JT’s fleet: Range Rover’s burst it’s bag

[J-T]

Listening to it, it doesn’t sound like it has a manifold blow but I’ll certainly have a much closer look. Well, pending the thing ever running again…

In a somewhat unexpected development, the car seems to have gone into complete electrical meltdown. I started it up this morning and the dash lit up like a Christmas tree. As well as the usual code above, there were codes relating to TPS, upstream bank 2 lambda, as well as CAN timeouts from engine ECU faults in several modules. Traction control light on, parking brake failure message, engine system failure message on dash, car will start but not rev. I’ll properly log the codes and post them up tomorrow, I didn’t have time today as I was meant to be going to watch the BTCC at Oulton Park in it and had to make other arrangements and after being there all day I’m knackered!

Between parking it up yesterday and starting it this morning, two things happened: I unplugged and re-plugged the upstream O2 sensor on each bank to check the electrical plugs were the same as I was planning on swapping them round. I didn’t actually swap them as I didn’t have a 22mm spanner so didn’t even attempt removal. Obviously, the car and ignition were off when I did this. There was also a rain shower of biblical proportions overnight. I’ve a feeling something electrical has gotten wet that shouldn’t and it feels like it could be the engine ECU.

I’ll check the battery voltage first as I know a low battery can play havoc but I don’t think I’ll be that lucky. Deep joy

[Crispian_J_Hotson]

You know by now what that probably is. Disco on that battery, discharge residual energy, charge and load test it out of the car. These cars are sods for eating batteries. I've got 3 in my shed, perfectly good to sling in your average ford cortina but no dice in these. Parasitic drains are an issue too, from door locks to seat motors. It'll be alright in the end, patience and 2 back up cars is all it takes.

[colino]

While I agree it may well just be a battery that is offering a mere 99% output that is upsetting it, just remember all cars are just big computers now and even mild fluctuations in temperatures, "confuse" them.  Just disconnect the battery terminals for ten minutes, after touching the cables to wishfully discharge any residual power, and reconnect.  Basically the first (and often only) thing I do now with any car with electrical gremlins.

[J-T]

Battery only reading 12.3 volts. Showing about 70% depleted on my highly* accurate battery charger. I’ve stuck it on charge and will see how it gets on but hopefully it’s just that. 

[Crispian_J_Hotson]

What battery is it fella? Sometimes cheap 017 gets chucked in with 85ah/ 680a. You should be up at 100ah/900a, not 'silver' or 'calcium' either, just plain Jane lead acid.

 

[J-T]

 

26 minutes ago, Crispian_J_Hotson said:

What battery is it fella? Sometimes cheap 017 gets chucked in with 85ah/ 680a. You should be up at 100ah/900a, not 'silver' or 'calcium' either, just plain Jane lead acid.

 

The one in it is 95ah/800a. I’ve ordered 100ah/900a yesterday that should arrive tomorrow.

I had the one in it on charge yesterday. It started it back up ok, although it took a few attempts whilst the ECU sorted itself out again. I drove it round the block (it then gave me a code for the upstream O2 sensor bank 2, but no lean running code), but then when I tried to re-start, it was back to dash-light bingo. I’ll see what codes it’s throwing once I know it’s got a decent battery in it but hopefully I’m getting somewhere.

[J-T]

New battery is now on and I’ve just taken it for a drive for 15 miles or so. Still getting a current and pending P1647 code for upstream O2 sensor, bank 1. As a result it’s operating in open loop fuelling so no fuel trims to read. I’m going to swap bank 1/2 upstream O2 sensors and see if the fault moves.

[Dave_Q]

Definitely sounds like slight progress.

It'll get sorted and we'll all be experts in S type running issues by the end of it and we can open our own garage where people can book in for a 20% chance of their car being fixed.

[J-T]

1 hour ago, Dave_Q said:

Definitely sounds like slight progress.

It'll get sorted and we'll all be experts in S type running issues by the end of it and we can open our own garage where people can book in for a 20% chance of their car being fixed.

Make that 5%. 

This thing would try the patience of a saint. 

