[series 3] Air Conditioner issues, hot air all time - testing procedure

Hi everyone! I’ve read as much as possible about the Delanair MK2 that is installed on my late 86 4.2 Sovereign, but I didn’t understand where to start to diagnose what is the problem that leads to the behaviour specified in the title of the thread.

In order: during the overhaul I commissioned after purchasing the car, the professional changed the parts required for r134 conversion and charged the system. According to him, after the charge the car produced icy air and he was quite satisfied of the work.

The day after I came to the shop to check the state of the work, he wanted to show the miracle but the car produced no fresh air, rather a very hot one! I left the shop with the promise that he would check again the system, particularly checking for leaks and for vacuum-related problems. The same evening he called me telling that the system was discharged and recharged again and the loss of gas he found was not noticeable, so he excluded any major leaks. Furthermore, he told me that the system started to work again.

I retired the car last Friday, with 37 °C here in northern Italy; I put the right knob in auto mode but after a minute or so an extremely hot air started to come from all the vents. :rage:

This is what I can state:

  • The mechanical part is somehow working, the system switches from lateral vents to central vents and both

  • the blowers go to maximum speed, stop, go to low speed if the knob is put on LOW, so I suppose that they work

  • the heater valve is open for sure, I don’t know if the amplifier tries to close it, for what I understand this happens only when maximum cold is required

  • when I bought the car last October, I drove it home by night and the temperature was quite low: my impression was that in heating mode the system worked well, the interior was not overheated

  • I didn’t made the drop on idle test, now I’m at he beach with my son but tomorrow I will do the test in order to understand if the AC clutch engages.

Now, I would like to understand more about the correct diagnosis procedure in order to identify the problem and know better my car. In this phase my goal is to check as much as possible dismounting as less as possible. Please, correct me if I am wrong in the upcoming checklist.

First of all, I would start to check the left knob (a potentiometer?) to check the resistance values, and the same for the temperature sensor on the right; someone suggested to simulate an extremely high atmosphere with a lighter, makes sense?

Then I would check the heater valve; the professional told me it was ok and that could be a vacuum line problem, but I want to be sure that is functioning properly. Is there a procedure to check the functionality without dismounting? If I’m not wrong several parts of the system depend on vacuum, so if some flaps are moving can I assume that at least the vacuum reservoir is working?

I understand that if the system fails it gies into DEF mode, but is this the case? Hot air all the time but in AUTO mode air from all the vents and changing.

About the AC compressor, I tried to check the temperature on the intakes but all the area was so damn hot that I didn’t understand nothing, I just burned my left hand like a chicken :expressionless:

Finally I understand that the core of the system is the amplifier, is there a procedure to test it, like checking the signal that engages the AC clutch?

I excluded fuses because fuse-dependent components seem to function (right?).

Thanks in advance to everyone!

I have owned XJ6s for about 19 years and keeping their air conditioning systems working properly has always been a challenge for me. In addition to the freon leaks (I have kept them running with R12 instead of converting them to the less efficient R134A) due to aging o-rings, and failing comoressors, I have also been plagued with A/C amplifier failures. When the A/C amplifier fails the system seems to have a mind of its own and does not follow control inputs on the temperature knob. This sounds like what you have encountered. The A/C amplifier is a common failure item in the Delanair MK II system and there are lots of posts in the archives about this, including many from me. The Jaguar part number for the A/C amplifier is C45402. If the one in your car is the original one, it is buried deep inside of the dash forward of the radio and impossible to access without a lot of disassembly.

When I replaced the A/C amplifiers in my XJ6s I just disconnected the wires and left the failed ones in place. As far as I know there is no simple way to test the A/C amplifiers except to replace a suspected bad one with a known good one to see if it fixes the problem.

Jaguar no longer produces the C45402. Used ones are usually avail able on eBay, but be sure to purchase a known good tested one that works. I purchased a few C45402s on eBay as spares from sellers that sold them as good, but unfortunately they tested bad in my XJ6s when I swapped out a good working one in a car with good air conditioning to test them. There are also some aftermarket ones available that show up on eBay or that can be found online. I posted about the different kinds of A/C amplifiers on Jag-Lovers a year or two ago with pictures of the various types of A/C amplifiers that I had in my possession. Unfortunately most of the OEM Jaguar ones that I had were bad. You can find that post and the pictures by searching the archives for something like “C45402 A/C amplifier pictures”.

