Logging Wideband AFR via Rear O2 ECU ADC Input
Aug 26, 2009, 08:44 PM
#31
Evolved Member
iTrader: (4)
Join Date: Feb 2008
Location: Edmonton, Alberta
Posts: 610
Likes: 0
Received 0 Likes
on
0 Posts
Done.
Here's a little (actually, not little at all) How To in case anyone else wants to do it.
https://www.evolutionm.net/forums/ev...ml#post7441923
Here's a little (actually, not little at all) How To in case anyone else wants to do it.
https://www.evolutionm.net/forums/ev...ml#post7441923
Oct 31, 2009, 04:50 PM
#32
IF V= 5 * ADC / 1023 doesn't it follow that V = ADC / 204.6 (none repeating decimal)??? or is there a problem in Evoscan using decimals for anything other than addition and subtraction?
IF the V=ADC / 204.6 IS valid then we can further reduce the formula thus:
AFR = 2.375 * (5 * ADC / 1023) + 7.3125 then AFR = (ADC / 86.15) + 7.3125 ?
The 86.15 IS a rounding up of a repeating decimal, so I understand why that may cause problems.... but in the first section to calculate V itself, 204.6 is NOT a rounded number... so why can't we use it?
Don't get me wrong, I will use the formula as stated to avoid any problems I don't know about... just curious as to what those problems might be
Keith
PS: We actually type "x" instead of "ADC" in the formula's correct?
Oct 31, 2009, 06:28 PM
#34
Now, something I really need is the address in the 2010 Ralliart ROM for the rear O2 sensor. Since I have it setup (appears set up correctly) and am getting garbage for data I have to conclude that it is not the same as the address for the 2009 Ralliart. Without the correct address the wire hookup and formula don't do much good
Here is my ROM:
http://norcalmotorsports.org/users/b...M_55590006.zip
Thanks,
Keith
Nov 2, 2009, 07:24 AM
#35
Using this formula: AFR = 2.375 * (5 * ADC / 1023) + 7.3125 I get 100% crap for results 8.xxx at idle with very little variation on a blip of the throttle... so I checked out the raw data (x alone) I was getting from the wide band sensor. At idle it hovers around 130!
Just on a hunch, instead of using V=5*x/1023 I looked at the formula for the stock rear O2 sensor, and it was x* 0.01952
Using the formula (2.375*x*.01952)+7.3125 resulted in 13.5 at idle, and reasonable variations on a blip of the throttle. I can't watch both the wide band gauge AND the computer screen at the same time very well, but they seem to track with Evoscan reading 1 point of A/F ratio richer than is reported by the wide band gauge itself.
Now how do I come up with a correction factor? Is it as simple as making the formula (2.375*x*.01952)+8.3125 instead of (2.375*x*.01952)+7.3125 or is that way too simple?
Keith
PS: Any clue as to why mine is so different please let me know!
Just on a hunch, instead of using V=5*x/1023 I looked at the formula for the stock rear O2 sensor, and it was x* 0.01952
Using the formula (2.375*x*.01952)+7.3125 resulted in 13.5 at idle, and reasonable variations on a blip of the throttle. I can't watch both the wide band gauge AND the computer screen at the same time very well, but they seem to track with Evoscan reading 1 point of A/F ratio richer than is reported by the wide band gauge itself.
Now how do I come up with a correction factor? Is it as simple as making the formula (2.375*x*.01952)+8.3125 instead of (2.375*x*.01952)+7.3125 or is that way too simple?
Keith
PS: Any clue as to why mine is so different please let me know!
Nov 2, 2009, 07:32 AM
#36
Using this formula: AFR = 2.375 * (5 * ADC / 1023) + 7.3125 I get 100% crap for results 8.xxx at idle with very little variation on a blip of the throttle... so I checked out the raw data (x alone) I was getting from the wide band sensor. At idle it hovers around 130!
Just on a hunch, instead of using V=5*x/1023 I looked at the formula for the stock rear O2 sensor, and it was x* 0.01952
Using the formula (2.375*x*.01952)+7.3125 resulted in 13.5 at idle, and reasonable variations on a blip of the throttle. I can't watch both the wide band gauge AND the computer screen at the same time very well, but they seem to track with Evoscan reading 1 point of A/F ratio richer than is reported by the wide band gauge itself.
Now how do I come up with a correction factor? Is it as simple as making the formula (2.375*x*.01952)+8.3125 instead of (2.375*x*.01952)+7.3125 or is that way too simple?
Keith
PS: Any clue as to why mine is so different please let me know!
Just on a hunch, instead of using V=5*x/1023 I looked at the formula for the stock rear O2 sensor, and it was x* 0.01952
Using the formula (2.375*x*.01952)+7.3125 resulted in 13.5 at idle, and reasonable variations on a blip of the throttle. I can't watch both the wide band gauge AND the computer screen at the same time very well, but they seem to track with Evoscan reading 1 point of A/F ratio richer than is reported by the wide band gauge itself.
Now how do I come up with a correction factor? Is it as simple as making the formula (2.375*x*.01952)+8.3125 instead of (2.375*x*.01952)+7.3125 or is that way too simple?
Keith
PS: Any clue as to why mine is so different please let me know!
