Overheating

1997 Overheating Survey Results
Overheating Generally
Manual Over-Ride for Fans
Griffs in Hot Countries

up to contents

1997 Overheating Survey Results

July 97
This is based on a questionnaire which I sent to Sprint in April. The accompanying letter said that the second issue of the Workshop Notes was available, and anyone who returned the questionnaire could have a copy. The main reason for doing this was that I'd been wondering how to encourage people who were asking for the notes to contribute something themselves. Offers to 'send something in' generally didn't materialise and all this one-way traffic wasn't quite what I had in mind when I started. I picked overheating as a topic because there had been a long exchange of emails on the internet about it, a few of which are included below. I am extremely grateful to those who did send the questionnaire in, a number of whom didn't want the Workshop Notes and were simply being helpful.

Profile of the Survey Sample

Sample size: 31, of which:
Griffiths: 17
Chimaeras: 14
400s: 12
4.3s: 11 (including two BVs and one 1994 4.3 HC)
500s: 8 (all 1994 or later)

Average mileage: 20,000
Average time with current owner: 18 months
Average mileage with current owner: 11,000
Bought new: 12
Fitted with cats: 21 (one with cat removed)
Fitted with aircon: 3 (one 500, one 4.3 and one 400)

All years were represented, but biased towards the older ones (ten 1992 cars down to three 1996 cars). I had intended to speak to a number of dealers, but after the first one confirmed most of the survey results I decided to stop at that point. The dealer's comments are included below.

Definition and Incidence of Overheating

People's idea of what was a normal running temperature for their car varied between 75 degrees and 95 degrees, the average being 85. The dealer I spoke to told me that they check temperatures off the ECU sensor and this should normally indicate 90 degrees, but that the temperature gauge (with a differently sited sensor) may be showing between 80 and 90.

I also asked what the temperature gauge was saying when owners felt their car was overheating. The answer varied between 95 degrees and 'off the dial', but the average was 100. The dealer's opinion was that the fans should cut in at 100 degrees on the ECU sensor. The fan otter switch (again in a different location) should cut the fans in at 98 degrees and switch them off again when the temperature drops to 93 degrees, but this may not be exactly what the temperature gauge is showing.

Twenty four of the owners' cars had overheated at some time, the other seven hadn't. Fourteen of the 'overheaters' didn't report any signs of the car being too hot apart from the reading on the temperature gauge. The dealer I spoke to services about 300 cars on a regular basis, and he guessed that up to a quarter of them had reported an overheating problem at some time. This really was a guess and needs to be interpreted in the light of other comments below.

With a couple of exceptions, overheating in the owner survey had been confined to hard driving or heavy traffic in hot or mild weather. The dealer reported a similar experience, with overheating confined mostly to town driving.

Classification of Causes
Thirty one faults were reported on the twenty four cars that had overheated. These are broken down by type below.

Maintenance Problems

I've used this term to include anything that wouldn't have happened if the car had been properly maintained. No offence intended to the owners here. Unless you've been told how often to check the coolant level and where to top it up to, how should you know? And you can't be held responsible for what the previous owner got up to.

There were thirteen faults of this type (40% of the total).

- Five incidents of overheating caused by thhe wrong pattern thermostat or fan otter switch.

- Eight incidents of low coolant level or aiirlocks.

The dealer believed that at least 80% of overheating problems fell into this category, the rest being spread fairly evenly over the other categories below.

Tips
- Check the coolant level every week (yes, rreally - if something's on the blink the sooner you find out the better).

- At the same time, look around for visible signs of leaks, particularly at hose junctions, and check for frayed or perished hoses and loose or pinched hose clips.

- Don't top up with plain water. The antifreeeze in an aluminium engine prevents corrosion and things can silt up pretty quickly if you let the concentration drop.

- Get the system pressure tested if you susppect there is a leak but can't find it.

- If the heater isn't working well, one or mmore hoses feel cool or soft when the engine's warm, or you hear a gurgling/wooshing in the passenger footwell, you may have an airlock in the system. Bleeding it is not difficult, but be very careful when putting the brass plug back in the rad as it's easy to cross the thread and it's a real pain to repair.

