ECU Editor - Data Logging and Auto Tune

[jkwool]

Hi,

Quick intro first: I have not really posted on here much but i have been around reading posts etc for some time. I have been working with Petri on ECU Editor for awhile now.

I have been working on some functionality for ECU Editor over the last week or so to allow people with a Boost by Smith ECU Harness and the stock oxygen sensor (in your exhaust pipe) to log and analyse engine data to your laptop for your Gen 2 Busa and Bkings

The idea for this is to allow guys who dont have access to a dyno tuner or a wide band oxygen sensor to be able to get an idea of how their bike is tuned by plugging in a laptop and going for a ride while logging data. I also plan to provide functionality which will suggest changes to the fuel maps to improve the tune of your bike based on the logged data.

The logs contain the following information:

RPM
TPS (Throttle Position %)
IAP (Inlet Air Pressure)
HO2 (ECU AFR reading)
IGN (Ignition Advance)
STP (Secondary Throttle Position)
GEAR (Selected gear)
CLUTCH (Clutch in or out)
NT (Neutral indicator)
CLT (Coolant Temperature 'C)
IAT (Inlet Air Temperature 'C)
BATT (Battery Voltage)

The first stage of this development is to get the logging functionality working and an accurate Air Fuel Ratio (AFR) conversion function. I have the logging working and I am now working to improve the AFR calculation.

This is where I could use some help Would any of you guys out there with Wideband Oxygen sensors on your bikes and logging ability be able to do some side by side logging with both your wideband and the ECU Data Logger. If you send me both sets of log files from the same ride I will be able to compare the results and improve my AFR calculation for the built in oxy sensor.

I have already improved the AFR calculation thanks to Greg for sending through some logs from his bike, but as with all things like this the more data the better results i can get!

When i have the above working reliably i will look to incorporate this directly into ECU Editor.

A test version of the logging application is available for free download here: ECUDataLogger.zip - Windows Live

Instruction for using the ECU Data Logger

1. Download the .zip file from the above link
2. unzip the ECUDataLogger folder onto your laptop
3. Plug your ecu interface into your bike and laptop
4. Click on the ECUDataLogger.exe application
5. Select the COM port for your ECU Harness
6. Start your bike (after a short delay you should see some basic info appear on the screen TPS, RPM, IAP)
7. Click "Create Log File" to create a new log file, save it where you can find it on your laptop hard drive
8. When you are ready to start logging click the "Start Logging" button
9. When you are finished logging click the "Stop Logging" button

Please send both your wideband log files and the 2 .csv files produced from the ECU Data Logger to the following email address so i can analyse the files and improve the AFR function.

ECUDataLogger @ gmail.com

Thanks

-Justin.

 

[Johnnycheese]

I could have sworn it is not a wide band ( stock one)

 

[jkwool]

I could have sworn it is not a wide band ( stock one)

Yes you are right the stock sensor is a narrow band Bosch LSM 11 with its centre weighting around lambda 1 or AFR 14.7 at around 600 mv. But it does have a measurable response, and corresponding signal from the ecu, down to AFR 12 and up to AFR 16. The sensor voltage output is non linear which is why i have requested some real world comparisons with widebands on bikes to get the conversion function from the ecu output somewhat accurate.

The stock sensor is not as good as a wideband O2 sensor but I am looking to offer an entry point for tuning for guys who dont have access to a wideband or a dyno and want to use ECU Editor to improve the tune on their bikes or at least to know that they need to look at getting their bikes professionally tuned.

Once this is implemented i plan to work on some tools in ECU Editor to help guys with a wideband O2s (Innovate, Zeitronix etc etc) to take their logged data analysis and adjust their fuel maps based on their logs and a user defined AFR map.

I am also toying with the idea of feeding the 0-5 v wideband O2 sensor signal directly into the ecu and using embedded code running on the ecu to auto tune AFR to a user defined target AFR map. One step at a time though!

 

[Professor]

Excellent work!!!

 

[ogre]

what about reflashing the ecu to utilize a standard wideband o2. then you'd have a more workable range if you wanted to auto tune. not being a computer nut, i don't know that this is even possible, but it would seem an ideal solution imho.

