Power Bank Testing Results

[John Papini]

So at the suggestion of David Thomas I bought myself a voltmeter/ammeter and have since been going HOG WILD. I’ve tested every cable in my house, bought more batteries and cables than I could ever use hiking 10 triple crowns, and generally have been driving my girlfriend crazy. Ok, maybe I can’t blame that last part on the voltmeter.

In any case, I wanted to present the results for four power banks I tested. My method was to drain each bank using a small usb resistor (https://www.amazon.com/gp/product/B00KLZHCEQ/ref=oh_aui_detailpage_o07_s01?ie=UTF8&psc=1) charge it to full, plug in this little guy (https://www.amazon.com/gp/product/B07FT6VZMZ/ref=oh_aui_detailpage_o08_s01?ie=UTF8&psc=1) and then drain the battery again using the resistor. I did this three times for each power bank and recorded the mAh discharged each time and took the average.

The first battery I tested is the seeming favorite among BPLers and Reddit Ultralighters, namely the Anker Powercore II 10000 (https://www.amazon.com/gp/product/B072MH1434/ref=oh_aui_search_detailpage?ie=UTF8&psc=1). On my scale it weighs in at 6.74 oz. The average capacity I recorded using the above method was 6,454 mAh. This is ~65% of stated capacity and works out to ~958 mAh / oz. The Anker has a single micro USB input and single USB A output.

The second battery I tested is something called the WINNI 10000 (https://www.amazon.com/gp/product/B074RHTL88/ref=oh_aui_search_detailpage?ie=UTF8&psc=1), which has a built in lightning cable and a built in micro usb. It also has two additional USB A outputs and a single micro USB input. On my scale it weighs in at 5.93 oz. The average capacity I recorded was 5,460 mAh. This is only ~55% of stated capacity and works out to ~921 mAh/oz. The numbers might be better if you are able to use its integrated cables and leave another cable at home.

The third (and best) battery I tested is the ZMI power pack, made in California (https://www.amazon.com/gp/product/B078S3QNM9/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1). It has a micro USB input as well as a USB-C input, but unfortunately they are so close together that they cannot be used at the same time. It has a single USB A output. On my scale it weighs in at 6.3 oz. The average capacity I recorded was 6,935. This is ~69% of stated capacity and works out to a whopping ~1,100 mAh per ounce. This one also has a neat feature where you can set it to low power charging so it will charge devices that don’t draw much power. I had an issue with my RavPower Savior where it wouldn’t charge a small penlight unless my phone was also plugged into it. This solves that.

Finally, although I want a 10,000 mAh battery, I found a 5,000 mAh battery weighing only 3.5 oz and I had to try it (https://store.ecoflow.com/collections/pocket-sized-portable-charger/products/river-rapid). It has a single USB C that serves as both input (12W) and output (18W) as well as a QC 3.0 USB A output. On my scale it weighs in at 3.55 oz. The average capacity I recorded was 3130. This is ~63% of stated capacity and works out to ~882 mAh per ounce. I don’t know how this compares to other 5,000 power banks, but I wanted to get it because all other 5,000 power banks I’ve seen online are 4+ oz.

Hope this is helpful. If you have any 10,000 power banks that you love and think would perform better than the above, let me know and I’ll probably buy it to test it out.

Cheers,

[Ian]

Not part of your study but how quickly does the ZMI recharge compared to the Anker, presuming you have a compatible quick charge plug?

[Gumbo]

Very informative — thanks, John.

I, too, would like to know fastest possible charge times.

[Nick Gatel]

Well, yes… the capacity is less than stated.

Those power banks use 3.7 volt batteries connected in parallel to get stated mAh specification. The stated capacity is correct for 3.7 volts. The discrepancy is that you are charging a 3.7 volt battery, which requires charging at 5 volts to overcome the internal resistance of the device’s battery.

For example the alternator in your car charges your car battery at 13.8 to 14.4 volts to charge the 12 volt car volt battery.

Anker’s website has this to say about it…

PowerCore II Slim 10000’s output efficiency is 60%-70%. Approximately 30% to 40% of the battery capacity is unable to be used because of efficiency complications that affect all electronic devices. These include power lost from circuit heat and voltage conversions from the battery, the charging cable and your device.

 

[BlackHatGuy]

Thanks for putting in the time and effort John, much appreciated.