Minor Component Failures

This refers to things that don't have a very long service life, like drive belts and hoses, or are easy to repair or replace.

Owners reported nine faults of this type (30% of total).

- One pressure cap leak.

- Five cases of faulty fan otter switches (mmostly on 4.3s and/or cars over three years old)

- Three cases of faulty thermostats.

This doesn't quite agree with the dealer's experience, who said that faulty thermostats were extremely rare.

Tips
- The correct thermostat is an 82 degree uniit with a large orifice. This is not a standard Range Rover part. The opening temperature should be stamped on the thermostat. You can test it in a saucepan with a thermometer to see when/whether it actually opens.

- If the fans don't cut in above about 90 deegrees, try the following:

- Sound the horn. If it doesn't work this maay mean the fan fuse has blown, although not all cars have the fan and horn on the same fuse.

- Take the wires off the fan switch at the bbottom of the swirl pot and connect the terminals together. If the fans don't start when you turn the ignition on, there is a fault in the fan fuse, the relay (which is on the main panel in the passenger footwell) or the switch itself.

- Wiring faults are very unlikely unless sommeone has been messing about with things.

- The fan (otter) switch is easy to replace but make sure you get the correct 98/93 degree unit and don't try to do it when the engine's hot!

Major Component Failures

This means things that shouldn't need to be replaced or repaired until the engine needs a major overhaul. Only one fault of this type was reported - a leaking radiator.

Tips
- A shrieking drive belt may only mean it neeeds adjusting, but it may also mean the pump bearings are on their way out.

- Radiator leaks are not always detectable iin a pressure test, but you should see the signs if you take the grille off as coolant collects there.

- The radiator, by the way, is a standard raange Rover part with a modifed bottom hose elbow.

Design Faults and Undiagnosed Faults

This means the car still overheats when everything is as specified by the factory, adjusted properly and in good repair. Owners reported five cases of fan fuses blowing repeatedly (four on 1994/5 500s).

- One of these was solved by the owner fittiing a fuse for each fan.

- One was solved with factory mod to staggerr the two fans starting up.

- The other three cars were still overheatinng.

One car's overheating was solved by fitting a bigger radiator.

One was solved with a manual fan override switch.

Tips
- The cooling system has been modified in smmall ways more or less continuously since the Griffith was launched in 1992. By mid 1994 when the serpentine engine came in, there was more than enough 'margin' to cope with the worst UK driving conditions. Attention since then has focussed mainly on tropicalising the cars for the far east market, where sitting in a traffic jam in mid 30s temperatures with the aircon running full blast places much greater demands on the system. Changes made for the Malaysian cars have been incorporated in the UK models as well.

- If your fan fuses keep blowing, ask a dealler what can be done about it. More powerful fans (with 40 amp slow-blow fuses) were fitted around late 1994 to early 1995, and twin fuses in 1996. It's possible your car has acquired one without the other.

- 1997 cars have a shroud around the fan (ottter) switch to protect it from exhaust heat. Otter switches aren't that reliable at the best of times, and maybe this would solve the problems owners have reported here.

- The factory did offer a twin relay wiring harness for the fans a year or so ago, which cut the fans in one after the other to reduce the load on the fuses. Later modifications may have made this unnecessary, but I would imagine it is still available.

Conclusions

According to the club, Sprint is read by close to 1,000 owners of Chimaeras and Griffiths. Although the questionnaire was linked to an offer of the Workshop Notes, I've already mentioned that a number of owners who didn't want the notes sent the form in anyway. I'm pretty sure that anyone with a serious overheating problem would have taken the opportunity to tell me about it, whether they wanted the notes or not. On this basis I think it is reasonable to conclude that an extremely small proportion of the 1,000 owners who read Sprint have ever experienced overheating problems.

This is certainly inconsistent with the amount of discussion about overheating that takes place within the internet mailing list, and with the perception a lot of potential buyers seem to have about problems they might expect with their first TVR. I can only assume that we are dealing with a prejudice carried over from earlier TVR models. Prejudice being what it is, people tend to look for things that support what they believe and ignore things that don't.

