PDA

View Full Version : Problem creatingformula for a new item to the .adx Values



bobdec
02-13-2013, 12:37 AM
Did a lot of searching but could not find the method to calculate the formula for a new .adx value I trying to add. I have a flash PCM 16188051 $EE files and I'm trying to add A/C Evaporator Temperature F* to the values in the .adx using ADX editor. I have found that the Evap temp byte is Mode, 1 Msg 0, Word 26 on the PCM data stream and in the ADX editor I'm using offset 0X19 (hex) 25 D on the Values, Item creation screen. By experimenting with another program I have found the following.... Temp sensor wire enters the PCM on pin D24, at 0v (ground) potential on D24 the scanner reads 90* F (32.3* C) and at +5V potential on D24 scanner reads 20*F (-6.7* C) . My problem is calculating the conversion equation to display temperature in degrees F. I've understand the 5V/255 fraction but am getting confused on how I factor that into the 90-20= 70/255 fraction needed to convert the voltage to degrees. The other point is the temp display value drops as voltage increases. If so then the fraction in the equation will be negative using start point or 90* F EG:( n* -fraction +90.00) this assumption could be my problem.

RobertISaar
02-13-2013, 12:52 AM
the fastest way to work around this with a linear equation is to create a lookup table and only enter two values: 0 A/D counts and it's corresponding real-world value and 255 A/D counts and it's real-world value. Tunerpro will interpolate for any value in-between.

non-linear requires more data points, but essentially the same process.

EagleMark
02-13-2013, 01:44 AM
I have found that the Evap temp byte is Mode, 1 Msg 0, Word 26 on the PCM data stream and in the ADX editor I'm using offset 0X19 (hex) 25 D on the Values, Item creation screen. That's only going to be in a F body Camaro bin file to output to data stream, it's not in a B body Caprice.


26 EVAPTEMP A/C EVAPORATOR TEMPERATURE
DEG F = N/2
DEG C = (N - 64)/3.6
27 ADACPRES A/D RESULT FROM A/C PRESSURE TRANSDUCER
VOLTS = 5N/255
PSI = 1.833N - 14.95

It's 26, but the list starts at 1, should start at 0, so just type in 25 and packet offset will fill in automatically. You got that, not sure what you meant by 25 D ?

RobertISaar
02-13-2013, 01:53 AM
25 D seems like it would imply 25 in decimal format, which would be correct.

anyways, if the B-bodies didn't use it, that leaves an easy option to pipe in a wideband AFR signal.

EagleMark
02-13-2013, 01:58 AM
Dam we were looking for that awhile ago... we ended up using 26

Cool, I didn't know about D, makes sense when your talking $hex.

RobertISaar
02-13-2013, 02:01 AM
i tend to do it myself.... 10h = 16d, stuff like that.

gregs78cam
02-13-2013, 03:08 AM
-0.27451 * X +90, I think that should work.

RobertISaar
02-13-2013, 03:20 AM
if 0 = 90*F and 255 = 20*F

90 - (X*0.2745) = *F

bobdec
02-14-2013, 12:19 AM
I was trying to get this working as a training exercise as I'm new to the TunerPro architecture and I enjoy some mind challenges. Apparently I picked a bad example and I truly don't want to waste others time on a self inflicted project. The above equations gave incorrect results so I backed off and built an equation F = 0 + (X* 1) " which I understand should provide the 'N' or 'X' value of the byte directly in the display field. Because I do not have a variable 0-5 V source I used some C-cell batteries. Results were: @ 0 volts X = 180, @ 1.5 volts X = 142, @ 3.0 volts X = 48, @ 4.5 volts X = 40 and @ 5 volts X= 40 . Notice the 4.5 V and 5.0 V readings are the same count. So to me it indicates the number range swing in the byte starts at 40 and stops at 180. Also important is it hits 40 somewhere between 3.0 and 4.5 volts, w/o a variable voltage source I could not find the voltage where it hits 40. But on a graph it looks like at around 3.5 Volts the count stops at 40 and never drops any more as voltage increases. I expect it will work after I find the actual high voltage stop point and then recalculate the fraction. I'll play with it a while , thanks for the feedback it cleared my mind and go me going again for a while.

RobertISaar
02-14-2013, 12:23 AM
hmm.... that's significantly non-linear... i wonder if it has a pullup resistor instead of a grounded A/D sampling circuit. it would make sense for a temp sensor.

bobdec
02-14-2013, 09:51 PM
Agree w/the temp sensor setup causing non-linear results. In the F-Body all the Evap Temp does is turn off the A/C clutch at 36* F and back on at 39* F (prevents freeze-up). So it's a small range that is used, I may concentrate in that range. For trivia.. Both temps are set to 0* F in the B-Body tunes. I also just noticed the $EE .XDF I'm using does do not contain the EVAP Temp enable/disable settings in the Scalars, got a feeling the B-Body .xdf's have been mixed into the F-Body.