Ground offset

How can you find out if analogue input to ecu is impacted by ground offset?

if the sensor is wired into the ECU it'll be tied into the sensor grounds so you shouldn't have an offset issue.

The main sensors I see offsets are from wideband lambda gauges, what you'll see if different values from the gauge in the software, once you bridge in the gauge ground to the ECU sensor ground as well as it's own ground it zero's out the offset

Atm i have prosport gauge and power and analoge output is all through ground of link but I have a discrepancy in read either due to wrong table provided by prosport or ground offset

it is lsu4.9 gauge, is there a general table for this sensor?

The sensor is irrelevant, the output is all done from the gauge.

Normally ballpark you'd expect 0-5v giving you 7.35-22.39 AFR (figures from an Innovate LC2)

I suspect the chap is looking at single wire analogue sender' connections, with the sender being earthed/grounded to chassis. This results in a potential offset between the sender ground and the ECU reference ground.

There is a control box which has 4 connections. One goes out to gauge, one power in, one out to sensor and one output to ecu

output to ecu has 3 wires, ground , wide band and narrow band

gauge ground as well as output ground are on the ecu expansion loom .

i have total of 7 gauges all powered from ecu ground but fed by power from battery direct and fused to 5 amp so it won’t fry ecu

i used calibration table prosport provided and it was off little, i applied 0.9 correction to entire value ad now gauge and ecu reading match up

to me this sounded like either calibration table is wrong which is setup using prosport table or the gauge had offset ground and i fixed it by applying 0.9 to calibration table across all values

do you guys not do cal table and use value of 0.5 v on ecu? How does ecu know what volt means what afr?

link below shows wiring and table

[url=https://cdn.shopify.com/s/files/1/0310/8198/8236/files/60EVOWBPREM.pdf?v=1607699861]https://cdn.shopify.com/s/files/1/0310/8198/8236/files/60EVOWBPREM.pdf?v=1607699861

Here is the wiring diagram of mine

All grounds fed from ECU protected by 5amp fuse. I keep reading and some state ground wire can be smaller than positive but that doesn't make sense as flow will be through both so unsure really. ecu expansion loom is 20 awg I think. see link below it states ground transmition is 14 amp and power is 1.5 amp for same 20 awg. I need to read more to understand why ground needs a lot thinner wire

https://www.powerstream.com/Wire_Size.htm

This is expected when using a wideband without a dedicated sensor ground. Good on you for recognizing the issue rather than tuning as if the data was correct.

CAN widebands provide accurate data without loss of accuracy since the information is transmitted digitally.

A middle ground options is an analog wideband with sensor ground i.e. Ballenger AFR 500v2, AEM xline.

I guess what I was asking is whether the cal table on prosport website is not correct. they use bosch lsu 4.9 which I understand majority of vendors use it including link and aem.

My skyline was setup by tuner and he is using calc table 4 and when I go there rather than a table, he has provided minimum and max value for lambda and voltage and let ecu work it out, how does ecu know voltage means what afr means what voltage unless it uses a liner scale of every x volt means x lambda change. See screenshot of how my skyline is setup with aem wideband

when I used the calc table the numbers are off even though I am using ground from ECU which is a wire available along with the wideband signal. i guess prosport left this to avoid ground offset( see table below) but when I use that calc table, the gauge and ecu number don't match so I asked if calc table can be wrong or is there a bosch 4.9 lsu calc table I could check. I can't seem to get much info on that online.

Do you have any thoughts on the article I included earlier which seems to suggest ground wires can be smaller awg gauge and carry more load than positive. I don't understand how this can be done as DC is a circuit so both positive and negative should be same size but maybe I am wrong.