You will discover that few if any of the problems that you have with your car are new. They may be new to you but members of this list likely encountered them decades before you and then posted about them. Some of them also posted about what they did to fix those problems and this treasure trove of valuable information is found in the archives. I use it regularly.


A few quick remarks. More later after Sunday chores

Right, valve open unless max cooling is called for.

No vacuum = open valve
Vacuum applied = closed valve

Have a helper rotate the temp control to full cold while you watch the valve. You should see the actuator arm move. If no movement no vacuum is being applied.

Run a hose from an engine vacuum nipple to the heater valve. It should close for sure with the engine running. If not, the valve is probably seized

The compressor should be engaged in all operating modes, even when heating is called for. If not, first check the fuses.

The compressor will turn off if A) the control is turned to ‘off’ or B) if the Ranco thermostat senses ice-up of the evaporator. You can easily bypass the Ranco for testing/experimenting

The compressor will not operate if the thermal fuse is blown, which would occur if the system is low on freon. This is a 3-prong fuse clipped to the compressor bracket…not one of the four regular fuses found in the cabin. From your previous remarks it sounds like the freon charge is OK, though



However, defrost mode defaults to max heat and max blower speed


Thanks everybody, I have read many threads (finding Paul many times) but everything seems possible and nothing clear :sweat_smile:

I understand that if the amplifier is the original one (and I think it is) is broken almost for sure, but before buying a Jag-Aire one or an used one (both from USA, not so handy) I would like to be sure that the rest is ok.

About Jag-Aire, I searched over old posts but I couldn’t find nothing specific about reliability: anyone mounting it without long-term problems?

Doug, so you confirm that if the vents are switching the system is not in DEF mode, despite the maximum heat and often the maximum speed?

Anyone knows where to find a schematic of the amplifier? I’m an electronics engineer and I am always curious about this old crappy electronics…

One of our Electrical Engineer list members took some OEM A/C amplifiers apart about 10 years ago to investigate the design and failure modes. He posted some pictures and his results including schematics I believe. His results will be in the archives. When the OEM A/C amplifier was manufactured, after assembly it was sealed in hard brown epoxy making non-destructive disassembly very difficult and repair almost impossible.

I have successfully used three types of A/C amplifiers in my two XJ6s with fully functional climate control systems, the OEM ones (looks like a small brown brick), an Australian aftermarket unit (looks like a white shotgun shell) and another aftermarket unit of unknown source (looks like a grey brick). I purchased a new Jag-Aire manual control unit a few years ago as a trouble shooting tool when I was stumped by a climate control problem but I never used it because I resolved that problem otherwise. It is now an on-the-shelf spare just in case I have another A/C amplifier failure. So I have nothing to report on use of the Jag-Aire units.


Be aware that just because the heater control valve seems to operate, the valve itself can remain OPEN when the control rod appears closed.

The internal valve can become disconnected and coolant will still flow when the valve is commanded closed.

I usually employ heater hose ‘pinch-off’ pliers to verify that the coolant actually stops flowing.




If the vents are switching the amplifier and servo are at least fundamentally operational; that is, they’re not ‘dead’. Correct operation is another matter.

With the temp control set for max cooling the system should direct almost all airflow to the dashboard center vent. The center vent opens only when max cooling is called for.

With the temp control set for max heating the system should direct most of the airflow to the foot well vents

If the system is in blend mode…neither max cooling nor max heating…most of the air will come from the outer dashboard vents and some from the foot well vents

In defrost mode almost all the air will come thru the defroster vents and, as mentioned, the system will default to highest fan speed and max heating

If heating is called for the system will not operate until the engine coolant warms a bit…so don’t be fooled.

If all of the above take place the the amp and servo are ‘probably’ OK…but you’ll still have to determine if the heater valve and compressor are being given commands and obeying them

It takes time to work through all the possible faults…and it isn’t unusual for the system to have multiple faults simultaneously. And the process can be made more frustrating without a ‘known good’ amplifier.

Anyhow, one step at a time. :slight_smile:


Because your mechanic said cold air came when he tested, I would start by trying if the servo works by running it manually. If it works including that the air flows change, it is probably the amplifier that is broken. I have used Jag-Aire myself for the past six months and it works, not cheap but the used ones often have some kind of wear and tear so they tend to break. Check the link that describes a manual test of the servo.

