Lead acid self discharge

real_aurgathor

A little while ago I started a test to determine lead-acid batteries self discharge rates since aside from some comments such as 2 - 5% month depending on temperature, I couldn't find much of anything.

I used an old (10+ years) DieHard Gold battery and Fluke 75 which is a 3 1/2 digit DMM. Blinking means that the display was alternately showing 2 different, but neighboring values, so in those cases I inputted the average. Both the battery and the DMM are in an unheated room, and for the most parts, we have a much drier and somewhat colder weather than usual. Freshly charged batteries have a good amount of 'surface charge' that presumably dissipates fairly quickly.

Here's the data:

01/05/2025:
01:45 12.60V
02:15 12.57V
02:45 12:55V
03:15 12.53V
03:45 12.52V
04:15 12.52V
13:15 12.45V
14:45 12.44V
16:45 12.43V
21:00 12.42V
23:00 12.415V (blinking)
01/06/2025
02:00 12.41V
04:00 12.41V
12:45 12.40V
18:00 12.40V
01/07/2025
00:30 12.39V
07:30 12.38V
19:00 12.37V
01/08/2025
04:00 12.36V
19:00 12.35V
23:00 12.35V
24:00 12.35V
01/09/2025
11:00 12.345V (blinking)
19:30 12.34V
01/10/2025
02:00 12.34V
11:30 12.33V
16:00 12.33V
22:00 12.33V
01/11/2025
02:00 12.32V
14:00 12.32V
21:00 12.32V
01/12/2025
02:00 12.31V
12:00 12.31V
21:00 12.31V
24:00 12.31V
01/13/2025
02:00 12.31V
15:00 12.30V
01/14/2025
01:00 12.30V
08:00 12.30V
20:00 12.29V
01/15/2025
01:00 12.29V
07:00 12.29V
08:00 12.29V
12:00 12.29V
19:00 12.29V
01/16/2025
01:00 12.29V
10:00 12.29V
20:00 12.29V
01/17/2025
00:30 12.29V
12:00 12.28V
19:00 12.28V
01/18/2025
09:00 12.28V
18:00 12.28V
01/19/2005
01:00 12.28V
12:00 12.27V
23:00 12.27V
01/20/2025
10:00 12.27V
15:00 12.27V
23:00 12.27V
01/21/2025
09:00 12.27V
19:00 12.27V
23:00 12.27V
01/22/2025
01:00 12.27V
10:00 12.27V
14:00 12.27V
01/23/2025
08:00 12.27V
12:00 12.27V
19:30 12.27V
01/24/2025
01:00 12.27V
13:00 12.27V
19:00 12.27V
23:30 12.27V
01/25/2025
07:30 12.27V
15:00 12.27V
23:00 12.265V (blinking)
24:00 12.26V
01/25/2025
17:00 12.27V
01/26/2025
01:00 12.27V
10:00 12.27V

The data is as expected until about 12,27V, but things are getting a bit weird there. For one, it stays there for unusually long; for two it seem to go back to 12.7 V after dropping to 12,6V., and this is a major head-scratcher.

Because of the DMM's inherent limitation due to being only 3 1/2 digit, a few mV fluctuations around 12,65 V can explain the readings I got, but still. Maybe the readings can be affected by temperature. I never really needed a DMM with more 3 1/2 bit resolution, but I guess this is a good reason to look into obtaining one.

Edit: looks like the forum SW eats tabs, so the formatting is royally messed up, and uploading .txt files is not allowed.

This is how it supposed to look like:

MacLeech

real_aurgathor so it looks like you confirmed those 2-5% discharge rates you previously found.

real_aurgathor

No. The initial drop for the first few days was the dissipation of surface charge.

However, where the surface charge ends is open to interpretation. Given the the last weeks's data, the self discharge rate for the last week is likely to be under 0.1%, I will post weekly rates based on the above data within a few days,

Splitpair

A lot of wasted time to prove nothing. :-(

real_aurgathor

Nope.

For one thing, I already learned quite a few things not usually published on web pages that deal wit lead-acid batteries.

For two, I'm planning to build my own charger and tester.

Splitpair

real_aurgathor

Starting with a ten year old battery is a fail to begin with and failing to do proper load testing before staring any testing negates your results as meaningless.

Splitpair

real_aurgathor

As for building your own charger, why waste the time and money, you are already out of your league to begin with.

real_aurgathor

I finished the first run of my self-discharge test and some of my conclusions are:

the self discharge rate of a good battery is negligible compared to the rate that is needed to keep the systems alive so that they don’t lose their settings. (in one of my car the current draw was around 28 mA while the tested battery’s equivalent self discharge rate was well under a mA after a few weeks)
temperature can have an easily measurable effect on the battery’s no load output voltage, the battery temperature need to tightly controlled (preferably well under +/- 1 degree) for accurate measurements
a 3.5 digit DMM is not sufficient to accurately measure output voltage change
the self discharge rate I observed was lower than it could be predicted from oft repeated quotes such as: “a lead acid battery being stored at 65℉ will only discharge at a rate of approximately 3% per month.”
Perhaps because it was colder during my test.

One important implication is when not using a car for a long time, disconnecting the battery can keep it in a good condition for a long time since its the constant drain by various not fully off systems (computer, alarm, etc.) that drains it. (Unless the battery is defective, but that’s a whole different ballgame)

Another is when using the oft argued solar charger – for the most part, the solar charger need to provide enough current to counteract the effect of the sum of parasitic draws since the self discharge is usually much smaller than the parasitic draw. (assuming no extra draws such as even a seemingly negligible one like the glovebox light left on, etc.)

I plan to repeat this test later when it’s warmer, but

want to automate recording
use a better volt meter
keep temps more or less constant.

battery-self-discharge.zip

1kB

randavis

The self discharge rate is higher with a dirty battery. You can use a dvm and check for case leakage by connecting to the negative post and using the positive lead placed at different points on the case.

real_aurgathor

randavis

The proper way to phrase that would be: may be higher if the battery is really-really dirty, I’ve seen picture where that may very well be the case, but that wasn’t the case here.

The battery I used has side terminals, and it was installed so that the terminals were facing backwards [least likely to get a dirt buildup], and I can safely state that it has no parasitic drain due to dirt.

real_aurgathor

I just measured the parasitic load on another car of mine, and it was 77 mA – over 2.5x more than another one (28 mA) I measured a few months ago. 🙁

Whether that’s normal or not – that I do not know due to a short a few weeks ago, which necessitated a fuse pull and manual recharge. However, that 77 mA is orders of magnitude higher than the battery’s self discharge rate after a week or two.