Questioning my calculation for AEM Wideband

[dykz34]

Hi guys,

I've got the AEM wideband in my car (model 30-4100) and I'm questioning the accuracy of the equation I'm using. I am setting everything up to track Lambda. In the manual, http://www.aemelectronics.com/Images...%2030-4100.pdf , on the last page, it is stated that the gauge range is 0.751 to 1.143. It also states that there is 0-5V output. So, my equation should be the following, right? =(V/(5/0.392))+0.751

I believe that to be correct, however, when I disconnect the sensor from the gauge (forcing the gauge to put out a steady voltage), my scanner reads 2.39V when it should be sending out 2.35V per the manual. This is odd because I've never read about anyone having a voltage increase across the lines, only a drop...

Secondly, when I connect the oxygen sensor back up to the gauge and turn the ignition on with the sensor in free air, the highest the voltage reading will go (in HPT) is 4.31v. This leads me to believe that the range of the device is NOT 0-5v, rather 0-4.31v. Am I thinking correct here? If so, should my equation be =(V/(4.31/0.392))+0.751 ?

Appreciate the input!

Thanks,
Dave

[dykz34]

After what could be a lightbulb just going off, I may have figured it out. Many guys use 10-20AFR for gasoline, but if you divide that by the gasoline stoich (14.64) the range of the device changes to 0.683 to 1.366 lambda. This then changes my equation to =(V/(5/0.683))+0.683 .... which matches the AEM gauge perfectly with the oxygen sensor unplugged!

The only two questions outstanding are the voltage range of the sensor and the voltage gain thru the scanner. Should the unit output closer to 5V when in free air? It seems odd that it would top out at 4.31v. And has anyone seen a voltage GAIN when scanning instead of a voltage DROP?

Thanks again!!!

[5_Liter_Eater]

This manual is slightly different from the one I have seen. I have used this one: http://www.schnitzracing.com/manuals/AEMWBK.pdf. That one actually lists the lambda output along with the other fuel types and while in the specifications at the end it says the lambda range is .751 - 1.143, that table on pages 5-6 shows a lambda of .683 at 0V and 1.365 at 5V.

That is weird that the voltage is higher than expected with the sensor disconnected. no explanation but I would subtract .04 to your offset to account for it. 4 hundreths of a volt is no biggie anyway.

HTH

[dykz34]

This manual is slightly different from the one I have seen. I have used this one: http://www.schnitzracing.com/manuals/AEMWBK.pdf. That one actually lists the lambda output along with the other fuel types and while in the specifications at the end it says the lambda range is .751 - 1.143, that table on pages 5-6 shows a lambda of .683 at 0V and 1.365 at 5V.

That is weird that the voltage is higher than expected with the sensor disconnected. no explanation but I would subtract .04 to your offset to account for it. 4 hundreths of a volt is no biggie anyway.

HTH

Interesting that there are multiple installation instruction PDF's floating around. Also even more interesting that the one from the AEM website is different than what vendors are posting. The one you linked to confirms my thoughts. The gauge range is larger than specified in the Tech Specs. Even the picture of the back of the gauges in the two manuals are different! The Schnitz Racing manual matches my gauge, the AEM manual does not.

Thanks for your help!

[pontisteve]

Even if the gauge only put out 4 volts, it's probably still based on the traditional 0-5v output scale. So you should always figure 0-5v.

I have seen voltage increases instead of drops. Since HP Tuners doesn't really leave us with much room to modify the formula, it's best to avoid these ground issues in the first place. Since the HPT interface has analog input and ground inputs, the wideband is being analog grounded to the same place as the PCM. This should remove most voltage drops. If not, perhaps grounding the analog ground of both the HPT device and the wideband to the engine block all by itself would help. I would probably just live with any offsets, since the HPT method is sort of built this way.

Get the range from the wideband manufacturer, in lambda. If that's not possible, you have to use the AFR range given and divide that by the stoich that the wideband manufacturer used in their conversion from lambda to AFR, not your own definition of stoich.

You could also try running an 8 or 10 gage wire from the PCM ground location to the engine block. This would reduce the PCMs ground offsets, if any, and therefore the wideband offsets would be reduced too.

[LSxpwrdZ]

The manual I've seen show's that 0v=10afr and 5v=20afr with the stoich ratio of 14.65:1. Using that info that calculates to 0v=0.683l and 5v=1.365l

For lambda on the AEM you need to use this function in the user defined custom PID:

Divisor: 7.325
Adder: 0.683

That should get your lambda reading's accurate in the HPT Scanner. Personally if you have trouble with it ditch the wideband all together cause they aren't very consistant!