From the other thread on fans a discussion came up about thermostats. I thought I'd start a new topic to keep this information more readily searchable.

As mentioned, the thermostats by SNG and others are made in batches and come and go in stock. I have purchased one of the SNG ones and the Mechanical Engineer in me wasn't happy with it. As I test fitted it to a spare 3.8 S Type intake manifold I noticed that there was a 5mm gap between the sliding element and the bypass port. From my experience with thermostats in V12 engines I know the thermostat disk only lifts 5mm off the bypass port when the thermostat allows the port to be fully open. Therefore, to me having a 5mm gap between the sleeve that is supposed to close off the port and the manifold is functionally equivalent to not having a sleeve at all.

My solution was to start with a "Series" Land Rover thermostat for the 2.25L engines. It needs some modifications, but it's not that difficult to do for someone with basic metalworking skills. The thermostats come in both 74 and 82 ratings, I choose an 82.

Step 1 is to remove the skirt on the outside and leave only the flat disk at the top. This is the same diameter as the S Type/ Mark 2 recess.I used a cutoff wheel in a dremel tool. Original LR on the left, skirt off on the right




Step 2 is to go to the hobby store and get a strip of 1/2" wide brass, form it to a semicircle and cut to length to have at about 180° of the thermostat's inner skirt covered. The object is to have the brass be a bit longer at each end than the bypass slot in the manifold. To determine the thickness of the brass, I measured the bore diameter, subtracted the skirt diameter to find total clearance, divided by 2 and then found the closest brass strip that was slightly thinner than the calculated clearance.

Step 3: Then solder this brass to the thermostat's inner skirt. Note the brass is soldered to the skirt only, not the top flange. This piece must move with changes in temperature of the coolant. Orientation is important, so when the weep hole is at the top the piece soldered on is in a position to cover the bypass slot in the manifold.





Step 4: test fit. I was able to achieve less than 0.5mm gap, By only putting the brass on half way around it allows some space for the skirt to move so it won't bind in the manifold bore. Refit the hose elbow and gasket as normal.
Since fitting the revised thermostat I have noticed lower running temperatures, close to thermostat temperature.





Bleedhole at top

 

Like this one ?

596225 - Thermostat for Land Rover Series - 4 Cylinder 2.25 | LR Parts

 

That's what I started with, yes.

 

It's pleasing that Jaguar owners are creative enough to do these things and disappointing that our parts suppliers too often aren't. The suppliers do a good job in making/sourcing components that otherwise wouldn't exist, but too often they are not quite perfect or a millimetre too small. Should we be grateful for having something or complain? When it's possible, I buy new old stock, secondhand, an equivalent for a modern car or make something rather than rely on repro.

 

I would stay away from new old stock brake seals, at least when it comes to silicone brake fluid.

 

The slightly loose fitting shroud of the Barratt thermostat is of no consequence if you fit a radiator that provides an over capacity of cooling like the Northhampton Radiators unit I recommended. (Jaguar's Alcohol shrouded thermostats used on the large bypass port heads were no better.) Changing the direction of flow provides substantial backpressure

Then the thermostat controls engine temperature. You are not running at close to boiling at any time with no spare cooling capacity to call on if required. i.e. you have excess cooling & the thermostat maintains engine temperature.

Tightening up the clearance was the WATJAG theory with borderline radiators. ( lack of cooling capacity)

 

I'm not sure I follow your logic. If the bypass port is open, then the coolant recirculates within the block only and never enters the radiator; so the size of the radiator is taken out of the equation. By closing the bypass, then all the coolant is forced through the radiator to be cooled.

I've certainly seen a dramatic difference in running temperatures in an XJS with the bypass blocked vs open by 5mm. Especially so in hot ambient temperatures with the AC on.

 

That is not my understanding at all & maybe I'm wrong. If so a lot of other people are wrong too. Always willing to learn. Some water always flows through the radiator unless the bypass slot is blocked, When the bypass slot is blocked by the thermostat opening then 100% of water/coolant goes through the radiator with a shrouded thermostat in theory. We accept it's not 100% effective unless you physically block the bypass like the racing guys do. After that the temperature is controlled by slight opening & closing of the thermostat dependent on operating temperature. e.g a waxstat is variable in it's operation. Jaguar is trying to achieve the most even heating of the block.

Or you can have the fancy Mercedes method but that requires an ecu, a heating element, & a number of parameters to control from the CAN Network. Plus a very expensive multi disk thermostat.

 

My understanding is that when the bypass is open, such as when the engine is cold, the coolant is taken from the intake manifold through bypass hose, into the water pump suction side and then pushed into the block where it circulates through the head and back to the intake manifold. The closed thermostat blocks coolant from flowing through the radiator. This promotes a fast warmup.

When the bypass is closed, then the "short circuit" from the intake manifold to the waterpump is closed, and instead the thermostat is open and pathway through the radiator is established. Then the water pump will draw coolant from the lower radiator hose as the source of coolant from the bypass is blocked. When the thermostat is open, then there is a connection to the upper radiator hose and coolant must circulate through the radiator.
If however, the bypass remains open, then the water pump can pull water from the intake manifold and thus the hot coolant exiting the head, rather then the cooled fluid exiting the radiator.

