I would like to make one point here - Scimitars did not overheat when they left the factory, and there were no mass recalls of Scimitars because of overheating problems. This suggests to me that the problems that these cars subsequently developed are not due to any inherent design fault, but to a deterioration in various parts of the engine and cooling system.
How hot should the Scimitar run ?
The temperature gauge reads high
Why isn't the fan running enough ?
Fitting BX fans and needing high-output alternators
This is probably going to be the largest topic in the FAQ
Firstly, we'll try to separate "really is overheating" from "seems to be overheating"
Your car really is overheating if...
It frequently dumps all the water out of the system from the pressure cap
or...
The gauge is showing well over 95 degrees all the time, the cooling system is pressured, and the fan is running constantly while under the control of the otter switch
You will notice that I haven't included anything about your car loosing water here. If it isn't blowing it violently or gradually out of the pressure cap then it's leaking from somewhere, or has a blown head-gasket, each of which are problems in their own right.
Most of the Scimitar heating problems come down to the cross-flow radiator, and getting the airflow through it. The radiator cores are horizontal, carrying hot water from the left-hand side to the right-hand side. When sludge begins to circulate in the system, it will start to collect in these horizontal cores, silting from the bottom of the core upwards. Eventually the silt will block up some cores completely. When this has happened, no amount of De-scaling compound is going to unblock the core when the radiator is mounted in the car. Hence the advice to remove the radiator from the car for reverse-flushing at least once a year.
Another unfortunate consequence of the cross-flow and the tendency to silt up, is that owners who do try to De-scale the car with a proprietary compound may actually make things worse, since the De-scaler will certainly loosen any sludge accumulation in the engine, but then this will collect in the radiator and possibly contribute to the further silting up of partially blocked cores.
I have experimented over the past few years with various concoctions (citric acid, Harpic loo cleaner with lime-scale remover, vinegar) added to the radiator with it removed from the car and turned through 90 degrees so that scale can hopefully drop from the cores under the action of gravity. I have to admit that I have had very little success. Once the radiator has scaled up it is time to fit a replacement. Also, once a certain amount of the finning has corroded and crumbled away the radiator efficiency will drop.
I believe that the only long-term strategy with any chance of success is to start with a recored radiator and the block flushed clean, and maintain antifreeze and/or corrosion inhibitor in the system at all times. I also think that the car will silt up less if it is used regularly.
See another thread How hot should the Scimitar run ? if your car "seems to be overheating", and we'll talk about ways of testing gauges and switches.
Back to the top of the cooling problems page
This is one of the hardest areas in which to be quantative. The amount of time that the fan must run is dependent upon the outside temperature, the air speed into the radiator, the driving pattern, the efficiency of the fan motor and blades, and finally upon the condition of the radiator. The temperature of the engine that you will judge it by is dependent upon the thermostat, the driving pattern, the radiator condition, and the fan control.
In an SE5A of mine (HTX) with a new re-cored radiator, good fan motor and blade, good otter switch, and correct mixture, I get the following temperatures and fan operating patterns
Cool day, driving fast, the fan will run briefly after waiting at a roundabout then accelerating up the slip road. The temperature gauge needle will be between the 9 and 0 when the fan comes on, and drops back to just covering the first part of the 9. If I slow down below 50 for traffic I expect the fan to come on and off regularly.
Warm day, driving fast, the fan will run for perhaps 2 or 3 minutes after waiting and then accelerating up the slip road. The temperature needle will still be between the 9 and 0 when the fan comes on, and may rise a little higher, cutting the first part of the 0. It drops back to the middle of the 9 when the fan goes off. I expect it to run briefly if I overtake traffic going up hill. If I slow down below 50 I expect it to run for a few minutes with some off periods.
Idling on either a cold or warm day, I expect the fan to come on and off with almost a 50% duty cycle.
I am running Rad-Hib Xtra Kool in the system, which does seem to reduce the amount of time the fan needs to run, so allow your figures to be a bit worse than mine when comparing.
