Useful UEGO sensor info.

http://www.nzefi.com/bosch-lsu-wide-band-airfuel-ratio-lambda-sensors-fail-often-aftermarket-performance-applications/

The key bit:

"Misleading Information

Bosch also tells us the following:

“The sensor must not stay in the exhaust gas stream when it is not heated or when the control unit is switched off.”

Most manufacturer’s of aftermarket wide-band Lambda controllers, pass this point on in their instruction manual in one form or another. It is indeed true that leaving a sensor in the exhaust system without it hooked up to a fully functional controller will kill the sensor. However, doing this for a couple of seconds isn’t a problem. In talking with many of our customers who are the end user’s of aftermarket wide-band controllers, we have often found that many interpret this piece of information to mean that the sensor needs to be fully heated up BEFORE starting the car. In reality nothing could be further from the truth. It isn’t a co-incidence that these customers are the same ones that are killing sensors frequently. After having a brief chat with them and setting them straight their ongoing sensor problems often disappear."

I include myself in those who read Innovate's instructions to mean "power up before startup".

A few points on that from firsthand experience.

1) There are some very specific situations where you want to have the heater warmed up before start. This includes certain cold starts where you are tuning how rich it gets right after start, basically the first 5-10 seconds of a cold start. It also includes engines with start/stop capability. These are more emissions related and not normally important for performance tuning. You still have to be careful of cracking the sensor due to condensation though.

2) Some wideband controllers control the actual heater better than others. I've never burned up a Bosch wideband sensor using the wideband system that they sell (under Bosch's ETAS brand), even without installing a special heatsink and without an engine speed input. This is with sustained torture-test style operation in an engine dyno. I've burned up a bunch of widebands with certain brands such as Innovate. I suspect it comes down to the way the heater is controlled. Obviously Bosch knows how to control their own sensors better than anyone else.

3) OEM applications of widebands have a way to determine the risk of condensation build up. If they calculate that there is a chance of thermal shock, they will start a timer and disable the wideband heater for a while until the condensation blows out of the exhaust.

Thoroughly agree with Ray here in that many sensor failures come down to how the heater is controlled. I have also had poor experiences with innovate products in this regard. When competing in FSAE we used a Link Fury with an inbuilt wideband controller, and the o2 sensor was abused to hell and back; it never failed. When a new one was swapped in, there was no significant difference in readings observed... Not a 100% test, but a good indication that the abused sensor was still reading just fine. The Link Lambda to CAN is another controller I've had great experiences with, and I believe its because it controls the heater accurately, reading engine rpm data over CAN and only heating it up once it detects the engine is running.

Give this a read too:

https://www.ecotrons.com/files/Bosch_LSU49_Tech_Info.pdf

Some good nuggets in there about the heater control.

Just to add in some info, ive been through 2 bosch LSU 4.9 sensors in a span of a couple of weeks:

Both time my sensors failed because the car was on (radio etc), I was doing some tweaking and I came in and started the car, cold exhaust hit the hot sensor boom, thermal shock.

Wire the uego to start only on ignition.

Never ever start the sensor after it has been running hot for long periods of time, that's what ive learnt.

What do you mean by the last sentence? By start the sensor do you mean power on the controller? Why would you be switching the controller on after the sensor has been hot for a long period? Surely the controller would already be on if the sensor had been hot for a long period.

I think he is cautioning against wiring the wideband heater to accessory power or any other configuration where it can heat up the sensor in a non engine running state. If you sit there with the key on and engine off, the wideband heater will activate. Then if you fire the engine, the accumulated moisture will touch the hot sensor and cause a failure.

Thank you for clearing that up raymond haha.

You gotta excuse my explanation skills they tend to gravitate towards the retard correlation at time :D

A clever wideband controller (ECU) will know when the sensor is heated up and then switch off the heater circuit

A dumb wideband controller (uego) will never know and keep heating the cunt forever!