It's been sat for a couple of hours. I've just gone back out to it to start it and again, I'm getting all the dash lights and multiple system failures intermittently when turning the ignition on (i.e. everything is in fault before attempting to start the car, so it isn't happening due to a voltage drop when turning it over). When this happens, the car will start but won't rev as it logs a TPS sensor fault, there are CAN error messages in most modules, the dash display cycles through 'engine systems fault', 'parkbrake fault' and 'DSC not available', EML, traction light and the orange and red light next to the led display are all illuminated. I cycled the ignition on and off 3 times and on the 3rd occasion, everything went away apart from the EML and it would start and rev up normally. I turned it off and cycled the ignition on/off a few more times but it did not go into fault again. So it can't have been the battery, as it now has a brand new, uprated battery fitted. It's something else. When it does start 'normally', it has a current and pending P1647 code for bank 1 upstream O2 sensor, no P0174 lean running code because it's in open loop fueling, so no fuel trim limits to exceed. The P1647 O2 sensor code had never appeared before doing this disco light party trick upon startup.

I checked the battery voltage and found the following:

Sat dormant: 12.8v

With ignition turned on but vehicle not running: 12.38v (vehicle wasn't in fault - I couldn't get it to fault again so wasn't able to get a reading)

Vehicle running: 13.94v

It's probably worth just re-capping the events immediately before this startup fault happening for the 1st time; the only notable activity between switching the car off in the evening and starting it the next morning (when the above fault occurred for the 1st time) was me unplugging and re-plugging both upstream O2 sensors to check the plugs were the same (ignition off) and very heavy, sustained rain shower overnight.

I'm starting to think I may have a bad earth or a faulty/water damaged ECU.  Obviously if anyone has other suggestions, please jump in!

 

[mk2_craig]

Wouldn't harm to give the electrical connections on those O2 sensors and plugs a damn good clean with the appropriate product, also just make sure that the cables on each are still connected to the contacts.

[J-T]

A lot of forum foraging has uncovered a fault on early Type R’s (like this) where the TPS gets waterlogged and displays these symptoms. I’ll have the scuttle off tomorrow and investigate.

 

[purplebargeken]

Good call re connections, earth, etc.

[J-T]

I've hopefully gotten somewhere today. After a lot of searching, I came across this very helpful post:

Throttle Position Sensor Fault (DSC Not Available...

It detailed that on some early STR's, water could get into the plug for the TPS on the side of the throttle body and cause the symptoms I was experiencing (same as low battery on startup and logged codes for TPS). So I had the scuttle and top part of the intake off for a closer look. Here's what I found:

 

 

Bit closer...
 

 

Someone has clearly been here before. It took me a minute to figure out, but everything in that photo is the male end of the throttle body electrical connection. The female loom connector is missing in action, so I'm assuming the wires are fired directly on to the pins and they've smothered some sealant over it. Which didn't work. Because it got wet in heavy rain. Sigh...so I'll have to try and chop the loom connector plug off a breaker and splice it in (does anyone know if that throttle body plug is shared across other S-Types?). The actual TPS sensor plug (LHS stood at the front of the car?) was ok. For now, I've dug out the old sealant, cleaned it up as best I can with contact cleaner, put some hylomar in there and around the wires to seal as that's all I had in that's suitable(ish), wrapped it in duct tape and tried to divert water away from the area by sealing up unused holes on the strut brace central support bracket. I also made sure the scuttle went back on properly as half the fixings were either missing or ineffective so it was making a poor seal against the glass. It's still not perfect but it's better.

This appears to have got things working normally again, but obviously needs sorting properly with the correct loom connector. So I read the codes again and was still getting current and pending P1647 relating to bank1 upstream O2 sensor. So I swapped bank 1 to bank 2. And the fault has followed with it, so it was now P1646. After a taking a closer look at the sensors, they were different makes, so I've googled the part numbers to check if they were correct. The working sensor is a Blueprint ADJ137030, which is correct for this car. The faulty sensor is a different make - Denso, and should be DOX-0428, however what's fitted is DOX-0426 which I think is for a facelift and has a green wire sheath as opposed to all the others which are black. So I've ordered a matching Blueprint sensor. Will I be lucky enough that this was the issue all along, or will I simply be back to the position I was in a week ago? I won't know until this sensor arrives, place bets now!

