Wideband Accuracy
Turbo Mini Forums

I have recorded EGT on a NA 1275 and have seen the centre exhaust port running over 100 Deg C hotter than the outer with equal AFRs. The wideband sensor was about 0.5m from the head, EGTs were at the head flange.

My simulations have also throw up similar numbers to that measured. It's mainly due to the way that thermocouples can only pick up cycle average temperatures and the centre port thermocouple gets a double dose, although there is more to it than that.

SC got similar data with the weber on the NA 1275/1380 before fitting their injection, but their interpretation of the results was different.

What we are concerned with now is the potential for the wideband sampling exhaust gas from the head flange is measuring the composition of gas that has not reached chemical equilibrium following combustion. We believe, from Rods post turbo AFR comparisons, that it is skewing both inner and outer cylinder readings richer. Whether the port temperature difference is making the centre port read even richer than the outers when the pre-ignition AFRs are the same is a further complication.

Edited by Paul S on 28th Oct, 2014.

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 had the hardware and software to separately adjust inner and outer AFRs based on the wideband data, how would you confirm the result? If the average of the AFRs didn't match a sensor further downstream, how could you prove the data is accurate? All I can think of is checking the spark plug colours.

What was SC conclusion to their EGT data?

I still want to test whether widebands in the runners will let us compare inners and outer AFRs even if they are offset from the actual AFRs as this has to be useful data.

Schrödinger's cat - so which one am I ???

how about 2 small turbo's ...one fed by the centre ,the other fed by the outers,then sensors after the turbo's.... 1241's or 1544's maybe?

alternatively ,put a series of bungs down the header tube on a na engine ,and compare the measurements as you move the sensor away from the head face .

Edited by robert on 29th Oct, 2014.

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

I'm going for 5 sample points, 4mm ID sample tubes and 3 widebands.

2 sample points at the head flange as usual.

With the remote turbo manifold, I can get 2 sample points about 450mm from the head flange but before the turbo. I could place them at 550mm but there would be danger of contamination from the other branch.

Then one post turbo.

Once set up on the downstream pre-turbo sensors, I can move the sample points to get three readings in-line on both inners and outers.

Luckily I'm at that stage of the build. The engine is about to go in and the next job was to finish the manifold/downpipe. Be next summer before it runs in anger though.

Edited by Paul S on 29th Oct, 2014.

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."

That sounds like a very good test which unfortunately I can't do for a comparison whilst still using the Metro manifold.

Depending on the timing next year I could probably lend you an additional two LSUs and, with a slight modification to a spare PCB to add additional addresses, a controller that would read all 5 together and their working temperatures as well.

EDIT - that's assuming you still have an IOx - end edit

In the meantime I'm going to investigate sample tube bore to see if slowing the gas down allows more time for equilibrium to be reached. Later this week if I can find my bag of injector "O" rings (one failed last week)

Edited by Rod S on 29th Oct, 2014.

Schrödinger's cat - so which one am I ???

I dug up this thread on afr vs egt I remembered reading a long time ago. uou have to ignore a few flat earther remarks . and just read the posts by David Redzus / redngold 86z/ brianp as they do actual work on this. i found it interesting and maybe it is of some use

http://speedtalk.com/forum/viewtopic.php?f...rt=15&hilit=egt

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/

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."

From my experience, if there is only partial combustion, then the wideband sensor will read leaner than if the mixture was fully combusted, this is for rich mixtures.

For lean, combustion happens very very rapidly so there should not be a reason for further combustion to take place outside of the cylinder.