Series II Air Con - Aaaaaghhhh!


Note that the AC amplifier’s sole, and vital, function is to position the servo, Patrizio

The standard amp test is to set the control to ‘Auto’ and set the temp control alternatively to ‘65’ and ‘85’. The servo should ‘buzz’ audibly, indicating that it is resetting - no servo motion indicates a faulty AC amp. Without a working amp further testing is pointless - though the servo can be set ‘manually’…

However; the compressor must engage for cooling; drop in idle whenever the function control is set out of ‘off’ confirms function. An inline fuse is sometimes fitted and may blow without affecting other functions…

Vacuum; defrost vents are held closes in all settings but ‘Def’. Centre vents are opened by vacuum when system is in full cold - open centre vents imply both that there is vacuum, and that the system/servo is in cold setting…

In ‘cold’ setting all air passes through the evaporator - the heater core is bypassed, so heater valve has little influence. There is a sight glass on the receiver/drier - to be used to check refrigerant state.

When the set cabin temp is reached; the cold air from the evaporator (AC working) is reheated/tempered by passing some of the air through the heater core. Without AC; ambient air is used for cooling…

These basic tests takes 5 minutes and can be performed sitting comfortably in the driver’s seat. They will show areas of concern, and must(!) be performed before individual components are checked based on results…

xj6 85 Sov Europe (UK/NZ)

All you need to know:


Thank you all, really a great community! Time to get dirty this night! I will keep you updated.

Don’t worry,

at the age of your car, its state of service and the position of the Delanair unit you should be able to do this job in you business outfit - the good thing about that kind of jobs: positions are humiliating at times, but labor is seldom dirty, oily, greasy etc.

Good luck though - sometimes easier to swap a u-joint than fault finding at the vaccuum system


75 XJ6L 4.2 auto (UK spec)

Ok guys, here is the results of tonight tests.

And @Jochen_Glockner, I’m dirty and so satisfied! It was many years since I didn’t waste my time in playing with cars, and is soooooo satisfying!

So here is the situation: immediately after starting the car, with the right knob in AUTO a warm air blowing at minimum or medium speed came out from lateral vents and foot vents. No air from central vents, like the last time. The air coming from the vents gradually heated but no change in distribution.

Putting the system in DEF mode, the air suddenly starts to come from dashboard vents and at maximum speed and heat, as expected I think.

The minimum test has given no result: the clutch doesn’t t engage. So I started to check the compressor, when suddenly the clutch engaged (I clearly heard the sound): I rushed inside the car just in time to feel a breeze of fresh air coming from central vents, then nothing again.

I decided to check the compressor, focusing on command line and heat fuse: this is what I found.

Ugly. The best has yet to come:

I don’t think this fuse is as it should be, someone decided to bypass the fuse, maybe is melt, but at this point the fuse can’t exclude the clutch from engaging. The continuity is awful, so armed with my Dremel I cleaned all the contacts with a contact cleaner and a little help from a brass brush.

Now the continuity is tested between the cable coming from the amplifier and the clutch. But no luck, the compressor doesn’t start to work.

Is this an erratic behaviour ( compressor engaging and disengaging apparently with no logic) that can point with no doubt to a broken amplifier? Or maybe the compressor is somehow worn thanks to this bypass?

Finally, out of theme but never let a work done without a result: after trying several different replacements from EU suppliers with no luck, I found a solution to my flat headlamps. The Lucas ones (JLM9618) seem impossible to source, the reproductions are quite different so according to what I read over various forums I gave this Hella lamps a try:

Surely a different spot, but they fit perfectly and need only a minor modification to the connections of the parking lamps. Next days I will go to the mechanic to perform a fine tuning of the spot and then I will test the car by night, you never know that this time I will be capable to see something in front of me…

You holding the thermal fuse, Patrizio - which is blown or indeed modified. And you need a replacement as it is not repairable - the unit is actually sealed, parts invisible…

However, such PO machinations are a bit tricky to assess; the reason behind the mod is unknown - as are the PO actions taken…

Bear with me - it is somewhat long-winded but important…:slight_smile:

