Data Logging the 12v Battery

[sor]

I’ve had this idea to put a data logger on the 12v battery for awhile now, after reading questions about whether the 12v is maintained by the HV (traction) battery. Seems like an easy thing to find the answer to.

I couldn’t find a DC voltage data logger that was both inexpensive and offered software that didn’t require Windows, so I hacked something together using a microcontroller, wire with a fuse for protection, and a 3d printed case using a leftover filament. The microcontroller monitors the power that feeds it, so the box only needs to have power terminals. There are alligator clips on the terminal ends. The paper towel is just there to keep the box from sliding around in the pocket.

The microcontroller sips power so it doesn’t really provide load. I’d need to wire up something else like a dashcam to see how the car reacts to battery drain. Have not done that yet but I’ve collected A 24 hr baseline and recorded my activity to get an idea of how it behaves normally. It can record about ten days with one sample every 180 seconds, keeping it all in RAM and making it accessible from a WiFi web server. Here are the patterns I see:

12.9v-13v at idle, unplugged
13.4v-13.5v while charging
14.9v-15.0v during cabin precondition
14.0v-14.2v while driving

I wouldn’t get too hung up on the absolute numbers here, I did a rough calibration but it could easily be skewed a few tenths. I’m just using the microcontroller ADC with a voltage divider.

If I find anything else interesting, I’ll post a follow up.

Sponsored

 

[ron_b]

Love the ingenuity. Thanks for sharing.

Curious, if you also put an oscilloscope on the 12 V supply to see if it was as noisy as your processor shows. Wasn’t sure if that was artifacts from the A2D converter.

A nice test would be to put some kind of a reasonable 12 V load on the accessory socket like let’s say an air compressor and see how quickly the DC to DC converter kicks in to keep things up.

 

[sor]

It looks the same on my linear bench supply, it has the resolution to measure hundredths but will bounce within about a tenth. Built in ADCs are not the highest quality. There’s a switching converter in the box to step down the input voltage 8-20v in) to 5v for the microcontroller and that could be adding some noise too.

 

[sor]

Also keep in mind these data points are 180 seconds apart, so while it looks like noise on the graph, I don’t think it’s anything high frequency like you might catch with an o-scope. Or maybe there is high frequency noise and it just manifests in the samples as well.

 

[alxman]

I’ve had this idea to put a data logger on the 12v battery for awhile now, after reading questions about whether the 12v is maintained by the HV (traction) battery. Seems like an easy thing to find the answer to.

I couldn’t find a DC voltage data logger that was both inexpensive and offered software that didn’t require Windows, so I hacked something together using a microcontroller, wire with a fuse for protection, and a 3d printed case using a leftover filament. The microcontroller monitors the power that feeds it, so the box only needs to have power terminals. There are alligator clips on the terminal ends. The paper towel is just there to keep the box from sliding around in the pocket.

The microcontroller sips power so it doesn’t really provide load. I’d need to wire up something else like a dashcam to see how the car reacts to battery drain. Have not done that yet but I’ve collected A 24 hr baseline and recorded my activity to get an idea of how it behaves normally. It can record about ten days with one sample every 180 seconds, keeping it all in RAM and making it accessible from a WiFi web server. Here are the patterns I see:

12.9v-13v at idle, unplugged
13.4v-13.5v while charging
14.9v-15.0v during cabin precondition
14.0v-14.2v while driving

I wouldn’t get too hung up on the absolute numbers here, I did a rough calibration but it could easily be skewed a few tenths. I’m just using the microcontroller ADC with a voltage divider.

If I find anything else interesting, I’ll post a follow up.

Can you explain your findings for the nontechnical members? Does the HV battery protect the 12v battery when the car is off? And, are these findings before or after the most recent update? Thx

 

[sor]

Can you explain your findings for the nontechnical members? Does the HV battery protect the 12v battery when the car is off? And, are these findings before or after the most recent update? Thx

So far I’m just testing the equipment. It is showing that while the car is charging, preconditioning, and driving the voltage on the low voltage battery circuit is higher that the base voltage of the battery when idle.

Because the voltage is higher than the base battery voltage, it indicates that the car isn’t relying on the battery to provide 12v power at these times. Similar to how a running ICE car shows 14-15v when the motor is on and turning the alternator to provide its electricity and charge the battery instead of using the battery.

Next step would be to add a small 5-10W load and see if we find the car doing the same even when idle when the battery needs help, or if it will just drain. I’m sure someone has already done this but I find the data logger interesting to see exactly how it works.

I got the car early September, so I don’t think I have the v28.13 or whatever new update landed in Nov, but I don’t know of a way to know for sure if this was available to factory or port at the time.

 

[ron_b]

It looks the same on my linear bench supply, it has the resolution to measure hundredths but will bounce within about a tenth. Built in ADCs are not the highest quality. There’s a switching converter in the box to step down the input voltage 8-20v in) to 5v for the microcontroller and that could be adding some noise too.

I assume you’re using some kind of voltage divider to get to the ADC and put in the range?
You could put a small RC filter on that to try to smooth out the noise introduced by your step down unit.

 

[alxman]

I have Apex Protect GPS vehicle tracking and management (provided complimentary by the dealership, will not renew it) which I believe monitors the 12v battery charge. It shows 12.6v when idle and 14.6v-14.9v when activating pre-climate or while driving. Have not checked when plugged or charginng. These values correlate with your findings. It also allows you to set an alarm for low voltage. I have selected 11v. The big question is whether the car monitors for low voltage and makes the HV battery “kick-in” whenever the 12v battery voltage is low. I believe the answer is no. However, a recent post by a forum member, I believe since deleted, stated that the new v28.13 update fixed that (?)

Sponsored