Omni 4bar map sensor Plug and Play
Mar 25, 2009, 06:45 PM
#91
Newbie
Join Date: Feb 2008
Location: socal
Posts: 3
Likes: 0
Received 0 Likes
on
0 Posts
The voltage conversion for 3,4, and 7 bar sensors is posted on our website in the instructions section. here it is below also, for the 3 4 and 7 bar sensors. the scalar and offset numbers are for hondata and the like systems for the honda community.
1bar=100KPA=14.5PSI=1000milibar
3 bar
Scalar: 617
Offset: 11
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.00318 X P - .00353)
Example: Atmosphere equals 1 bar or 100 KPA = 5 X (.00318 X 100 - .00353) = 1.572 volts @ 1 bar or 100 KPA
4 bar
Scalar 838
Offset 31
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.002421 X P - .00842)
Atmosphere equals 1 bar or 100 KPA = 5 X (.002421 X 100 - .00842) = 1.168 volts @ 1 bar or 100 KPA
Crome Custom
Min: 34.7 mbar
Max: 4000 mbar
AEM
MIN=0
MAX=4.8
LOAD OFF SET=0
LOAD SCALAR=4.0
THE 3 AND 7 BAR ARE THE SAME LOAD OFF SET AND MIN JUST USE THE MAX VOLTAGE FROM THE CONVERSION TABLE AND 3.0 AND 7.0 FOR THE SCALAR
7 bar
Scalar 1582
Offset -351
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.0012858 X P + .04)
Example: Atmosphere equals 1 bar or 100 KPA = 5 X (.0012858 X 100 + .04) = .843 volts @ 1 bar or 100 KPA
1bar=100KPA=14.5PSI=1000milibar
3 bar
Scalar: 617
Offset: 11
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.00318 X P - .00353)
Example: Atmosphere equals 1 bar or 100 KPA = 5 X (.00318 X 100 - .00353) = 1.572 volts @ 1 bar or 100 KPA
4 bar
Scalar 838
Offset 31
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.002421 X P - .00842)
Atmosphere equals 1 bar or 100 KPA = 5 X (.002421 X 100 - .00842) = 1.168 volts @ 1 bar or 100 KPA
Crome Custom
Min: 34.7 mbar
Max: 4000 mbar
AEM
MIN=0
MAX=4.8
LOAD OFF SET=0
LOAD SCALAR=4.0
THE 3 AND 7 BAR ARE THE SAME LOAD OFF SET AND MIN JUST USE THE MAX VOLTAGE FROM THE CONVERSION TABLE AND 3.0 AND 7.0 FOR THE SCALAR
7 bar
Scalar 1582
Offset -351
Voltage to pressure (KPA) conversion
P= pressure in KPA 4 bar = 400 KPA
Voltage out/signal = 5 X (.0012858 X P + .04)
Example: Atmosphere equals 1 bar or 100 KPA = 5 X (.0012858 X 100 + .04) = .843 volts @ 1 bar or 100 KPA
Mar 25, 2009, 10:27 PM
#95
I am sure thats right, but the thing I have noticed in the 3 bar is that saturation alledgedly happens at 3.63 volts (255). I am drawing myself a map right now, literally, and when I trace out the voltage vs Kpa plot from the pic I posted I get 4.86v at 33.2psi which is the highest I have logged (Bryan as well and I am sure others). This jives with the formula there as well.
So I am trying to decide if something in the scaling, our interpretation of that table, or something else isnt correct. I will post a pic of my graph in a second
So I am trying to decide if something in the scaling, our interpretation of that table, or something else isnt correct. I will post a pic of my graph in a second
Last edited by JohnBradley; Mar 25, 2009 at 11:18 PM.
Mar 25, 2009, 10:32 PM
#97
For what it's worth, I rigged up a little voltage divider on the MDP input and basically input a voltage from 0 to 5 volts into the signal line.
0 counts < 0.02 volts
255 counts > 4.98 volts
EDIT: I think I misunderstood what you were saying.
What table says 255=3.63 VDC?
0 counts < 0.02 volts
255 counts > 4.98 volts
EDIT: I think I misunderstood what you were saying.
What table says 255=3.63 VDC?
Last edited by 03whitegsr; Mar 25, 2009 at 10:36 PM.
Mar 25, 2009, 10:45 PM
#98
The JDM MAP Scaling table references 3.63v as 255 yet it continues to climb.
https://www.evolutionm.net/forums/ec...tructions.html
Originally Posted by Mrfred
After modifying and saving the ROM xml file, launch ECUFlash, open your ROM, and find the new entries. Note that they have their own category. It may be necessary to set ECUFlash to "Developer" mode in the ECUFlash options (and relaunch ECUFlash) to be able to see the tables. Here is what needs to be changed in ECUFlash:
0.84 11
1.37 20
1.89 31
2.42 40
2.95 48
3.48 57
4.00 67
4.53 80
4.98 255
Change it to the JDM values:
0.51 20
0.90 40
1.29 57
1.68 78
2.07 97
2.46 117
2.85 138
3.24 158
3.63 255
0.84 11
1.37 20
1.89 31
2.42 40
2.95 48
3.48 57
4.00 67
4.53 80
4.98 255
Change it to the JDM values:
0.51 20
0.90 40
1.29 57
1.68 78
2.07 97
2.46 117
2.85 138
3.24 158
3.63 255
Mar 25, 2009, 11:02 PM
#99
I think I have been overthinking it. We can reverse engineer the EvoScan formula to calculate out what each value should be more or less.
