We are using this 3.6V Lithium battery (ER14505) to power the micro controller circuit in one of our devices. The device is capable of reporting the battery level. Since the controllers brown-out threshold is at 2.7V we use the range from 2.9V-3.5V to report 0-100% (battery should be replaced @0%).

While testing the device, we noticed some sporadic drops in the range of 20-40% that we could not explain. During our investigation, we noticed that the battery voltage varies up to 200mV depending on the orientation of the device!

So we grabbed a single battery and a multimeter and did some tests:

  • With no load, the voltage stays the same
  • With 150 Ohm load (22mA@3.3V), we were able to confirm that the voltage is changing nearly 200mV from best case to worst case orientation
  • "Fresh" batteries are not affected as much as cells that have be in use for some time (still about 3.6V with no load)

Batt 1

Batt 2

Batt 3

I tried Google to gather some information regarding this phenomena but wasn't able to come up with something useful.

Does anyone have information about what exactly is happening, what types of batteries are affected or if we are missing something obvious?

  • \$\begingroup\$ I've never seen that before, but have also never used that particular chemistry battery before. This is clearly a question for the manufacturer. \$\endgroup\$ – Olin Lathrop Oct 6 '14 at 12:17
  • \$\begingroup\$ @OlinLathrop: I managed to find some information, see my answer. However, I also just mailed the manufacturer in question about providing some details with respect to their cells. \$\endgroup\$ – Rev Oct 6 '14 at 12:48
  • 1
    \$\begingroup\$ They are liquid electrolyte... is it literally sloshing from one end to the other? \$\endgroup\$ – Bryan Boettcher Oct 6 '14 at 17:15
  • \$\begingroup\$ @BryanBoettcher, bought some of these recently and can confirm, I can hear liquid sloshing around if I shake it side to side (I have two "14500" (AA sized) cells, and they both sound identical). \$\endgroup\$ – user98663 Jan 24 '20 at 17:39

Ok, after some additional research I think found some useful information from other manufacturers. I will just quote some points of interest.

From Technical Brochure LTC Batteries

Depending on mechanical cell design and system properties, there is a certain dependence of available capacity on cell orientation during discharge. The effect is caused by the tendency of the electrolyte to move towards the void and inactive space of the battery if the orientation deviates from the preferred direction. The capillary effect of the cathode and separator pores acts against this tendency. As a result, the orientation effect is smaller for thin cathodes than it is for thick ones and is not even observable when discharge currents are very low or when batteries are moved during discharge [...]
At the high current end, available capacity of big cells (C, D, DD) is affected if the batteries are discharged upside down. Therefore this orientation should be avoided if possible.

From Handbook Primary Lithium Cylindrical Cells - VARTA

Under upside down installation, the capacity of smaller size (1/2 AA, AA) is less affected whether discharge current is high, nominal or low. However, the capacity of bigger size (C, D) especially at higher discharge current is affected. Under upside down installation, the lithium and cathode is located in a fixed area whereas the electrolyte falls to the bottom in this case. At the top of the cell there is a space leaving an area of the anode and cathode, not covered by the electrolyte. Bigger size cells have a bigger empty space, so the capacity decrease in upside down installation is higher than in cells of smaller size. (About 20~40% of its capacity at same higher discharge current.)

While we are not "at the high current end" and are using a AA cell, the measurements from above confirm that "upside down" is the worst orientation.

So orientation does absolutely matter when using Lithium-Thionyl-Chloride cells. Learning never stops...

  • \$\begingroup\$ I have to be honest that I'm a bit sad I'm too late to give this answer ;-). "Back in the day" I had a hunch like this and did some private acid-cabinet-based research dismantling a cell. All for naught now :'( ... ;-) \$\endgroup\$ – Asmyldof Oct 6 '14 at 16:48
  • 1
    \$\begingroup\$ @Asmyldof: Well, there is room for another answer here ;) \$\endgroup\$ – Rev Oct 6 '14 at 18:06
  • 1
    \$\begingroup\$ @rev1-0 Yes, but yours covers all that there's to know about directionality of wet/wet-ish cells. If only I knew about SE before the internet was here! Oh wait... \$\endgroup\$ – Asmyldof Oct 6 '14 at 20:03
  • \$\begingroup\$ this is amaizing... \$\endgroup\$ – Denis Oct 8 '14 at 1:13
  • \$\begingroup\$ Ironically, one common application for this kind of cell is for use in flashlights. Often we point flashlights at the ground causing the cell to be almost inverted - and losing capacity all the while, although walking motion might mitigate the effect. An interesting but annoying thing! \$\endgroup\$ – user98663 Jan 24 '20 at 17:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.