View All Posts
Want to keep up to date with the latest posts and videos? Subscribe to the newsletter
HELP SUPPORT MY WORK: If you're feeling flush then please stop by Patreon Or you can make a one off donation via ko-fi

Discover the efficiency of a DIY 20,000mAh USB-C power bank compared to commercial versions, as a control board and lithium batteries charge up an array of devices, including a soldering iron and mini-hotplate, while also considering the project's overall cost.

Related Content

[0:05] Well, thank you He-Man for that excellent intro.
[0:08] We do indeed have the power.
[0:10] Or do we?
[0:11] Watch on to find out.
[0:12] I’ve been accumulating a lot of USB-C devices.
[0:15] It started off with a mini hotplate and now I’ve also got a TS-80P soldering iron.
[0:20] Both of these devices have a USB-C input and take advantage of the various quick charge
[0:25] and power delivery options to get the most out of their power supplies.
[0:29] I have got this really nice gallium arsenide 65W USB-C power supply, but it would be nice
[0:34] to be a bit more portable, particularly with the soldering iron.
[0:38] Now I could go out and buy a power bank.
[0:41] That would be the most sensible thing to do.
[0:43] But I’ve had these 10,000mAh lithium cells lying around for a year now, so I thought
[0:47] I might as well buy a charging board and go with them and make a DIY version.
[0:52] Let’s see if this actually makes any sense at all.
[0:55] The board I’m using states that the batteries need to be able to supply up to 10 amps, so
[0:59] I thought we should probably check that first.
[1:02] Wiring up the battery to my electrical load doesn’t give a very promising start.
[1:06] We only get 3 amps before the battery is cut out.
[1:09] Even with 2 of the batteries in parallel, we’ll only be able to get 6 amps in total.
[1:13] Nowhere near enough.
[1:15] My initial suspicion was that this was due to the battery protection boards being a bit
[1:19] too protective.
[1:20] Fortunately, along with the 10 amp requirement, we also have these instructions telling us
[1:24] to remove any battery protection as it’s all built into the board.
[1:28] It’s pretty easy to do this, we just need to unwrap some of the Kapton tape and then
[1:32] the battery protection board can just be desoldered.
[1:35] It comes off surprisingly easily, which does make me wonder how well it was attached in
[1:40] the first place.
[1:41] Removing the battery protection board also lets me solder on some much thicker and more
[1:45] flexible wires.
[1:46] The ones that came with the battery seem to be pretty thin and stiff.
[1:50] Wrapping this back up with Kapton tape, it’s the perfect crime.
[1:53] It looks perfect, like it was made to be like this.
[1:57] With the battery protection board removed, we can easily get up to 10 amps from a single
[2:01] battery.
[2:02] We can even go much higher if we wanted to, but 10 is plenty.
[2:06] With 2 batteries in parallel, we’d get a whopping 20 amps.
[2:10] Connecting the batteries to the board is pretty simple.
[2:13] You can safely wire up lithium cells in parallel, provided they have the same capacity.
[2:17] They will just self-balance during charging and discharging.
[2:20] However, I did make sure before connecting them that I had discharged them to about the
[2:24] same voltage, so there won’t be any big current surges between them.
[2:28] The control board does have a resistor to set the battery capacity.
[2:32] Fortunately, my board is already configured for 20,000 mAh, so there’s nothing for me
[2:36] to do.
[2:37] With it all wired up, it charges nicely.
[2:40] The control board is pretty nice and it takes advantage of the various fast charging options
[2:44] to take the input power up to 21 watts and speed up charging.
[2:48] If we measure the voltage across the battery terminals, we can see that it is definitely
[2:51] charging them up.
[2:52] The voltage is increasing.
[2:54] With it fully charged, we can test the real-world capacity.
[2:57] But first, a quick word from our sponsors, PCBWay.
[3:00] PCBWay have been sponsoring the channel for a while and we’ve had a bunch of PCBs manufactured
[3:05] by them over the past few years.
[3:07] They also offer 3D printing, CNC machining and much, much more.
[3:11] Check out a link to them in the description.
[3:13] They are really great.
[3:14] So I hooked up my electronic load.
[3:17] The maximum current I can pull by default is limited to 3 amps.
[3:20] Trying to go above that triggers the board to shut down.
[3:23] However, I can use my USB power monitor to switch into Huawei’s Supercharge protocol,
[3:29] which gives us 5 amps output, not bad at all.
[3:33] Unfortunately, after the discharge is complete, we only seem to have measured 12,000 mAh.
[3:39] That’s quite a bit lower than I expected, but it could be just the power monitor is
[3:43] affecting the results and taking away some of the amps.
[3:46] So I’ve recharged the batteries and we’ll try again.
[3:49] As I’ve left out the USB power monitor, we won’t be able to trigger any of the fast
[3:53] charge modes, which means we’re limited to the default 2 amps output at 5 volts.
[3:59] Assuming we do have 20 amp hours at our disposal, this should take almost 10 hours to discharge.
[4:04] Unfortunately, this has produced some very similar results, just under 14,000 mAh over