 

[busa2001]

good work Justin.

 

[jkwool]

what about reflashing the ecu to utilize a standard wideband o2. then you'd have a more workable range if you wanted to auto tune. not being a computer nut, i don't know that this is even possible, but it would seem an ideal solution imho.

This could be possible, and i think i mention this towards the end of my original post above, writing auto tune in embedded c code to run on the ecu is quite an undertaking, this may come later, one step at a time

 

[ogre]

cool **** bro. have you thought about collaborating with petrik?

 

[jkwool]

cool **** bro. have you thought about collaborating with petrik?

yeah i am working with petri and Jasa on ECU Editor, Petri works on Gen 2 functionality, Jasa on Gen 1 and myself on B-King. There is a lot of common stuff between Gen 2 and B-King including this ecu logging functionality.

 

[ogre]

oh, so you're doing this for the second gen busa? damnit.

 

[smithabusa]

15 minutes of road data sent, some idle in garage while i suited up with laptop, some cruising around up to 50 mph or so on the country roads, slowing down for about 100 bicycles out riding today, then some back road boosting

Assuming this goes how I think it will, its going to be very beneficial to a lot of folks, myself included. Though I am really going to push hard for wideband, tuning iap map on cruising etc i would feel okay with, for power tuning i would really like to see wideband only being used for that.

Awesome work Justin.

 

[jkwool]

15 minutes of road data sent, some idle in garage while i suited up with laptop, some cruising around up to 50 mph or so on the country roads, slowing down for about 100 bicycles out riding today, then some back road boosting

Assuming this goes how I think it will, its going to be very beneficial to a lot of folks, myself included. Though I am really going to push hard for wideband, tuning iap map on cruising etc i would feel okay with, for power tuning i would really like to see wideband only being used for that.

Awesome work Justin.

Thanks Greg, you have been a big help.

 

[OB_Boss]

Agree with Smitha regards narrow band for light cruise and idle, but in saying that with narrow band in closed loop mode it's controlling the A/F anyway ? Sooooooooooooo given then that the narrow band is doing it's job ............. to then leave it connected after any adjustments have been made almost becomes a mute point as it will then simply revert settings back to obtain it's default A/F's ? (in closed loop conditions) Those settings of course being to help the cat do it's job in meeting EPA regs and also the always sought economy factor, both of these factors being a far cry from power and performance that most seek over the latter. It is for this reason why when tuning I always disconnect the o2 sensor when mapping (and leave disconnected) as having put custom settings in to obtain the desired A/F readings I seek, to then allow the o2 to over ride them would to me defeat the whole purpose of the exercise ................... being tp obtain correct fueling for the application as so desired.
Now being a curious little fellow that I am, I'm also curious as to how one can or will obtain data that requires a wideband with a narrow band sensor ???
This concerns me in that engine damage is highly unlikely to ever occur in a light throttle or cruise environment but then very quick to bite under high load conditions. Just thinking out loud here ???

 

[jkwool]

Hi Pete,

You are right, In closed loop mode the narrowband adjusts fueling to what ever suzuki needed to pass emmissions so you are quite right to disable it if you are doing custom mapping. This allows you to choose what afr you want in the different map regions as opposed to the ecu automatically adjusting.

I am also disabling the closed loop in the ecu, and I am just using the measured AFR from the sensor. This allow you to measure and adjust the AFR in the map the same as you do on the dyno with your wideband.

The narrowband is not an on off switch, it has a measurable response curve from AFR 12 to AFR 16 so it is definitely usable. The response curve is non linear, i.e. it has a lot of resolution in the center around around lambda 1 or AFR 14.7, and less resolution around AFR 12 and AFR 16 but it still does measure the AFR in these ranges. A wideband has a linear response with the same accuracy throughout its range of measurable values.

For example I have found the stock sensor measures AFR 12.8 and gives an ecu value of 45, AFR 13.4 gives an ecu value of 41 AFR 15.2 gives an ecu value of 6. i.e. its an inverse relationship, lower AFR give a higher ecu value. So by logging RPM, TPS, IAP and the converted ecu value for AFR you can determine how the bike is running in the various map regions.