Please shoot holes in my understanding if you see some!

You're right, the racing guys block off the bypass completely, or put in a plug with a 1/16" hole in the hose allow for a slightly faster warmup. The plug is to ensure all the coolant goes to the radiator. My modification emulates that, but allows for cold recirculation within the block to give a fast warmup in normal driving.

As I have fitted the Megasquirt EFI system, I can datalog everything in 1/100 second increments while driving, and my bulk coolant temperatures are now near thermostat temperature after doing the modification; previously they could be 10-15°C above thermostat temp on a long grade or in standing traffic. It's made a very noticeable difference to my car with no other modifications. The need for the thermostat to actually reach the bypass port to close it is well known on the XJS forum, that's usually the first cure for an overheating V12- change to the proper thermostats, as the OEM Jaguar ones a dealer will sell you are wrong.

 

The original system in the Mk2 was very stable. The temperature on the gauge rose to 70 degrees and sat there constant pretty regardless. The only situation I recall mine doing anything different would be stopping after a long hard drive when it would wander up a little. I don't think I've had another car with such a stable temperature needle.

 

Jagboi ~ I don't know whether you are right or wrong. All I can say is that there are a large number of people out there that think differently. Me being one of them.

I'm not shooting holes in anything. See my response to Bill Mac on the other thread. I'm here to learn.

My response to Bill.

"Bill ~ you are supposed to be the specialist here.

My understanding has always been that the bypass is there to stop any hotspots/uneven heating forming in the block/head during warmup to prevent stress. Simple as that.

You are spot on with uncorroded water rails becoming rare. I searched high & low for mine. Many have been welded multiple times.

Can the car run without the bypass ~ Yes. Been done in racing for years. Is it good for the block & head ~ probably not."

Hotspots are known to build up around Cylinder 1 (rear) of the XK engine.

 

I suspect that how well you need to close the bypass depends on the local fluid mechanics, that is the tendency for the fluid to go straight on rather than round a corner, and the back pressure from the two directions. The local fluid mechanics will be messy and unpredictable as none of the parts are what you might call streamlined. The back pressure from the radiator is also uncertain especially if it becomes partially blocked. The V12 cross flow radiator is extremely susceptible to becoming blocked as the system seems to produce a lot of sediment to the point that some owners have installed filters in their top hoses. Vertical flow radiators as in the Mk2 might well be a lot better as sediment is likely to sit in the bottom tank. Still if you want to be sure the flow doesn't go in an unwanted direction, it's good to close that direction well. At the same time, it's good to avoid clearances so tight the system jams.

A modern bypass thermostat in the top hose might be the easiest solution for anyone (like me) not dedicated to originality.

 

"I suspect that how well you need to close the bypass depends on the local fluid mechanics, that is the tendency for the fluid to go straight on rather than round a corner, and the back pressure from the two directions."

And there lies the rub.

 

It's interesting you should say that, my car is the same. I've added an additional temperature sender for the EFI and I can watch the temperature on my laptop on the seat beside me as I drive. Up to about 75°, the dash gauge and the laptop agree. As the temperature rises however, the dash gauge will stay at 75 while the computer reported temperature keeps on rising. I have checked the hoses with an IR thermometer and it agrees with the computer. I've had it into the 90's and the dash gauge still says 75. It appears the dash gauge is an early form of idiot gauge? It always stays in "normal range".

 

I'm used to the American V8's, where the coolant circulates in the block until it gets warm enough to open the thermostat, and that is what opens the flow path to the radiator. The idea was to get the block warm fast and that in turn takes the choke off, when cars were carburreted. The Jaguar system looked functionally identical to me, and it made sense to do the same purpose, as it's always good to get an engine up to temperature as quickly as possible.

I agree with you, the bypass is to warm the engine quickly and evenly.

 

That sounds disturbing. Does your gauge never go above about 75deg? As I recall (my car's been off the road for a long time), mine stopped at 70 and went above only after a long period of high load (mountain or fast motorway) and then stopping without a short cooling down run. That would push the temperature up, but not past 80. Has the sender been replaced on your car?

 

I have added to Bill's reply from me ~ just had a 2 hour rolling blackout here.

Hotspots are known to build up around Cylinder 1 (rear) of the XK engine.

 

No, it seems to stop at 75. I have changed senders, no effect. I have not changed the gauge though. I did have the car go up to 95 once, and the gauge didn't go over 75.

I have put on the pusher fan from a Mercedes 560, the ECU controls that. I had a relay fail and the fan didn't come on, so I didn't notice the high temperature right away.

 

I'm pleased you put in the full ECU control. The ECU controls the fan controller (usually mounted on one leg of the fan). If you don't do this the fan blade tips can break through the sound barrier & kick up a hell of a noise running flat out.

 

It's a reasonably quiet fan, but moves lots of air. I picked it because it's diameter is the width of the S Types radiator. It was actually fairly easy to fit, only had to make one bracket.