Thought about individual wbo2 per cylinder

[PierreL]

Hello Everybody,
I have a question/change proposal that maybe could be helpfull for everyone that use individual wbo2 per cylinder.

I think about this because i have a problem to get a more wbo2 reliable information on a single cylinder engine. All my tests showvery fluctuating AFR values

I'm considering if the wbo2 data should not be analysed only once per engine cycle.
I mean: if i'm right, the megasquirt gets a continuous voltage from the wbo2 sensor. Whatever if gases are flowing trough the exhaust, or not.
Wouldn't it be better if the megasquirt only captured that o2 data only at a particular moment, when the exhaust valve is open ? Rather than getting the data in an unsynchronised way .
What do you think ?
Many thanks,
Pierre

[jsmcortina]

There are other factors to consider:
- how quickly does the wideband sensor respond and what is the lag to the controller providing that?
- does the exhaust gas change composition as it sits?

In your case it sounds like reversion may be a bigger issue.

James

[PierreL]

There are other factors to consider:
- how quickly does the wideband sensor respond and what is the lag to the controller providing that?
- does the exhaust gas change composition as it sits?

In your case it sounds like reversion may be a bigger issue.

James

Well James, i think that i should record how the lambda data is (eventually) changing during cycles to answer your questions.
I will try to record it on an oscilloscope.
Or maybe is there a particular logger in my (regisred) tunerstudio that could help me to do a high frequency record of the AFR ?

Thank you for your advise,
Pierre

[jsmcortina]

The high speed loggers can't presently record AFR, so using the scope is probably best.

James

[ol boy]

I'd like to hook up a MAF to a single and scope the output in relation to the map reading and see if the measured MAF corresponds to the map dip. Watching the wb o2 and the other 2 inputs could give a good idea as to when to sample the wb o2. First we need scope shots.

[PierreL]

L1030944bis.jpg

Hi Ryan, James,
I had the chance to test the Lambda output (Innovate LC1) on a scope. (see picture here attached)
The engine was idling at +- 2000 rpm.
You can see on the screen shot the trace of the lambda sensor and the signal of the 36-1 wheel.
It covers a full engine cycle, so you can see the missing tooth in the column 2 and 9 of the grid.
The lambda trace is the (slightly jagged if you zoom in) yellow line at the bottom of the 3rd row of the grid, starting from the top.
The average voltage was 2.08v, which means on a LC1 +- 13.6 AFR.
The difference between the higer and the lower values of the yellow trace is around 0,4v, which is around 1.2 AFR.
So the AFR was changing between 13 and 14.2 during an engine cycle.
The goal of the test was to observe the changes in the AFR curve during a cycle, and try to find if i could find any relation between the shape of the curve and the angle of the crankshaft.
From what i see, i just can conclude that there is no relationship between the value of AFR and the moment when that value was read. The frequence of the change is too high.
Do not hesitate to comment!
Have a nice weekend,
Pierre

[Xillur]

Hi Pierre,

Eagerly following your trials

Can your scope trigger on pulse width? if yes, then I would try to trigger on the missing tooth and then set the span to show ie 2 full motor revolutions.
The VBO voltage I would ac couple for starters and zoom in voltage wise. This would easily reveal any repetitiveness in the vbo signal correlated to a full revolution.

/Kristian