There aren't any rad mounts as such, the rad has a couple of pins on the bottom that fit into rubber grommets on the crossmember and the top is held in place by a couple of shaped 'lumps' that slot into the steelwork at the sides. It might just be that the alloy radiator is conducting the heat away into the bodywork where the standard plastic topped one wouldn't.

It might be worth getting an infra red thermometer and seeing just how hot things are getting.but if the running temperatures are OK, I can't see it being a problem.

If it is running at 87C and only getting up to 93C when stationary, there's nothing wrong with the cooling system, that's about spot on for a GEMS (Thor tends to run slightly hotter for some reason). and almost exactly what mine shows. A blocked radiator would show up as the temperature rising under high revs or load. Not sure why the front wings are getting hot, unless that is purely heat from the sun (is the car a dark colour?).

A high capacity, high flow, radiator is available from https://www.gravityperformance.co.uk/55mm-high-flow-radiator-range-rover-p38-4-0-4-6-v8-94-99/, although one of my cars has one of these and the other has a standard Nissens and there's doesn't seem to be much to choose.

Yes, they are all the same except for the connectors. Very early cars have a socket on the latch itself, from around early 97 they have a flying lead with two connectors on it, a 6 way and a 1 way, while later ones have a flying lead with a single 8 way socket with only 7 ways used but if you have one type and need to fit to a car with the other type, the sockets can be swapped.

Driver's door latches have 3 microswitches while passenger latches only have 2. A driver's door latch from an MGTF (FQJ102262PMA) is identical to the 6+1 P38 door latch except it has an extra grey wire (so has 6+2) which can be chopped off an ignored. Chances of finding a used one of these that has had far less use than a used one from a P38 is better but make sure you get the correct part number as some versions of MG didn't have central locking so they don't have the microswitches.

Once you've done it a couple of times, it's a 20 minute job but if you've never done it before, it'll probably take a couple of hours. Once it is unbolted and the rods and cable disconnected, the hardest part then is getting it out of the door. They have a spurious lump of steel on them that does absolutely nothing other than get caught on the window runners so it may be loose but won't come out unless you get it at exactly the right angle. I always cut it off which at least makes getting it back in easier (and out of it has to come out again).

From the top. Take the door panel off and fight your way through the membrane to get to it. The rod from the door handle disconnects easily enough, you twist the green plastic clip and that releases the rod so it just pulls out. Undo the bolt holding the door handle in place (the steel block between the inner and outer door skin will probably drop into the bottom of the door so you'll need a magnet on a stick to fish it out and put it back into place when you come to put things back together) and slide it towards the front of the car which will release the rear of the handle so, if you pull on the handle as if opening the door that will allow it to be pulled out of the door enough that you can get to the rod that attaches to the back of the door lock. Very carefully unclip the rod from the back of the lock barrel lever (the lever is only mazak and easy to break). Lever the outer of the interior door handle cable from the fixing and twist the cable so it unclips from the latch. Unscrew the button from the end of the sill locking rod, unplug the cable(s) to the latch and remove the 3 screws that hold the latch in place. Then it will be floating around inside the door and it needs to be lowered and twisted as you do so to get it out of the door. The first time I did it, I must have spent the best part of half an hour trying to work out how it came out until suddenly, it just dropped out of the door.

As they say in all the best books, reassembly is the reverse of the above (although I've never worked out how you unhammer something back on...)..

This thread prompted me to contemplate doing a video for Youtube on how the latches actually work and why they fail. I agree, I've got replacement microswitches but the button on them doesn't protrude from the body as far as the original ones, so they would need to be mounted slightly higher. But, as you say, the tolerances are pretty tight to get them to operate reliably. One thing I have noticed though is that the keyswitch only operates right at the end of the travel when unlocking (being pushed down) but reliably when locking. This is what causes the problem with people being stuck with the immobiliser when always locking and unlocking with the key. A lot of this could also be down to wear in the nylon insert in the lever so when the lever is being pulled up (to lock), it moves the full amount and operates the switch, but when being pushed down, it doesn't move quite far enough for the switch to operate. My thought on this is that if the lever was very slightly shorter, it would move further and stand a far better chance of causing the switch to operate. I suspect making a replacement, shorter, lever would be easier than trying to line up replacement microswitches.

Not today but Sunday. When we first got the red 4.0SE it came with just a single dumb key, no remote. Not liking the idea of only having one key, I tried to get a spare cut but it appears that Timsons, who cut me one for my car, no longer do them as they have wasted so many blanks when the ones they cut didn't work. Went to a lock specialist who checked and found that the blank keys are NLA but they had 4 blanks for the flip key on the shelf so they cut me a couple. All I needed then was a remote. Fortunately at some time in the past, someone has been inside the BeCM and it was unlocked which meant I could change the key code in the BeCM to match a remote from a different car. Marty was able to supply me with a remote without a key blade and one of his fob filters as it had a blue spot, Gen 2, receiver in it where the blue antenna wire had been snipped off.