From the responses received, I can find very little evidence that a well maintained car with the correct parts fitted should suffer from overheating. If it does, it is most likely to be the sort of problem that would be picked up during a routine service.

There are some pointers towards 4.3s being a little more prone to overheating, but this may be as much to do with their age as anything else. Common sense suggests that the first versions of any car might not be as well sorted as later ones, and that the more highly tuned versions might be a bit less forgiving in heavy traffic or if their cooling system isn't in top-notch condition. The survey doesn't provide any information on the effects of air conditioning, but it seems obvious to me that the stess imposed by this is not going to be kind to a cooling system that's a bit below par.

top

Overheating Generally

Internet Mailing List Oct 96
We seem to be getting another round of questions about overheating, so I thought I'd pitch in with some observations. I don't have a problem with my Chim, but my MGB V8 was a real pig in this respect and no sooner was one cause identified than another one developed.

1. Is the car really overheating, or just displaying a high temperature on the gauge? It's worth getting a garage to check this, because I've had a faulty gauge in the past and got all upset about nothing. We'll assume that the gauge is accurate.

2. If the fans are not cutting in until the gauge is into the red, but the temperature does go down once they are running, this suggests that the otter switch is operating at too high a temperature. There could be two reasons for this:

a. The switch is faulty. In this case replace it with one of the original pattern.

b. The switch is operating within spec but the spec. is no longer appropriate. On the Chimaera the otter switch is in the swirl pot (millimeters from the exhaust) while the gauge sensor is in the block. Something may have happened to change the relationship between the temperature at these two locations, so that the poor old otter switch is doing its job properly but it no longer has the desired effect. There are some thoughts on this possibility below. In this case you will need an otter switch with a different temperature range.

3. If the fans are cutting in when they should but they don't pull the temperature down, then you have a cooling system which can't dissipate the heat being produced by the engine. Getting the fans to cut in at a lower temperature isn't going to cure this. Again, it could be because the system is malfunctioning, or because a different system is needed for current conditions.

a. The cooling system could be under-performing for a variety of reasons, all of which should be fairly easy for a garage to diagnose and remedy. The most obvious ones are:

i. Tired fan motors not shifting enough air. The motors themselves may be worn out, or there could be poor electrical connections. Try a direct connection to the fans from the battery and see if they run faster. See if someone else's fans run faster than yours.

ii. Leaves or similar obstructing the air flow through the radiator.

iv. Gunge in the radiator core, cylinder block or hoses. Get it flushed. If still unsure, get the radiator dismantled, as blocked rads are very common.

iii. Minor leaks (hose clips, radiator cap, radiator itself) causing a drop in coolant pressure. This allows the water to boil and froth can't carry the heat away properly. Might not be enough to produce noticeable coolant loss. Get it pressure tested.

iv. Clapped out water pump or drive belt slipping. Listen for squeals and rumbles.

v. Thermostat stuck closed. Shouldn't be possible, but test it in pan of water.

vi. Too little antifreeze (lowers the boiling point again). Get it checked.

vii. Air locks in the cooling system. Bleed it.

b. Different conditions could have occured because:

i. Modern fuels produce more heat. A lot of classic cars have overheating problems for this reason. The result might be general overheating, or localised hot-spots in the cooling system (eg. near the heads and exhausts) which the otter-switch can't pick up.

ii. Driving conditions have changed. New cars (even TVRs) are designed with long traffic jams in mind, but earlier models might not have been.

iii. The engine is not running properly (timing wrong, weak mixture, head gasket leak, etc., etc.). Serves you right.

