Work around for a nonlinear a/d voltage input in 3.xxx

[wstaab]

Anyone have a work around for a nonlinear a/d voltage input in 3.xxx? Supported in 2.24 but not in 3.xxx.

<CustomAUX>

<add key="1" value="AFX W/B,AFX,120,130,0.657,9" />

<add key="2" value="LR Sensor,,50,50,1,0" />

<add key="3" value="RR Sensor,,50,50,1,0" />

<add key="4" value="Oil Temp,Oil
Temp,17,10,0,0,0,0,0,60,66.667,73.333,80,84.286,88 .571,92.857,100,101.429,10
5.714,110,120,125,130,135,140,144,148,152,156,160, 164,168,172,176,180,185,19
0,195,200,206.667,213.333,220,226.667,233.333,240, 250,260,270,280,292.5,305,
320,335,0,0" />

[Keith@HPTuners]

There is no easy way.

What sensor are you using?

[wstaab]

STACK oil temp gauge with a nonlinear voltage output vs temperature.

<add key="4" value="Oil Temp,Oil
Temp,17,10,0,0,0,0,0,60,66.667,73.333,80,84.286,88 .571,92.857,100,101.429,10
5.714,110,120,125,130,135,140,144,148,152,156,160, 164,168,172,176,180,185,19
0,195,200,206.667,213.333,220,226.667,233.333,240, 250,260,270,280,292.5,305,
320,335,0,0" />

Supported in 2.24.xxx

STACK.jpg

[Keith@HPTuners]

Your best option, at this point, will be to create a transform for the working range you use, and understand that the value is approximated.

[shroom93]

sorry to hijack this thread, but I have the same issue on a 30-2310 AEM wideband O2 sensor.. is their anyway at all to use it now with 3.XX vcm scanner? I would have created my own thread, but thought since the same thing was discussed here, I would add here.

edit- Just curious, why not leave the support for nonlinear inputs in for 3.xxx

[wstaab]

Your best option, at this point, will be to create a transform for the working range you use, and understand that the value is approximated.

OK I have read the help files twice with no luck. How do I set up a linear transform starting at .5 volts = 60 degrees F ending at 4.8 volts = 335 degrees F with .1 volt resolution?
Also I need to create a linear transform for voltage vs linear distance for suspension sensors. 0 to 5 volts with .1 inch resolution.

[schpenxel]

sorry to hijack this thread, but I have the same issue on a 30-2310 AEM wideband O2 sensor.. is their anyway at all to use it now with 3.XX vcm scanner? I would have created my own thread, but thought since the same thing was discussed here, I would add here.

edit- Just curious, why not leave the support for nonlinear inputs in for 3.xxx

That sensor has a linear output... and the formula is clearly put right in the manual, so I don't quite understand the issue on this one?

Page 4:

AFR = 2.375(V) + 7.3125

http://www.aemelectronics.com/files/...Controller.pdf

[schpenxel]

OK I have read the help files twice with no luck. How do I set up a linear transform starting at .5 volts = 60 degrees F ending at 4.8 volts = 335 degrees F with .1 volt resolution?
Also I need to create a linear transform for voltage vs linear distance for suspension sensors. 0 to 5 volts with .1 inch resolution.

This is what I would do. I have not personally done this on an oil temp sensor but have used it on a methanol pressure sensor and it worked as expected:

1. Add the input that the sensor is connected to to the channels list (MPVI 1/2/3/4)

2. Right click on the MPVI input you just added, hit transform

3. Go to the user defined option, put in the formula that most closely matches the linear portion of that formula. The formula I came up with is V / 0.015636 + 28. To come up with the slope you would use (335-60) / (4.8-0.5) = 63. (Y = Mx + b format here..). HPT wants 1/x, so 1/63 = 0.015636. For the offset part you plug in what you have know. So, let's use 0.5 Volts. 60=0.5/0.015636+b. Solve for b and you get ~28

And there ya ago. y = x / 0.015636 + 28 (x is voltage). Enter these values in the Function section.

4. Click beside of "Parameter", search for oil temperature, double click on one of them, hit no when it asks about generic sensor.

