Exhaust Gas Temperature
Turbo Mini Forums

I'm hoping to test the Siamesed Port code by Christmas on my NA 1275.

I've got two widebands, one for each leg of the LCB. These should provide sufficient information to develope the timing and advance of the injection pulse.

I've been thinking of how we can then apply the code to my 998 turbo.

It is not practical to fit two widebands before the turbo due to temperature, so i was thinking of fitting some thermocouples to the exhaust ports, close to the head.

We could then determine the realtionship between the AFR and EGT for an outer and centre port. This would clarify if the centre port carrying twice the flow has an impact on EGT.

The EGT could then be used to set up the timing and advance on boost.

I can monitor and log EGTs on my LM-1 with Aux Box.

Any thoughts?

Would this thermocouple be OK?

http://www.maplin.co.uk/Module.aspx?Module...e=14&doy=search

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

what about using a K type probe that is just 2wires joined at the tip, this could then be fitted between the manifold and gasket with a bit of exhaust paste. try looking at rswww.com

I mentioned this some time ago - in Carl Austin's nat-asp racer with the open pipes - the centre pipe is a lot hotter than the outers.

Also, the thermocouples have a noticable 'lag' in their remperature reporting, by virtue of being enclosed in a case - which itself conducts and retains heat.

With this in mind - it makes sense to me that even a perfectly set up engine would show the centre braches as being hotter than the outer?

I was just going to run a few psi (like 2/3) and use that to set up the fuelling strategy. This would have the effect of taking the 'efficiency' of the intake system to the 'boosted' (rather than nat-asp) region - no idea if this makes a real & significant difference, but keep the EGT's and injector duty low enough so as not to ruin the widebands and to give plenty of variability to play with the injection window and not run out of fuel.

A single K-type at the inlet to the turbo would serve as a warning for when getting close to the danger point of the wideband temp range.

And as regards the back-pressure having an effect - well - yes it will - but to how much? And if the pressure change is equal over the two, then it should still show even / un-even fuelling?


Here is an idea - why not set up the 1275 with O2 AND thermocouples? That way you'll at least have a real indication if they do track (scuppering my thoughts- LOL) or if they do deviate, by what percentage - maybe plotted against RPM?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

If you don't want the manifold itself to skew the readings, then some form of 'insuator' between the manifold and the temp sensor needs to be found I reckon... The compression fittings for the K-types I'm familar with would easilly pass a lot of heat into the sensor. There MUST be a sensor out there, maybe in a K-type frame - but where the sensor tip (and measuring part) is isolated from the body?

I think that if we get the tip of the thermocouple in the middle of the air stream we will be OK.

The sheathed type thermocouples have the measuring element in the extreme tip, so should react fairly quick if fitted correctly.

Something like this:

http://www.spectite.com/pf.htm

Edited by Paul S on 29th Oct, 2007.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

If the center is running hotter it is running richer. advancing the ignition on the center cylinders will give more complete combustion resulting in a cooler gas.

I suspect however that the center cylinders running hotter are actualy run a higher VE than the outer cylinders due to the valve over lap, not necisaraly leaner or richer, basicaly the center cylinders are able to draw in a larger charge volume of the correct AFR between cylinders. How do we get round that, Scatter Pattern cams

I use 3mm diameter, 250mm long, Type K thermocouples from RS components with a max temp of 1100c

http://rswww.com/cgi-bin/bv/rswww/searchBr...heID=uknetscape

There are stainless compression glands for these but i cant find them on the online catalogue. I chose the 3mm diameter one for a little extra mechanical strength over the 1,5mm version buts still respond reasonably quickly due to its small diameter. Type K are slow anyway.

WBO2 sensors dont like fluctuating backpressure so putting it infront of the turbo is a bad idea, calibration will be a nightmare.

I dont think heat soak on the sheath of the thermosouple will matter as long as the tip in in the gas flow. As long as it returns a value that changes with fueling, it wont matter how acurate the reading is as long as it is repeatable

Edited by Sprocket on 29th Oct, 2007.

generally lean means hot, rich means cooler .
a lean running engine can make a much higher temp than a rich one .

i think ring shaped thermocouples on the spark plugs is about the only way to go on this .

with ref to the 2and 3 being hotter , i think this is due to the heavier part of the charge biasing towards the outer cyls ,and so the inners run leaner ,due to manifold design .

another reason the inners read as higher in temp , is that if all the cyls are running the same temp ex gas ,and 1 and 4 have a certain rate of cooling the gas as it shoots down the pipe ,then that sensor will read the gas temp minus the cooling of the gas by its environment , 2 and 3 on the other hand ,will have the same environment ,but twice as much gas quantity ,and so read a higher temp .

just theory of course .

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I see where you are coming from Robert.

I was under the impression that the highest combustion temp is actualy on the rich side of Stoichometric, may have been fed duff info



one thing i have noticed is, obviously, the EGT rises as you go to full throttle, the AFR is richer yet the temp rises. The cylinders are working harder, more charge is being drawn in, cylinder pressures are higher and combustion temps rise. All i was saying is that on a siamese port onec cylinder starts drawing as the other has finished, valves are still open. is it not reasonable to suggest that the cylinders starting to draw on the port will draw some of the charge out of the other cylinder, thus reducing its VE? you would then get a temperature difference between pairs of cylinders as two are working harder than the others. It has nothing to do with AFR, just cylinder filling. Its physics rather than chemistry. Scatter cams were developed to reduce this effect, and im surpised no one is even contemplating using the Kent MD274SP for this reason!!

Isnt it only the tip of the thermocouple that actually reads temp anyway ??

I had a play with the TC-4 on my volvo td, and I thought response from the 3mm sensors I was using was fairly good.
Not, it isnt instant, but it isnt that slow either.