There is no point in fitting non-standard components in an attempt to mask a fault in the cooling system, so the first step is to check the items in 3a. If (and only if) the system is performing as is was designed to, you may have to consider modifying it. The easiest mod is to replace the coolant with the pre-mixed stuff that racers sometimes use, as this will conduct more heat to the radiator. The next easiest is to get more airflow through the radiator, by moving obstructions or by fitting bigger fans. Bear in mind that the airways to a radiator are sometimes designed with constrictions to speed up the airflow, and removing these might make things worse. The third option is to fit a high capacity radiator. Once the cooling system is capable of dissipating the heat under all conditions, you can start fiddling with thermostats and otter switches to RESTRICT the operation of the system.
Peter Beech

Internet Mailing List Oct 96
I am getting paranoid about how hot my S3 apparently gets in traffic before the cooling fan cuts in. It doesn't get above the white line and never into the red, but I'm sure I read an article some time ago saying the Ford V6's prefer running cool. Also the engine feels a bit "woofly" when it is this hot. Does anyone know if there is a different fan switch that I can use? Reading the Chimaera/Griffith notes that Peter Beech produced it is apparently called an Otter switch and alternatives are available for the V8 engine - is it the same switch on the S3?
Mike Jennings

Internet Mailing List Apr 97
Is the otter switch next to the weasel filter or the stoat flange?
Andrew Cliffe

Internet Mailing List Oct 96
It sounds that your S3 is running fine if it stays at the 90 degree temperature level and does not move into the red. Both the S1 and S2, I had stayed at 90 degrees religiously with the fan coming in and out every few minutes.

I fitted an override switch to my S1 as an insurance policy and found that switching the fan in earlier did not make that much difference. The reason was that the temperature that the dial measures is that of the water circulating in the block and not the radiator. The thermostat essentially controls this temperature and only allows water through the radiator at about 85 degrees (this is off the top of my head). Having the fan switch in earlier will reduce the water temperature in the rad but not necessarily in the engine block. The cooler rad will help when the thermostat opens but the benefit may not be worth it. If you want to run your engine cooler, you will need to change the switch and the engine thermostat as well.

The best solution is to use one of the variable thermocouple type stats. Lotus used them on the Elan and I got one several years ago for about £25. These allow you to change the on off temperature and are pretty reliable. The cheapest option is to put an override switch and light inside the car so that you can see when the fan has switched in and switch it on at any point. By using piggyback connectors, the original wiring can be left as it is. The Sprint article I wrote gave the details. You can then play at being a human thermostat and work out if you need to change the stat.

By the way, you don't have to change the thermal switch. You could wire a second in parallel - the first one to reach the opening temp will start the fan. If this then fails the second one will kick in at a higher temperature. And if you have the switch as well then you can override everything. Just because I'm not paranoid doesn't mean they aren't out to get me . . .
Steve Heath

Internet Mailing List Oct 96
Thanks for your thorough and very helpful response! I agree that my cooling system seems to be working to spec., it's just that horrible feeling in the back of your mind when you glance down and see the temperature gauge is higher than you thought. I fitted a supplementary fan switch to my Caterham as this used to run slightly hot (Caterham subsequently doubled the depth of the rad, so I think they were admitting that it was marginal). I also found that a shroud between the radiator and the nosecone (rather than big gaps) directed the flow of air through rather than around the rad. I guess the above is a bit of a divergence from the TVR, but perhaps it explains my concerns - I would rather have a "safety margin". Also, are there lower temp thermostats available, and who from?
Mike Jennings

Internet Mailing List Oct 96
As for changing the stat, a good motor factor should have a selection . . . but I am really not sure that it is worth considering to give you the safety margin . . . The reason is that the thermostat will already be fully open and therefore the running temperature will be set by the thermal equilibrium - as Peter Beech said in his contribution - and will continue to be 90 degrees. The only real way of gaining some benefit would be to increase the efficiency of the radiator by increasing its size, cleaning it and pushing more air through it - i.e. fit a second and/or bigger fan. Kenlowe do make bigger and faster fans by the way.

The most important aspect is to ensure that the whole system is in A1 condition and fit the override switch and fan light to give the safety margin and the feedback that the fan is actually working. If the temp starts to go up over 90 degrees and the fan has not come on, then you have a problem and a few minutes to do something about it . . .By the way, the V8S runs cooler at about 70-80 degrees but still goes up to 90 degrees when stationary in traffic despite having bigger air intakes in the bonnet, a bigger radiator, a bigger fan and pints more coolant
Steve Heath

top

Griffs in Hot Countries

Oct 96
This is an exchange between Kenny Heng in Singapore and Ken Dobson in Cyprus about their 1994 Griff 500s. Ken Dobson's comments are in italics.