[J-T]

“Hello, is that I R Lyer car spares”

”Yes”

“Oh hello, yes I contacted you yesterday about a lambda sensor that you definitely had in stock and could supply next day, hence my ordering from you; well I ordered it and I’ve not had any confirmation, could you let me know if it’s coming”

“Ah yes I can see that order now. Yes we had to order that in, but it’s here and it’ll get collected this afternoon”

“Oh right. So it wasn’t in stock then”

“We had to order that in”

“Sigh…right. So I’ll be getting it tomorrow then?”

“It’ll be Monday now, I’ve put it on 24hr courier because we had to order it in”

”Yes you said. I was told it was in stock when I ordered it, that’s why I…

“We had to order that in”

 

Meanwhile it sits goading me on stands

 

“Fix me? LOL, many have tried lad. I’ll have you in fucking tears yet just like the others”

As you were.

[Crispian_J_Hotson]

That's a rough lash fella. Your car hasn't been to a specialist near Birmingham has it? I had similar wiring wank on B2 injector harness. Tossers.

It'll pay to get dielectric grease. Whenever I disco a connector, it gets doused in that stuff. Take some advice and buy it straight from online and don't go to 23 shops that don't know what it is.

[J-T]

New upstream O2 sensor has arrived. I've just swapped out the non-working sensor for the new one (a Blueprint sensor which is an exact match for the other one fitted that it likes). P1646 code which  is upper O2  sensor bank 1 did not clear. I've tried swapping the new sensor over on to bank 2 and the code has moved with it (now P1647)! I know this indicates that the new sensor is faulty but I just find that very hard to believe.

I can't think of much else I can check. There are are numerous wiring tests that the jag manual suggests if the code doesn't clear, which involve getting the ECU out the dash from behind the glovebox but I don't see how it can be a wiring fault when the fault code moves according to which sensor is fitted.

[J-T]

Turns out I was being thick and had bank 1 & 2 the wrong way around, so I had changed the good sensor, not the bad! With the correct sensor dispensed with, it got rid of the fault codes for the HO2 sensors.

Initially, it looked like this had been the issue all along; a failing O2 sensor giving strange readings (reporting bank 2 as lean), then failing altogether. With the new sensor fitted, fuel trims on both banks were good (within +/-5%) at idle and at 2500rpm.  This was last night and I didn't have time to take it for a drive.

I've done that today, twice for about an hour each time. After the first drive, things were fine, I had live data on, fuel trims were behaving and did so after a long idle when I got back.

I've just been out in it for the 2nd time and took it for a hard drive. The car has 7 diagnostic monitors for stuff like EGR, cat efficiency, misfires etc which it does a self check on and will report back 'ok' on the scan tool (it can take a while for this as certain driving conditions have to be met) . The monitor for O2 sensors was still saying incomplete. The workshop manual tells you how to try and complete a drive cycle for each monitor. For the O2 sensors, you have to get the car up to temp, drive it between 3-4k rpm in 3rd, then coast to a stop x 2, then let it idle for 11 mins. I tried this a couple of times but it still didn't complete, so I took it out and gave it a good ragging. 

I plugged the scan tool back in and noticed the fuel trims on bank 2 had crept up again, sort term trims were now up around 15%. I didn't like the look of this, so I turned it off, let it rest for 30s or so, then took it back out for a gentle drive around the block, with live data on. Back at home, this was what it was showing:

At idle:

Bank 1:
STFT 0.8%
LTFT 5.5%

Bank 2:
STFT 25%
LTFT 18%

At 2500rpm:

Bank 1:
STFT 9.4%
LTFT 0%

Bank 2:
STFT 25%
LTFT 5.5%

FUCK. I read the fault codes again and sure enough, I now have P0174 (bank 2 lean) pending (on these, the conditions have to be met twice to turn the EML on, if it hits it once, it's stored as pending, so basically next time I drive it, the EML will come on again). So, really I'm back to where I started.

However, I'm struggling to understand why would it take 100 or so miles and 2hrs driving for this to show up? Surely if there is an air leak it would show immediately, yet the trims were fine at first, then slowly went out of range on one bank? This seems to be what it's done every time.

