Hi all
As I’m now installing two new electric cooling fans I wish to know what would be the optimal temps to switch on the fans.
The radiator has been cleaned out and I have two new 82deg C (180deg F) flow thermostats in place. The main fan will operate via the temp sensor in the water pump inflow (non adjustable) and the auxiliary fan will operate via the AC and also a pair of adjustable secondary temp sensors (dry thermistors) attached to the font of the top of the heads.
So what I wish to know is what level temp sensor should I buy for the main fan (my options are 167, 170, 174, 180 and 185deg F) and what should I be setting the two adjustable sensors at?
Trev
PS the auxiliary fan will switch on for two minutes via a timer after the ignition is turned off. This will have an overriding switch on the dash incase of an engine fire.
Hi Trev
I am just playing with these temperatures as well!!
I am busy trying to get my engine running with a megasquirt…so I can see the coolant temperature on my laptop…as well as the reading on the gauge.
But…if your thermostats are 82deg C then I think you need to select a temperature above that level a little bit. I have changed my fan switch ( in the pump inlet) from 85 deg C ( which is the original jag labelled part with 85C stamped on it)…to another which is on at 92 off at 87. It’s a std Tridon switch. My MS temp is on at 90 and off at 85…and at idle the MS controlled fan keeps the temp below 90. So one fan running can keep the system below 90 C
But…I have also added the Lutz mod …inserted variable size orifices in the coolant pipe connections to force more flow to the back of the engine…I’d be tempted to stick to the standard temp switch setting of 85 deg C because of the short circuiting that can take place on the standard systems. This probably keeps the rear temps at a better level…
Regards
Matt
Thanks for that Matt.
You are waaaaay ahead of me with the megasquirt thingy but it sound cool (pun intended). I’m inclined to agree with your suggestion of 85deg C as if it were cooler then the fan would keep going all the time unnecessarily and the engine has to have some heat to run reasonably.
I measured the head temps with a laser after the last couple of runs and found the hottest readings at the bolts that hold down the engine lifting rings at the front of the heads. They were the hottest spot at 92 when I pulled up and 98 five minutes later I chose to turn the motor over at this point and flush through some radiator coolant rather than risk the valve seats (keep in mind that I have had a 50% radiator block fixed since). I found no increase in temp at the back of the heads (like you might expect) at all. This is why I plan to place my sensors (for my auxiliary fan) on these bolts at the front.
The dual sensors (known as “Engine Guard”) are controlled by a very small computer on the dash and I can elect when either an alarm goes off (not going to do that) and/or have a relay start the auxiliary fan.
Thought I would do the front crank oil seal while it was reasonably exposed… which led to changing the oil filter bracket gaskets … then oil line rubbers etc, so I’m in deep about now…
Trev
If Jaguar used an 85ºC switch (I thought it was more like 94ºC?) then that explains why my original aux fan is running virtually all the time with 88ºF Jaguar thermostats! Annoying and unnecessary. Looking into this has been on my “to do” list.
For those interested here’s a listing of Tridon switches.
Do you remember what thread size the Jaguar switch is? (Sorry for my laziness…I’m sure the info is out there. But if you remember off-hand please let me know)
A 95/90 switch sounds about right for me.
Thanks
DD
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
5
That’s supposed to be the COLD spot in the coolant circuit. I think I’d
probably go with the 167. The fan might run whenever the engine is
warmed up, but I’d be OK with that.
My '83 wasn’t controlled this way, but in the end my main fan ran pretty much
whenever the engine was warmed up. You’d think that it wouldn’t be needed
on the highway, but apparently it was; you could see the engine temp start to
creep up when that main fan wasn’t on, even at highway speeds.
– Kirbert
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
6
Everybody here seems to be laboring under the misconception that the
coolant remains at a constant temperature all the way around the circuit.
Nothing could be further from the truth. It should be coolest at water pump
inlet and hottest at the thermostats. If the coolant temp at the water pump
inlet is at 85°C, your car is overheating big time, there is no capacity for heat
removal from the engine before it gets back around to the thermostats.
