Hi Rodri,
The important concept underlying oil analysis is trending. A good start would be to send your brand new oil from the barrel to the lab and establish a baseline for the oil properties. Once you have the start sample then you can start to establish a trend based on future samples.
Here is a good introductory article from Machinery Lubrication:
http://www.machinerylubricatio.../oil-analysis-report
The decision to change oil is not solely based on TAN/TBN. You need to take into account the whole picture. There are several other important parameters on an oil analysis report:
- viscosity
- ISO code (particle contamination)
- moisture contamination
If you keep particle and moisture contamination under control then the oil should remain stable for a long time (assuming you are not running at high temps.) When you let particle and moisture contamination remain in the oil this speeds up oxidation and increases TAN.
If you can control contamination in the oil then the viscosity and acidity will remain stable for a long time. Based on the trending of your oil analysis when you DO start to see a change in viscosity/acidity then you can consider an oil change.