Marty also gave me the fob code from the key so I was able to use the Nanocom to change it in the BeCM and, once synced, the remote with the newly cut key blade has been working perfectly. Then the dash started coming up with a Key Battery Low warning so decided it was time to do something about it. On an unlocked BeCM, the Nanocom can display the EKA so checked that worked. Opened the drivers window, closed the door and locked the car with the remote, waited a couple of minutes then unlocked the drivers door with the sill locking button and, as expected, only the drivers door unlocked.and when I tried to start the car, found it was immobilised. Entered the EKA with the Nanocom, all doors unlocked and the immobiliser was turned off. Great, all working as it should so time to do the rest.

Soldered the cut antenna wire back together and fitted the fob filter, then took the fob into the house to change the batteries. Got the back off, old batteries out, new ones in and checked that the LED flashed when I pressed the buttons so took it back outside. As the owners handbook says that if the batteries are removed for more than 1 minute, then the fob will need to be resynced and I'd had then out for at least 10 minutes, expected to have to do this. But no, pressed the unlock button on the fob and the doors unlocked, pressed lock and they locked, so all working as it should without having to sync the fob.

No more key battery low warning on the dash and with the antenna now connected, the car can be locked and unlocked from within the house, no longer from within a few feet.

If it is, as has been mentioned and is the primary suspect, the wiring to the lumbar pump, it goes via C0974. All the other wires go from the seat outstation directly to the motors. If you unplug C0974 you won't be able to adjust the lumbar support but it should stop it from intermittently blowing fuses. It's a small 4 way connector with only 3 wires in it here.

They aren't there to cool the engine but to cool the AC condenser, so only run at high speed when the AC refrigerant pressure gets high. There is no connection between them and coolant temperature as the viscous fan can shift far more air than they ever will. The AC on a P38 runs all the time, even in cold weather. If you look at the evaporator temperature on the Nanocom, you will see that it always sits at around 5-6 degrees C and the compressor clutch engages if it starts to get higher than that. Obviously, in very cold weather it won't cut in as the evaporator is already down at that temperature but ordinarily it always runs.

If by engine cooling fan, you mean the big one on the front of the water pump, it has a viscous coupling. The coupling is filled with a silicone fluid that gets thicker as it get hot. So when cold the fan is just free-wheeling but as it heats up, the fluid in it causes the fan to be driven at engine speed. It will also engage when the engine is initially started from cold until the fluid is spread around inside the coupling. You can usually hear the roar from the fan which gets quieter as the coupling releases the drive.

I would say that would come under 5.3.1 d(ii), a Spring modified so the suspension is inoperative. See https://www.gov.uk/guidance/mot-inspection-manual-for-private-passenger-and-light-commercial-vehicles/5-axles-wheels-tyres-and-suspension. However, if it is a very old vehicle that never had suspension in the first place, it would still pass (even though it would no longer require an MoT these days).

davew wrote:

PS: While you have the MOT Testers book handy what does it say (indicate) about chassis corrosion on P38s ?

I don't have it handy, I just open it up if I need to check something, it's here https://www.gov.uk/guidance/mot-inspection-manual-for-private-passenger-and-light-commercial-vehicles

Your note isn't a specific fault but a free text one put in by the tester because he can, so anything that doesn't have the exact section number is a pedantic tester (usually done so gullible customers will pay him to put right faults that aren't actually faults). I've seen ones like, "Sump rusty but not leaking" and? The section on the chassis is https://www.gov.uk/guidance/mot-inspection-manual-for-private-passenger-and-light-commercial-vehicles/6-body-structure-and-attachments and the appendix (https://www.gov.uk/guidance/mot-inspection-manual-for-private-passenger-and-light-commercial-vehicles/appendix-a-structural-integrity-and-corrosion) points out to thick testers what bits he should look.

Bolt wrote:

Section b seems to be referring to "components" that may have been modified.
If the whole suspension has been changed, it would not apply.
(Oh my, what if you modify a spring!?)

I think you are probably right as the whole system has been changed, so it is the vehicle that has been modified, rather than an individual component. I've seen a car (not a P38) that had had the suspension lowered by clamping the springs and heating them up with a welding torch. That, I would assume, would be noticed by the tester (I noticed it as soon as I looked under the car) and frowned upon but if whoever did it had simply cut one turn off the coil, as long as it remained in place when the suspension was fully extended, it would be no different to fitting a shorter spring.