Right now the 3 bar is 0.19347*x-14.5 (1 bar/100kpa)
X cannot be greater than 255
I have observed 33.2psi with that setting (coincides with my boost gauge at all points with the above scaling btw), so running it backwards:
33.2+14.5=47.7/0.19347=246.8 counts
Following the testing that Mrfred did when he came up with that JDM MAP vs USDM MDP chart and my formula a few posts back (V=0.1028p+1.456) means 246.55=4.869v
33.2*.1028=3.41296+1.456=4.8689
Setting the math to some whole numbers 4.8v/4.8689v is a correction of .9858. This times 246.55 is basically 243 (243.06)
If the Omni is 43.5psig (43.5+14.5=58 then 58/243) at 4.8v then each count is worth .238683 or so. It is advertised as linear so we can just grab a few different voltages, run the maths and apply it to the table, then check it vs a boost gauge and fine tune.
...or I have made some mistakes and we'll have to do something else
Right now the 3 bar is 0.19347*x-14.5 (1 bar/100kpa)
X cannot be greater than 255
I have observed 33.2psi with that setting (coincides with my boost gauge at all points with the above scaling btw), so running it backwards:
33.2+14.5=47.7/0.19347=246.8 counts
Following the testing that Mrfred did when he came up with that JDM MAP vs USDM MDP chart and my formula a few posts back (V=0.1028p+1.456) means 246.55=4.869v
33.2*.1028=3.41296+1.456=4.8689
Setting the math to some whole numbers 4.8v/4.8689v is a correction of .9858. This times 246.55 is basically 243 (243.06)
If the Omni is 43.5psig (43.5+14.5=58 then 58/243) at 4.8v then each count is worth .238683 or so. It is advertised as linear so we can just grab a few different voltages, run the maths and apply it to the table, then check it vs a boost gauge and fine tune.
...or I have made some mistakes and we'll have to do something else
Last edited by JohnBradley; Mar 26, 2009 at 07:47 AM. Reason: originally entered 33.25 when my note clearly says 33.2
Mar 25, 2009, 11:12 PM
#100
He says it above:
(4bar)
or: volts = 5x(0.002421xPascals - 0.00842)
so if you want to work out the left hand of the formula you just work it backwards:
1) volts/5 = (0.002421xPascals - 0.00842)
2) (volts/5) + 0.00842 = 0.002421xPascals
3) ((volts/5) + 0.00842)/0.002421 = Pascals
now we can replace volts by 5*ADC/255 (because 8bit is 0-255)
4) (((5xADC/255)/5) + 0.00842)/0.002421 = Pascals
which nicely cancels out the 5/5
5) ((ADC/255) + 0.00842)/0.002421 = Pascals
Of course you could break that down futher - but best to leave it.
So Evoscan formula would be:
Eval="((x/255) + 0.00842)/0.002421"
That would give you KiloPascals...
ok so to test:
Lets say 255 (max voltage) = 416.530 KILApascals...
Seems right to me
(4bar)
Voltage out/signal = 5 X (.002421 X P - .00842)
so if you want to work out the left hand of the formula you just work it backwards:
1) volts/5 = (0.002421xPascals - 0.00842)
2) (volts/5) + 0.00842 = 0.002421xPascals
3) ((volts/5) + 0.00842)/0.002421 = Pascals
now we can replace volts by 5*ADC/255 (because 8bit is 0-255)
4) (((5xADC/255)/5) + 0.00842)/0.002421 = Pascals
which nicely cancels out the 5/5
5) ((ADC/255) + 0.00842)/0.002421 = Pascals
Of course you could break that down futher - but best to leave it.
So Evoscan formula would be:
Eval="((x/255) + 0.00842)/0.002421"
That would give you KiloPascals...
ok so to test:
Lets say 255 (max voltage) = 416.530 KILApascals...
Seems right to me
Last edited by tephra; Mar 25, 2009 at 11:16 PM.
Mar 26, 2009, 07:27 AM
#101
If I correctly understand what jcsbanks has learned about the MAF implementation routines, there is a fall back routine that allows the Evo 9 (with a factory MAP sensor) to run under a very limited SD mode when the MAF is not working properly. The MAP sensor scaling table that you show is the scaling used by the ECU for this limited SD mode. For some reason, the last point in the factory JDM MAP scaling does not follow the other points. Try using the second to last point in that table to set your slope, and then you'll see that things match up well.
Mar 26, 2009, 07:42 AM
#102
Interesting M. I knew that 9s will run with the MAF unplugged on a completely different set of calculations. Cruise timing looks like WOT timing, it wont go into boost without going extremely lean, etc.
White, you were correct when I went through the math (or differed only ever so slightly) and all that.
I am curious though, mrfred, dont we need to know what the values are there (255 aside) so that when we log we get the right value? Or since Evoscan uses the raw ADC we just needed to know what the split was per count?
White, you were correct when I went through the math (or differed only ever so slightly) and all that.
I am curious though, mrfred, dont we need to know what the values are there (255 aside) so that when we log we get the right value? Or since Evoscan uses the raw ADC we just needed to know what the split was per count?
