Here is an extract from a post I wrote a couple of years back:
There are quite a few differences between the LM-1 and other widebands.
1. The LM-1 can datalog internally, no need for a laptop while logging.
2. You can use an Autometer gauge, voltmeter, or any other analog AFR gauge as remote display. Even the one PLX sells. The PLX gauge is a voltmeter, calibrated to their analog outs, not a full digital gauge. With the programmable analog outs of the LM-1 you can adapt it to anything you like.
The upcoming XD-1 display is fully digital and reads directly the digital datastream coming from the LM-1, independent of analog voltage drifts and ground issues.
3. The LM-1 firmware is downloadable. New features can be added by just reflashing the unit with the included software. New firmware can be downloaded for free from out web-site.
4. The biggest difference is actually the (pat. pend.) measurement principle of the LM-1. Different from all other widebands it does NOT use the pump current as AFR indication. Instead it uses the sensor to form with the LM-1 circuit what's called a delta-sigma analog to digital converter. The difference is that the analog signal in this case is not a voltage or current, but directly the exhaust gas composition. This allows it to:
A: react extremely fast with no settling or overswing
B: be independent of electronic parts tolerances and drifts
C: compensate for sensor drift due to aging every time you do a free air calibration.
D: calibrate for the actual sensor characteristic independent of the factory calibration resistor, which is only correct when the sensor is new.
E: is much less susceptible to exhaust back pressure.
Another advantage of that measurement principle is automatic compensation for 'rich gas loading'. This is an effect most WB manufacturers do not compensate for or even know of. When a WB sensor is operating in a rich gas for a prolonged time (minutes), it's cells 'load up' and slowly drift, requiring more and more pump current. This will indicate richer and richer than it actually is. If the ECU is WB controlled in closed loop, the engine would actually run leaner and leaner to compensate.
The LM-1's measurement principle is not susceptible to that.
The key point to realise, is due to the signal conditioning abilities of the innovate controller, it is totally independent of the calibration resistor found in the sensor. It's a fact that this calibration resistor drifts with time altering the reading. Big money sensors are essentially identical sensors, but with precisely know resistor values - this guarantees accuracy for the first 12 hours of use or so when used for mapping. Doing away with the resistor all together obviously gives significant advantages.
Rick.