Obviously if you have access to a wideband you will possibly get more accurate AFR readings, but the purpose of developing this is to allow guys who dont have access to a wideband and/or dyno tuner to be able to log and measure their AFR with some accuracy and make adjustments, or at the very least to be able to know that their bike is not running in the correct AFR range so they can go get it tuned at a dyno or buy a wideband and do it themselves!

Cheers

-Justin.

 

[OB_Boss]

No probs Justin ................... understand all that, and have to ask given readings your quoting .............. what was probe temp at the time those readings were taken ? (ie Was probe temp the same (constant) as each different A/F value was recorded ?) Also what will compensate for the change in them as the probe temp alters ? Especially given another factor to consider is that the influence of temp (temperature dependent) on voltage is less when on the lean side than it is on the rich side ?
This is one of the reasons I feel you may struggle to get the accuracy lease: needed to adjust to and with using a narrow band input ?
Is it possible as I think some one has allready suggested to go the same principle but incorpate a wide band sensor instead ? Adding here I know very little about writing code (in fact I know bugger all about anything ) just bumming along learning as I go.

Innovate have a great adaptor available that would allow a wideband sensor to be screwed directly into standard position allready available.

 

[Draco1340]

Love to help but my A/F is kept right around 12.0 so its too low for a narrow band. Also, I don't think you will find too many people running the stock narrow band AND the Bosch wide band?? The stock pipe is the first thing thrown in the trash. Good luck though.
Tom

 

[smithabusa]

I have both sensors on my turbo but I'm an oddball old

I installed both for things like this LOL

I'm in same boat under boost 12:1 would be absolute max, I'm sending jkool a LLC as I really like this concept but with wideband: )
Posted via Mobile Device

 

[jkwool]

I believe there is a heating circuit in the sensors that maintains the correct operating temperature for the senor, similar to the wideband. So for the purpose of what i am trying to achieve with this, and given the resolution away from AFR 14.7 i dont think the sensor temp will be too much of a factor. I may be wrong on any of this though But the testing that Greg has been doing for me has given us some good results around AFR 12.6 - AFR 13.5 with the narrowband reliably giving the same AFR as the Innovate wideband give or take AFR 0.1

This range is where you want some reliability in reading with a non boosted engine (you taught me that Pete ha ha ha.) If the AFR is higher than this range and TPS is greater than 11% you know you are too lean and you need to add fuel, if AFR is less than this range then you are rich and need to take some fuel out. For a non boosted stock motor you dont need to measure below 12 cause you would not tune it in that range, if it read 12 using the narrowband you know its too rich.

I dont see anyone with motor work or boost etc using the narrowband for any sort of tuning, thats not really the reason i am working on this, its more for giving guys with stock motor and a few add ons an easy way to check their fueling to see if they are in the ball park or need to go to the next step and go to a wideband.

Using a wideband to feed into the ecu or ecu logging application is phase 2 of this little project and Greg has kindly offered to donate an Innovate LC-1 to me to use so i can get this up and running. Hopefully we can feed this into the ecu and feed it out the other end into ECU Editor and log the data so we will have the best of both worlds with all the engine data directly from the ecu and an accurate wideband afr signal.

I have not seen those adapters before but I reckon I will just take the pipe off and get the bung welded in so its directly in the gas stream and i can retain the stock sensor for further testing!

 

[OB_Boss]

on that Greg
Looking around for his FREE LM1
Yeah Justin I also thought of the accuracy re the adaptor, but figured if Innovate produce them themselves then figured they must not effect accuracy or I should imagine they wouldn't go that way. I'd say they could also have an advantage of working like the heat sink Innovate put out also protecting the sensor from "shock' I used one of those for years with no ill effect.
Whatever whichever ........................... there is certianly a lot more to "flashing" now than just dropping ya dacks

 

[smithabusa]

Im not made of money LOL, LC1 is all I can afford to donate at the moment