For trips up to 5 days, I’m a fan of the Anker “Anker [Upgraded to 6700mAh] Astro E1 Candy-Bar Sized Ultra Compact Portable Charger, External Battery Power Bank High-Speed Charging PowerIQ Technology” charger. Weighs 4.3 oz on my cheapo kitchen scale and does what I need it to do with charge left over at the end of the week. That’s partially because I use a Mophie charging case on my iPhone.

[Ian]

Nick,

I won’t speak for the OP but I think most of us understood to expect about 30% loss during power transfer.  These numbers help weed out products like the WINNI bank that demonstrated  well over 40% loss compared to Anker and ZMI which were close enough for me that I would ultimately make my decision between to two based on recharge times, reported reliability, and weight.

Should’ve stated it in my first response but thanks to the OP and my condolences to your SO as it only goes downhill from here as one day she’s going to find you watching water boil with a stopwatch and scale

 

[Nick Gatel]

I won’t speak for the OP but I think most of us understood to expect about 30% loss during power transfer.

When these things first came out I was given one as a gift. I didn’t expect only 60% – 70%  “efficiency.” I expected close to the “rated” capacity. When that didn’t happen, I did a little research.

 

Yes, it’s useful for John to do this. Now if John really wants to test the household domestic tranquility, I suggest purchasing a Kill A Watt meter to document how much energy is being used by all the appliances and other consumers in the house :-)

I had to hide mine because the wife threatened to throw it away :-(

[Jerry Adams]

Good test.  I like the unit that measures mAh and mWh.  I have one that just measures instantaneous current and voltage.

What Nick said – the mAh rating is of the 3.7 V Lithium battery.  You’d think maybe they’d give the rating of the power bank that includes the 3.7 V to 5 V conversion plus any inefficiency, but no one does that.  At least every manufacturer is consistent.  The 3.7 V to 5 V conversion is 74% – that’s what you’d expect if there was no inefficiency and the battery capacity was exactly as spec’d.

Your best one is 69%, so that would be 69 / 74 = 93% efficiency – not too bad for an electronics device.  Your worst is 55% = 74% efficiency – I’d expect better than that.  It could be the battery capacity is under spec too.

I’ve been using an older Anker 10,000 which has worked well.  7.85 ounces.  I could buy a new one and save more than an ounce : )

[Jerry Adams]

I have one of those Kill-A-Watt units.  Wife doesn’t know about it.

[David Thomas]

I have three of the Kill-a-Watt meters (one actual Kill-A-Watt Brand, and two others for their slightly different features.  The day my wife met me, I brought a hot tub on a backpacking trip, so she’s long known what she was getting into.  Since she got a plug-in Prius, she’s been playing a similar game.  There’s 1600 miles on the current tank of gas and the tank is still 3/4 full – it almost always functions as a 100% electric car.

John, thanks for your work on that!  It gives us very actionable data.

I quickly learned all the manufacturers were lying about their Ah capacities.  Now we know which are lying the most.  Auto engines used to be rated in gross horsepower instead of net.  Those stupid “peak horsepower” ratings on cheap power tools and shop vacs (“5 peak HP” with a motor that pulls 9 amps of 120 VAC = no more than 1.4 hp).  Sure, my 1988 Toyota Corolla made 1000 “peak hp” if I just revved the engine and dumped the clutch.  I’m a Director of an electric utility.  Do you want to be charged for the 3413 BTUs of energy in a kWh we deliver to your house, the 3650 BTUs that comes off our transmission system to get that kWh to your house, or the 8100 BTU of natural gas we burn to generate that electricity?  What is the point of delivery – the 3.7-volt battery inside or the USB outlet?

[Jerry Adams]

I don’t think that’s lying.  They just spec the size of the 3.7 V battery.

If one manufacturer spec’d the USB voltage capacity, they would under rate mAh compared to other manufacturers.  As long as manufacturers are consistent, it’s okay.

Besides, who has much of an idea of what USB voltage capacity they’d need?

