Hackaday Prize 2021 entrant [Philip Ian Haasnoot] has been building a well-polished power bank. But this is no ordinary little power bank the like you would throw in your rucksack for a day out. No, this 2.5 kW luggable power bank is neatly encased in a tough, waterproof Pelican 1550 case, and is suitably decked out with all the power sockets you could possibly need for a long weekend of wilderness camping and photography.
This box sports USB-C and USB 3.0 connectors for gadget charging, as well as 12 VDC cigarette lighter and XT-60 ports for high-drag devices. Also it provides a pair of 120 VAC sockets via an integrated inverter, which at 1.5 kW could run a small heater if you were really desperate, but more likely useful to keep your laptop going for a while. Now if only you could get Wi-Fi out in the desert!
[Philip] doesn’t actually talk much about the solar panels themselves, but we know the box contains a 600 W MPPT boost converter to take solar power in, and feed the LiPo battery pack in the correct manner.
The battery pack is custom-made from salvaged and tested 18650 cells, as you would expect, which we reckon took an absolute age to make by hand. The whole project is nicely finished, and looks like something we’d be happy to throw in the back of the car before heading out into our local wilderness.
As [Philip] says in the project description, it’s a tough job to carry enough power and keep all his drones, cameras and lighting equipment charged, not mention helping prevent the campsite occupants from freezing overnight during the chilly Arizona nights.
Many power bank designs have graced these fair pages over the years, like this rather polished build, and long may they continue to do so.
Lotta details not in the build notoes. Like why 24 volt lithium ion and then throwing in the loss of using a voltage dc to dc converter to get back down to 12 volts? Also the underside of that board has a lot going on that is not listed in the details or parts list. If your loads are primarily 12 volt outputs why did you choose lithium ion over lifepo4? It is better mated to 12v applications and in most instances decidedly safer. Not throwing shade on your project. These details and why they are chosen are interesting in all these DIY builds. Stuff like using anderson power poles for the solar input instead of MC4 is a design choice. I get that. But I have found that anderson power poles suck at strain relief. So a recessed mc4 connector, which is pretty standard for solar panels, may make more sense. But I also see you have a finite amount of vertical clearance underneath. Nice layout. Love the custom panel. Love the formfactor. Need to start mine.
Absent a response from the project creator, I can provide some speculation on the answers to some of those questions. In general, I’ve found used Lithium Ion cells easier to source than LiFePO4 cells. They’re easier to find and cheaper to buy per Wh and they also store more energy per unit of mass. In doing a bit of searching recently for myself, however, I haven’t located a cheap inverter (I was looking for pure sine wave inverters) that seemed like they would work at the voltages for a 3S or 4S lithium ion battery. A 7S lithium battery overlaps pretty well with the voltage range for a 24V nominal lead acid system though and 24V inverters are not quite as common as 12V ones, but still relatively easy to find. The same can be said for solar charge controllers as well.
Pretty common. For 12V cases 3S voltage is too low. 4S voltage is too high, or you cannot charge to full capacity. And ya, full sine wave inverters are considerably more expensive. Definitely comes under “buy right the first time and you only cry once”. 24V inverters should be common enough. And there is certainly some efficiency to be had there, not to mention manageable wire gauges.
LiIon seems to be because 18650s are plentiful, the notes do say they salvaged at least most of the batteries. The notes do talk about the voltage, too, they seemed to choose 7S to allow for direct input voltage to 20V-capable USB-PD chargers. Much easier to buck down to two different DC voltages than to buck down to one and buck-boost to the other, and the USB-PD would likely be the highest-power DC output in practice, anyways, as they’d likely charge their phone off it too, and maybe their laptop too.
fiddlingjunky Correct, 18650s were ultimately selected not just for the cost of the cell (Salvaged cells being extremely cheap, but a crapshoot), but the cost of the supporting parts as well. I was trying to avoid 3D printing cell holders and custom cutting 2P/4P custom nickel strips for the first version. I’d like to get more adventurous on the neck pack I build however.
I was also concerned with buck/boost topology for potential noise issues when trying to run my astrophotography setups.
I generally like to start out on a new project by building something with as many OTS parts I can find, and manufacturing the rest to get the prototype working. Venturing further into fabricating all of the parts myself as I build future versions. I understand that not all of the parts are OTS on this build, but the PCB mounts can be easily 3D printed, the top cover can be produced from a template with a drill/jig saw. I felt this was a good first version that could be built using a range of budgets and tool availability.
My intent with the project moving forward is to design the PCBs myself in stages. I am working on a single board to combine the Buck/Boost solar input (With MPPT), Multi-chemistry battery charging, HMI, and 12V power supplies. IC availability is a huge headache on that effort currently.
Using Li-ion 18650 cells in 12V applications just doesn’t work, since at full charge a 4S pack puts out 16V, and a 3S pack only puts out 12V, whereas at minimum, they are down to 12.8V and 9.6V, respecitively. Neither of these ranges is suitable for 12V devices made to run off of lead-acid batteries. The use of 24V is probably somewhat arbitrary, but it allows a buck converter to get down to the 13V you really want, and buck converters are genrally more efficient and simpler than boost or especially boost/buck converters. As you say, LiFePO4 is the obvious solution, since its slightly lower cell voltage allows for direct substitution for lead-acid, but these aren’t cheap, especially if you are considering using used cells, which aren’t nearly as plentiful.