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

On 29th of Oct, 2007 at 07:58pm Mini Sprocket said:
I see where you are coming from Robert.

I was under the impression that the highest combustion temp is actualy on the rich side of Stoichometric, may have been fed duff info



one thing i have noticed is, obviously, the EGT rises as you go to full throttle, the AFR is richer yet the temp rises. The cylinders are working harder, more charge is being drawn in, cylinder pressures are higher and combustion temps rise. All i was saying is that on a siamese port onec cylinder starts drawing as the other has finished, valves are still open. is it not reasonable to suggest that the cylinders starting to draw on the port will draw some of the charge out of the other cylinder, thus reducing its VE? you would then get a temperature difference between pairs of cylinders as two are working harder than the others. It has nothing to do with AFR, just cylinder filling. Its physics rather than chemistry. Scatter cams were developed to reduce this effect, and im surpised no one is even contemplating using the Kent MD274SP for this reason!!

Edited by robert on 29th Oct, 2007.

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

Best lambda for torque is around 0.86, thats roughly 12.5 AFR, going richer doesnt always produce more power, what it does do is cool the combustion temps down, hence why many tuners use this method to reduce the possability of detonation, but going less then a Lambda of 0.7 you are starting to get into the visable black smoke and rich missfire area. Your wideband wont read at these levels either, it certainly wont be an acurate reading.

I think the reason lean mixtures show up with the engine overheating is due to the fact that the flame burns quicker resulting in the hot gasses being in the cylinder longer, transmiting its heat to the combustion chamber for longer, less heat is ejected into the exhaust and the cooling system picks up the excess, rather than the combustion temps being higher

Lean burn engines can run as lean as 20:1 AFR, but you will also notice they run oil cooling jets on the pistons, have aluminium heads and a mahoosive rad to get rid of the excess heat.

Heat and temperature are different things but related

??

Edited by Sprocket on 30th Oct, 2007.

our lean burn is 30:1 gas turbine though!

i've got access to loads of K-types, they typically have about a 3second delay in them but to be honest i wouldnt of thought it would matter much as you would be holding the engine in positions whilst tuning for longer than that.

Dont worry about the stem etc, its only the tip that matters.

I need to investigate to see if i can "borrow" some shorter TC's. I can get hold of 1mm TC's so could instrument the head chamber. but i've got to find the time for projects like this!!

The Maplin thermocouple is no good as it does not do the temperature range.

The RS units are good but expensive, they also do the glands. Total cost about £62.

This looks like a viable alternative:

http://www.rc-log.co.uk/rclog/index.php?ta...s&product_id=40

It's amazing what they do with models these days!

The TC-4 with cables etc is going to be about £130.

So total cost about £200 - gulp.

Now if I hadn't bought that M24 supercharger or that K100 head last month, it would be alright.

Might have to delay the 10" wheels for a bit.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

have a look at picoscope paul.

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

What does Picoscope do ? ( have a similar, but cheaper PC scope )

I thought the TC-4 was pretty good value actually !!!! I couldnt find an EGT controler for similar money, let alone one that does 4.
I paid around £80 for mine.....contact Bill Shurvinton.

Although I was led to believe it gave a 0-5v output I could use for logging. It doesnt appear to. It does have a 5v output for supplying external sensors though.

If anyone knows of an EGT controller that will output 0-5v for logging, Id be very interested in one.

Edited by stevieturbo on 31st Oct, 2007.

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

they do various data logging stuff steve , iv got the 8 channel i think at home so i will have a play with it when i get back in dec .

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

To use the TC-4 for logging, you have to connect it up using the serial port to a laptop running Logworks.

In fact the serial port is the only output on the TC-4.

No problem for me as I already have an LC-1, LM-1 and the LMA-3 Aux box.

This will allow me to feed the following parameters through to logworks:

Wideband 1
Wideband 2
EGT 1
EGT 2
RPM
MAP

Plus a lot of spare channels for 0-5v input.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I have access to FeKo sensors if they are any use to you? I know we use them up to 300 C and im sure they are capable of withstanding higher temps.

All the FeKo sensor does is send a mV signal out, the higher the temp the higher the voltage.

https://www.facebook.com/pages/Fusion-Fabri..._homepage_panel

www.fusionfabs.co.uk



1/4mile in 13.2sec @ 111 terminal on 15psi

We need something that will work up to 1000 degrees C !

Sounds like the FeKo is similar to a themocouple because they also put out a mV signal.

The expense is still in converting the sensor into something to datalog.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I've bought two thermocouples from TC-Direct.

http://www.tcdirect.co.uk/deptprod.asp

They gave me a ring to check my requirements, said it would be with me Monday and it was. Good Service.

Cost £42 including two stainless steel compression fittings.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

The Innovate TC-4 that I'm going to use to log the Exhaust Gas Teperatures has four inputs. As I'm only using two, I have two spare.

I was thinking about using these inputs to measure the actual cylinder head temperture.

I remember someone on here posting about using washer type thermocouples under the spark plugs. I've searched in vane.

These seem to be good value:

http://shop1.actinicexpress.co.uk/shops/pa...bes_and_Sensors

I'm just not sure if we are going to see a fast enough response to obtain any useful data.

Any thoughts?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

On the topic of EGT's and logging...

0-5v output from an EGT probe...

http://www.o2simulator.com/index.php?main_...&products_id=13

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Fitted the thermocouples this weekend. Not a lot but too cold and feeling iffy. Plus sturgeo decided to fit Hi-Los today and I ended up helping him, as usual.



I've also ordered some spark plug thermocouples to measure the cylinder head temp (CHT).

Hopefully we should be able to correlate EGT/CHT with AFR using the two widebands. Then I'll be using the EGT/CHT thermocouples to set up the injection timing in the siamesed code as well as a single wideband.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."