Forum » General Tuning Discussion » EGT sensor placement

EGT sensor placement

General Tuning Discussion

Discuss all things tuning in this section. News, products, problems and results. 

Page 1

how should I determine the best placement for an EGT sensor in general?

My application is slightly diffrent to most on here, it will be implemented on a 125cc single cylinder 2 stroke engine, and I'm hoping that the EGT sensor placement theory is similar to a 4 stroke and that im not going to have to worry about back pressure waves from the expansion chamber intafearing with the readings. A custom pipe will be made and the perfect EGT temperature for that exhaust will be known hence why I would like to come as close as possible to that figure in real world (by altering carburation/afr)

I would imagine that heat radiation from cylinder could alter the readings as well as possibly flame front reaching the sensor if it were placed too close?

What is the application for this engine? Are we talking go kart, dirt bike, generator...? Do you expect it to run mostly a steady throttle or steady rpm most of the time or will it be constantly changing like a normal automotive application?

Can you install a wideband into the exhaust pipe or use a 5 gas emission analyzer if you have access to it? An EGT is going to give some relative change as it gets richer and leaner but you won't get much insight into how it's burning beyond that.

@Arghx7 This engine is mainly used as a development engine in a motorbike to see how much the design can be improved, how much extra horsepower can be gained on pump petrol and to then see how much power can be made on alternative fuels such as ethanol. It will also go out on track but that's not it's main pupose. I was told that Lambada sensor was a no go in a 2 stroke due oil intafearing with Lambada readings, as well as unburnt oil clogging the sensor up. I'm responsible for the electronics/data logging side of things.

Achieving the correct EGT is in the real world is important to me, and I will know the target temp. The first reason is so that I know if the AFR is optimal (higher than target=lean, lower=rich) (but from what you're saying it seems that egt sensor will not do a great job at that). As for the second reason, the exhaust will work it's best at its target EGT, as temperature affects the speed of preassure waves, which is important in a 2 stroke (due to lack of exhaust valve) so that the preassures wave going back to the cylinder "blocks" the exhaust port so that unburnt mixture is not lost, improving the efficiency. It also alters how the power is delivered.

For the purpose of testing on the Dyno (limited time keep in mind) I am guessing that there will be a lot of steady throttle runs, but on the track it will be constantly changing. The exhaust is made for top of rpm range in mind.

The egt sensor I was planned to use is a mainline EGT sensor with cold junction calibrated guage.

I have not considered a gas analyzer so far. Would I run into the same issues that I would with lambada? Also I'm not sure how much the things I was told about lambada use in 2 stroke are true or not. Maybe there's a way to compensate for oil in the mixture and get correct readings.

In 4 stroke applications/cars, how far away from the head is the sensor installed and what are the reasons for that distance?

Attached Files

You can put the thermocouple as close as you can access. The temperature range on that sensor you listed is way hotter than your engine is going to get. For example, a modern turbo direct injection engine will run around 1050 C before the turbo. If you have a target temperature it's going to be more useful.

A gas analyzer is absolutely going to give you the useful readings, but they are not going to measure quickly in fast transients. The CO concentration tells you rich or lean, the CO2 and O2 concentration tells you combnation efficiency, and HC tells you misfire. For example, lean it out and your CO goes lower and lower. High CO and high HC is rich misfire, low CO and high HC is lean misfire. CO2 concentration tracks similarly to air fuel ratio and can vary with ethanol. High O2 or high CO in near stoichiometric conditions can imply poor burning or mixing.

From a cost perspective it's better to rent than buy a gas analyzer.

I have a colleague who used to be a test engineer develop small engines for Keihin (Honda), although I am unsure of the details of his role. I'll see if he has any insight about widebands or instrumentation.

Talked to my colleague; his experience was with 4 stroke small engines only. He used wideband o2 and a gas analyzer (CO %). For wideband o2, well I would install one anyway. There are tons of rotary engines running around with 2 stroke oil and high oil consumption running widebands. Sure, it will degrade the sensor faster, but I don't think it renders it useless.

So I would say, gas analyzer if you can, with CO% being most important, and wideband o2 with a fresh sensor.

Good point about rotaries. I will do some research on the use of wideband sensors in rotary engines

I've tuned plenty of rotaries, we run widebands all the time. Rx-8 also has a factory wideband (Denso sensor, which works a little differently from the Bosch ones used in aftermarket).

Intresting. So I assume they last longer then just a month?

I I'll take a look at the stock lambada rotary sensors

Hi Mateusz,

I have been involved in a two stroke project for quite a while which runs LSU 4.9 lambda and EGTs. The lambda sensors have not been an issue apart from the first set which failed in first ~10 hours of running, but the tuning was extremely bad. Since then the bike has about 3500 km of road riding with the same sensors and I am confident they are still reading correctly. It is a post mix oil system running around 40:1.

I was sceptical of lambda sensors having much use on a two stroke, but have found them useful in tuning.

Hi RSlogger , have you considered under plug temp sensor? , we used them when i was racing 2 strokes a while back , we used this as a safeguard against seizure , we also had a guy working with us that used an Australian engine simulation program....,

he designed the pipes for our engines and i made them , really great results as well . Are you running carbs or FI ? if FI then you will need as already said gas analyser and or wideband.

Keep us posted with progress i love 2 strokes..