elbekko Posted November 21, 2012 Share Posted November 21, 2012 About a month ago I bought a '99 P38 with a Bosch (Thor) 4.6 V8. It wouldn't keep running when we went to pick it up, after changing the plugs and draining some sketchy fuel, it fired up but was running very rich. Eventually we tracked it down to the MAF (after replacing the engine wiring loom because the LPG integration wasn't done very well), as it drove a lot better with it disconnected. Also testdrove with a friend's MAF, which made it run very well. Clues were: plugs black as night, backfiring massively in the exhaust. So I drove it without a MAF on LPG for a few weeks while waiting for a new MAF to arrive, all seemed fine, but a bit rich as it was managing about 21l/100km. Last week the new MAF arrived, I plugged it in and drove it for a week on petrol, seemed to go very well and was adapting its tune. On Sunday I started running low on petrol, so switched back to LPG. During that drive it started feeling worse and worse, at first I thought it was just learning to cope with the LPG, but eventually I had a backfire on LPG and it was almost undriveable, even on petrol. I unplugged the MAF again, but this time there's no difference. So I tried taking the LPG out of the loop, which is just a matter of switching over a connector in the engine loom (a result of my rewiring it), but that also made no difference. The lambda's showed bank 1 running rich, and the rovacom showed a lot of roughness and a bit of misfiring on that bank, so I tried pulling plugs 5 & 7. Both made no difference in running... Took out the plugs, they looked a nice grey colour, not black from overfuelling. Yet you can smell the richness. We're pretty much out of ideas, anyone have any clue what the problem could be? Tonight we're going to check the wiring to the injectors (I don't fully trust my wiring), and if that's not it it may be time for the first P38 with MSv3 and CAN messaging Quote Link to comment Share on other sites More sharing options...
FridgeFreezer Posted November 21, 2012 Share Posted November 21, 2012 Is it getting sparks on those cylinders? Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 21, 2012 Author Share Posted November 21, 2012 I felt it spark when I unplugged #5 by hand (not doing that again), so yes. I also think the plugs shouldn't look like it ignited well if there's no spark, or am I wrong in that? Quote Link to comment Share on other sites More sharing options...
FridgeFreezer Posted November 21, 2012 Share Posted November 21, 2012 You may see a wet plug, but if the engine's hot it can evaporate fairly quickly. Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 21, 2012 Author Share Posted November 21, 2012 Could it be a rocker shaft on one bank? Quote Link to comment Share on other sites More sharing options...
Hybrid_From_Hell Posted November 21, 2012 Share Posted November 21, 2012 Buy a noid tester and sort out if injectors getting power Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 21, 2012 Author Share Posted November 21, 2012 I could test those quite easily without taking the manifold off. My multimeter is supposed to be able to measure frequency, any idea what frequency we should be looking at at idle? Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 22, 2012 Author Share Posted November 22, 2012 To elaborate a bit (was posting from my phone yesterday): we tested the injector circuits, from the fusebox to the ECU, and everything is fine. So the only way the injectors could be acting up is if the Bosch ECU is dead. The theory for the rocker shaft is this: The shaft broke, causing the valves to open less on that bank. The ECU starts trimming the injectors to correct for it, but after a while it's trimming so much it says "this can't be right" and goes open loop, so it's back to its original roughness. It's a bit far-fetched, but so far this is the only thing I can't really rule out logically. Quote Link to comment Share on other sites More sharing options...
FridgeFreezer Posted November 22, 2012 Share Posted November 22, 2012 Well it's not hard to take the rocker covers off, but it sounds quite unlikely to me. May as well check the valve springs while you're there. Best way to measure injectors is just to look at the voltage across them, if it's varying with load/revs. If it's stuck at 12v or 0v something's wrong. Could also be stuck shut or clogged. Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 22, 2012 Author Share Posted November 22, 2012 Well, it is somewhat hard on a Bosch, as the intake manifold and all the junk that comes with it has to come off. But it's definitely a last resort check. The injectors themselves being the problem I find a little hard to believe, I checked all of them a few weeks back when searching for the other problem, put 3.5bar of pressure on them and actuated them manually with a battery. Nice pattern, no sticking, instant response. I can check the voltage/frequency the injectors are getting due to my LPG wiring, I put in a connector so I can easily take the LPG system out of the loop, and in that connector I have the switched side of the injectors. So no need to take the intake manifold off for that. Quote Link to comment Share on other sites More sharing options...
