Eyeballing the temperatures

Monitoring of the temperatures on an engine can provide some useful information on whether it is operating in the most desirable range to achieve a certain parameter, be that for longevity, best fuel consumption, or best power.  On water cooled engines observation of the water temperature (referred to as WT) is commonly used.  For non- water cooled engines the cylinder head temperature (CHT) is useful, and the exhaust gas temperature (EGT) can be used on either a water- or air-cooled engine.

WT is easily measured via a sensor positioned within the water jacket on the engine.  CHT sensors are typically shaped like a ring and are normally trapped between the base of the spark plug and the top of the cylinder head so as to get a good contact i.e. they would replace the gasket of the spark plug, and are not normally used in karting.  Sensing the EGT requires the mounting of a probe into the header pipe of the exhaust close to the exhaust port.  To achieve this, one needs to weld a nipple onto the header at a distance of between 50mm and 80mm from the ball joint on the header, and the sensor’s probe would then be screwed into this nipple.

The Kid Rok, Mini Rok, and OKJ classes have adopted the principle of keeping things as simple as possible and no temperature sensors are allowed at all, but on DVS one is obligated to use a WT sensor when competing at Regional or National levels.  For the Rotax series one is allowed to use up to two sensors, and they are the WT and the EGT as shown in the picture above.  If used correctly, the readings one gets can provide some useful information on what is happening inside the engine, so it really boils down to a question of why not use all of the tools at your disposal.

Types of sensors

The construction of the WT and EGT sensors differs somewhat and also has a lot to do with the actual value of the temperature being measured.  The WT sensor is usually a resistance type of device where a fine wire is encased in a container or ‘bulb’ which in turn is fitted into the water jacket.  As the water temperature (and hence the temperature of the bulb as well) changes, so does the electrical resistance of the fine wire inside it.  So, what the pointy heads do is to pass an electrical current through the wire, measure its resistance, and from this the temperature can be calculated – clever hey.

The EGT sensor is measuring a much higher temperature and therefore needs to be pretty robust.  This type of sensor uses two wires made of different materials, e.g. Chromel and Alumel in a type-K thermocouple, with their ends welded together and then assembled into a casing.  When the temperature around the casing changes, a very small voltage is generated by the wires which is proportional to the temperature, and when connected to a Mychron or Unipro can display a very accurate temperature value (as is the case with a WT sensor).  A type-K thermocouple is inexpensive and can measure temperatures up to about 1,100°C which is well above our requirements in karting, but most importantly, it responds rapidly to temperature changes.

As mentioned previously, a CHT sensor that fits under the spark plug is seldom used.  Generally they can provide an accurate output of constant temperatures but they tend to be a lot less robust than either the WT or EGT sensors.  On a kart where there is a lot of vibration, this can lead to early failure and this is probably one of the reasons as to why they have fallen from favour.  Because the temperature ‘wave’ first has to pass through the cylinder head material before it reaches the sensor, there is a definite lag in their response compared to an EGT sensor and any sudden peaks in the temperature readings largely tend to have been ironed out by the time it shows up on a gauge.  Lastly, as they are effectively exposed to the ambient air passing over the cylinder head then the lower the air temperature, the lower the CHT reading.

Water temperature

Most internal combustion engines run best when the WT is fairly cool, and the water cooled engines we use in karting are no different.  The reasons for this are easily understood if you have a look at the generalised graphs below.  These portray the internal frictional losses and also the mechanical efficiency of an engine at varying engine speeds.  Note that there isn’t a value on the vertical axis of the graph on the LHS as the frictional losses are a function of the engine size and its power output.  That said you can also see why Rotax, for example, recommend a WT of around 50°C because not only are the frictional losses at a minimum, but the mechanical efficiency of the engine is also at its highest.  For those running in the Micro and Mini classes, you won’t be able to run at these low temperatures (you’ll be up in the 80’s or 90’s) as a result of the size of radiator being used, but the principle still remains valid.

The water temperature can easily be controlled by judicious use of the flap or curtain that is usually fitted to the radiator.  One can also apply duct tape around the radiator core to raise the temperature if required but there is sometimes a stipulation in the rules that it should not be able to be easily peeled off during the race.

