Dual Wideband Datalog
Turbo Mini Forums

Latest log attached.

Same spec.

10 degrees, 975 mBar, 68% humidity.

Even with the grounds all at the same point, similar results.

The four stages of WOT on the graph are as follows:

1 & 2 as above.
3 got it into top gear.
4 round a corner and back into third.

Only difference this time was 15 stone of Sturgeo on board.

But I forgot to top up the dashpot, not that it makes much difference.

Attachments:

22.08kb

Edited by Paul S on 2nd Dec, 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."

I think that part of the AFR difference may be related to wall wetting. Most of the fuel evaporating from the manifold wall will go to the inner cylinders because of charge stealing. At low RPM, there is more time for fuel to evaporate so the AFR difference will be bigger as shown in the log. At higher RPM, the difference becomes less because there's less wall wetting.

http://www.jbperf.com/

It could also be due to the inner cylinders getting more fuel at low rpm due to the charge stealing.

Then at high rpm, the outers get richer due to the inertia of the fuel round the bend of the manifold.

The AFRs do get closer later on in the log, maybe due to the engine getting hotter. I'll do another extract.

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."

This is the last bit of full throttle and the engine should have been well and trully cooking.

However, not much change.

There seems to be a fairly consistant offset to the readings which makes me suspect the validity of the data from the LC-1.

I'm waiting for some serial leads to hook the LC-1 up in an MTS chain so that I can log raw AFR data rather than converting it to volts and then back again. Something may be incorrect in the conversion.

I'll also get the sensors out and do a free air calibration next weekend, then run another log. As long as my knuckles have recovered!

Attachments:

13.63kb

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."

Ignore my last email then as you had worked that out :)

Intercylinder numbers certainly look believable and are close to others I have seen from an A series

Charge stealing with a carb shouldn't happen. It does happen due to either wall wetting or fuel inertia. And these 2 effects will affect the AFR difference in different directions.

Having said that, it does seem like you have some sort of offset and the only way to know for sure will be to switch things around.

http://www.jbperf.com/

could you not set the lc1 up to say put out exactly 2.5v no matter what the AFR is, and then see what logworks records through the aux-box? i.e.e see that it is actually 2.5v?


p.s. even though iv no interest in the a-series any more, its still a very interesting subject! cheers for sharing info!

edit again::
for the free air calibration could you not jsut go out to the car the morning after the night before, and withhout cranking or anything fire up the 2 lambda's and run a free air calibration? i do mine in place, but cause i have injection i run the engien up to the red line, go full throttle and shut it off, pumping air into the exhaust, i then let it sit for an hour or so and then calibrate :) not ideal but saves me getting my hands dirty

Edited by evolotion on 2nd Dec, 2007.

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

I have a gauge on the LC-1 as well as feeding 0-5v into the Aux Box.

The outputs are 2v at 8:1 and 5v at 20:1, so multiply the volts by 4 to get the AFR.

When the gauge is showing 16:1 then the logworks is showing 4v, so everything looks right.

I could do a calibration without removing the sensors, but I'm inclined to swap the LC-1 and LM-1 around and do a calibration at the same time. Then we will know for sure if the offest is correct or instrument error.

I might appear to be getting too hung up on detail, but these instruments are going to be relied upon very heavily in setting up the EFI, so need to be just right.

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."

damn right paul good attitude!

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

What is the customer support for the innovative products like? Isn't this something they should be assisting with?

Is there any way of running both sensors in the aux box - or maybe a third sensor that is - so you have one main output for the main display, and a visual for the two others?

Alternativly - is it possible that the slope of the curve of the output is not constant between the two? I guess swapping them over will reveaql this for sure.

Edited by Paul S on 3rd Dec, 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."

Paul,

How about an overview of how you use the LM1 and LC1 together?

I've not understood - and can't easilly find - what it is that is actually recording the data. Is it done inside of the LM1?

If so - How does the LM1 interpret between the two channels as far as the display and output is concerned?

And what of the other sensors - what is the actual need of the aux box at this time as I thought you could tap the extra sensors into the LM1 as it was (it lists "Up to 6 Channels of data logging" on the website for the LM1).
Or is it needed to get the extra LC1 input into the LM1?

(as you can tell - I'm a little confused - and am even more so having been on their site and forum for a half-hour LOL)!


Incidentally...
When i spoke with innovative a loooong time ago regarding this kind of stuff for the A-series, they said the best way to go was the DL32 and two LC1's as this had the best sampling rate and bought all data in at the same time. I never looked much further into this though as the AEM sensors appeared at a great price, and I figured i could log the outputs from these if i so needed.

Looks like they do now do a kit for the DL32 and a few sensors - aimed at v8 guys for a sensor in each 'bank'

http://www.innovatemotorsports.com/xcart/p...41&cat=0&page=1


As regards the sampling rate - what is the sample rate of the raw data out of the LM1? I know it says 12bits / sec on the site - but I'm guessing the raw data has a significnatly higher resolution???


oops, sorry for the tons of questions! My bad! :)

Edited by Paul S on 3rd Dec, 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."

Great and simple answers. Cheers. :)

Would you agree that it is reasonable to expect that the LC-1 and LM-1 should give comparable readings?

We shall soon see!

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."

Interesting stuff Paul, watching with interest!

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."

Well, I just found out that the log output on the AEM gauges suck big time.... So my plan to simply log the outputs from these on my maxQ unit is not going to be of any use.

Incidentally - it seems the guages (52mm display) themselves are accurate - it's just the output is not representative of what is on the 52mm display when you hook it up to an ECU or logger.

So, looks like I might be into some innovative stuff after all. The customer support on the forum is mighty impressive for sure!

I'll wait to see what Paul establishes first however LOL!

The serial leads turned up this morning, so I'll be able to daisychain the whole lot and not rely on the aux box for the LC-1 input. Although I will run them both at the same time to start with just to check the accuracy.

I've been looking at the Megasquirt plug-in for the Innovate Logworks software. It means I can actually log any of the Megasquirt parameters at the same time.

So as well as the AFRs, I can log temperatures, TPS, Injector pulse, ignition advance and others. Now that must be worth doing.

Just need to add another serial port to the laptop. Might get one of those PCMCIA dual serial port jobbies as I'm not sure if you can use two USB/serial converters at the same time.

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 serial port jobbies are definately the way to go.

Cheers for the log file BTW... :)

LOL at Daves Sig

I just saw across this on minimania.

http://www.minimania.com/web/threadid/8885...3/msgthread.cfm