K operating temp

[Rob Walker]

In the latest LF Minister suggest that 70-75C is the optimum operating temp under load and that they would not like to see 85-90C ever. Well the K is fitted with a 88C thermostat as standard which starts to open at 88C and is fully open at 100C. In practise I have seen over 90C water temp on my K on track days, I have noted similar temperatures on other members cars. Minister also suggest 100C is the ceiling limit for oil tank temperature, again I regularly exceed this. The question is are we all running our K`s too hot and is there a case for rushing out and fitted 78 or 82C thermostats or are Minister talking hot air?

[Roger King]

Personally, I'm with Minister on this one, although I like to see operating temp slightly higher at between 75 - 85 degrees.

 

So yes, I'd advise a 78 degree thermostat.

[andy oliver]

Where can you buy a 78 degree thermostat?

[CHRIS CLARK]

So Roger, would you say that my 'K's operating temp is OK? It runs at an indicated 75-78* normally and the fan ( operated by a Kenlowe adjustable stat) I have cut in at an indicated 88*. I always felt it should run at around 85*, but it doesn't get there unless stopped in traffic!!!

[R500 CYA]

I'm not convinced by the accuracy of the temp gauge anyway!

 

My K series runs at a constant 90-95 degrees, which is 5 degrees higher since changing the sender unit, but making no other alterations. Odd!

 

The fan only cuts in on hot days in heavy traffic, and since this works from a seperate temp sensor, I can only presume the gauge is inaccurate.

 

Everything seems to work just fine, so I think my head in the sand philosophy is working out wonderfully!

 

Nick

[Rob Walker]

Thanks Roger. Andy you can get 78C thermostats from QED and 82C stats are standard fit on dare I say it the Rover Diesel engines.

[Mike Bees]

The gauge isn't an accurate guide Chris. My ECU reckons the temp. is 10deg higher than that shown on the gauge, and there's no guarantee that the ECU is right either...

 

Mike

[CHRIS CLARK]

I remember you mentioning that before Mike. Maybe I should just remove all the gauges and save the weight! My old chief engineer (aircraft) used to call 'em worry gauges. Why ? Well you look at them and worry!!!

[Peter Carmichael]

Chris,

 

How are you measuring the temperature? If the thermostat opens at 88 degrees then there is stuff-all chance of it running below that... stands to reason.

 

The Minister interview was a journalistic article rather than an engineering treatise and I suspect that Graham's words are good advice which could be backed up by solid reasoning. The reasons for wanting reduced coolant temperatures could vary from prevention of pinking to issues around the stability of the coolant (cavitation), maintenance of the bulk oil temperature and thermal expansion. It all depends on which failure condition is most marginal and what you are trying to fix.

 

I would love to know how Minister are achieving stable low temperatures. You cannot do it without a low temperature thermostat and my casual enquiries were unable to find anything less than an 82C thermostat in the QH catalogue. Drilling holes in the thermostat actually has the opposite affect to the one desired and should be considered to be an old wives' solution.

 

The cooling package in the Caterham is extremely capable. The radiators have sufficient heat transfer capacity, a good flow of air (apart from when slipstreaming) and the pipework has good short, clear runs from the engine to the radiator. The only thing in doubt with the k-series is the thermostat setting (and siting) and the tendency for the radiator switch to stop working. Additionally, the oil temperature can get a bit high.

 

I would recommend an 82 degree thermostat (available from Lotus dealers as part of their Sport pack for the Elise) as a first measure. I am going to run the radiator fan switched by the ECU. The oil temperature is a more difficult matter - an oil/water heat exchanger will speed the rise of oil temperature and will attempt to tie it to the water temperature. It might be that the extra heat transfer will exceed the cooling system's ability to dump heat, but that remains to be seen.

 

 

 

[Peter Carmichael]

Aha! 78 degrees from QED.

[clarkey]

I remember a thread a while back about the temperature reading on Superlight Rs - mine rarely reaches 60 degrees, and has only been above this once or twice. Is this due to where the sender is positioned, or is this the optimal operating temperature? Seems pretty cool to me.

[jonhill]

It was I who started the thread on SLR temps., as on the road my car hardly seemed to get properly warm. Since then I've been on the track with it, and have seen the water rock steady at a smidge over an indicated 80 degrees no matter what, with the oil exceeding that, but not by that much, and not getting to 100 (I assume it's 100, as it's the next mark).

 

I'm currently running 0W40, but I am thinking that for mainly TD use 15W55 would be a better bet? Minister said this too, but I can't recall whether they were talking about the VHPD or the R500 (I don't think it was exactly clear in LF).

 

Jon

[stevefoster]

I calibrate the gauges I have had on my 7 by popping the connected sender in to a kettle of water as it is boiling. If the gauge shows 100 then you know that at least 100 is 100.