What is lambda

There is nothing better than a visual analogy to aid the understanding of certain principles, so let’s try and use one that many of the petrol heads out there might be familiar with.  An oxy-acetylene torch will burn with a wide variety of A/F ratios.  Generally when you light the torch, the mixture will be extremely rich because it has too much fuel (viz. the acetylene) for the amount of oxygen that's flowing.  The flame will be yellow, produce a lot of smoke, and relatively speaking it won’t be all that hot.  Then as you start to lean out the mixture by opening up the oxygen valve, the yellow flame and smoke start to disappear, the flame turns to a bright blue, and the flame temperature goes up pretty dramatically.  However, as you continue to increase the oxygen flow, the flame begins to shrink and the flame temperature actually goes down even though the mixture is leaner.  By continuing to open the oxygen valve you will eventually reach a situation where the flame simply goes out i.e. the mixture is too lean to support combustion.

This analogy is identical to what happens inside your engine and is shown in the diagram below.  The commercial pump fuel that we use in karting can be ignited by the spark plug and burnt in the combustion chamber over a fairly wide range of A/F ratios.  The best utilization of the chemical energy present in the fuel occurs when the A/F ratio is in the region of 14,7:1 but the best power is made when you are running a slightly richer mixture than this.  The ‘actual’ A/F ratio compared to the ‘ideal’ A/F ratio is commonly referred to as ‘lambda’ (denoted by a Greek L = λ).  Rather than saying the jetting is rich or lean, tuners often prefer to quote the lambda value.  So, as an example if you were running a rich mixture of say 13,5:1 then, λ = 13,5/14,7 = 0,92.

Power, EGT and lambda

Looking at the diagram above, one can see that there appears to be a fairly predictable relationship between the power curve and the exhaust gas temperature.  Therefore you’d also be tempted to thinking that it can be used as a tuning guide for the correct jetting and you wouldn’t be far off the mark with that thought process.  However, a word of caution.  The absolute best way to tune an engine for maximum power is on an engine dyno, and whatever EGT reading you get there is fantastic to be used as a comparative number when the engine is making maximum power at wide open throttle under the same atmospheric conditions.  Once off the dyno and strapped to the kart, a similar A/F ratio can be established at a later date by altering the jet and needle clip position to achieve the target EGT.  Bear in mind that the main objective here is to get the A/F ratio correct and not the EGT reading.

Most engines will make maximum power at an A/F ratio somewhere between 11,5 (λ = 11,5/14,7 = 0,78) and 13,0 (λ = 13,0/14,7 = 0,88) but the EGT may vary pretty dramatically dependent on a number of other factors such as the altitude, air temperature, relative humidity, etc.  Maximum power for the two-strokes we run in karting usually occurs at a lambda of about 0,86.  If the λ is much less than this, the exhaust gas will contain unburnt fuel and unused oxygen, plus conditions in the hot exhaust may produce combustion resulting in a ‘popping’ sound being heard by the driver.  Once again, the Micro and Mini classes are the odd ones out here and make their best power at lambda values approaching 1,0 or even a bit more i.e. they need to run a bit leaner than one would expect to be the case.

Note that at the point of maximum power, the EGT reading has a certain value which is below the maximum on the curve, and that is because the excess fuel present cools off the combustion gasses.  An identical reading of EGT can also be achieved on the ‘opposite’ side of the curve when you are running a lot leaner and making a lot less power in the process, so be well aware of this phenomenon.  Of course, you won’t be running a full lap at wide open throttle trying to make the best power, because unfortunately for the speed demon dragsters out there, the track designer added in a few tricky bits where you need to tap off, slam on brakes and use some driving skills so that others don’t end up overtaking you.  As a result, if you could plot the lambda readings over the whole lap, you’d end up with an average much closer to 1,0.  Just a word of caution if you are comparing your EGT with that on another engine – the temperature reading is dependent on where the probe is located both in the distance it is fitted from the exhaust port and also how far the tip of the probe is located into the stream of hot gasses, so you may well be comparing apples with oranges.

What’s the best

Remember that keeping an eye on the temperatures is just another tool in your tool-bag to assist you in understanding how to get the best out of the engine, and by implication, closer to getting your grubby paws on the silverware at the end of a day’s karting.  Altering your jetting using relative air density (discussed in another chapter) is another good tool, just as is the case with correct chassis setup, or tyre setup, etc.  Ultimately, there is only one gauge that always tells the truth, and that’s the stopwatch.

Emile McGregor - MSA Technical Consultant