Possible wide band issue

I have been targeting .95 lamda at idle, it idles fine and is not erratic but seems to smoke pretty bad. I'm using the bosch lsu 4.2 sensor wired directly into my ecumaster classic. Also it will occasionally get stuck and will not change lamda. When it is working the reading varies between .01 to .02 lamda at idle. The sensor is calibrated. Is this characteristics of a bad sensor or maybe another issue. I have a dyno day scheduled and do not want to get there and have issues.

The fact the lambda reading 'gets stuck' is a little concerning. Typically you can expect the lambda to move slightly as the engine runs and this is normal even when it's properly tuned. Normally when the sensor is on its last legs you tend to see it go into error or the reading to drop off completely rather than it just becoming slow to change or freezing. It's always a good idea to double check if you're in any doubt though with a separate wideband. You'd be surprised how much variation there is between different widebands even when the sensor is brand new.

Any chance you may have poisoned the lambda with a lead containing fuel, or a silicone sealer on the inlet manifold flanges?

Any chance you're using a ground or voltage source that is affected by something else using the same source/ground and introducing an offset?