Escape Posted November 22, 2012 Share Posted November 22, 2012 The problem occurs both on petrol and LPG, so it's unlikely to be the petrolinjectors themselves. However, as the LPG takes it's injectorpulses from the Bosch ECU, a problem on that end would affect both fuels. Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted November 22, 2012 Share Posted November 22, 2012 My similar problem with the same engine but in a different vehicle (Disco II) was solved with a Nanocom and adaptive fuelling reset. Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 22, 2012 Author Share Posted November 22, 2012 Hmm, but from all I've heard/read the fuel trims aren't used in open loop, so that can't be it either, right? Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted November 22, 2012 Share Posted November 22, 2012 That was not my experience. Does it run ok above 3000 rpm? Mine did, which is when the ECU goes open loop. Mine chucked black smoke out the back, terrible rough running at idle and all way thru to 3k i changed plugs and leads looking for it. Night and day when reset. still get occasional stumble and o2 sensor errors, but I haven't changed them yet My tale here with link to Bosch ECU description which was really useful for me. http://forums.lr4x4.com/index.php?showtopic=79375#entry679453 Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 22, 2012 Author Share Posted November 22, 2012 No, it runs badly all over the rev range, and backfires if you let off the throttle. It's worst at idle, but definitely not at all good over 3000rpm. When the ECU is saying it's open loop, I'm inclined to believe it, and I see it both in the Rovacom (which I don't really trust) and Torque. I may have to buy a Nanocom after all, just really annoying that the Rovacom can't and won't support the Bosch properly! Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 22, 2012 Author Share Posted November 22, 2012 Right, so we did more troubleshooting today. Started out reading the frequency of the injectors (tried duty cycle, but for some reason that gave 98% with only one lead connected, so something's off there). After that also measured the voltage, from the battery + to the switched injector ground (which I had available in my LPG connector). Measured values: Also tried unplugging the lambda's, no change. Then bit the bullet and digged into the rocker cover, almost thought we had it because all rockers except one lined up perfectly, but on turning the crank they all moved, so that's not it either. Rocker shaft looked to be in good condition too. So now I'm fully out of ideas, may have to bite the bullet and buy a Nanocom so I can reset the adaptive values and/or exchange the ECU (can't do that now because the Rovacom doesn't support syncing it to the BeCM). Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted November 23, 2012 Share Posted November 23, 2012 are you able to monitor IAT and Coolant temp using the rovacom? If it has run fine without the MAF, the IAT is in the MAF, that could be it. Have you been able to check the coolant temperature voltage from the curve given in http://p38arover.com/rover/p38a/Engine_Management/BOSCH_521_Systems.pdf? I am pleased with my Nanocom but they are expensive pieces of kit. I hope you get to the bottom of it, it got me down for a couple of weeks till the Nanocom arrived. According to the LR ECU doc it can run with nearly all sensors disconnected, have you tried disconnecting the Coolant temp? Quote Link to comment Share on other sites More sharing options...
Escape Posted November 23, 2012 Share Posted November 23, 2012 Coolant and IAT readings seem normal. It used to run acceptable without the MAF (we tried before the new MAF arrived), but now it doesn't seem to make much difference. Quote Link to comment Share on other sites More sharing options...
elbekko Posted November 26, 2012 Author Share Posted November 26, 2012 So does it actually still use its fuel trims in open loop? I'd expect not, but you never know... Quote Link to comment Share on other sites More sharing options...
elbekko Posted December 2, 2012 Author Share Posted December 2, 2012 So the problem was indeed the adaptive values. Had them reset at a shop, after buying a Motronic license for a friend's Nanocom Evolution didn't work (argh!). Now, the problem started slowly coming back when driving on LPG, so switched back to petrol for the time being. Need to find out what's going wrong there. As for the Nanocom, after first getting the wrong code (good show BBS), it now says "vehicle authentication failure". WTF? Surely a system such as Motronic that's used on multiple models doesn't require that model's cable? Does anyone have a P38 cable they can open up and post a picture of? I'm intrigued as to what the difference could be, didn't find anything in the wiring schematics between the D2 and P38 that could really give it away. Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted December 3, 2012 Share Posted December 3, 2012 Glad you managed to get to the bottom of it. With regard to the P38 and disco II cables they do show as different at Nanocom. (blue and green cables although I don't know the difference) From the site http://web.nanocom.it/default_i.cfm "Please note that if purchasing P38 codes to add to an existing Nanocom Evolution where you have a BLUE OBD lead you will also need to purchase the GREEN OBD LEAD (NCOM23) from the Blackbox Solutions Web shop priced @ 49 Euro + 17%VAT (Where applicable). Shipping not included." Quote Link to comment Share on other sites More sharing options...
elbekko Posted December 3, 2012 Author Share Posted December 3, 2012 Yes, I know, and I find it utterly ridiculous I'd have to buy an overpriced cable to access a system that's used across models and is on the same communications bus. So all it is is a weak attempt at DRM, which I'm none too happy about to be honest. I could (somewhat) see needing a different cable to access the EAS, but not one of the common systems! Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted December 3, 2012 Share Posted December 3, 2012 I really don't think that it is. Maybe land rover fault for making the p38 non standard with in its own marque. Remember P38 was brought in whilst obd II was being standardized and has quirks compared to later models. All of the systems are not on the same bus, dependent on the age of the vehicle from what I have read, with srs a particular problem. Is fuse 33 good? As this supplies power for comms. Quote Link to comment Share on other sites More sharing options...
elbekko Posted December 3, 2012 Author Share Posted December 3, 2012 We've gone through the wiring diagrams, it IS on the same pin and bus, only the BeCM requires a second line. The nanocom gets power from the plug fine, so it can't be that. Quote Link to comment Share on other sites More sharing options...
RangeyRover Posted December 3, 2012 Share Posted December 3, 2012 This topic gives the P38pin out? https://www.scantool.net/forum/index.php?topic=1171.0 Is the wake up one significant? As the disco II seems not to have it from this thread http://www.landyzone.co.uk/lz/f8/obdii-computer-reader-80369-2.html Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.