[Dave_Q]

I guess you have seen this thread with some good info on fuel trims?

https://www.jaguarforums.com/forum/general-tech-help-7/faq-fuel-trims-thoroughly-explained-quiz-49317/page2/

It is odd that it was fine for a while then started going up.

I guess you need to focus on why that bank is going lean (or reportedly lean) after a while or when hot. 

There isn't much that can only affect 1 bank if we now believe the lambda sensor, an air leak of some sort, a misfire/failing coil pack or an injector that is under injecting.

Interestingly in the pdfs in that thread it mentions the case of a misfire or disconnected coil pack, this does cause fuel trims to go up which is counter intuitive. The lambda sensor measures oxygen rather than lean/rich, so if a cylinder is not firing and chucking fuel down the exhaust this still shows as a + fuel trim as the unburnt O2 from that cylinder is what the sensor is measuring.

Are coil packs expensive? Or could you swap bank 1 coils to bank 2 to rule them in/out? Maybe even the same for injectors?

[J-T]

24 minutes ago, Dave_Q said:

I guess you have seen this thread with some good info on fuel trims?

https://www.jaguarforums.com/forum/general-tech-help-7/faq-fuel-trims-thoroughly-explained-quiz-49317/page2/

It is odd that it was fine for a while then started going up.

I guess you need to focus on why that bank is going lean (or reportedly lean) after a while or when hot. 

There isn't much that can only affect 1 bank if we now believe the lambda sensor, an air leak of some sort, a misfire/failing coil pack or an injector that is under injecting.

Interestingly in the pdfs in that thread it mentions the case of a misfire or disconnected coil pack, this does cause fuel trims to go up which is counter intuitive. The lambda sensor measures oxygen rather than lean/rich, so if a cylinder is not firing and chucking fuel down the exhaust this still shows as a + fuel trim as the unburnt O2 from that cylinder is what the sensor is measuring.

Are coil packs expensive? Or could you swap bank 1 coils to bank 2 to rule them in/out? Maybe even the same for injectors?

It’s worth a try if only to rule out. But I’m not getting any misfire codes and it never feels as though it’s missing.

I’m going to try and rig up a home made smoke tester a la @cort16. I’m only ever getting the code on bank 2, so if it is an air leak, it can only be in the region of the charge cooler on that bank as everything else on the inlet is before it splits and would presumably affect both banks. Unless it’s a cracked exhaust mani. I might have to buy an endoscope as you can’t see shit down there. 

[mercedade]

On 22/06/2023 at 19:39, J-T said:

After a taking a closer look at the sensors, they were different makes, so I've googled the part numbers to check if they were correct. The working sensor is a Blueprint ADJ137030, which is correct for this car. The faulty sensor is a different make - Denso, and should be DOX-0428, however what's fitted is DOX-0426 which I think is for a facelift and has a green wire sheath as opposed to all the others which are black. So I've ordered a matching Blueprint sensor. Will I be lucky enough that this was the issue all along?

If I've read this right, you confirmed that you had an incorrect sensor, and (at the time) believed it to be the faulty one.

Since then, you've discovered that the one you thought was broken was in fact fine, and the incorrect one was actually working - the 'correct' one was broken, and you've now swapped it with an duplicate.

So the incorrect sensor is still in place?

[J-T]

8 hours ago, mercedade said:

If I've read this right, you confirmed that you had an incorrect sensor, and (at the time) believed it to be the faulty one.

Since then, you've discovered that the one you thought was broken was in fact fine, and the incorrect one was actually working - the 'correct' one was broken, and you've now swapped it with an duplicate.

So the incorrect sensor is still in place?

It's me not wording things very well. I tested the heater circuit on all three O2 sensors. One was open circuit, two were not. The one that was open circuit was the dead one, which is no longer fitted to the car. As soon as it was replaced the P1647 fault code went off and the car went back into closed loop fuelling, so the 'bad' one is definitely off now.

Following on from @Dave_Q's suggestions, I've had another mess with it today:

As suspected, as soon as I started it up today I got an EML, restricted performance and P0174. Fuel trims on bank 2 are more or less permanently 25% short term and 19.5% long term at idle. The long term trim will drop to around 6% when held at 2500rpm.

