Single-battery failures in multi-battery arrangements: diagnosing selective cell derangements

Why does one battery (or a few) in a multi-battery pack always seem to drain faster than others, and how does this outcome affect both its siblings and the system they jointly power? Read on.

A recent teardown noted a surprisingly (at least to me) common occurrence that I’ve repeatedly experienced: a tendency for an Amazon Echo smart speaker (or other similarly powered device, for that matter) to functionally fail due to the demise (specifically: droop or other DC output voltage compromise under high load, I’m assuming) of an easily replaceable AC power adapter:

I’ve had a few second-generation “Dot” devices’ external power supplies fail in the past; the end result is either a flat-out refusal to start at all or a perpetual repetition of partial boots followed by abrupt restarts. In those cases, the consistent “fix” was straightforward and non-wasteful. Since the AC/DC converter with USB-A output was distinct from the USB-A to microUSB cable that fed the device, I could just swap in a replacement for the former and be up and running again in no time. Every time I did this, by the way, I wondered how many Echo Dots prematurely ended up in the landfill due to typical-consumer ignorance of both the exhibited issue’s root cause and simple resolution solution.

An oldie-but-goodie

In this post, I’ll cover another situation that I come across a higher-than-expected percentage of the time, whenever a multi-battery-powered device goes down for the count. To begin, I’d like to introduce you to a long-time, frequent-use friend of mine, my BT-168 battery tester:

I was happily surprised, while researching the BT-168 online just now while writing, to come across a link to a colleague’s review and teardown of it from a few years back:

Within his writeup, T.K also briefly mentioned a digital display-based successor, the BT-168D, whose existence I wasn’t aware of until now but which is apparently less accurate than my “old school” original analog version due to a comparative applied-load deficit:

I have no idea how long I’ve owned it, or for that matter, how it originally came into my possession. That said, it’s still available for sale (variously company-name branded by multiple retail sources) at Amazon and other distribution intermediaries, as is the follow-on BT-168D.

What’s this got to do with “single-battery failures in multi-battery arrangements”? Well, whenever a two-AA-powered remote control, for example, or a three-AAA-based bathroom scale:

or an LED flashlight, or even (an extreme example) the six-AAA-each (!!!) LED illumination-augmented automatic salt and pepper grinders we recently received as a gift:

functionally fades, I never reflexively slot all the batteries in a charger for refresh or toss ‘em all in the trash (depending, duh, on whether they’re rechargeable). Instead, I sequentially stick each of them in the BT-168 and see what remaining-charge level each reads. Invariably, one is significantly more “dead” than the other(s), even if they were all brand-new when originally installed. Replacing only the drained one more cost-effectively (for non-rechargeables) gets the gear going again, not to mention a reduced landfill payload…until the next one inevitably fails.

Organization determines compromise-outcome specifics

Why, though, does this operating-life inconsistency occur at all? I’d long been aware that batteries’ initial from-factory charges, therefore measured voltages, were predominantly-to-completely a function of their inherent chemical processes. To wit, so-called “precharged” rechargeable batteries are fundamentally just a marketing-driven relabel of low self-discharge, therefore longer-than-otherwise shelf life, battery chemistries and internal architectures.

But I admittedly didn’t fully realize until researching this writeup just how inconsistent battery-to-battery internal resistance can be, even within a common chemistry-and-architecture combination, both manufacturing batch-to-batch and even within a given batch. To be clear, Ohm’s Law, which I learned way back in my first semester of electrical engineering at university, has long informed me of the effects of higher-than-normal resistance: greater “waste” heat output, reduced current output and lower voltage, especially under load. And I also had some inkling of the fact that for a given battery, resistance also evolves over time and use, typically increasing (unless, of course, the battery develops an internal short). But notable battery-to-battery variability even fresh from the factory? That was, I confess, news to me, although in retrospect I shouldn’t have been surprised, especially for off brand, “cheap” battery options.

The resultant effects of internal resistance variability on multi-battery combos, as suggested by my research results along with another set of fundamental electronics laws, this time from Kirchhoff, are intriguing (IMHO, at least). For multiple batteries connected in series, as I’ve recently editorially inferred by analogy to solar panel connections, the outcomes of a higher-than-spec internal resistance for one of them are reduced aggregate output voltage along with bottlenecked peak current flow. Speaking of current, and on the other hand, batteries connected in parallel—where incremental peak current output potential is one key motivation for this organization, along with increased aggregate charge capacity—are hampered in both of these regards when one of the batteries is high resistance-compromised.

Multi-cell battery pack structures that connect their contents both in serial and parallel are increasingly common, both to boost the effective voltage (serial) and increase overall system runtime (parallel). As I was writing this post, for example, I came across editorial coverage of a YouTuber’s (modestly successful) project to power a (modestly equipped) desktop PC motherboard using only AA batteries:

You’ll see that he has four rows of 16 batteries each. Do the math and you’ll conclude, as I did (unless my methodology was flawed, which is always a possibility; if so, let me know in the comments) that each 16-battery bank is serially tethered (to generate ~25 V) and the four banks then connect in parallel (to boost capacity, therefore runtime). Battery degradation anywhere within the series/parallel cluster will thus result in both voltage and capacity compromises.

The YouTuber’s commentary, elementary as it may be, also makes important points about the importance of robust wiring and connectors, both topics which an excellent white paper (PDF) I came across in my research, published by Victron Energy, discusses at length. While both wiring and connectors, along with the batteries themselves, have resistances typically measured in dozens to hundreds of mΩ (that’s milli, not Mega), none is a perfect conductor. Use, for example, excessively thin wire, and you’ll end up with performance-degrading current flow constraints (along with maybe a fire). The same goes for a corroded battery contact, as anyone who’s dealt with a geriatric vehicle battery likely already knows. And each wiring run’s length is also a critical factor; if one span of a multi-battery parallel configuration is notably longer than the other(s), the resultant (slightly, but still) higher resistance will act akin to higher-than-average resistance in the battery itself.

More to say (but not today)

With brevity in mind, I’m only focusing here on the more common case of higher-than-average internal battery resistance (initially and, especially, over time). That said, as I already alluded to with my earlier “internal short” comments, resistance can also both inherently exist and evolve over time in the opposite (lower) direction. Such a situation is, perhaps obviously, particularly problematic in a multi-battery parallel configuration, both for the affected battery, the others in the parallel bank, and whatever they’re commonly powering.

For similar brevity reasons, I’m also covering today only situations where the installed batteries are either non-rechargeable or are removed for recharging. Multi-battery packs recharged in situ (while installed inside a portable power unit, for example) translate to an even more complicated scenario involving, among other factors, the critical importance (and difficulty) of balancing the various cells within the likely series/parallel cluster. The earlier-mentioned Vitron Energy white paper also explores this topic at length. More generally, I also found the various resources at Cadex Electronics’ Battery University quite helpful. And I’ll likely have more to say about these topics in future posts as well. Until then, and as always, I welcome your thoughts in the comments!

—Brian Dipert is the Principal at Sierra Media and a former technical editor at EDN Magazine, where he still regularly contributes as a freelancer.

Related Content

• Inside the battery: A quick look at internal resistance 
• Cell balancing maximizes the capacity of multi-cell batteries
• Active balancing: How it works and what are its advantages
• Internal Resistance Tester for Batteries

The post Single-battery failures in multi-battery arrangements: diagnosing selective cell derangements appeared first on EDN.