I wanted to verify the accuracy of the NGK gauge display against what HPTuners reported. I am using the NGK AFX wideband with the NTK sensor, not the typical Bosch LSU4 adapted to the NGK. This sensor is supposed to be more accurate and used with the more expensive ECM LambdaPro (from what I have read). I rigged up the HPT to read the voltage output a well, I only logged AFR Cmmd, AFR, AFR Error% and Raw Voltage. I also used the VCM Bi-Directional controls to target an AFR from 8-17. My test vehicle is tuned to the tee, fuel trims are 0% on average throughout the rpm range. Interesting to see that the NTK sensor either wouldn't read higher then 15.8 or the vehcile used was unable to go any leaner. Also the NTK would read as rich as 9.8, but that was the richest the vehicle in question would go (pulse width limit ). I allowed at least 10 seconds for the fuel to stablize before recording the gauge display data. It actually was off no more then 1% from what HPTuners was logging. Pretty impressive sensor, considering the Bosch LSU4 (using an NGK, AEM, and Innovative controllers) has given me results varying largely on the extreme ends of the range, while being .2 off in the 14.5-14.7 range.
Next time I have time with the "test vehicle" I will have to log pulse width and determine if the vehicle couldn't go leaner/richer then commanded ... or if the sensor couldn't report a lower/higher voltage then specified.
Disclaimer: This viewpoint is solely based on my experience. It is not intended to be a scientific analysis of the Bosch LSU4 versus the NTK sensor. This was by no means a scientific experiment, but a simple study. Your results may vary.