This of course may change over time and it isn't a cert. that 75 is a true 75 of course...

 

[Sid Spalding]

I'm concerned about temp as well as I'm doing a round the Alps tour at the end of June - "The Italian Job". The temp is likely to be "slightly" higher than UK ambient temp (plus the pressure will be lower at the top of the passes !) When cruising in top at 5500 I maintaining about 85 degrees. So what we're saying here is swap the standard Thermostat which cuts in at 88 degrees for a non-standard that cuts in at 78 degrees and everything will be Hunky Dory ? So are Minister Spec'ing that for the Engines they configure ?

[Peter Carmichael]

So what we're saying here is swap the standard Thermostat which cuts in at 88 degrees for a non-standard that cuts in at 78 degrees and everything will be Hunky Dory

 

er.. not quite.

 

The 78 degrees thermostat will attempt to keep the coolant at ~78 degrees. If ambient conditions and airflow mean that the radiator cannot ditch sufficient heat to keep the temperature down, the thermostat will be fully open and the temperature will rise until the temp differential to ambient increases the heat transfer. The temp may settle above the fully open temperature for your standard thermostat. If conditions meant that with an 88 degree thermostat your temp would run at 95 degrees, with a 78 degree thermostat your temp will still run at 95 degrees.

 

If your radiator does not have enough capacity, the thermostat makes b*gger all difference. FWIW, I think the Caterham radiators generally have sufficient capacity.

 

The logic for fitting a lower temp thermostat is to maintain a stable temperature lower than standard.

 

The problem you stated is exactly the problem that will not be fixed by a lower temp thermostat. Fortunately it is not a problem you will actually face because the cooling capacity of the radiator is adequate. The only thing to watch out for is when the air flow through the radiator is reduced. The thermoswitches in the radiator have been known to fail and an override switch is a good reassurance.

[Rob Walker]

I have just spoken to QED their 78C stat as per their catalogue is in fact 74C. They advised me that they also sell a remote Thermostat housing to get the stat away from the engine. Their recomendation for the Caterham was to fit the 82C.

[TorAtle]

So all this boils down to getting the thermostat from the mentioned Rover diesel engine??

 

[pfreeman]

My R500 is now running with a mod which turns the fan on at 75 degrees via the ECU - although I'm sure that the thermostat is still the standard 88 degrees one which doesn't make much sence to me !!

[murph7355]

Would not running a thermostat be a problem? Permanently open then?

[Peter Carmichael]

I should have given acknowledgement to Rob Walkers point about the standard thermostat being fully open by 100 degrees. i.e. All thermostats operate over a range of temperature.

 

The speed of response is an important characteristic of a thermostat. The mixing of the radiator and bypass coolant flows as they pass through the thermostat is important. The siting of the thermostat (on the cool side of the radiator) is important. The flow rates are important.

 

In the following it is worth knowing that the Stack sensors under-read by a small amount.

 

My datalogs show the temperature of my 250bhp k climbing from a thermostat controlled 82C to a stable 93C with the fan on in traffic. Once on the move again, the temperature quickly settled back to 82C. The temperature started to drop within 10 seconds of the car running at 30mph. The rate of temperature drop was pretty constant right down to 82C and then the thermostat behaviour caused a slight bounce to 83C.

 

The coolant temp continued to drop back to 82C as the oil temperature stabilised from 87C down to 70C, meaning that the 82C seen on the bounce was in the thermostatically controlled region.

 

A bounce that closes the thermostat fully is something to avoid and something that Elises have a reputation of doing.

 

The bounce indicates that the cooling system behaviour was just marginally underdamped in this circumstance. This gave a quick settling of the temperature.

 

If the themostat temperature is too low, the engine will spend a lot of time running with the thermostat wide open. This will result in significant temperature fluctuations depending on how hard the engine is being worked. Imagine a situation where you can hold all other things equal...

 

e.g. the difference between averaging 80mph round a track and a rock steady 80mph on the motorway.

 

The air flow is approximately the same but the cycle of accelerating and braking means that more power is being produced and more reject heat is being produced. If the coolant settles at a temperature within a degree or so of the steady state temperature then it is still under thermostatic control. This is certainly the case with my Supersport, but I don't have figures for the 250bhp engine becuase it didn't survive long enough in its first incarnation to get to a track.

 

If this isn't the case and you are unhappy with the temperatures being reached, you need to look at the radiator configuration. If the temperature fluctuates but does not reach unacceptable levels, then the thermostat is set about right and the radiator configuration is about right. If the temperature never fluctuates then you could happily move to a lower temperature thermostat and achieve consistent lower temperature running (this is the situation with the Supersport).