5. Set the units to *F and hit OK

Now to use it in a chart or graph, use the same "oil temperature" parameter that you selected in step 4 and it should work just fine. Fortunately what you're using is pretty close to linear so I think it will work fine.

[wstaab]

Thanks. The help files do not have much info on formulas for transforms. All my histograms from 2.24 are broken in 3.0.
So for a 1 to 4 volt input on a suspension travel sensor same type of formula if 1v is full droop and 4v is fully compressed and say the travel is 4 inches?
Also used a histogram for VE error based on MAF flow in 2.24.
e38 ECU, Custom 2 bar operating system with a real VE table.
MAF g/s / MAFve g/s x 100 - 100= %. Does not work in 3.0. generates strange numbers.
Example 57/61x100-100=-6.557%. This value what would populate cells in 2.24. 3.0 generates numbers in the 50's. Any thoughts?

[schpenxel]

Thanks. The help files do not have much info on formulas for transforms. All my histograms from 2.24 are broken in 3.0.
So for a 1 to 4 volt input on a suspension travel sensor same type of formula if 1v is full droop and 4v is fully compressed and say the travel is 4 inches?

You need to define two points. What's the suspension level at 1V? What is the suspension level at 4V? Using that info and the method I posted earlier you'll be able to come up with a formula.

Also used a histogram for VE error based on MAF flow in 2.24.

I'm not sure what you're saying here. Are you wanting to fix a VE table or a MAF table?

MAF g/s / MAFve g/s x 100 - 100= %. Does not work in 3.0. generates strange numbers.
Example 57/61x100-100=-6.557%. This value what would populate cells in 2.24. 3.0 generates numbers in the 50's. Any thoughts?

Not sure what you're saying here either

It might be better if you just explain what you're trying to do

"MAFve" doesn't make a lick of sense to me.

[wstaab]

2.24 has a PID for VE airflow in g/s and a PID for mass air flow g/s. MAF transfer is tuned in open loop with a wide band. Now the real VE table needs to be tuned. Need to measure the difference in MAF airflow vs VE airflow in g/s and express the value in + or - %.
Thanks for the help.
VE_ERROR.jpg
VE_Table.jpg

[schpenxel]

Most people just tune each table independently with the other disabled

Regardless.. In V3 VE airflow is simply called volumetric efficiency airflow

MAF airflow is obviously mass air flow

Volumetric efficiency is volumetric efficiency...

To make a math parameter that compares the two, make a math parameter that does whatever you need

Here is what it came out to on mine.. PID #'s may be different on yours. Obviously the formula needs tweaking, I just wanted to get the values on one screen, you can add whatever math you'd like it to do
VE.PNG

[wstaab]

Thanks I will try your suggestion.
Cammed LS7 track motor.
MAF transfer was tuned in open loop against a wide band with VE disabled on a load control dyno.
So MAF only is pretty accurate.
Tunning VE only with 1024 cells would take a week.
Easier to log the error between MAF and VE in normal driving and adjust the VE coefficients in the cells.
VE is disabled above 4000 rpm.

[wstaab]

What about logging the 0-5 volt output from one of the MPVI inputs without a transform?

[schpenxel]

Pretty sure you can just use MPVI 1 in a chart or graph but I've never tried it

[wstaab]

Not sure what I did wrong. Oil temp displays 208f in channels when the actual gauge shows 140f and -40f on the gauge in the display.

[schpenxel]

See what the temp is reading, then delete the channel, add it back, and see what the voltage is. Just trying to compare voltage to temp here to see what is going on

The gauges are easy to fix, I'd guess you're just using the wrong parameter. Needs to be accurate first though.

[wstaab]

Problem solved. Your formula was correct.
Layout was imported from a 2.24 gauge file. PID for oil temp sensor (dry sump oil tank) was active in 2.24. and is not active in channels in 3.0.55. Generic channel for oil temp sensor was in the channels list from the import from 2.24. It is not active in 3.0.55 and shows a reading of -40. My user defined oil temp works but the scanner was looking at both the generic oil temp sensor (-40) and the user defined oil temp sensor with the correct reading and trying to combine the two. Deleting the generic oil temp sensor from the channel list solved the problem. Lots of little things in 3.0 that makes the learning curve tough.