What's it have to say about "modified vehicles"
Surely changing the whole design of the suspension would qualify as a modification?

There's only the odd mention of modified vehicles, for instance, if a car has been fitted with an engine older than the car, the emissions limits are those that apply to the engine as limits have been tightened over the years. So if I was to ever achieve one of my fantasies of fitting a P38 V8 into a Prius, the emissions limits would be for the P38 engine, not those that would apply to the Prius. Most modifications are allowed here as long as items still work as expected. I know of one car that is virtually all hand built using a pair of 6 cylinder Jaguar XK engines in tandem so it is an 8 litre straight 12. That was built on the chassis from an old FX taxi (the traditional London black cab) and, as it retained the chassis and suspension from the original vehicle, it is regarded as the same car.

The only reason I was looking was because I wanted clarification on whether a spit ball joint boot on my other half's Merc would be a fail or advisory. Seems that a boot that is deteriorated but still keeps dirt out of the ball joint is an advisory but if it is split and would allow dirt in, it's a fail. Which is a real bugger as I've got to change a track control arm and re-align the steering because the boot is split and not because there is any wear in the ball joint......

Surely under (b) the suspension system has been modified? That is how I would read it anyway.

While looking something else up on the MoT testers manual, I came across this section

5.3.5. Gas, air and fluid suspension

Defect Category

(a) A gas, air or fluid suspension system inoperative - Dangerous

(b) A gas, air or fluid suspension system component damaged, modified or deteriorated in a way that:
(i) it would adversely affect the functioning of the system - Major
(ii) its function is seriously affected - Dangerous

(c) An obvious leak from any part of the system - Major

That suggests to me that a P38 on coil springs should be an automatic fail under 5.3.5 (b)(i)

You won't get an error as the ECU doesn't know the difference between failed sensors giving 5V or a lean mixture giving 5V. Don't forget that GEMS is only just OBD compliant as it became mandatory from 1994 in the US market, it only changed when OBD2 was required from 2000 onwards (hence the change to Bosch Motronic).

PaulD4 wrote:

So now I know the LHS motor should run low speed as soon as there's an AC Request HVAC to GEMs,

No, they should both run low speed when there is an AC grant from the GEMS ECU.

and the RHS motor should run low speed too if the X315 switch goes over 21 Bar,

No, the RHS should run at low speed in series with the LHS fan. If the X815 switch goes over 21 bar, then both relays 13 and 14 operate and they both run at high speed.

If I drive the fans from the Output Options on Nanocom both fans run at low speed - not sure why both run

They should, they are in series. You need to check if relay 18 is operating and that is controlled by the GEMS ECU.

To be quite honest, when running at low speed they do naff all so I wouldn't worry too much about them not running then. It is only important that they run at high speed when needed.

They aren't controlled by the HEVAC, but by the engine ECU. The diagram isn't the clearest but they should run at low speed whenever the engine ECU sends a AC grant signal (which it should unless it is getting too hot). Power comes from fuse 31 and goes directly to the LH fan motor. The other side of the motor goes to centre contact on changeover relay 13 which is not operated unless the AC refrigerant pressure exceeds 21 bar (which it won't unless the AC is having to work really hard). From the normally closed contact at relay 13, it goes to the contact in relay 18 which should be closed by the AC grant signal from the ECU. The other side of that relay contact goes to the normally closed contact in relay 14, another changeover relay that operates in parallel with relay 13, and from there to one side of the RH fan motor with the other side grounded so the two fans operate in series so at low speed.

If the AC refrigerant pressure exceeds 21 bar, both relays 13 and 14 are operated so the connection from the LH fan goes via the relay contact in relay 13 directly to ground and power from fuse 36 goes via the now closed contact in relay 14 to the RH fan motor so they both run at high speed.

Are you sure both run at high speed? I noticed that mine didn't seem to run at low speed and that was down to one of the fan motors having burnt out so it was open circuit. Like that they won't run at low speed but the remaining one will run at high speed.

The O2 sensors shouldn't read a constant voltage, they should flip flop between 5V and 0V about once a second. 5V means it is running lean so the ECU will put more and more fuel in to richen the mixture until they read 0V. If they stay reading around 5V, they are dead so never will drop to a lower voltage.

Marty has a lot of work with the day job, so may take a few days to reply. He sent me a filter and key fob for the red car and postage from NZ to UK took 6 days.

Head gaskets, plural, suggests a V8 so your D2 will have the Thor as fitted to a similar age P38. Take the serpentine belt off and try spinning all the idlers, the PS pump, alternator and water pump. I've known a water pump to squeal but only on initial start up not all the time and the alternator is also suspect. Any roughness should be felt and that might be a better way of identifying it that listening for it.