If your temperatures are lower than my quoted ones, consider that the otter switch may be faulty, and is not turning OFF when it should, or that someone in the past has taken the thermostat out or bodged it so that it is no longer closing and maintaining the correct minimum temperature.
If your temperatures are higher than mine even when driving fast, I would suspect the radiator has begun to scale up.
After some cold November mornings I can update the driving patters as follows
Driving fast in cold weather, the temperature gauge needle is one thickness below the 9 digit, and I suspect the thermostat is hardly open. Stopping at traffic lights or queues, the fan will run briefly, and when idling, the fan is running for no more than 30 seconds, with the temperature gauge needle between the 9 and the 0
I suspect that I should change to an 88 degree thermostat if the cold weather is going to be prevalent. This will be an easy job, because I use copper-slip on each side of the thermostat gasket, and can remove the housing without tearing the gasket or leaving half of it stuck to each surface.
Back to the top of the cooling problems page
The gauge is a hot-wire device that measures current flowing through the temperature sender (a germanium crystal) in the inlet manifold. The current is controlled by the voltage regulator, a bi-metallic strip device which makes and breaks to try and give an average supply of about 10 volts to the temperature gauge and fuel gauge.
If the temperature gauge reads high, but the fan is only coming on and off intermittently, then the sender, the gauge, or the voltage regulator could be at fault.
There is a simple test is to see if the car is running hot. You will need a small thermometer of the type used to test joints of meat whilst roasting. Start the car and let it idle until the fan cuts in and out. Push the probe tip carefully under the edge of the rubber hose where it is clipped on to the thermostat housing, and watch the temperature as the fan cuts in and out. I estimate that there is an 8 degree C drop through the aluminium, so if you are seeing readings of between 84 and 90 degrees then the engine water temperature is normal, despite what the gauge says.
If the voltage stabilizer is faulty then the fuel gauge reading will also be innacurate.
Since the temperature gauge and fuel gauge are identical instruments, you can try swapping the sensor connections between them, and seeing if the fuel gauge reads half full when connected to the temperature sender. If you have an analogue voltmeter (the car one is too slow, it's another hot-wire ammeter ), you can look at the switching action of the voltage regulator as seen on the "I" terminal with respect to ground. It should be moving about between 10.5 and 8 volts. If you increase engine revs it should start to switch more frequently.
Of the three suspected items here, the temperature sender and voltage stabilizer are cheap and easy to replace, so start with the one that you feel is most likely to be faulty.
Back to the top of the cooling problems page
Above 40 MPH, the fan is rarely required, unless the radiator is heavily scaled, or you have just overtaken someone or powered up a long hill. At slow speeds, or when stationary with the engine idling, the fan will cut in and out under the control of the otter switch.
If the car overheats when idling, but has a normal temperature when driving, the fan is most likely at fault. Check that
- the fan will run when powered (pull the terminals off the otter switch and join together, keeping them away from the earth points. The fan should run )
- the fan motor is running fast enough (it should be audible above the engine noise). How often have you lubricated the fan motor spindle ? Is there a dead hedgehog being dragged around in the blades ?
- the fan and motor are running in the correct direction, (pushing cold air through the radiator into the engine compartment). It is a common fault to wire them up the wrong way round.
- enough air is being forced through the radiator (there should be a palpable air flow out from the radiator cowling inside the engine bay ). Is the radiator blocked with dead flies in the gill fins ?
If all of the above checks pass, it is probable that the otter switch is at fault, and is not cutting in early enough, or is switching off too soon. Try a new one. There has been a craze for going to scrap-yards to get otter switches out of Allegros. Sadly, although these look like the Scimitar switch, they are situated in a different part of the radiator, and switch at a higher temperature.
Back to the top of the cooling problems page
Modern cars run hotter than the Scimitars, and have a different on/off characteristic fan cycle, so don't try and judge if your Scimitar is behaving correctly by comparing it to a Peugeot, for example. The fan should come on when the water at the lower portion of the radiator exceeds a preset temperature, indicating that the radiator and any airflow due to car motion is not able to dissipate the heat that has come in to it from the engine outlet. The fan control is carried out by the "Otter switch"
The otter switch, mounted on the lower right side of the radiator in GTE's, (but the lower left in SE6B and GTC's), should cut in and out when the water in the bottom of the radiator has risen to about 90 deg C.
