AFM Temp Units??
 

I logged some temperature data during a recent dyno session, transferred the pdb files to an SD card and reviewed the data in RX7PLViewer.

I noticed that, while everything appeared in US units (as desired) in PocketLogger on the PDA, the raw data exported with the Java app were a bit different.

Most data values were converted to their metric equivalents (e.g. airflow was in kg/hr (raw) vs. lbs/min (pda)), which is fine. But the AFM temp units didn't seem right. They didn't match up to the PDA when I converted them to Fahrenheit (assumed they were Celsius) and they were trending in the wrong direction. The greater the *F shown on the PDA, the lower the unit shown in the raw data. It was a kind of reverse temperature scale.

I did a linear fit of the raw data vs. expected values for AFM temp, IAT, and Pin 3F (Zeitronix wideband). The fits gave me the expected equations for converting *C to *F (IAT) and the correct transfer function for the wideband AFRs. But the AFM Temp gave a fit that doesn't look like any temp conversion I know of (see table/chart pic, or zipped xls attached).

In addition, if I plotted AFM temp and IAT against time, they were trending in opposite directions, which didn't make much sense. (green curve is engine RPM for three successive dyno pulls).

This is for an RTek 2.0 installed in N351 ECU for an s5 n/a application.

Anyone experience similar issues with AFM temp reading?

Yeah, this looks like a bug. We changed the AFM air temp index for the N351 and the PLViewer probably never got updated.

How old is your Rtek? Have a serial number?

Installed 11/12/2009. Version 1.1.1. S/N = 1648.

Thats the latest version. Are you sure you have the latest version of the viewer?

2.05g
 

I have versions 2.05g and 2.05f. Both show the same funky units.