 

I should have paid more attention in my control engineering lectures.

 

[Peter Carmichael]

Not running a thermostat is a last ditch attempt to increase the total cooling capability of the system by increasing the coolant flow rate. You will get temperature fluctations depending on duty cycle of the engine and road speed. It should only be considered when the track temperature is getting too high for comfort. There are better things you can do.

 

You could always plumb in two thermostats in parallel.

 

i.e. output from first feeds the bypass input of the second.

 

The QED mods sound like they are primarily intended for the Elise. In the Elise application, the return temperature of the radiator circuit is too low because it has been piped all the way to the front mounted radiator and back again. The large difference between the temperatures confuses the poor thermostat. Running a lower thermostat means that the thermostat is more often open and the thermostat is not bouncing off its closed position all the time.

 

If you mount an appropriate thermostat in an appropriate housing on the exit side of the engine, the coolant temperature will be more perfectly mixed as it passes through the thermostat. (The housing and thermostat have to be configured to split one flow into two rather than merge two flows into one.) The flows can then be recombined after the radiator at some distance from the engine to allow proper mixing to further reduce thermal shock.

 

My previous advice is in line with QEDs. The cooling pack in the Elise is too efficient for a high temperature thermostat. In the Caterham, the cooling pack is more closely matched to the engine requirements, although it still has an excess capability for standard engines with an 88C thermostat.

 

If you cannot achieve consistent temperature behaviour with a modified engine, you need to look at the radiator, not the thermostat.

 

If your temperatures are suspiciously consistent you can run cooler with a lower thermostat, with all sorts of benefits.

 

Edited by - Peter Carmichael on 7 Jun 2001 13:23:40

[Peter Carmichael]

I have just thought of a cunning mod...

 

... I don't know if Minister thought of it first.

 

Remove the thermostat.

Run the fan in reverse.

Switch the fan off when the temperature goes above 75C.

 

The coolant flow rate through the radiator will be greater because there isn't a thermostat getting in the way. When the coolant is below 75C, the fan will work against the airflow, reducing the effectiveness of the radiator and so the engine will get a chance to warm up. When the coolant gets above 75C, the fan will switch off allowing the airflow to do its stuff.

 

The problem is that at standstill the fan will switch off and the temperature will rise unchecked. You could probably get an ECU to switch the fan, based on rpm and temperature.

 

The warm up period will also be slower because all the water in the radiator will need to come up to temperature whereas with a thermostat the cold water would gradually be bled into circulation once the bypass circuit gets up to temperature.

 

Also, the fan will draw power under circumstances where it might otherwise be unnecessary.

 

Anyone think this could work??

[Rob Walker]

Peter running my fan in reverse by accident on my Esprit increased running temperatures by 20C even at high speed. This situation occured when I fitted pattern made fans, I didn`t know they ran the opposite way.

[TorAtle]

I'm not totally with you Peter. You would like consistent temperature behaviour. You talk about thermostat opening characteristics vs radiator size and the importance of them being matched. You seem to dislike the idea of a 78 (actually 74) degree thermostat?! Then you launch the idea of a reversed fan and no thermostat - this idea must surely make the temp go up'n'down with your driving style unless you cruise in 6th at low revs in cold weather.

 

So is controlled, consistent temperatur more important than as low as we realisticly can get it, even if it will vary a little?

 

I like the idea of a 74 degree thermostat. This will bring the temp up fast, keep it at minimum 74 degrees and as low as the radiator can manage. The only downer I can see is that it restricts flow somewhat, how much I don't know. If the temperature is still too high worst-case, then we need a bigger radiator, period.

 

Or am I talking bollocs??

 

Tor Atle

PS I like your newthinking ideas btw, the fact that no-one has thought of it before proves close to nothing :-)

[Peter Carmichael]

Tor,

 

I'll admit that in the jumble of information I was perhaps not clear. Let me try to clarify.

 

We are aiming for:

 

- low consistent temperatures under load (~75)

- quick warmup

 

If we change the thermostat, it may help or it may not help, depending on whether the other parts of the cooling system are up to the job. Looking out for the onset of inconsistent temperature behaviour shows you when some other part of the cooling system has run out of capacity.

 

The only reason for disliking the 74 degree thermostat idea is if the radiator isn't up to the job and you get inconsistent temperatures.

 

Another consideration is that we know there must be an optimum temperature and we are lending credence to the Minister view that 70-75 is *the* temperature. Just to clarify by exaggerating: we can imagine that 0 degrees would be too low because of thermal shock and 130 too high because of coolant boiling and all sorts of nasties. Why is 70-75 degrees the right answer? We have no justification for that particular figure apart from Minister's experience. We don't know what mode of failure we are trying to fix by keeping the temperature down.