I am putting my RCP into a Diesel race car. I need to verify that the Stock EGT's that I can log with my lap top are accurate, the tuner thinks they read low, which means he is hesitant to turn the fuel up any more, and I need to push the limits with this car, it's in STU and gets the smack down from the Bimmers.
Anyway, I ordered this sensor today, it'll be here Wednesday and I want to be ready for it:
http://www.aemelectronics.com/Images/Pr ... 0-2050.pdf
So, here we have a chart of Temp over resistance, pretty straight forward. So I add in a pull up resistor, I get it into the range of 0-5V, but raw voltage data does me no good. I need to know what the temperature actually is. I'm blanking on that part, what math do I do in order to have RCP spit out a temperature instead of a voltage? I know that I have done this before with other systems, I just can not remember how to calculate this.
Thanks for any help!
EGT I get the pull up resistor part, but. . .
Moderators: JeffC, rdoherty, stieg
OK, I think that is not actually the right question, I can figure out, what the voltage is that corresponds to the temp. That's actually pretty easy. But, So I have this Voltage number, and then I have a temp number is the temperature value what I enter into the"Scaled" row underneath the votlage? Then when I get my data I get "scaled" data based on the voltage values that I entered? Sorry, I think I'm kind of thinking my way through this as I write the post, 
-
dimondjack
- Posts: 101
- Joined: Tue Jan 15, 2013 1:37 pm
I think the answer is "yes".
You need to enter the voltage that the RCP will be reading and the corresponding temperature in each of the five boxes. Pick values that will create a good trend around the temperatures you want to measure, as it will yield the greatest accuracy (use the two end values as "bookends", and the other three values to create precision).
If you feel like doing more math, you should import your data to Excel and create a transfer function which will calculate the exact temperature. This will be significantly more accurate than a 5 point linear approximation.
You need to enter the voltage that the RCP will be reading and the corresponding temperature in each of the five boxes. Pick values that will create a good trend around the temperatures you want to measure, as it will yield the greatest accuracy (use the two end values as "bookends", and the other three values to create precision).
If you feel like doing more math, you should import your data to Excel and create a transfer function which will calculate the exact temperature. This will be significantly more accurate than a 5 point linear approximation.
-
dimondjack
- Posts: 101
- Joined: Tue Jan 15, 2013 1:37 pm
A transfer function is a fancy name for a look up table which converts one value to another. It can get a lot more complicated, but for RCP it works like this:
Lets say you fill out the calibration table to look like this:
Value 500 750 1000 1250 1500
Voltage 1V 2V 3V 4v 5v
These values would be based on what you calculate based on your circuit. You want to pick values where the middle three "values" are close to the temp you want to read.
So, lets say that RCP reads 3.5 volts. It looks at the table you put in and does a linear interpolation (http://en.wikipedia.org/wiki/Linear_interpolation) between the point above and the point below. The formula would be:
((1250-1000)/(4-3)) * (3.5-3v) + 1000 = 1125
Lets say you fill out the calibration table to look like this:
Value 500 750 1000 1250 1500
Voltage 1V 2V 3V 4v 5v
These values would be based on what you calculate based on your circuit. You want to pick values where the middle three "values" are close to the temp you want to read.
So, lets say that RCP reads 3.5 volts. It looks at the table you put in and does a linear interpolation (http://en.wikipedia.org/wiki/Linear_interpolation) between the point above and the point below. The formula would be:
((1250-1000)/(4-3)) * (3.5-3v) + 1000 = 1125