MAF sensor calibrations to pipe size

[turboice]

I have done some data logging of a Denso MAF sensor comparing the voltage signal and the ECU's interpretation of airflow to that voltage signal. The function appears to exponential or log. Does anyone know a good source for sensor specs in different pipe sizes or a way I could estimate the airflow of the existing MAF in a different pipe size?

Ed.

http://www.turboice.net

 

[mpg9999]

You are correct, the Voltage is a non linear representation of airflow. I know nissan uses VQ maps to convert the MAF voltage to a percentage of flow. Here is a very good explanation of it:

http://www.sr20forum.com/showthread.php?s=&threadid=25488

 

[turboice]

That is a very informative post on the functioning of ECUs that are primarily MAF driven for fueling and the MAF themselves. While I knew a good deal of it already it will be a good source for me to fill in the gaps and it is complete and comprehensive rather than dispersed.

However, and I may not have asked the question properly, it did not get to the question I had at hand.

If you have ever seen a VQ map laid out we can tell immediately it is not linear and most likely logn given the its attempt to measure air flow based on the cooling of the hot wire and so many natural phenomena being a natural log function.

My car comes with an equivalent to the VQ map which is based on the IAT, MAF sensor calibration and the size of the pipe with the MAF housing lets just say for arguments sake it is 52mm i.d. Obviuosly if I replace the pipe with the map housing with a 76mm i.d. the VQ will read the resulting voltages as a mass of air signifcantly less than is actually passing the MAF. If I can not edit the VQ table or do not know the correct values to enter into the VQ table because I have not bench flowed the MAF output in the new size tube - but I have a way to apply a factor to the voltage would I be able to trick the ECU and its VQ table?

Is the nature of the MAF such that at a given voltage say 1.0V that the appropriate air flow is linear to the cross section of the MAF housing pipe? In my example above the cross sectional area is a bit more than twice as large - would applying a factor of 0.468144 to the MAF output voltage result in the correct airflow being selected from the VQ table? I understand that I couldn't necessarily report more air than the VQ table would hold since my new max voltage input to the table would be 2.34.

The question is not asked so much as to directly apply the idea but more to further understand the operation of hot wire MAF sensors and the effect of pipe size on calibration. Is the increase in airflow past the MAF linear to the cross-sectional area of the pipe? Maybe my question is really one of gas physics than MAF operation...

Ed.

http://www.turboice.net

 

[turboice]

Alternatively if I could rewrite the VQ table in my example above could I apply a factor of 2.136 to the existing air flow values based on voltage inputs as a recalibration of the VQ table to my new pipe size?

Ed.

http://www.turboice.net