How to Log Data Using an OBD Reader

Hello, TacomaWorld. Today we're going to talk about how and why to log data on your Taco.

Why would you want to log data?

The short and nerdy answer is to trend metrics over time to see how the truck is performing or diagnose a problem. For example, I recently had my truck tuned and I increased the alternator output to ensure the AGM battery was being supplied a minimum of 14.4V as specified by the manufacturer. So, I set up the logger and let it run on my back from Denver which was about an hour worth of data collected on a one second interval. Post-processing and analyzing the data gives me insight into the alternators voltage output and the battery charge state over time.

How to log data

You only need two pieces of equipment to log data: an OBD reader and a tablet or computer. I use the OBDLink MX+ and a cheap Android tablet, but you could also accomplish this with a different reader such as the devices listed on the OBD Fusion website. You'll also need software such as the OBDLink software that works exclusively with the MX+ or OBD Fusion as mentioned above. Finally, you may want to buy the Toyota-specific PIDs from the app--I think mine cost $10.

Once you have the OBD reader installed in the OBD2 port and the app set up on your device, you're almost ready to start logging. You'll need to find the specific logging settings in your application--in OBDLink these settings can be located in the "Settings" page under "Logging." Just enable the PIDs you want to trend, connect the OBD reader to the tablet, and start driving. I would suggest test driving for about 5 minutes and stopping the collection. In the app, the data points are written to a comma-separated value (CSV) sheet that can be read with Excel or Google Sheets depending on your OS. Locate the CSV and open it. As long as there is data in there, and the headers match the PIDs you selected, you're good to go.

How to analyze the data

In OBDLink, the CSVs are automatically written to a Dropbox folder that I can then download onto my computer. The easiest way to analyze your data is to open the CSV in Excel or Sheets and save the data to a format that is compatible with charts--in Excel this is a .XLSX file. It's imperative that you save the CSV to a different format as the CSV format is not compatible with charts, special formatting, etc. Once that's done, you can highlight your data and insert a chart to view the data over time.

The tricky part is trying to understand the data. How do you know what you're looking at is good or bad? The understanding part comes from a "reference" which is a known good set of data. For example, if I suspected that the alternator is failing on my truck, I might ask a friend with a 3rd gen to collect some data--which is the reference data in this scenario--on his truck for a few days. I can then compare my data against his and attempt to draw a conclusion.

An example analysis

For the battery SoC analysis, I know that I need the voltage to be a minimum of 14.4V as the "cycle charge" specified by the manufacturer is between 14.4V and 14.8V. I also know from previous analysis and a basic understanding of electrical engineering that the alternator output is always going to be higher than the battery state of charge (SoC).

So I collected a few weeks worth of data and I noticed that the alternator output is about 0.2V higher than the battery SoC and I had the tuner set the alternator output to 14.5V. This means that the SoC should average 14.3V and I wanted to confirm that before ramping up the alternator output to a higher voltage. After analyzing the data, I calculated a mean battery voltage output of 14.39V ± 0.03V at 1 sigma standard deviation. The standard deviation is the average variation of 68% of the data points surrounding the mean. The standard deviation gives us a stochastic measure of the data which is more valuable than a deterministic measure where we just look at the mean. For example, if the standard deviation was much higher than 0.03V, that could point to an alternator that isn't supplying a constant voltage thus potentially indicating early alternator failure. We might miss that hint by just looking at the voltage output with deterministic methods.

Conclusion

For those who like to geek out on data, for about $250 you can capture data from your vehicle, perform your own analysis, and obtain a micro-level view into your truck's health and performance. And this analysis is exceptionally useful when you start modifying your truck from the factory standard. We know from crashnburn80 that the AGM batteries most of us put in our truck necessitate a higher cycle charge and while we can ask our tuner to up the alternator output, how can we be sure that they actually did what we asked? This was my case. I had the truck tuned with OTT and after reviewing the data I found that the tuner did not alter the alternator output as I asked. So, I had to drive back to Denver to have him apply the correct settings and now I know that battery is almost receiving the specified cycle charge. Besides being fun, this analysis will help me sleep at night knowing that the expensive Northstar battery is being kept happy with the correct cycle charge.