Thread: TD5 MAF Recalibration

Hey all.

We've been chasing a performance issue on a D2 TD5 after an engine swap. Old engine had coolant loss to cylinder 1
It's been a long battle against multiple issues; Leaking fuel pressure regulator, lost 2nd gear on the auto, split belt tore the harmonic balancer in half, etc etc.

We've finally arrived back at the performance issue. After analysis of a trip log from the nanocom using information available on the DiscoTD5 website, I've determined the issue is an under reporting MAF sensor. We've cleaned it and it looks OK but still not right. We will have one on order later in the week.

In the mean time I decided to try and characterize the new behavior of the MAF against the rest of the engine's sensors and have been able to successfully create a tune for this degraded MAF.

The results I was able to achieve seem pretty good and I am looking forward to characterizing the new MAF with stock sensor settings against this calibrated MAF.

Screenshot_20241007_142136.jpg
The first graph shows the original airmass in blue. It shows the MAF is severely under-reporting airmass by up to 35% when compared with MAP+IAT calculations.
In a perfect world, the plot should be a solid 45 degree line, but there are a lot of factors that cause mild variance either way.
The calibrated airmass in orange is significantly better and both sensor sets now agree with eachother fairly well.

Screenshot_20241007_142640.jpg
This second graph shows the calculated MAF error% across the MAF airflow range.
Again in blue is the original MAF's error plot, now in purple is a line showing the averaged error offset. The MAF is reading between 20% and 35% low.
In orange is the post calibration error plot. The green line shows the calibrated MAF average error does not exceed +-3%

After this calibration the car performs slightly better with the MAF plugged in than it does with it unplugged.

I'm curious to know any thoughts on this. It may be viable method for me to extend the useable lifespan of a weakening MAF, for only a little bit of data logging and spread sheeting every so often.

Firstly, well done on the creative thinking! I appreciate it.

I've done this numerous times for my own entertainment.

Yes it will work.

The MAF will continue to degrade unfortunately, so continuous logging will be required to see when it needs to be done again.

A number of AFM's under report by different amounts based on the temperature of the air, this is impossible to calibrate for.

I've had 1 AFM fail by over reading, every other one has been under reading.

What this means in real terms is that you now can't get really high air charge figures to read, so you are limited for performance at the top end.

This only worsens as it degrades.

You also need to make sure that the MAP/IAT is bang on before you start to try and calibrate.

From memory I got it to the point that it was really quick to calibrate by using the returned MAF MV readings, but I never pursued it further.

If you are just doing it for yourself, I think it's a decent mod, but I'd never sell it, it's just a bandaid solution that could easily come off in the shower....

Other than that, it's probably handy if you know what you are doing.

Cheers
James