If it goes faulty and decides not to cut in until 95 deg C, you will notice the temperature rise on warm days when climbing hills, or in slow-moving traffic, or when idling. Replacing the switch with a new one might cure the problem.
Also, there has been a craze of raiding scrap-yards to get the similar-looking otter switch out of several British Leyland cars. These switches are in the top part of the radiator, and are designed to operate at a higher temperature.
Most Scimitars you come across have had a switch wired in to manually bring the fan on and over-ride the otter switch. Reliant didn't think it necessary, but I do in this case think it is a sensible addition. If you want to put such a switch in, the best way is to mount a toggle switch somewhere in the dashboard, and run a pair of wires to it to the otter switch itself, splicing them into the wires close to the terminals. If you have the modern otter switch with spade terminals it makes sense to crimp the switch wires and loom wires into new insulated female terminals.
Back to the top of the cooling problems page
Back to the top of the cooling problems page
I had a few enquiries last year from people wanting to fit an up-rated alternator to their SE6's. Why did they need an up-rated alternator, I asked ?
"Because the BX fan draws a lot of current when the engine is idling and the alternator can't keep up with it"
Why had they fitted a BX fan, I asked ?
"To stop the car from overheating"
I have a BX-fan in my green SE5, which I fitted a few years ago when I put in a high-power John Wade engine. I assumed that the extra power would need a lot more cooling effort (wrongly, as John himself will tell you). And yes, at idle the fan is gulping current.
However, my other current road car benefited greatly from my fitting a recored radiator, so in a spare moment last week I fitted a recored rad to the SE5 and had a good long run down to Cornwall and back. What a transformation ! About 5 degrees knocked off the temperature when cruising "at the speed of the prevailing traffic", and the fan runs briefly at slower speeds after climbing hills or stopping at lights, etc. In fact, I shall probably remove the BX fan and put back in the original one.
Cost of an up-rated alternator - anything up to £90.00 depending on where you go
Cost of a BX fan from a scrap-yard - £25 upwards
Cost of a recored radiator - £105
And there's a hidden cost with the up-rated alternator - extra fuel, you don't get something for nothing.
Sorry if I'm labouring this point about radiators a bit, but it really is the simplest way to cool your Scimmy down.
Back to the top of the cooling problems page
I have had a few conversations on the phone about this which have made me realize that not everyone has realized the significance of the Don Pither modified cooling system for the SE6.
Don proposed, and marketed parts for, two changes to the SE6 setup. The first was adding a header tank to the radiator, the second was opening up the blanked off water outlet at the back of the inlet manifold.
Various people have rung me to talk about adding an expansion tank to the SE6, and have either suggested that it be plumbed into the lower water pipe in the manner that a central-heating header tank is, or that it should be plumbed into the top water pipe or thermostat steam pipe.
Both of these ideas are incorrect. The tank is not needed to provide a head of water for the pump, there is half a radiator full of water to do that. The tank is designed to give steam and trapped air somewhere to settle out without forming pockets either in the radiator or the engine, with consequent overheating. Steam and trapped air leave the thermostat at quite a speed when the engine is revving, too fast for the bubbles to make their way into the tiny steam pipe. The bubbles get carried down into the radiator by the velocity of the water, and tend to start collecting at the top of the radiator. When the engine speed drops, some of these bubbles might make their way back up the top pipe against the reduced water flow, and exit through the steam pipe. However, at full pump speed, the bubbles will tend to hang around in the top of the radiator, rather than drop with the cooling water and go round the pump again. The way to get these bubbles out of the cooling system is by having an expansion tank outlet on the top of the radiator with a sufficiently large enough pipe to encourage them to rise up out of the main cooling circuit.
And there's an answer to the thing that's been puzzling me for years about Don's second modification - opening up the rear water outlet in the inlet manifold. I can understand that with the heater valve open, trapped air and steam will indeed be pulled out of the manifold and back to the pump, but the SE5A heater valve shuts off the connection to the pump completely when it is closed, so how come the modification still works when the valve is closed?