Kirby
The thermostats should control the outlet temperature flow to try and maintain somewhere around 82 deg C flow temperature of the head outlet temperature. This is the heat extraction temperature into the radiator. Then this part of the system adjusts the heat removal by adjusting the flow via the thermostats. On my modern cars this running temperature hardly moves from summer to winter…but the Jag seems to wander around a bit…but you are right…the temperature drops through the radiator and then increases thou the engine…but you want this particular switch to be an “emergency” switch…so set at a level where things are a bit beyond maximum design temp…and my MS set up allows me to watch the outlet temp…which I have switched at 90 deg. And one fan can hold 90 deg outlet temp. When I switch off I want this “emergency” switch to remove the heat soak energy…so I chose 92degC…but I haven’t tested this yet. I’ll have to disable the MS fan and read the temp when the emergency fan comes on to get an idea of temp difference . Other problems to sort out before this though…
Regards
Matt
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
9
Then this part of the system adjusts the heat removal by
adjusting the flow via the thermostats.
Actually, no, it does not. The flow is dependent only on water pump speed,
not thermostat operation. The thermostat merely redirects the coolant either
through the rad or a shortcut straight back to pump inlet to control the system
temp.
On my modern cars this running
temperature hardly moves from summer to winter…
On modern cars the gauge is fake. The sender is a series of resistors, and
as long as the temp is anywhere within reasonable limits, one particular
resistor is engaged and the gauge is held at a fixed point.
…but you
want this particular switch to be an “emergency” switch…
That’s debatable. IMHO, you want that fan to come on BEFORE things
reach the level of an emergency. Again IMHO, you want the thermostats to
be controlling the engine temp, not the fan switch. Ideally you’d like the fan
to be off at highway speeds, but when sitting at a stoplight you’d want the fan
to come on BEFORE the thermostats are full open and the engine is starting
to sweat.
Thanks for all your input…
After reading all the replies I’m having a re think.
Move the adjustable dual temp sensors (dash read out) from the heads to the rad inlets. This will give me both an idea of the head/s temp/s and allow me to adjust the main fan control.
Then I can move the auxiliary fan control back to the pump inlet. Then that would raise the question… what temp sensor to use there now. What would be the appropriate temp at the water pump inlet for engaging the auxiliary fan?
Trev
Here is an alternative temp sensor for the auxiliary fan I picked up. It’s an adjustable temp sensor that tucks into the rad hose. I could put this in the rad out flow hose but I’m not sure about sneaking the thermal line down beside the hose connection. I would be concerned re leakage. Has anyone used this method for getting a tem sensor into the coolant?
Trev
The existing auxiliary fan’s thermostat is stamped 85C which I bench tested and it switches on at 85C and off at 75C…which is what it is supposed to do … it never came on. even while the rad was blocked by 50% crud.
Trev
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
13
Don’t understand. If the cooling system is working properly, the temp at the
rad inlet is constant at 180°F. That means it’ll be difficult to get any
meaningful control from it. The rad OUTLET, by contrast, should vary all
over the map depending on airflow and water pump speed, meaning it
should be able to provide consistent control.
if the cooling system always worked properly we wouldn’t need control, meaningful or not… but it doesn’t, and that is why we have gauges and temp sensors so that we watch what our thermostats, radiators and such are doing.
Monitoring the temps of the rad inlets tells me if she is straying from that ideal 180F. The original dash gauge doesn’t do that…quoting you Kirbert… “it only measures the A bank”… and I believe you also said that we could overheat the B bank and drop a valve seat if the upper third of the radiator was compromised and that the dash gauge may not show it.(a tad paraphrased) but individual temp sensors would.
By monitoring both radiator inlets (both banks outlets) we …firstly, we allow the unit to switch on the main fan whenever either bank’s outlet reaches the desired temp…, secondly, to assess the efficiency of the cooling system (for both banks) in real time and not in dropped valve seats retrospectively… and thirdly, to be able to make seasonal adjustments for us that drive our cars in both extremes.
Having the outflow of the banks temps measured will allow for an overall assessment and not the erratic swings in rad out flow temp. What does one do if there is a slight variation in rad out flow temp that the auxiliary fan is not going to do? If it is a significant problem wouldn’t the banks outflow temps reflect that?
Trev
Kirby
The engine coolant flow is always dependent on the pump/engine speed…but if you restrict your considerations to max values you get a glimpse of how it is supposed to work at the extremes.
So in essence the flow thru the radiator is controlled by the thermostats…that’s how the engine coolant temperature is regulated. The flow that is diverted doesn’t add to the cooling effort. The coolant pump flows 355 litres per minute at 6000 rpm…so say 300 litres per minute for a worn pump…
At 240 kW power output the engine is throwing about 340kW of heat into the cooling system…roughly…so the temperature loss in the radiator needs to be about 16 deg C at max power…and this is what the designers HAD to allow for…add 16 psi pressure and the boiling point goes up to about 125 deg C …plus the antifreeze mix adds a bit…so say 120 deg C to be conservative…and the thermostats are designed to regulate the flow temperature to 82 deg C
So…you can see that the system is designed with a reasonable amount of “spare” capacity at max performance. Now…I am assuming that at lower speeds the power is less and the coolant flow is less…but how these parameters depend on engine speed and relate to each other I don’t know. You would need the pump flow curve to get a feel,for this.