Some devices that you’d want to charge specify the capacity of their internal 3.7 V battery, so in that case, specing power bank Li Ion battery capacity would be directly comparable (minus any inefficiency)

It would be nice to know the actual efficiency, which varies between 74% and 93% from John’s data.  Any decent device would be closer to that 93% in which case it’s not very important, close enough to 100%

[Ian]

Well since three contributors from the RV Thread Of Justice are here, I’ll mention how that the usable amp hours of a flooded battery is typically 50% of the rated hours.   Even the lithium ion ones can’t be drained 100%.

Yes I know that this is an apples to oranges comparison, but I think exageration/misleading is a well established industry wide practice for batteries, regardless of application.

[Dan @ Durston Gear]

Thanks for this. Much appreciated.

[Nick Gatel]

Well since three contributors from the RV Thread Of Justice are here, I’ll mention how that the usable amp hours of a flooded battery is typically 50% of the rated hours.

Are you sure?

 

[Ian]

Nick,

Technically one can skydive once without a parachute, but from a practical POV, I’m not dropping my battery below 12.2 volts.

Probably better continue this in Chaff on that RV thread, but you get my point and know that a 12v 100 ah battery is not going to deliver 1 amp for 100 hours or 100 amps for one hour.

[Ian]

Sort of related to the OP…

I travel a fair amount and the various quick charge options really appeal to me.  My gen 1 10000 Anker takes most of the night to recharge.  The current version, linked in the OP, is reported to recharge in four hours?  If so, that’s great and something travelers and through hikers would certainly appreciate when trying to top off in an airport, laundry mat, etc.

Anker is now advertising a yet to be released PowerPort Atom that’s supposed to be 40% smaller and 2.5x faster than the quick charger they currently offer.  From looking at the images, I guess that it’ll be of similar size to Apple’s 5w charger.

https://www.anker.com/deals/powerport_atom?search=masterbanner&keywords=1_powerport_atom_us

[Graham F]

Yes ZMI is Xiaomi.

They are renowned for their quality. Massive, maybe the largest power bank company in China. Most of the Xiaomi and ZMI have the double tap feature on the button for headtorches and such. Thanks for your work.

[Jerry Adams]

I always limit my lead acid battery use to 50 amp hours if it’s a 100 amp hour battery.- 12.2 V or so.

I think it reduces it’s lifetime if you go more than that.  I accidentally left a light on once and used up the 100 amp hours.  Maybe it has less capacity than that 100 amp hours now, I can’t really tell.

Mostly though, I kill lead acid batteries by leaving the charger on for months, then the water level goes below the lead plates.  That will kill it.

I wonder if USB power banks with Lithium batteries have this same property – they’ll last longer if you recharge when they’re only half discharged.

[Nick Gatel]

Jerry,

From what I have read the Lithium batteries used in consumer electronics will have a battery life (number of recharge cycles) based on the depth of discharge (DOD). If a battery is allowed to always discharge 100% it will have fewer recharge cycles before it fails. It appears that a 50% depth of discharge might be the best strategy — as per this article from Battery University. The same article showed the graph below where the battery capacity (mAh) decreases with the number of charge cycles.

The same results for a true deep cycle lead-acid battery used in golf carts and RVs. You can discharge 100% but it will have fewer charge cycles. The chart below shows a Trojan T-105 will get approx 500 cycles at a DOD of 100% versus over 1,000 cycles with a DOD of 50%. Since I have solar my DOD is rarely greater than 20%. My last pair of deep cycles (not Trojan brand) were 5 years old when I replaced them a few months ago. (This time I did get a pair of Trojan T-105s). I could still charge the old batteries to 100% state of charge (12.8 volts) and the specific gravity in each cell was good at 1.288, but the capacity in AH was starting to drop off — just like the Li-Ion chart from Battery University above.
<p style=”text-align: center;”>TROJAN T105 6 volt golf cart battery</p>

 

[Ian]

Nick,

This is one of those flame worthy topics that results in lively debate.  One of the selling points for Lithium batteries is how it’s cheaper in the long run (but with a higher up front cost).  With a greater cycle life and an advertised 80% DOD, they are reported to do more with less.  I’ve no personal experience with the larger ones to say and my experience is limited to the 10000 mah and smaller banks.

Graphs are great but I anecdotally see this play out in the disposable battery world in headlamps and UV pens (insert Roger’s steripen article here) where the lithium batteries deliver better life and are likely a wash in cost, if not cheaper.

