It will be worth having your AAV rebuilt. I chose to do the ‘fix’ stated in the book, got a sledgehammer and socket, and pounded the inside cylinder inwards, so it closed all the way when warm. Fixed my high idle when warm. Only downside, my car idles rough for the first couple minutes, especially in cold weather, because it cannot open all the way.
John, after doing this ‘fix’ to mine, is it still re-buildable?
If the body is not cracked then yes. I would say every third AAV I receive have this ‘fix’ done 
Boeing_Driver,
Yes, there is fuel in the sump tank with the engine running or not. The fuel drains by gravity from the main tank to the sump tank.
There is a drain plug bolt on the bottom of the sump tank accessible from beneath the car by removing a rubber plug. Make sure that you have a large enough container to receive all the fuel that could be in the tank. Your fuel tank gauge is not a really accurate way to judge the amount of fuel you will have pour out of the sump tank drain plug after you remove it.
Paul
I’ve done that fix! If you’re trying to keep an XJS on the cheap, then it’s entirely appropriate. Those days are waning… and I’m glad you can rebuild them properly now.
I’ve got a coolant temperature sensor ready to fit but I can’t find anything in the Haynes Manual describing the removal/refitting.
Guidance/description appreciated along with any ‘gotchas’ to be prepared for.
With regard to the AAV it’s off for refurb (hopefully safely en route to its destination).
Thanks
Probably not in there because it’s quite simple.
I would spray a bit of PB blaster (or your favorite rust breaker) on old one first.
Make sure engine is cold, or you’ll be relieving pressure under your nose!
Simply take it out with wrench or deep socket, clean up the threads a bit, put on a tiny bit of anti-seize, hand thread it all the way, then tighten snugly.
Done - came out without a murmur and new one in. If only all XJS jobs were that simple!
Have disassembled your AAV and I must say it is in a relatively good state. Piston is not stuck and there is no sign of galling, except that the thermal bulb is shot. I have seen much worse 
. Here is the main reason for your high idle:
Old wax-bulb (right) next to the new wax-bulb (left) at same room temperature:
Or do what I did, hammer the whole thing in a bit. Of course, my idle sucks/is too low until the engine gets above 160F.
Problem is not only that the pin sits lower,
but also it is more “slugish” and extend less compared to new bulb when hot. So hammering your AAV doesn’t fix it… it just makes it broken in another way 
It fixes the problem that the idle is too fast and cannot be adjusted any slower. The result of this “fix” is that the engine tends to idle too slowly when cold, but oddly enough this is not a serious issue with the Jaguar V12; it’ll happily idle at 300.
Well, mine does not happily idle below 700rpm at any temp.
Update - New Year’s Eve. Coolant Temperature Sensor replaced and also the AAV, after a super refurb by John_John, has been refitted and tested. The RPM post start is reducing as advertised and I’ve set the idle at 700rpm in P.
I think that those two elements can now be discounted as I took her for a run this afternoon (engine already at operating temp) - exactly the same symptoms: 5 minutes of flawless XJS motoring then the first almost imperceptible ‘miss’ then within the next 2-3 minutes full ‘kangaroo petrol’ mode - RPM unable to exceed 2000 then falling back to around 1500, ease off the accelerator, apply throttle and repeat.
So, attention now turns to the fuel pump/filter/sump tank unless anyone can suggest anything else at the front end to check.
the next things I would check up front are the coil and the ignition module. I’ve had bad coils act funny once they get hot (just after a couple minutes of running). Same with the control module.
But as you say, it could easily be fuel system. First thing is just replace fuel filter, as it’s super easy. Sump tank means draining fuel tank.
If you can get it to idle erratically, testing fuel pressure would tell you if that’s it. But there’s no easy way to test without splicing in a shraeder valve (which is what I did on the return hose from injector fuel rail to FPS)
One last thing to check, TPS. Make sure it’s .32V -.34V at idle. Test it again once car runs badly. Another possible electrical issue, once hot
I had the same symptoms, but in reverse…
The car would take for ever to start and when it did it would run in the same full ‘kangaroo petrol’ mode but 5 minutes later it would miraculously cure it’s self and run perfect.
It turned out to be a bad ECU, a cold solder somewhere or a faulty component, a heat related fault.
Nevertheless, I would first check the fuel pressure, the coils and change the GM ignition module before moving deeper…
Seven months on and I thought I’d give you an update. I’m pleased to report that all is well. 
After replacing the AAV and ETS which did manage to get rid of the surging, I decided that the rear fuel system, including the surge/sump tank had to come out.
I used the drain on the bottom of the surge tank to drain the car’s fuel after loosening the tank’s 3 retaining bolts (fuel pump had already been removed). The much hyped 11/16 spanner wouldn’t budge the nut after 35 years - so a same size socket was introduced up through the drain hole and with a big pull, it loosened.
The tank was removed and this was the state of the filter which pretty much mirrored the interior of the tank, There was no oily/fuel gunky reside clogging it up but what I can only describe as very fine, limestone-like dust resembling lime-scale.
This came out of the filter.
I used a solution of salt and vinegar to clean the tank interior and with a clean vent tube and new filter
all was reassembled.
Getting the tank back in to align the bolts with the threads on the flanges on the underside of the battery tray was ‘challenging’ (never, ever allow anyone to mention ‘precision engineering’ and ‘Jaguar’ in the same breath) but with everything back in place, HT lead back on the coil, ‘85’ back in the relay and battery connected the moment of truth - turn on the ignition, expect short delay before the 3 second run of the fuel pump and…nothing, just the engine cranking!
I was sure the pump was wired correctly so I went to the relay. ‘85’ was difficult enough to get out and I gave it some more ‘shove’ to make sure it was fully located. Back to ignition, whirr of fuel pump, and around 10 seconds later - all 12 are firing - hoorah!
But had the problem been cured? I’ve taken her on 2 test runs and she has behaved faultlessly: everything is working perfectly. So whether it was a dodgy fuel pump or a fuel pump made dodgy by the clogged filter? I’d suggest the latter.
Take from this saga what you will. Cheers everyone!
You had the larger pickup screen to begin with! Earlier cars such as my '83 came with a pickup screen half that size.
Your photos illustrate why I recommend removing the pickup screen in entirety and instead installing an inline filter between pickup tube and pump inlet. Any crud in the surge tank would get sucked into the pickup tube and caught in the inline filter. Then just replace the inline the filter and all of that crud is removed from the system entirely.
Famous last words!
The lack of power, kangaroo stop-start, lack of acceleration - call it what you will - seems cured. However…I took the car on a round trip of around 20 miles today. The outward leg was faultless.
On the return leg, it was a normal start, normal idle RPM and ‘ops normal’ until about 5 minutes into the trip. I had to stop at a road junction and the RPM sat at 1200 with the gear in D: I later put the car into P at another junction and the RPM shot up to 2000 before settling back at 1800.
I’ve already fitted a CTS and a refurbed AAV so any suggestions as to where to start looking for this, yet another, Jaguar aberration.
Thanks
Easy answer would be that the throttle isn’t properly returning to idle. That sometimes happens when the butterflies get dirty, notably the B bank one where the PCV is. Next time it happens, open the hood and manually push the butterflies toward closed. If that drops the idle speed, we know how to proceed.
There are zillions of other possibilities, but that one’s easy to check, so get it out of the way first.
Thanks - I’ll take a look there.