Well, it's quite simple. Even with the valve closed, there is a loop of 5/8" bore pipe rising up from the inlet manifold to allow bubbles to rise up and collect out of the main cooling flow. Of course, this short length of pipe can't accommodate a large volume of trapped air, but if your cooling system contains this much air then there's something else fundamentally wrong that no amount of extra pipes, radiators, or BX fans are going to be able to cope with. However, it is enough to take out the odd bubble or two. If it's a steam bubble it should reduce in volume as it cools. And of course, you know that when you refill the cooling system you should have the heater valve fully open for a while to get any trapped air out, don't you ?
As an experiment to test this last idea, I shut off the heater in my tatty red SE5A when I towed the trailer to and from Drayton Manor, and saw no change at all in the running temperature or the fan pattern. (And my feet felt a lot better for it too)
In case you're wondering how I learned so much about about the trapped steam and header tank, I first ran my Sabre Six with a different radiator, and had to get an expansion tank onto the system to accommodate the change in volume between cold and hot. I found that a small bore pipe setup didn't work properly, the water went into the tank but very little air did, and plumbing a tank with a larger pipe into the bottom hose didn't work either. Finally, I decided to get the original radiator re-cored, and found that it took 20 degrees off the engine temperature. I originally couldn't believe that such an agricultural-looking lump of brass would be able to work properly, and tried a modern aluminium-cored VW radiator. Now of course I know better
Yes I know, I keep harping on about this, but the simplest and cheapest way to stop your Scimmy loosing its cool is to make sure the radiator is up to spec. Just up to spec, no need for super cores, BX fans, second radiators alongside the first, holes in the body to force more air in. All just a waste of time. Keep it clean, and keep it simple
Back to the top of the cooling problems page
You will find in other posts here I have documented my experiences running an SE5A with a scaled-up radiator for a few months, and then fitting a reconditioned one. It is easy to look at an old radiator that has lost much of the finning and has green weep marks from pinhole leaks, and decide that it is no good. It is not so easy to decide if an apparently good radiator is in fact scaled up and useless. One test is to remove the otter switch and look at the few cores you can see to assess how much scale has built up.
This is harder with an SE5 radiator, which does not have the otter switch hole. I have recently had to assess such a radiator in my SE5 "wreck to runner" project. Superficially the radiator looks sound, with no external corrosion. There was some loose scale inside which I flushed out.
I decided to carry out an experiment to see if I could measure the difference in flow rates between the old radiator and a reconditioned one from stock. Each radiator was tested in turn by pouring a measured 7 litres of water into a funnel in a piece of rubber hose attached to the radiator top inlet pipe, starting a stopwatch as the water was poured in, and stopping the watch when the water from the bottom outlet ceased to flow. The water was poured from a watering can so that the spout acted as a flow control to prevent variation in inlet pressure.
The figures, in seconds and hundredths, are
| New | Old |
| 26.97 | 31.91 |
| 27.06 | 31.07 |
| 26.90 | 31.50 |
| 26.35 | 30.69 |
| 25.57 | 30.78 |
| 25.84 | 31.96 |
Average 26.45 31.32
The average increase in time between the new and old radiator is of the order of 18%
As to how this radiator would behave in a car compared to the new one, I cannot say. However, with the loss of nearly 20% throughput, I would expect this to translate to a lot more fan running time in an SE5A, and possibly some increase in temperature. I will use the old radiator just to flush out the remaining sludge in the engine, and fit a new one when the car takes to the road.
I hope that this will dispell some of the myth that surrounds Scimitars regarding overheating. It would seem that cross-flow radiators scale up much faster than the vertical-cored radiators in other cars, and that the Scimitar will start to overheat when a certain amount of cooling capacity has been lost. You can of course continue to run the car in this state, by fitting a much more powerful cooling fan, but you then will find that the alternator will not supply enough current at idle to keep the fan happy, and so you then need to up-rate the alternator ...
It is probably cheaper to fit a recored radiator
Back to the top of the cooling problems page