Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Calibration or Verification of AF meter with Bosch LSU 4.9 sensor

Understanding AFR

Forum Posts



Tech Articles

Discussion and questions related to the course Understanding AFR

= Resolved threads


Good Evening HP Academy.

I have a few questions and thoughts about the calibration and verification of lambda and / or oxygen sensors.

Over wide industry applications when test instruments are being used - typically these will come with a factory calibration certificate commonly valid for a year. Weather this be a process calibrator to test process variables (pressure / temperature) or electrical meters to test electrical circuits, even calibrated mechanical tools such as torque wrenches.

However it is common industry knowledge that measurement accuracy will “drift” or measurement uncertainty will increase over time. Which prompts the need to have this equipment verified and any calibration trims applied where necessary to meet a specific threshold.

It seems to be common industry practise in the automotive performance industry that while engine builders are expected and trust calibrated equipment like torque wrenches (and not their elbow)... tuners do not seem to have much care in the world about verifying their oxygen sensors... and seem to take an attitude of “when it fails I will just replace it” - without any concern whatsoever ever about the impacts of measurement accuracy drift of the equipment and the impact it can have when tuning an engine. Sure some will also go by the engine note and “feel” as well regarding what they believe the engine wants. However there is some part to me that thinks - is this somewhat negligible ? It seems a wide variety of industry’s regulated the need for calibrated test equipment... but not so much in this specific circumstance.

Obviously some lambda meters can offer a free air calibration and while this does offer a correction at atmosphere it does not factor in any measurement linearisation drift over the whole measurement spectrum. Ie the meter may show 20.9 at atmosphere but have drift at the rich end of the spectrum.

Does HP academy know of any suppliers that can produce certified petrol exhaust test gas at a variety of mixture. Say like a 9:1 and a 14:1


any thoughts about how to make a inline component between the sensor and the lambda controller that would allow to apply a measurement correction or trim to the sensor signal prior to hitting the controller. Similar concept to how a piggyback ecu manipulates signals before landing in a stand-alone ecu

Lool forward you your feedback


Light up with Lyle

Hi Lyle,

I completely understand where your coming from; when I was engine building all of our measuring equipment including setting rings, calipers, mics etc were calibrated once a year by a UCAS certified company (usually norbar) and even still we had a desktop wrench checker to use between those times.

Now when it comes to sensors at a motorsport level they will have been tested and offsets provided for drift etc; they also would be checked in house via the engine dyno systems and then the dyno calibrated annually (or between projects)

Lambda sensors however where replaced if they were seen to be faulty or had any errors. We did have an offset calibration for them from the supplier but as all things in motorsport would be 'lifed' before errors would present themselves

Road cars, oem calibration is done using rather large emissions analysers which are calibrated with pure gases each day or before a specific task. The lambda sensor that would be fitted to the exhaust would be left in until an error occurred (I never replaced one... They (within a reasonable margin) would read the same as the analysers

Aftermarket, yes I agree that it's something not many people think about and I have taken my car to a mot station to check my lambda sensor against the calibrated emissions testers they use. I run a ntk unit and free air set my sensor before any extensive mapping. In the road world it isn't to critical (small drifts are ok) and this can be said also for mild modifying also. But yes I agree that it's not widely spoken about not tested.

If anyone has experience of having any sensors calibrated in the real world I would be keen to hear everyone's experiences.

Hope that gives you an insight to oem and motorsport set ups.

Good question, and not just for the lambda sender itself - the gauge end is constructed of electronics with tolerances and, even with a couple of test points, may be off a little away from those points - which will already have tolerances built into them.

There are a few companies in Australia that seem to offer a calibration service, if you give them a call they may be able to assist?

The PPG gas detection equipment we used to use were air calibrated before entering enclosed spaces, and also were regularly checked by an outside company but that was something the office arranged, so can't help even if not in NZ.

While I'm not certain that using a purge gas at a specified AFR is quite necessary for field validation I'm definitely of the opinion that people assume far too much with their 'prosumer' grade electronics and sensing elements.

To keep myself as honest as I can and limit my own liability I do a free air calibration before any tuning session and keep sensor usage down to less than half of Bosch's recommended use life under leaded conditions (pretty sure its 40-50 hours, I retire them or put them in one of my street cars around 20 hours). I also have the luxury of keeping tabs of realistic EGTs vs AFR on the in-house vehicles I take care of so I'm able to put everything through some checks and balances that many others wouldn't be able to, so that's another 'check' that could be applied to an engine management/tuning program.

We usually reply within 12hrs (often sooner)

Need Help?

Need help choosing a course?

Experiencing website difficulties?

Or need to contact us for any other reason?