Connection for Bosh Sensor?

[MMGT1]

Anyone got a connection to get a NTK WB sensor? Hoping someone on here knows where to get a good deal on one.

Thanks guys,

Paul

P.S. Have the controller w/ a BOSH sensor, would like to upgrade

[eficalibrator]

Going from the NTK wideband sensor (L1H1/L2H2) to a Bosch (LSU 4.x) is NOT an "upgrade". Quite the opposite in my experience.

[MMGT1]

Shit Greg, I posted that backwards!!!
I "HAVE" the BOSH sensor and would like to "GET" a NTK sensor

oh man.......

[5_Liter_Eater]

See some infor here:

http://www.wbo2.com/lsu/sensors.htm

[MMGT1]

Mine is the BOSH LSU 4.2(the 7 057 sensor)

Thanks for the info Bill

Bill, from what that site is claiming, I can get the BOSH sensor #13246 and it's actually a NTK sensor??

[5_Liter_Eater]

I'm not really sure.

[MMGT1]

Found one for $160(the NGK/NTK #24302). Anyone have a better price?

[5_Liter_Eater]

Whats the reason for the switch? What WB controller are you using?

[MMGT1]

The NTK AFX package. I'm not trusting the sensor because of my low end trims. I have a "fairly" new NB, and it's saying I'm off by 3% when I tune with the WB, then re-enable the NB. I do believe that the NB is accurate at 14.7 and would like to try it out to see if this is really the case. I've now tuned the trims @ stoich using the NB.
If this is right, would my PE then be out by 3% having tuned it with the WB?

[eficalibrator]

Keep in mind that the NGK AFX sensors displays "14.56" at stoich, NOT 14.68 like your PCM does. Both are still lambda=1. I usually convert both to lambda units to avoid this confusion in the first place.

[MMGT1]

Greg here is my PID:
((([PID.2811]/.714)+9)-[SENS.121])/[SENS.121]*100 for error

and [PID.2811]/.714+9.02 for WB through EGR

In the vcm my stoich is factory set to 14.70020<<can this be changed to 14.56 to match what the WB is looking for?

[EC_Tune]

Don't change Stoch in your calibration. Just realize that the NTK uses 14.56 as stoch. If you're outputting Voltage vs AFR it shouldn't matter anyway. 14.56 AFR on the NTK=14.56 AFR

[SilverG]

Don't change Stoch in your calibration. Just realize that the NTK uses 14.56 as stoch. If you're outputting Voltage vs AFR it shouldn't matter anyway. 14.56 AFR on the NTK=14.56 AFR

Keep in mind that the NGK AFX sensors displays "14.56" at stoich, NOT 14.68 like your PCM does. Both are still lambda=1. I usually convert both to lambda units to avoid this confusion in the first place.

Let me see if I have this straight. Assume 93-octane gas with no ethanol. Greg’s post says that if I were to monitor my CL NBs with my NGK AFX monitor instead of seeing 14.68 I would see 14.56, correct? ie both the NBs and the WB are reading stoich.

But EC_Tune says that if I use the custom PID formula AFR = 1.4 * Vout + 9 that should read 14.68 in HPT, correct? and the Voltage should be close to 4.06v for stioch?

[SilverG]

I know this question isn't really in the right section but can I get some input?

TIA

[GregRexAdams]

You have to adjust for it.

The AFX reads lambda but displays it as AFR. 14.56 on the unit is lambda 1.0, or for pump gas 14.68. So you have to adjust HP Tuner to show it as .12 higher in the scanner suite than it's actually reading on the display.

[eficalibrator]

All widebands are really lambda sensors, not "AFR" sensors. They just multiply the lambda reading by some number (typically 14.64) to give less experienced users something they're used to seeing. The real solution is to just work in units of lambda for both the ECM side (commanded ratio) and wideband side (measured, delivered ratio) and calculate any error based on these instead of letting the stoich point nomenclature confuse you.

As long as the calibrated stoich value in the ECM matches the chemistry of the fuel in your tank, you shouldn't have any trouble with straight units of lambda.

[SilverG]

As long as the calibrated stoich value in the ECM matches the chemistry of the fuel in your tank, you shouldn't have any trouble with straight units of lambda.

Greg, can you explain what you mean by this statement?

I am working on PIDs and histograms to try to work with Lambda. Too bad GM uses equivalence ratio for PE, yet another calculation.