SM-AFR Wideband O2 Module Issues

Hi all,

Finally had the chance to actually try tuning on a dyno last week (rather than street tune), but to my dismay it ended up as a waste of time when I realized my wideband readings were all wrong.

I have a PLX SM-AFR module with a Bosch LSU 4.9 sensor and a DM6 gauge, which has an analogue 0-5V output to the Haltech Elite 1500. Car is an FD3S RX-7 which I just switched to E85 (hence the dyno time). Air pump has been removed.

Symptoms were that the sensor would hesitate to read anything besides 14-16 petrol AFR while the car is under load. Under full boost and high rpm, the dyno wideband (which was stuck in the back of the exhaust) would be reading 10.5 while the PLX/Haltech would still be reading 14.5. It'd read around 14.7 while idling (and rotaries definitely don't idle so nicely that lean).

The strange thing was that under a fast, transient throttle blip, it would actually briefly read a rich value. Under decel cut, it would read lean. So clearly, the sensor is working to some degree.

The Haltech and the gauge were aligned and in sync (so it's nothing to do with the wiring between the two).

I replaced the old wideband with a new LSU 4.9, but the same problem persisted. So it's not the sensor either.

The wideband setup has worked perfectly fine for 1.5 years, but the car sat around for a while until I did the fuel system and some wiring clean up in the engine bay (but not any wires connected to the wideband).

Any idea on what these symptoms indicate? Is the SM-AFR module broken? Is it a wiring issue? Exhaust leak? Something else?

Did you check the 12v supply and ground to the module? Maybe a voltage drop could cause this issue.

G'day Rahul.

First thing to be aware of, is that when you're running E85 fuel, the stoichiometric ratio for complete combustion is ~9.8:1, not 14.7:1 as for pure gasoline. It's one of the reasons we like to talk in terms of Lambda usually, as Lambda 1.0 is stoich, no matter what fuel is being used.

Was the value in the Haltech matching what was being shown on your DM6 gauge?

You've already performed the first test I also would've, which was to try another LSU 4.9, but I think you're going to have to pull the unit from the car to test it, to isolate whether its a problem with the wiring in the car, or the unit itself. Have you got any other vehicles you can test it on?

Particularly with wideband O2 controllers, voltage offsets are a real issue, as most of them generate their analogue outputs referenced to their power ground, which may be at a different level to the sensor ground of the ECU, which is what it gets compared to when the ECU makes its reading.

@Reinout - will be checking all wiring issues this weekend. I wanted to make sure that these symptoms weren't related to the module as I'd need to order one to ensure that it arrives in time (car is being raced at Vic Time Attack next week).

@Zac Perston

Yeah, aware about the E85 stoich level. Currently have the gauge configured for petrol AFRs still since I'm a terrible, terrible person and am not used to working in lambda yet. I will switch to lambda readings soon, just wanted to work with what I'm used to in terms of setting AFR targets, etc.

"Was the value in the Haltech matching what was being shown on your DM6 gauge?"

Yes.

"Have you got any other vehicles you can test it on?"

Not really, but I can probably find a friend willing to let me test it.

"Particularly with wideband O2 controllers, voltage offsets are a real issue, as most of them generate their analogue outputs referencedto their power ground, which may be at a different level to the sensor ground of the ECU, which is what it gets compared to when the ECU makes its reading."

Given that the DM6 and the Haltech match each other, would this be the case?

Thanks for the ideas all.