I'm working on a 2005 S-type 3.0. It came from another shop where the Bank 2 Upstream Oxygen sensor was replaced in response to a P0051 (B2S1 heater control circuit low).
The code was cleared but returned right away. They told the owner they did not want to get any deeper and did not want to throw parts at it.
I confirmed proper sensor location: driver side (US) upstream, and quality of the part: it is a Denso 234-9030. I measured the resistance of the sensor's heater element: ~1 ohm
which is correct. B1S1 measures the same. Relay and fuse are good; there is power to the heater element. I checked the wiring to the ECU which completes the ground side
by variable pulse width: good. I also loaded the circuit with a sealed beam. Having accessed the ECU plug for testing, it was easy enough to pull the ECU for a closer look: all OK.
I sent the ECU to a rebuilder for testing and assessment. They sent it back saying no problem found. I swapped B1S1 and B2S1 but P0051 returns after second start.
I have been told that a new O2 sensor must be initialized or configured to the car. Does anyone know if this is true? I believe this was the case with the X300.
If it is required, is there a device that will do it other than WDS or IDS? I have both but neither is presently working. My WDS died years ago and IDS fails to boot up these days.
I'd appreciate any advice. Thanks in advance. Patrick

 

No config needed.

What's "sealed beam"?

I can't see anything you've done wrong & it feels like it ought to work so I'm missing something (same as you, sorry).

The resistance (1ohm etc) will vary I expect as it heats up.

 

Never heard of anything like that. The only thing I've seen is if you replace a sensor that is out of whack, the computer will have adjusted fuel trims to compensate. Install a new sensor but the computer doesn't know that, and several drive cycles may be required to learn the new input data and hopefully return fuel trim to normal. You can speed up the learning process by disconnecting the battery to reset everything to preprogrammed initial values. In essence, you're skipping a slow relearn and starting over from default. Typically, any adjustments from that point are minor and accomplished quickly.

Back to the current fault. For giggles, try swapping the two downstream sensors and see if the fault follows. See if this hunch makes any sense. Wiring diagrams here, see figures 03.1 for the sensors and 01.7 for power distribution:

http://www.jagrepair.com/images/Auto...cal-2005on.pdf

All four sensor heaters receive power via relay R2 under the hood. If this relay had a problem, it should affect all 4 sensors.

Downstream of the relay, fuse F20 feeds the 2 sensors on bank 1. Fuse F35 feeds both sensors on bank 2. I'm wondering if the downstream sensor on bank 2 is hogging too much power on this circuit. It's not bad enough to set a code specific to the downstream sensor, but is affecting its partner on that circuit.

It would be easy enough to rule out. Swap the two downstream sensors. It won't cost anything but a little bit of labor. They are out in the open and easy to access.

Another thought is to inspect the sockets in the fuse panel at F35. Look at the prongs of the fuse for discoloration caused by overheating. A loose socket can cause arcing, without blowing the fuse. Break a spare fuse in half and check that each socket grips well. Maybe some extra resistance here is dropping voltage to the two respective sensors, but the upstream sensor is the only one to set a code.

One last thought is to backprobe the connector at the sensor. Turn on the ignition and check for voltage under load. If checked with the plug disconnected, the circuit is obviously unloaded. A marginal connection (such as a loose or corroded fuse socket) can let just enough power through to show normal voltage with a meter, but fall down under load. I've seen this trip up many mechanics, self included. Years ago, I learned the hard way that a good voltage reading on an unloaded circuit is not conclusive that all is good.

Keep us posted.

 

Well it looks like I was on the right track all along.
After getting a clean "bill of health" on the ECU I sent out for testing I proceeded to double-check everything.
I even monitored the current through both fuses (each fuse feeds the O2 sensor heaters on a bank). The current
was the same on each bank.
After many hours of this, I still suspected the ECU. I sourced a used one with identical part number, plugged it in,
reset immobilizer, and fired it up. No check engine light, no codes (other than P1000 indicating incomplete monitors)
even after several starts. Previously it would log P0051 on second start after clearing. I ran two drive cycles to
complete the monitors (nearly 50 miles) and passed inspection. Replacing the ECU fixed it. Car is back to owner.
Now the question is: does anyone know of a company that can reliably test and repair such a module in the future?
I used to send them to BBA Reman near Boston but they seem to have closed. There is ModuleMaster in Idaho
but when I called there early on, they said they "didn't have much luck with those modules".
It also appears that it s not necessary to initiate a replacement Oxygen sensor. This was done on the X300.
Anyway, thanks again for the advice.

 

Nice work! I generally try to discourage people from automatically blaming the computer, but it was obviously the culprit in this case. And a YUGE thanks for coming back with an update.

Could you please elaborate how you reset the security system? I don't remember all the details, and maybe it's only applicable to earlier models, but I thought once an ECU was installed on a vehicle, it could not be swapped. Something about VIN-specific programming that could not be overwritten, and the security system would not allow a mismatch. You could install a brand new (never programmed) unit, but couldn't transfer a used one. I'm sure somebody will be along quickly to correct me...

 

Hello Karl,
The immobilizer reset is a function available on an IDS/SDD system. I have one that I bought used a few years ago.
It's a bit clumsy to use but that may be because I use it infrequently.
This is only the second time I have swapped out an ECU with a used unit. Previous time was in a 2006 XJ8 about 5 years ago.
I don't remember what the specific problem was, other than it wouldn't start, but I do remember the feeling of relief when it cranked and fired up.
I did read some previous posts related to the issue of VIN specificity. I wish I could add some certainty to the topic but I only have first-hand
experience with these two cars. In these cases there was not a problem beyond immobilizer reset.
Even the terminology can be vague and misleading: configuration, reconfiguration, programming, initializing, resetting, etc.
These terms have precise definitions but are used to describe various, sometimes overlapping, operations.
Then there's the variety of tools to communicate and manipulate various modules on a given year, make and model.
I agree that being very sure you have eliminated all other possibilities and condemning a module can be difficult (and expensive).
That's another good reason to have the original tested, if possible and reliable. This has the additional advantage of not having to worry about the VIN.
Thanks for the interest and advice. Hopefully someone will benefit from some of this information in the future.
Patrick