeing oil in the coolant as well.
Cheers (you forgot to mention Corrado ) There is co adjustment on the afm. Good steer on the isv pipework, my hunch was that's the way it should be. Yes it's complete 2e. I did a 40 mile run today with the lambda attached. The car guzzled fuel, stank of petrol and suffered at mid range revs. So when i got home, i detached it again. I took the cover off the afm and found that the car idled a lot smoother if i knocked it 3 notches anti clockwise ( i thought that made it run richer!). After that, i wound the co screw out 4 turns and the idle got pretty steady at 950ish. That's an improvement but it still takes a few seconds to return to idle. I'm wondering if that isv pipe may be involved or if it's a sticking throttle body butterfly or badly set TPS. It isn't a sticky cable cos the car does it with the cable detached, that much i do know.
If it helps, I have bits off a 2E that I'm not using (not sure exactly which bits you need as I am clueless with digi stuff) but basically a full air intake doofer with some wiring and same with exhaust. I basically bought a 2E complete and have stripped the head and digi stuff off to graft a carb head onto it.
Cheers Funky. I guess i could really do with a lambda probe, Air flow meter, throttle body (for the TPS), idle stabilization valve and it's pipework. In theory i reckon one or all of those are involved. Loom might be good as well .Drop me a message if you fancy parting with some bits.
dont bother with a 2nd hand lambda, new bosch ones can be had for 40-50. corrado is a possibilty but so rare to find a 2.0 8v one not really worth mentioning. trouble with fiddlign with the afm spring is as I say there is no co adjustment to check if its set properly, i have the co checking values in the elsawin thread so that might be a clue. but personally id just swap it for another one thats not been frigged with. it could be the car was set mega lean, and the lambda is tellign the ecu to dump in a ton of fuel to correct it, or could just be the lambda is shafted. best to get it on vag-com, see if ecu has logged any faults. plus can check the measuring blocks to see what tvalues the ecu is seeing for coolant temp and throttle position
Yeah i was thinking along the lines of the lambda compensating for an over lean setting but without it being fubar i can't quite get how it'd result in a stupidly rich mix. Surely it would be it's own policeman so to speak (if it's working fine). At least i now understand (a bit) why the last owner disconnected it! Last car i had was 2e carbed and after many hours reading the carb bible i fully got my head around it. I think i need to find a digi bible now. For now i've let my ears and the car's better running judge that i've got the afm set somewhere near right but i take your point that a known good unfettled afm is probably the best way from here.
if the lambda is buggered it'll return incorrect values to the ecu, which could result in too much or too little fuel injected, i know our polo was really crap with a buggered lambda. the ecu temp sensor could be throwing things out too, which is why its best to get vag-com on it before you go any further to see what values the ecu is seeing from all the sensors
Yeah i think i'd be inclined to agree that a diagnostic is in order. Haven't even managed to find the port yet though! Jeepers knows where it got stuffed when the conversion was done. I'd also be inclined to agree that the lambda is shot. Like i said, i can't see how it would cause bad overfuelling if it wasn't. As far as the temp sender goes, i'm not convinced that's ok either. The car has the 4 pin combined ecu/dash sender. The dash gauge only ever get's to 90 degrees, even when the mfa gives the oil temp as being much hotter and the fan kicks in. I have tried another sender from a scrapper though with the same results. Maybe a coincidence that both have the same fault, maybe not. Makes me suspect the loom a bit, would vagcom be able to pick up a loom issue though?
diagnostics port may not have been wired in., they often get forgotten. look for a grey/whitre wire behind the fusebox with a brown plug
I've just been doing a bit of googling about lambdas and decat zorts. People seem to say that the lambda probe won't get hot enough without a cat and will then send bad messages to the ecu. Others say that with the lambda disconnected the ecu will resort to limp mode. I don't think that's happening because it still pulls like a train all the way up to 6.5k. So what's the real story with lambdas and decats? And what's the deal with having the lambda disconnected? Also where does a decat stand legally on a pre 92 car with a post 92 engine? Is it purely down to emission levels? Lot's of questions i know sorry
there are 2 different lambda probes available depending if there is a cat equipped or not, the diffence being the heater in the non-cat lambda is a higher wattage presumably due to then heat issue you mention. it wont go into limp without the lambda, but it goes to a kind of base map. I ran my abf without a lambda for a week and it ran fine. once I fitted the lambda it wasnt a night & day differnce, but it did feel slightly better. its hard to explain but it just seemed to drive ever so slightly smoother
Ok, i've just taken a quick snap of the fuse box. As you can see, there is the brown connector with a grey/white wire. There's also a green 12 pin connector and the one with the puple wire. Couls anyone tell me which is what and what i need to do to be able to get vag-com onto it? Cheers.
the grey/white wire plug you have there looks like the vag-com port end rather than the engine end. best thing to do is drop the fusebox and follow these wires to where they come from. its not a good plan to just go by the wire colours, as sometimes the colours are different. if its a signle grey/white wire coming out with the engine loom you can be pretty certain its for the diagnostics though
Cheers, assuming it is the diagnostic, would it then have to plug into anything else or straight into vag-com?
Have got as far as fitting a known good afm (a 600 brake bill has diverted my money ). The car does now seem to be running a bit better but it's still not ideal. The revs now drop down smoothly when you lift off but they drop down to 800 and obviously the engine idles lumpily as a result. From cold the idle is around 1000rpm but still lumpy. Power delivery seems pretty smooth though. A new lambda probe and temp sender are on the list. I'm not sure about the temp sender, i've tried two used ones and the car behaved the same with both, also the dash gauge seems to function as you'd expect (it's a 4 pin sender). I'm just going to replace it as they are a cheap bit of peace of mind. Is the lambda probe involved at idle?
