Bylund Automotive Support

Hi, I have had a quick look at the Arduino Lambda Sheild which I think is great and well done. But (And I could be wrong) the circuit seems to be missing the 100k biasing resistor to VCC and UN which is only required for the LSU4.9 and not the LSU4.2. There is a brief mention of it in the CJ125 data sheet (9.2 External elements).
Also the formula for the calculation of the lambda value is only applicable for the lean side of measurement (lambda >1)and results will be overly rich for (lambda <1).

I admit I have only had a quick look, as I was looking for dyno contollers when I stumbled across this, and I could be entirely missing something but it looks to be a nice little unit and I liked to offer what may be improvements or not.

Hi Andy, thank you for the interest in our Lambda Shield for Arduino.

Andy wrote:...the circuit seems to be missing the 100k biasing resistor to VCC and UN which is only required for the LSU4.9 and not the LSU4.2. There is a brief mention of it in the CJ125 data sheet (9.2 External elements)

This is actually an interesting topic. Unfortunately I can not give you the complete reasons behind it, but I can give you an explanation of the missing bias resistor. Bosch actually published two datasheets for the CJ125 controller. One named Lambda Probe Interface IC - CJ120/125, I believe this is the one you are referring to. The other one, only named Lambda Probe Interface IC - CJ125. Both are dated 04/2006. Basically there is one design that shows the CJ125 as backwards compatible with the CJ120 where the only adjustment needed is the bias resistor between VCC and UN for use with the LSU 4.9. So I assume that for the LSU 4.2, with that design the CJ120 and CJ125 is interchangeable.

Bosch only refers to the CJ125 specific reference design on their website, and this is what we have based the Lambda Shield on. There is some minor adjustments based on our experience from using the CJ125 in our dyno equipment.


Comparison The seemingly backwards compatible design to the left. And CJ125 specific design to the right.

Andy wrote:...the formula for the calculation of the lambda value is only applicable for the lean side of measurement (lambda >1)and results will be overly rich for (lambda <1)

You are absolutely right. For the educational purposes of this device it gives a good understanding of the functionality in my opinion. A lookup table based on the LSU 4.9 datasheet is a much more accurate way of converting the output voltage to the lambda value.

I believe that there is no such resistor because at bosch document they are at "proposal" section not "necessary".