So I just upgraded to 3.6 and lost all my stuff. I know at one point there was a ground offset table someone made (Used to be a histogram) where it monitored a few parameters and calculated the actual ground offset and voltage that was an average i believe. I cannot find it nor recall the exact method. Can anyone help me out and perhaps point me to it?
Thanks
Dave
Last edited by TTVert; 08-20-2017 at 12:01 AM.
Anyone? It isn't an excel spreadsheet either.
Dave
You have to be careful with ground offsets. They are, generally, a result of the high-current heater load of the wideband sensors. That load changes with EGT and exhaust gas flow. So, an offset that is good at idle is, probably, not right at WOT, etc.
Be that as it may I need to do something as most have ground offsets unless they're grounded to where the ECU is correct?
Dave
All widebands that do not have a differential output suffer from ground offset error.
Below is a MILD example of the offset error's sensitivity to RPM/EGT/LOAD. The yellow trace is, more or less, proportional to the voltage offset.
So, you can see that setting it so that it is right at idle will be way off under load.
heater-load.JPG
That looks nasty. Its been a while since I had to setup my WB, but if I remember correctly the voltage offset was used to try and calibrate the reading in the scanner to match the gauge reading. The readings would be different if a different ground source was used for the MPVI then the gauge. I guess the one thing I would think some people might over look if they are using the same ground source for both is the voltage drop across the wire to the MPVI (wire to long or thin).Originally Posted by dr.mike
Feel free to correct me if I'm remembering incorrectly Dr. Mike, you are the O2 Doc. On a side note serial or CAN are good options, unfortunately I have not been able to get my Innov. LC-2 to work via serial in HPT 3.6.26 (works in Logworks).
Serial, or, especially, CAN, are a much better option that 0-5v analog.
The point of the graph is that, adjusting the offset so that the logger and gauge read the same at idle does NOT mean that it will be right under load ( when you need it to be right ).
The voltage offset is based on the high-current draw of the sensor's heater; which varies from 0 to about 3 amps. The wideband controller has to keep the sensor temperature constant at about 700C. If the EGT, at the sensor, is higher than that, the heater SHOULD get 0amps. If its colder and low velocity, you might get 1amp. If the exhaust flow is high, the heater needs a lot of current to maintain temperature.
If you are pulling 1 amp at idle and 3 amps at WOT, the offset will be about 3X higher at WOT. Even though the logger and gauge will match at idle.
The LC-2 is a particularly hard case because it shares the same ground wire for the heater and 0-5v signal. So, grounding is at the ECU can't compensate for the difference.
Good info....That is why our Tutorial tells the reader to run in closed loop, while logging a user defined parameter Stoich AFR/WB AFR which defines the error between commanded stoich that the PCM is striving to attain vs. the actual WB output. Drive at like 20-30 mph in a high gear (50-70 kPa) to come up with a nominal offset which is a compromise between idle and WOT. Use this error to adjust the WB Transform offset value to get the two to match. Not perfect, but should suffice.
Yes the Serial and OBD II strategies are the way to go today as long as your vehicle/OS and Wide band are supported....
Ed M
2004 Vette Coupe, LS2, MN6, Vararam, ARH/CATs, Ti's, 4:10, Trickflow 215, 30# SVO, Vette Doctors Cam, Fast 90/90, DD McLeod, DTE Brace, Hurst shifter, Bilsteins etc. 480/430
ERM Performance Tuning -- Interactive Learning ..from tuning software training to custom tunes
HP Tuners Dealer- VCM Suite (free 2hr training session with purchase), credits and new Version 2.0 turtorial available
http://www.ermperformancetuning.com
http://www.facebook.com/ERMPerformanceTuning
[email protected]
Sorry for the hijack.
My wb02 2C0B has a differential output but loses accuracy above 4V. Is there any literature that states how to perform an offset strategy at a 'knee' ?
Currently I have programmed the 0 - 5V to run 0.8 - 1.1 lambda, thus it is accurate where I really need it, and lean is just lean - but I wouldn't mind logging it accurately.
Also, how receptive have HPT been to third party serial strings ? - the wbo2 does spit out streaming data on an RS232 port
The output amp is, likely, running out of "headroom". i.e. it only has 5.0v to work with. So, if it needs more than that to compensate for the offset at > 4.0v, it will run out of gas. The "knee", is likely the limit of the output swing of the amplifier at the "rails". It is, usually, temperature and ( electrical) load dependent.
If you can, program your full range to 1v - 4v. That should give plenty of headroom to stay linear. 0.5v - 4.5v is usually good enough.
Reply With Quote