With the OP’s linked battery banks all coming in at under $50, there isn’t much financial motivation to baby them the same way as one would with a >$1000 battery.  I’m sure limiting discharge to 80 or 50 percent would be ideal, but I routinely discharge mine 100% and it’s still performing fine, albeit with no testing to see if there’s any actual performance drop.

Some more reading here where the authors note a standard 80% discharge on lithium batteries.

https://www.batterypoweronline.com/wp-content/uploads/2012/07/Lead-acid-white-paper.pdf

 

[Jerry Adams]

Good article Nick

Your plot shows the capacity for 11 different batteries.  You can see the variation from battery to battery.  And how the capacity drops from about 90% of spec when new, to about 80% of spec after 250 charge/discharge cycles.  The 11 batteries were between 73% and 83%.

That was done with a 100% discharge for each cycle.

The point of that plot is the variation from battery to battery, and how the capacity drops with number of charge/discharge cycles

“Although a battery should deliver 100 percent capacity during the first year of service, it is common to see lower than specified capacities, and shelf life may contribute to this loss. In addition, manufacturers tend to overrate their batteries, knowing that very few users will do spot-checks and complain if low.”

“Table 2: Cycle life as a function of
depth of discharge.

Depth of discharge  Number of cycles (NMC/LiPO4)

100% DoD ~300 / 600
80% DoD ~400 / 900
60% DoD ~600 / 1,500
40% DoD ~1,000 / 3,000
20% DoD ~2,000 / 9,000
10% DoD ~6,000 / 15,000”

Table 2 shows how the number of charge/discharge cycles you can get depends on how much you discharge.

NMC and LiPO4 are two types (chemistries) of Lithium Ion batteries.  There are others too.  This is very confusing because you don’t normally know which chemistry you have and there’s such a big difference.  There is a factor of 2 or more difference, in number of cycles you can get, between those two battery chemistries.

Look at a couple of examples.  NMC chemistry battery, 100% discharge, you get 300 cycles.  With a 10% discharge, you get 6000 cycles.  But for 10% discharge you’d need 10 times as many cycles to get the same amount of usage out of your device, so that’s equivalent to 600 cycles of complete capacity.  So if you only discharge 10% your battery will last twice as long.

For a 50% discharge, interpolating the table, you’d get 800 cycles, which would be equivalent to 400 cycles of complete discharge.  So, your battery would last a little longer but small enough you might not notice.

So, the take-away is that limiting discharge to a small amount like 10% your battery will last twice as long, but if you only limit it to 50% it only makes a little difference.

The article also said “Lithium-ion suffers from stress when exposed to heat, so does keeping a cell at a high charge voltage. A battery dwelling above 30°C (86°F) is considered elevated temperature and for most Li-ion a voltage above 4.10V/cell is deemed as high voltage. Exposing the battery to high temperature and dwelling in a full state-of-charge for an extended time can be more stressful than cycling.”

So, maybe that’s the most important thing, don’t let your battery get hot for an extended period, especially when it’s fully charged.

[Ethan A.]

John thank you for investing the effort and sharing the results. This was very helpful and timely for me.

Also appreciate the explanations of the results.

Ian with those specs the PowerPort Atom would be so fast. Any idea when it’s supposed to come out?

For anyone looking for chargers, cables and other power accessories, it’s helpful to compare offerings from Anker, Aukey and Monoprice (in addition to Amazon/general search). They all make similar items, but there are variations in size, weight and speed that can be meaningful.

[David P]

I have an Anker Powercore 20,000 mah powerbank with QC 3.0. Weighs 12.7oz on my scale. I’m assuming with the about 35% power loss I’m only getting actually about 13,000mah of juice to power my Note 8 phone? (3,300 mah internal battery)

I guess that makes sense since I get about 4 charges out of it (enough for a 4 day backpacking trip using the Gaia GPS app all day) instead of carrying around a solar panel

[Ray A]

This is the power bank we’ve used for the past couple of years, bought at Fry’s. I’ve never had to recharge my iPhone 7+ more than 3 times, but it’s always had juice left over. (We never let the phone get below about 20%, so it’s a recharge from that point). Curious what the efficiency of this one is because it’s the lightest 10,000 I’ve ever seen. (4.8 oz).

[Greg Mihalik]

^^^

Load testing is the only way to know.

Send it to the OP along with return label and postage?

[Author]

Posts