Glad to hear from you and its nice to be able to share information with someone with the identical car (at least as far as manufacturer and model are concerned!). I'm also copying this to Peter as it does contain information that will be helpful to update the Workshop Notes. My experiences here are based on two cars in Singapore both 1994 Griff 500s and therefore pre-serpentine, and also some from TVR Malaysia's (TVRM) development of the Chimaera, including a Chim 500HC. You should know that the owners of all 3 cars are the best of friends (still!) and we therefore share a lot of information.

Cyprus shares a similarly hot climate to Singapore i.e. up to 40 deg.C in summer & down to 10 deg.C in winter (except this year; actually down to 0 for first time in about 40 years).

We officially have 28 deg C all year round but the fluctuations within a day can be from 16 to 38 deg C.

My main ongoing problem is overheating though I should explain that it rarely expels water from the relief valve but there is clearly too much heat to be healthy for the car as the temp. gauges (yes, I fitted a second opinion one which actually goes crazy after 100 deg. C) go from 90 deg. in traffic to just in to the red with the A/C off and past the red 110+ with the A/C. on. Obviously the time when I need A/C is the time I cannot run it! I have tested the temperature accurately with a digital probe and the TVR gauge is only a few degrees lower.

It is a good idea to fit that second gauge. Its interesting to note that our cars behaved exactly like yours when we first got them. The first thing TVRM did was to monitor the temperature using a thermocouple and we found surprisingly that the gauge was accurate. That is except when the a/c fan and/or the lights are switched on, when it reads a good 5 deg more (we joked about this being a safety feature). There is an inherent electrical wiring fault in the earlier Griff 500s like ours which causes this which has since been fixed.

Are you able to run your A/C. without overheating problems and if so, what year & month is your car and do you have the one large belt driving the water pump and the A/C.? Mine has separate belts which according TVR's design dept. makes a difference).

Both cars we have here are one chassis number apart and the date of manufacture is circa July 1994. Separate belts run the a/c and the water pump. I have eventually got my car to run cooler with a/c on, which is 85 deg C on the motorway, a shade below 90 deg in slow traffic, and 100 deg in bumper-to-bumper crawling. Without the a/c, I run at 75 deg, 85 deg, and 90 deg respectively. Amazingly, without a/c, I can actually see the temperature being brought down from about 80 deg (when the fans come on) to 75 degs! In order of what I can deduce made the most difference (bear in mind this is difficult as some were done separately, and some at the same time) here is what I did:

1. Coolant mixture and additive: have a mixture of 80% water and 20% glycol, and add a bottle of Red Line's WaterWetter. This is amazing stuff that works and its explained at http://www.redlineoil.com/wwti.htm . Actually, if you read the theory of it, there is no need for glycol at all. However, since it raises the boiling point and lowers the freezing point of the coolant, it's worth the 20% for a safety margin.

2. Coolant circuit: goes without saying but cannot be emphasized enough. Check hoses, radiator, caps, gaskets (oil in coolant?), water pump, otter switch etc to ensure that the system does not leak and is holding the pressure. I had a loose hose, a damaged O-ring seal on the aerator tank cap, and a defective header tank cap on separate occasions and the resultant loss in pressure each time meant that the boiling point of coolant was lower and hence loss of heat transfer properties. On the last occasion, a hairline crack on the header tank was the culprit and bravo to TVRM who found it (may have been there since day one). Air pockets do develop and so bleeding the system occasionally is important and essential when the system is flushed and refilled. I have a simple way of doing this but with caution please (insulate hands with a cloth). After a drive and with the engine off for about half-an-hour (still warm but not hot), unscrew the header tank cap slowly to release pressure which you can hear. This may be intimidating to do but so far I have yet to see any coolant come up through the top. Just a lot of noise and the odd sprinkle as is bubbles. Leave it overnight without tightening the cap. The next morning you will be surprised to find a lot of the water sucked back into the aerator tank. Fill up the header tank with water and tighten the cap. You may need to repeat this for as many as 3 more times. I am satisfied when I don't hear any bubbling and the level does not drop too much the next morning. It means that there are no more air pockets in the system.