[4:10] 7 hours.
[4:11] Measuring the voltage of the batteries, it looks like there’s still some juice left
[4:14] in them.
[4:15] We can also check that the battery protection has not cut in by measuring the voltage here
[4:20] and it looks fine.
[4:21] So I’ve hooked the batteries up to the electronic load directly and discharged them.
[4:25] And sadly, it actually looks like there’s not much juice in them at all, after all.
[4:30] Maybe the batteries are not 10,000 mAh, there is quite a suspicious review on a similar
[4:35] listing on AliExpress for the same batteries.
[4:38] So I am now doing what I should have done in the first place, I’ve fully charged the
[4:41] batteries and I’m measuring what the actual capacity is.
[4:45] So let’s kick it off for the first battery.
[4:47] If it’s 10,000 mAh, then at 3 amps it should take around 3 hours to discharge.
[4:53] And here’s the result for the first battery, it’s bang on the money, 10,099 mAh.
[5:00] So let’s check the second battery, we’ll reset the stats and kick off the test again.
[5:05] I’m starting to suspect that it’s going to come in at around 10,000 mAh as well.
[5:11] And the results are in, it’s 10,039 mAh.
[5:15] So it turns out the batteries I got from AliExpress are actually really good.
[5:19] I owe them an apology.
[5:21] So we’ve got 20,000 mAh capacity, but we’re only getting 14,000 mAh out.
[5:27] Does that mean our power bank is only 70% efficient?
[5:30] If true, that would be really rubbish, especially as the datasheet claims up to 96% efficiency
[5:36] on the boost output.
[5:37] Well, it turns out that as always, things aren’t that simple.
[5:41] When looking at a power bank efficiency, we should be looking at the energy and not the
[5:45] amp hours.
[5:46] I’ve put a link to a great website that explains all this in detail in the description.
[5:51] Our two batteries give us 20 amp hours, and they have a nominal voltage of 3.7 volts.
[5:57] That gives us 74Wh of stored energy.
[6:01] Eagle-eyed viewers will have seen that as well as the milliamp hour value on our discharge
[6:05] monitor, we also got the watt hours.
[6:08] We measured 69Wh of energy on the output.
[6:11] This means our efficiency is actually over 93%, that’s a pretty good result.
[6:17] Our power bank is actually very good.
[6:19] Now this shouldn’t really be too surprising.
[6:22] The actual control board is pretty amazing.
[6:24] There’s nothing really on it at all.
[6:26] All the heavy lifting is taken care of by this one chip, the SW6206, so you would really
[6:32] have to mess things up to make it not work properly.
[6:35] The IC supports a completely bonkers number of output options, absolutely amazing.
[6:42] These other ICs are just MOSFETs for controlling the outputs, and this single chip battery
[6:46] protection IC, the XB8089A, is the reason we can remove all the protection from our
[6:51] batteries.
[6:52] Let’s test out some of the output options using my USB power monitor.
[6:56] On the USB-C output, we get these options, and on the USB-A outputs, we get these options.
[7:02] I did have to apply a bit of load to the power bank, or it shut off automatically.
[7:06] You can put it into a constant output mode by holding the button, but then the power
[7:10] delivery options don’t seem to work.
[7:12] This might get annoying after a while.
[7:15] Trying out the TS80P soldering iron, it works OK.
[7:18] On the USB-C output, it only uses 9 volts, which is OK, but it’s not fantastic.
[7:22] And on the USB-A outputs, we get up to 12 volts, which is the maximum voltage the TS80P
[7:27] can run at, so that will work nicely.
[7:29] The mini-hotplate is not as promising.
[7:32] With my gallium arsenide power supply, it will draw 20 volts and heats up really quickly.
[7:36] But with my power bank, it will only draw 12 volts, so it heats up quite slowly.
[7:41] Now, of course, this is not that surprising.
[7:43] It matches what we saw with the USB tester, and it also matches the spec of the control
[7:47] board.
[7:48] It does only go up to 12 volts on the output.
[7:51] So on the question of DIY versus buy, was it really worth it?
[7:56] Including shipping, the cost for the batteries was around £17, and the charge control board
[8:01] was around £4-5, so our total cost was around £22.
[8:06] This comes out slightly cheaper than some of the commercial versions on Amazon, so I think
[8:10] it’s a bit of a toss-up.
[8:11] It’s a fun project, but you would probably be better off just buying one.
[8:16] There are some 65W power banks available that would be able to run the mini-hotplate at
[8:20] full whack.
[8:21] They are slightly more expensive, but maybe an extra £10-20 also would be worth it.
[8:26] I haven’t found any 65W bare PCB control boards, so for now, the DIY route is out of
[8:32] the question.
[8:33] Let me know what you think in the comments.

HELP SUPPORT MY WORK: If you're feeling flush then please stop by Patreon Or you can make a one off donation via ko-fi
Want to keep up to date with the latest posts and videos? Subscribe to the newsletter
Blog Logo

Chris Greening

> Image


A collection of slightly mad projects, instructive/educational videos, and generally interesting stuff. Building projects around the Arduino and ESP32 platforms - we'll be exploring AI, Computer Vision, Audio, 3D Printing - it may get a bit eclectic...

View All Posts