I wrote computer code in the long ago, so I checked with more current programmers and developers and confirmed what most of us would suspect:
The Volkswagen emission scandal (diesel emissions are only controlled during emission testing) could have been caused by several possibilities, including these:
- The code in the emissions control computers/chips was never intended to work. While this seems unlikely, Bosch has gone on record as to having told Volkswagen not to use one version of this code in the real world.
- The code in the emissions control computers/chips works, but has ‘decision points’ in which the code decides to emit after all. Such conditions could include steering inputs, abrupt acceleration, et cetera.
In the first case, a fix will be a real chore to produce. Emissions reduction depends on reading many sensors, vehicle conditions, et cetera, which are possibly just not allowed for in the code. This is the worst case scenario and has, imho, the largest potential for damage to Volkswagen, as this emission ‘system’ has been in production for a long time. Not having it work, more or less at all in real driving, is a huge liability with potential for market share loss as well.
In the second case, it’s a matter of how the code is implemented, exactly. I have seen and known people who patched in-chip code and it can certainly be done. Catch is, how re-programmable is the chip(s) involved? are they EEPROM? Do they have room set aside for patch code? (Normally this is the case for this kind of hard/soft code, as patches are likely to be required.)
It will be interesting to watch this play out and see what the effects are. If Volkswagen can fix the code, a revolving recall should be enough: either replacing or patching the computer chips involved.
If Volkswagen cannot quickly fix the code, and needs time to create a ‘real’ emission control version, I think they are in deep doo-doo as governments and users will conclude that they are fundamentally unreliable vehicle manufacturers.