I think this may be because the address I have for the rear O2 sensor is single byte instead of 2 byte, is this correct? Do you have the 2 Byte address for the rear O2 sensor on the 2010 Ralliart? The single byte address is 23808681.
Thanks,
Keith
Last edited by Fourdoor; Nov 2, 2009 at 07:48 AM.
Aug 4, 2011, 06:01 AM
#37
logging wideband via rearO2 formula
Im trying to get my wideband to read correctly through the rear O2 sensor. My formula in evoscan under the tephra v7 mod ROM, shows up as "WBO2" 10.00+(16.99-10.00)*x/255... for some reason at idle it is logging around 13.3-13.5 and I know its supposed to be around 14.7. Can someone help me with my formula? I have a glowshift wideband gauge that reads from 20-10 a/f so would I have to change my formula to 19.99-10.00 instead of 16.99-10.00? Or do I need to change the scale of my 255 to 350 since I have the big maps conversion. Thanks
Aug 5, 2011, 08:18 AM
#38
Im trying to get my wideband to read correctly through the rear O2 sensor. My formula in evoscan under the tephra v7 mod ROM, shows up as "WBO2" 10.00+(16.99-10.00)*x/255... for some reason at idle it is logging around 13.3-13.5 and I know its supposed to be around 14.7. Can someone help me with my formula? I have a glowshift wideband gauge that reads from 20-10 a/f so would I have to change my formula to 19.99-10.00 instead of 16.99-10.00? Or do I need to change the scale of my 255 to 350 since I have the big maps conversion. Thanks
...should get you closer, but ideally you would want to verify that no noise is throwing off the reading a bit by logging a digital signal from the gauge if it has one. You could also try to feed it voltage to confirm, but thats a bit more complicated.
Changing the 255 to 350 is definitely not correct... not sure where you got that from. The x/255 represents a bitwise value of the full range that can be represented by the input. This number should *always* be 255 so that 0 volts = 0 (0.00)and 5 volts = 255 (1.00). If the range is linear that means that [base] + (range * percentage of range) will be the correct AFR value.
Aug 6, 2011, 07:15 AM
#39
Thanks for the help! I tried the formula and it seems to be reading correctly... I dont have a digital output on my wideband gauge, but for now I will use this... 10.00+(20.00-10.00)*x/255
Thanks again!
Thanks again!
Aug 11, 2011, 09:05 AM
#40
Newbie
Join Date: Aug 2008
Location: Miami
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
Hy Dsm Riders... I have this problem with my evoscan an Aem Uego reading...
Im trying to get my wideband to read correctly through the rear OEM O2 sensor imput, i cut the white wire on my ecu from the OEM rear O2 pin 73 on my evo 9, and tap to the white wire from the uego, i have teprha V7 88580714 ROM and i enable the rear O2 simulation option and flash it back, the rear O2 is still in place, no removed yet, no mut table chance since its getting the reading from the Oem O2 sensor mut 3C table 8409, i use evoscan V 2.7 0090 and i use this formula 10+(20-10)*x/255on the WBO2 eval and the reading its off by aprox 2.6 afr, so what do i need to check it out? I tried sevarl times with different formulas and i cant log it just right
Anyone who can point me in the right direcction would be great...
Thanks in advance
Im trying to get my wideband to read correctly through the rear OEM O2 sensor imput, i cut the white wire on my ecu from the OEM rear O2 pin 73 on my evo 9, and tap to the white wire from the uego, i have teprha V7 88580714 ROM and i enable the rear O2 simulation option and flash it back, the rear O2 is still in place, no removed yet, no mut table chance since its getting the reading from the Oem O2 sensor mut 3C table 8409, i use evoscan V 2.7 0090 and i use this formula 10+(20-10)*x/255on the WBO2 eval and the reading its off by aprox 2.6 afr, so what do i need to check it out? I tried sevarl times with different formulas and i cant log it just right
Anyone who can point me in the right direcction would be great...
Thanks in advance
Aug 11, 2011, 09:47 AM
#42
Evolved Member
iTrader: (3)
Join Date: May 2009
Location: Autocrossing Somewhere
Posts: 501
Likes: 0
Received 0 Likes
on
0 Posts
0-5v analog output Gasoline values from 10 to 20:1 AFR
0-5v analog output E85 values from 6.6 to 13.2 AFR
0-5v analog output Ethanol values from 6.14 to 12.3:1 AFR
0-5v analog output Methanol values from 4.4 to 8.8:1 AFR
0-5v analog output Lambda values from .68 to 1.36:1
0-5v analog output E85 values from 6.6 to 13.2 AFR
0-5v analog output Ethanol values from 6.14 to 12.3:1 AFR
0-5v analog output Methanol values from 4.4 to 8.8:1 AFR
0-5v analog output Lambda values from .68 to 1.36:1
Aug 11, 2011, 09:56 AM
#44
Newbie
Join Date: Aug 2008
Location: Miami
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts
The uego its sending 1.8v with the igtion switch on, and with the engine runing its sending 1.1 and 1.2 volts
Thanks in advance...
Aug 11, 2011, 09:58 AM
#45
Newbie
Join Date: Aug 2008
Location: Miami
Posts: 8
Likes: 0
Received 0 Likes
on
0 Posts