Tried to do a bit more ruling in/out.  I swapped all coil packs from bank 2 to bank 1 and vice versa to see if a bad coil pack would move the fault to bank 1. Other than discovering that 1 coil on each bank has been replaced (googled the part numbers and they are correct), there was no change. There is some oil in the plug wells so it probably needs the cam cover gaskets changing, but I don't think that's affecting anything. I pulled a spark plug on each bank. They look relatively new, are NGK IFR5N10 which I think are correct. Only thing to note was bank 2's plug was very fuelly.

I got some brake cleaner and sprayed it all over the seal between bank 2's charge cooler and the manifold that splits the air flow to see if it would reveal an air leak. No change to the trims. I sprayed a load of brake cleaner towards the base of the charge cooler to try to access the gasket between it and the head. No change. I sprayed brake cleaner at the crank case ventilation tube on bank 2. No change (I also later pulled this out and checked the O-ring which still looked good and wasn't flattened). I couldn't see anywhere else that an air leak to that one bank could occur.

I took the MAF out. It looked fine but i gave it a very gentle spray with electrical contact cleaner and re-fitted it. It made no difference to the trims, but I initially thought I had killed it, as on live data, the air flow rate and intake air temp were both reading 0 - I had just forgotten to plug it back in! Once plugged in, these values returned to what they were (incidentally, I though if the MAF was unplugged, it would change to open loop fuelling, as it would not know the volume of air to expect, but it didn't). No change to the trims, although I now have a pending code for MAF and Intake Air Temp!

I jacked it up and had a good look at the bank 2 exhaust manifold to cat joint in case there was a leak. There definitely is no leak there; there is no evidence around the joint and I was able to feel around it with my hand with the engine running and it's not blowing. I couldn't see the manifold itself as the heat shield completely covers it. The bolts are very rusty and I don't think it'll come off.

I'm going to try and cobble together my own smoke tester to rule in or out the air leak. The only place I can see a leak now being likely is the gasket between the charge cooler and the head, as you can't really get access to it with brake cleaner with everything in place. If the smoke test does not reveal an air leak, then I think injectors are the only thing left? 

 

I'll post up the full live data stream below with the car at idle, in case anything jumps out that I'm not seeing that could be another clue:

 

 

 

 

 

 

 

 

 

 

 

 

[Dave_Q]

Not much to add, just that if the plugs are looking fuelly on bank 2 then could that bring you back to there being something wrong with the lambda sensor for that bank? 

As in it's reading the O2 wrong (too high) and it's chucking fuel in to compensate but its still reading too much O2 so gives the lean code?

It just seems unlikely to have a leak that is big enough to cause this level of fuel trim without it really chuffing in/out. 

Wiring to the sensor? Sensor relay or something? I know it should bring up codes for the sensor being off but maybe it isn't somehow? 

You measured heater resistance on the sensors, are there any multimeter measurements that can be done at the ECU plug to see if both sides are the same from there?

Does @cobblers know if there could be an internal ECU fault causing the lambda sensor to read too much O2? Cars are not his main bag but the inside of ECUs is probably fairly similar.

[J-T]

There is a relay for the O2 sensors and 2 fuses for the sensor heaters. The fuses are fine and I swapped the relay with a matching one in the fuse box yesterday and it made no difference.

There are some pinpoint tests in the workshop manual to do at the ECU but I’ve been putting them off because 1. I don’t have multimeter leads long enough to reach both plugs and 2. It means taking all the scuttle apart again as well as removing the heater intake to get at the ECU, but it’s looking like I’m going to have to. I just hate doing electrical testing because I don’t understand it properly and don’t know what I’m looking for but I suppose I can’t go that far wrong if I just follow the manual.

The sensor itself is new, so I don’t think it’s that. 

[Dave_Q]

Yeah agree it's unlikely to be the new sensor, just trying to fathom what else might cause the good sensor to read wrong.

