v01canic

iTrader: (1)

Looks like I'm just running into issues back to back this week...

After getting my cage installed my wideband was stuck at 22, assuming the guy try to start the car without the gauge installed. Last night i pull it off to recalibrate per the instructions on Innovate's website (http://www.innovatemotorsports.com/s...Manual_1.1.pdf) and left it in open air for 15 minutes or so.

I reconnected everything and started up TS and compared it to the wideband. The MTX-L was was now reading the correct AFR's around 13 on E85 but TS was only showing 10.

Double checked my voltages set at .22v = 7.35 & 5v = 22.34 on TS but the two gauges were still way off.

Here is my original thread when i first set this up: https://www.miataturbo.net/megasquir...eadings-61178/


Has anyone had this issue after a re-calibration? I called innovate and they said as long as there is no error code i dont need a new sensor. Kinda at a loss.

Thanks again

stefanst

iTrader: (1)

Measure the voltage feeding into the MS. If it corresponds with the reading on the MS, your problem is on the MTX-side. If it corresponds with the reading on the MTX, your problem is with the MS.

v01canic

iTrader: (1)

is there some kind of volt to afr matrix i can use. Meaning, how do i know x volts equals what afr?

stefanst

iTrader: (1)

You already posted the info:
Interpolate in between.
To make it even easier:
5V-0.22V = 4.78V
22.34-7.35 = 14.99

So a change of 4.78V is equivalent to a change in Lambda of 14.99.
14.99/4.78V = 3.136/V

Measure Voltage (let's say you measure 3V)
deduct zero-offset voltage of 0.22V
-> 2.78V
Multiply with 3.136/V
-> 8.72

Add zero offset of Lambda
8.72 + 7.35 = 16.07

This is how, according to the values you posted, the lambda value corresponds to voltage.

v01canic

iTrader: (1)

so vis-a-vis (matrix reference) when i measure 3v i should have an AFR of 16.07

stefanst

iTrader: (1)

Yes.

And here's an EXCEL formula too.
=(A1-0.22)*3.136+7.35

Paste the text above in cell A2 on a spreadsheet and enter your voltage in cell A1.
I really can't make it any easier.

v01canic

iTrader: (1)

per your very handy formula i got the following

1.14 volts measured between ground and signal
equals 10.2 AFR in the formula and was correctly displayed on the TS

Wideband reads 13.5.

Do i just continue to dick around with the correction table till everything matches? If so do i mess side of the calculations do i start with, the .22v = 7.35 & 5v = 22.34

v01canic

iTrader: (1)

If someone wants to see what to correct their voltages to in Tunerstudio given you know what the measured voltage is at the wideband, you can use the attached Excel spread sheet.

A huge thanks to Stefan for doing ALL of the leg work.

Man i really hope i figure this out.

skinny

it's even easier if you use settings 0-5v = 10-20 AFR. you don't need the outer edges of the AFR range.

i don't know if you can do it with the MTX-L but with the LC-1 you can get it to output set voltages and then you can see what the megasquirt reads to check it matches up.

Braineack

iTrader: (62)

How does this help anything? Now you've reduced resolution by reducing the amount of measurable AFR by 30% and you still haven't accounted for any voltage offset--which is exactly this issue discussed in this thread.

skinny

how exactly have you reduced resolution when your 0-5v fixed output is now describing a smaller AFR range?

voltage offset is accounted for how i described. get your wideband controller to output a fixed voltage, see what voltage you get at the ECU. if it's different you have an offset. do it for a sweep across the voltage range and correct accordingly

Braineack

iTrader: (62)

Okay I see what you're saying, but you're basically rehashing what stefanst and v0lanic are already discussing, just throwing in use of a 10-20 scale instead of 7.35-22.4 into the mix.

Instead of outputting a fixed voltage, they are using the known output scale to determine what voltage the controller should be outputting, measuring it at the MS, and then comparing the two.

So right now V0lcanic is measuring 1.14v at the MS, seeing 10.2AFR on his MTX-L gauge, but seeing 13.5AFR in TS. Using said "look-up table," 10.2AFR is the correct AFR for 1.14v. Doing this just eliminates the need to open up logworks and rescale the output, but it's essentially the thing you are suggesting.


However I will agree I was wrong about the reduced resolution--you'll gain resolution in those areas--but lose 30% of measurable range. I admit I don't need the lean scale, but being able to read under 10:1 is pretty important when tuning boost initially as your maps should be pig rich when first starting out and 10:1 is too close to comfort if you're tuning for 11.5:1. I think an ideal scale would be something like ~9-20:1, but having all that range in the first place is a benefit that makes the lc-1 and mtx-l unique.



Now with that all being said, OP if you're that far off in TS, something's amiss, as they should match now.

stefanst

iTrader: (1)

Don't change the table in MS. Since you have now determined that MS is giving you the right display for what you programmed in it, we must assume that the problem is that the MTX-L is not sending the voltage signal you are assuming it's sending.

From the manual:
There's two things jumping out at me:
1. The manual says that the voltage range is 0V to 5V on Output 1 and not what you had quoted earlier. Where did yo get those numbers?
2. For an AFR <=14 Output 2 will show 1.1V and not change at all. Since you are measuring approx. 1.1V, are you sure that you have the MS hooked up to output 1 and not output 2?

skinny

yeah, the 10-20 AFR just makes the conversion from v to AFR simpler (2.5v = 15AFR) but i guess it's unnecessary with the formula the OP now has. can you not look at actual O2 sensor volts input on TS rather than back-calculate from AFR?

there are two ways to set the output voltage on the O2 sensor - but only one worked for me. you can either set it to output a given voltage whatever the actual afr reading (so a flat line response), but for some reason my LC-1 didn't like that. The other option is to go into advanced settings i think, and tell the LC-1 to give a certain output whilst it warms up, and that worked like a charm for me.

One thing i don't think OP has mentioned but i'm sure already knows - lambda 1 is not 14.7 AFR for E85.

v01canic

iTrader: (1)

so the dumb **** who did my cage fucked up my wiring and connected the MS to the narrow band wire instead of the wideband cable (the brown cable from the innovate opposed to the yellow cable).

Needless to say, im annoyed, specially since all he had to do is disconnect one little clip apposed to ******* with my crimps etc.