So…back to the issue. …I suspect Jaguar set the 85 deg C switch at a compromise level …not wanting it on unnecessarily and making sure it would help stave off an unsightly steam puffing Jaguar!!! It represents a theoretical hot side temp of about 100 deg C which is nicely below the steam puffing level …and is above the normal running temperature so shouldn’t come on after a normal shut down if the coolant temps are from 82 to late 60’s
And…as to strategy…if you have added full electric fans…I’d probably come down on the side of reading the water temp exiting the heads…if only because that is the closest spot to the hottest place in the cooling system…
Matt
That is just a random pic of the type of unit I mean. I fit these whenever I add an E-fan to an older car (I usually remove the mechanical fan). The inline units I source here in Switzerland come with a wiring harness and relay, and can easily be modified to suit positive earth. They don’t do fan speed control, but that’s not an issue with the cars I work on.
You can also get the AL inline fitting with a thread to suit a standard temperature sensor or switch. If you’re running Megasquirt I think you could fit a sensor inline and use a 2 Megasquirt outputs to run the fans with a bit more control than the simple on/off/bit of hysteresis gap that the switches do. I’m not all that familiar with Megasquirt, but from what I’ve read you do have some free I/O and the ability to write code to control them.
Thank you Andrew. I feel privileged to have the moderator chime in but please note that was not a random pic but rather the same unit sitting on my bench waiting to be fitted to the girl and there is nothing random about her.
Trev
Heh… like it says in my “About me”, I’m just a guy.
By “random pic” I meant the sensor which I posted a link to.
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
19
I think we’re talking about a difference in operating philosophy. You are
interested in the aux fan operating ONLY when all else has failed and the
engine is beginning to overheat. I’m interested in the aux fan operating
during normal operation and PREVENTING the engine from overheating. I
suppose either scheme is workable, and many production cars (such as my
Honda) use your method, not bringing the fans on until the engine temp is
well past the thermostat control range.
– Kirbert
Kirbert
(Author of the Book, former owner of an '83 XJ-S H.E.)
20
…but if you restrict your considerations to max values you
get a glimpse of how it is supposed to work at the extremes.
Excellent analysis of the max power extreme. But I think we’d also need to
look at the OTHER extreme, idle stuck in traffic. The coolant flow is waaaay
down, arguably dropping faster than the waste heat rejection. In fact, I have
opined before that one possible cause of overheating in Jaguar V12’s is an
idle set too low. The idle being too low will impact both coolant flow and the
performance of the belt-driven fan.
…so the temperature loss in the radiator
needs to be about 16 deg C at max power…
That sounds like a good number. I’m betting the number at idle in traffic is
HIGHER.
…Now…I am assuming that at
lower speeds the power is less and the coolant flow is less…but
how these parameters depend on engine speed and relate to each other I
don’t know.
Me neither, but it should be possible to make some assumptions. It’s a
centrifugal pump with backward-angled vanes on the impeller. You can find
performance curves for similar pumps, and their curves all look basically the
same, varying only by the diameter of the impeller. You could come up with
a reasonable guess at the pressure generated at an engine speed of 750
rpm, and plug in a presumption that the coolant circuit is unchanged from the
6000 rpm model, so a typical flow/pressure drop relationship applies. I’d be
willing to bet that the coolant is barely moving at idle.
…I suspect Jaguar set the 85 deg C
switch at a compromise level …not wanting it on unnecessarily and
making sure it would help stave off an unsightly steam puffing
Jaguar!!!
Same philosophy my Honda Civic uses. Works fine on a Honda Civic. On a
Jaguar V12, not so much. That’s why my preference is to bring that aux fan
on sooner.
IMHO, the big problems with this cooling system are coolant flow and airflow.
Both are fine if everything is OEM and in good working order. But the rad
gets obstructed with all that sand internally, which plugs up the coolant flow
– often impacting the B bank more than the A bank. And the rad fins get
plugged up with leaves and dead bugs, plugging up the airflow. Both
problems are difficult for the lay owner to diagnose because you can’t just
pop off a radiator cap and see the crud in the tubes, and try as you might you
can’t see the obstructions in the fins without pulling the rad out.