3. Installed Cerbera fans (2 of them) which are of higher rating and of a different blade design, and they sure move some air!. They fit very well into the existing shrouds (which are also essential and presume you have them - earlier Griff's did not). The reason this does not rank higher on this list is that when it was installed, it did not stop the car overheating. It just took longer to get there. This was what led us to conclude it was something more fundamental. In fact the other cars have not bothered and they are running fine too. What is important is to make sure that the fans are running well. This has necessitated re-wiring of the fans' electrical circuits and utilising individual fuses instead of the original one. I believe this is a standard factory/dealer mod to solve cars which blow their fan fuses repeatedly,and therefore fans stop working and engine overheats. This has happenned to almost all cars here and therefore will have this mod as standard. Strange thing is that they can't manufacture it this way because it is not an approved manufacturing process by TVR Enginering!

4. Oil cooler: I have a 16 row element fitted. This did not actually lower the running temperature but it was thought to provide good insurance for engine longevity by keeping the cylinder head temperatures cooler. The Rover V8 was not exactly well designed for cooling in this area and more so with the TVR Power modifications. So, this is more to supplement the cooling rather than for cooling oil itself to keep it stable(not needed in this day and age of super-synthetics). The engine did feel heathier with this.

5. Air-flow: some improvements were obtained by removing the front number plate (and having painted ones at the nose). However, this was only helpful with car on the move which is not when we have this problem. A chin spoiler, located under the car just in front of the axis of the front wheels, helped likewise. I have slots drilled at the back edge of the bonnet to allow fow of air out of the engine bay. None of these can be said to reduce running temperatures but helped bring it down when on the move. I have since replaced the number plates at the front because of the legal requirement and also the painted one (actually decal stick ons) spoilt the cars flowing lines.

Since I got this sorted out, I have been confident enough to drive it daily to work and that I could arrive in a decent state with the a/c working. By the way, it happens that way. Solve the engine cooling problem and the a/c starts to work too. Reason is that R134a gas (a/c coolant) boils at 90 deg. If your radiator ever gets beyond that, R134a in the adjacent condenser is rendered useless for cooling.

The other Griff runs equally well now, with all of above except 3. For reference, the Chim 500HC runs at similar temperature too, but does take longer in traffic to get from 90 to 100 deg. This can be down to the fact that it has its radiator and condenser in the conventional vertical position, and that it is a serpentine. The Chim 400 runs at 90 deg or below, regardless, with a/c on.

I would be anxious to hear of anything you have tried. I would still like to see improvements as it still does run hot. Not likely to cause defection to a Suzuki Cappuccino, but nonetheless with too little margin. The Chim 400 just about has it right at 90 deg and I understand many normal cars like Mercedes Benz and BMW design their cooling systems to hold the engine at 90 deg.

top

Manual Over-Ride for Fans

Sep 96
The standard fan otter switch (at the bottom of the swirl tank) seems to cut the fans in when the ECU is reporting an engine temperature (taken from a sensor in the block) of 100C. At the same time (on the car I was watching) the dashboard gauge was showing only 90C. I am told that the dashboard gauge is not only inaccurate but also slow to react, and engine temperature may have reached critical levels before it goes into the red. Advice is to stop the car if the fans have been running continuously during normal driving, and let it cool down.

Otter switches are generally unreliable and if they pack up the fans won't come on at all. A get-you-home fix is to unplug the connector and join the two spade terminals together with a short piece of wire. This fools the fan relay into thinking that the otter switch is on, and the fans will run continuously. It might be worth doing this for a whole journey if the weather is exceptionally hot. V8 MGB owners routinely run longer versions of this wire to a manual over-ride switch on the dashboard and flick it on at the first sign of a rise in temperature. The reasoning behind this is that a marginally effective cooling system can often stop the temperature going up any further but may not be able to actually bring it down. The driver can react to the rate of change in temperature (and predict overheating) whereas a 'stat can only react to what has already happened.

top

Hosted by www.Geocities.ws

1