Inside the bit, (the thermal switch assembly) you are holding; between ‘S’ and ‘C’ should be a heating coil, and between ‘B’ and ‘C’ the actual meltable fuse. The actual power to the compressor clutch is delivered by green/brown to ‘B’, passing though fuse to ‘C’ - then green wire the compressor clutch. Which has a permanent ground, black wire, from the clutch…

Green/brown is powered when the function control is out of ‘off’ - but is routed through the Ranco switch at the evaporator - cutting power to clutch if temp in the evaporator drops below +2C. It is normal for the compressor to declutch at intervals - ambient temps plays a part. All air passes first through the evaporator, and while that cools the air - the air also heats the evaporator…:slight_smile:

The thermal fuse is a safety devise, protecting the compressor against low refrigerant levels. A superheat switch is fitted at the rear of the compressor - a black wire connects it to ‘S’. If too hot, nominally lack of refrigerant, the superheat switch grounds ‘S’, the heating coil heats up and melts the thermal fuse. Breaking clutch power, and the compressor disengages. Proper procedure; check refrigerant levels, rectify as required an replace thermal fuse assembly…

Two possibilities; PO has modified fuse assembly; nominally connecting ‘B’ and ‘C’ - which may have failed. This compromises compressor safety, but (between you and me) is not necessarily fatal…:slight_smile:

Alternatively; a PO may have replaced the compressor, and/or fitted the more modern safety devise - a pressure switch. Which is also fitted at the rear of the compressor, but disconnects clutch if low pressure is noticed in the compressor.

This set-up differs in wiring from the thermal fuse. With the pressure switch; the green wire from the Ranco thermostat is now connected directly to the compressor clutch. The ground wire from the clutch is connected to the pressure switch - which normally is grounded.

When low pressure is detected, the pressure switch breaks ground - and the clutch disengages. The advantage of this system is that the pressure switch automatically resets without any action required.

Theoretically; the varying pressure, like due to low refrigerant, may intermittently declutch the compressor, causing intermittent clutch engagement. So checking refrigerant is still pertinent.

However (again between me and you), some have permanently bypassed the pressure switch, permanently grounding the clutch - without serious consequences.

Obviously I do not approve, but standard tests for a non-engaging clutch is to both apply power directly to the clutch or grounding it. Simply to probe ‘why’ the clutch does not engage - or disengages out of turn…

So you have to check whether the pressure switch is fitted - rendering the thermal fuse you are holding redundant. Or a PO has done a ‘funny’…:slight_smile:

xj6 85 Sov Europe (UK.NZ)

I think that it is the “standard” layout with three wires and the superheat switch; as soon as I come back home I will take a picture of the compressor, buti it seems the standard one.


You can replace the Superheat Fuse with a more “modern” schema, using a relay (credit to Alex Cannara).
When the Low Pressure Switch trips the relay cuts the power to the compressor clutch.
I added two LEDs so you always know at what state the relay is.




1 Like

To check the clutch just connect 12V and ground to the clutch connections and you should hear, and see, the clutch engage.

By default the AC is activated all the time, (search Bernet and Napoli modifications if you want to play around) so either your system has not enough refrigerant and the low pressure switch is tripping , or there is a problem with the wiring somewhere.

A non operative servo when you change the Temp dial usually means a faulty AC Amplifier.
I strongly suggest that you also download Kirby Palm’s “XJS Help” (from this site), it has a big chapter dedicated to the Climate Control among the plethora of other useful info and mods.


Standard layout, yes. I think Frank meant that the previous owner bypassing it was non-standard

Until very late in Series III XJ6 production the superheat switch with thermal fuse was the standard arrangement. Late cars used the HSLP switch (high side, low pressure) arrangement. Different scheme, no thermal fuse needed.

Sometimes, if a compressor is replaced, the superheat switch type is replaced by the HSLP type, with appropriate wring changes. You can’t use an HSLP switch in a superheat compressor, or vice versa. There’s is a very subtle difference in the rear plate of each.

(As an aside I’ll add that your thermal fuse is probably very old, possibly original. The number ‘6551258’ is the original GM part number. GM hasn’t sold these fuses for a very long time. GM began phasing-out superheat arrangement circa 1977. In GM parlance it was know as a “Thermal Limiter”.)



But, first you’d have to determine if you have a low pressure switch (normally closed)…as opposed to the superheat switch (normally open)