[J-T]

I’ve got my home brew smoke tester set up, planning to have a go with it tomorrow. Does anyone know if there a way of keeping the throttle butterfly open when doing the smoke test? I’ve heard it’s a bad idea to move them manually (it’s a fly by wire throttle obvzz)

[J-T]

Had a go at smoke testing it today. First I tried it with the smoke entering through the PCV line in the intake hose and capped the end off with a rubber glove.

I had a few attempts but couldn’t find any leaks. I managed to get a bit of pressure in there, but nothing.

I had the ignition on with a brick on the throttle but I don’t know if that was actually opening the throttle butterfly so I had another look for somewhere else to test past the throttle body. I found this little vac line on the back of the charge cooler manifold, so found some smaller diameter tubing and connected to that (pic is upside down).

 

At first I though I’d found a leak on the gasket between it and the top of the supercharger (albeit on bank 1 side) but I think it was just smoke tracking under the inlet pipe that was coming out of the vac line stub in the photo above that I’d connected to as the seal wasn’t perfect. Other than that, I could find no leaks. And I tried several times.

I had a play about with some items on bank 2 with the car running and the scan tool attached. I lifted the dipstick, made no difference to bank 2 fuel trims as they are already maxed out but bank 1’s increased, then dropped back when it was pushed back in. The O ring on it was quite perished so I replaced it but no change. I unscrewed the oil cap with the engine running which made it run slightly rough and again made bank 1’s fuel trims increase. Once it was back on, bank 1’s trims fell back to near 0, so I can assume it’s not that (bank 2 still maxed out). I pulled the little vac line off in the pic above with the engine running and had the same results as the oil cap, with bank 1’s trims increasing, then falling back once it was back on and bank 2 remaining maxed out throughout (what is this vac line for?).

I’m at the point of giving up with it now, I’m not prepared to go through the battle of pulling the supercharger when I can find no obvious fault. I can’t really think of anything else to try.

[J-T]

I’ve left this alone for a bit as I’ve just stated a new job and didn’t have the mental bandwidth!

Just been out and had another mess with some brake cleaner and a straw trying to find this air leak. It’s only affecting bank 2 so can only be the upper manifold gasket or gasket between charge cooler and head, if it’s an inlet leak rather than exhaust. The only time I got the short term trims to deviate from between 24.2 and 25% was if I aimed the straw in between the supercharger and chargecooler on bank 2 as far down as possible and kept my finger on the trigger for about 10s. This would cause the STFT to drop to about 20%, then creep up again once I stopped spraying.

This would fit with an air leak between the charge cooler and the head, however I would have expected this to have a more dramatic effect on the short term trim, as in I would expect it to nose dive if it managed to draw in the brake cleaner. I repeated this 3x and each time got the same result as above.

What do we think?

[Crispian_J_Hotson]

Mate, I think... I know your pain. It's a toughie and a sack draining issue, just be wary of red herrings. If you've repeated the same check 3 times and got the same change, well, I'd investigate that further after exhausting the obvious things. You know the chances of there actually being an issue there could be slim, that's how it works.

I can offer to get icarsoft OBD live data at the same time you do it to compare, but you know my sensor readings are... lets say boarder line through unknown CAT bodgery and manifold blow (hence the upcoming rebuild), so it could throw you way off. I do know that my wiring and EGRs are ok.

Just do me a favour, next time you are out there, just see if you can smell fuel in the oil.  It's a thought that has entered my head that I can't shake off. In fact, I've just seen that I have some idle live data on my phone from a while back, you want me to post it up? 

 

 

 

[Dave_Q]

It sounds like that could be your leak, doesn't quite match if its leaking enough air to get to +25% trim but then it's not sucking enough juice in to send it miles the other way but if its making any difference to the trims and nothing else is probably need to make it not leak to rule it out.

[Crispian_J_Hotson]

Well, I've just compared them and there are some close readings. The biggest difference I can see is your at operating temp and the intake air temp is 25, mine is 43.  These cars do experience heat soak and pulling in hot air at idle and that's why you have that doobry with a wire coming out of it at the bottom of the air box, it actuates a flap to close off hot air from the engine. I would suggest that either mine isn't working (likely) or you were in the cold on that day. These solenoids appear to be rocking horse shit... great*. This is the reason why cone filters are no good on these cars.

*writes on mile long list... need to improve cool air intake.