I wish PowerPole connectors would become more of a standard, since their hermaphroditic nature makes it difficult to get the polarity wrong (a significant problem with the widely-used double-bullet connectors, for example), and means you don’t have to do deal with male and female connectors (or worse, m-m or f-f adapters) when making extension cables.
I use PowerPoles in my RV, and have been for about three years, with no problems whatsoever, not even strain relief issues. I keep a PowerPole-to-lighter plug and PowerPole-to-lighter socket on hand for plugging in other people’s appliances, or plugging mine into other people’s vehicles. I’m also putting together a backpack power system for field video productions I can’t park close enough to to run extension cords from my RV’s inverter. This one uses one to four 20Ah (250Wh) LiFePO4, mainly because I didn’t want to do the DC-DC conversion to feed my inverters, and had no stomach for dealing with used 18650s at any price. I am using buck/boost DC-DC for charging, though, since this makes it easier to get power from various and even multiple sources.
Not completely dismissing APP’s. I am a amatuer radio guy. I have them everywhere. I just do not like them for larger gauges where they are standing straight up out of a connector or hanging suspended from the side of something. Inevitably, connecting and reconnecting in those orientations causes the insulation to move from the connector and exposes bare wire over time. In relatively static conditions, or inline, they work fine. Just my own personal observation over the past few years in my various go-kit builds and other mobile projects.
I would agree that MC6 is the right connector to use directly with PV panels, I was not able to find a suitable bulkhead style connection for panel mounting to the case. I made an adapter whip with the APP on one end and the MC6 on the other. I believe your main cause for concern with the APPs has more to do with the specific wire installed in the connector than the connector itself. I have had failures in the past with the wire jacket chaffing or cracking over time around the APP due to the sharp corner of the plastic where the conductor exits the connector, adding additional heat shrink tubing alleviates that issue somewhat. My preferred method is to use watertight heatshrink as the hot melt glue laden shrink tubing adds an additional layer of protection.
The bulkhead APP adapter I used is pricy for what it is, Perhaps a project based around APP improvements is in order. Some easy to manufacture strain reliefs and printable panel adapters would make projects like this easier and more reliable.
Because 12V isn’t sufficient for most “12V” operation. Generally speaking, “12v” operation is operating within the environment of a 12V buss automobile. The 12V buss commonly sits well above 12, many people have chosen to model it at 13.8V.
At 12V or less, most electronics barely operate. By down-converting, you can always keep the voltage at 13.8, even as the batteries are near empty.
I will make sure to reference the decision process for the battery voltage selection debate in the build notes shortly. I waged war internally for several months before landing on 24vdc for the battery pack voltage. The short answer is a mixture of power requirements for the USB-PD/inverter vs. performance vs. Cost. The physical space available within the confines of the Pellican case also factored into the decision.
LiFePo4 is fun to play with but still much more expensive, and not readily available in the waste stream. I was able to salvage all of the batteries. I may explore LiFePo4 for my overlanding trailer build as that system is intended to be run all on 12v.
What’s the difference between this and a Jackery battery with solar panels attached?
Cost, USB-PD and doing it yourself.
Don’t know about Jackery specifically, but a friend bought a Rockpals clone, and the result was complete loss of her van by fire, on a camping trip. Luckily, none of the nearby trees caught.
Point is, although people are sometimes reluctant to DIY high-energy projects, it’s not always safer to trust what comes from Amazon. If nothing else, if you build it yourself, you know how to quickly take it apart and yank the wires out of whatever is smoking or spitting sparks.
Is another reason I question the use of Lithium Ion in these projects. I belong to several DIY battery groups and have seen more than a fair share of “this part of my home is now gone” pictures when using reclaimed 18650’s in these projects. Lifepo4 is ever so much safer.
Fun with batteries includes lots of smoke, a dented metal wall and more recently having to throw a 18650 pack into the garden to finish burning. Mechatronics projects are fun :)
> Also it provides a pair of 120 VAC sockets via an integrated inverter, which at 1.5 kW could run a small heater if you were really desperate, but more likely useful to keep your laptop going for a while.
Pfft… any laptop worth buying has an option for a power brick that runs directly from 12V, or itself runs direct from 12V sans power brick.
Even an Apple MacBook can be convinced to run on such a supply with a pair of side-cutters and some soldering work.
The last laptop I had that could run, unmodified, directly from a 12V battery was my Panasonic CF-270, from 1991, that used a 12V NiCd. Everything since then requires some kind of mediating brick, even if it’s just a regulator. You’d think 12V supply could substitute for a 3-cell (11V) battery, but no. The darned things are too smart for that, wanting to talk to the battery, and (without hacking) will demand 14V or 19V from the DC input to work. At least a Dell, Toshiba, and a half dozen Thinkpads I’ve tried do. I’m allergic to Apples, so can’t comment on those.
Do you have a suggestion on a modern laptop that will drink 12V directly?
Why is it called a ‘solar generator’ it is not generating anything. At best it is a battery in a box that can be charged by several methods. Call it a power pack or something else. Being called a generator is misleading.
It’s generating alternating current electric power from the chemical energy stored in its lithium cells.
Is that really that much different from generating the same electricity from the chemical energy stored in a liquid fuel?
That this kind of stored chemical energy can be so easily recharged by a variety of methods is a nice bonus, even though it sucks that this battery form of stored energy is 10 times heavier than, say, gasoline.
So instead of saying “battery box with inverter”, some twit from marketing got to spin up “An Amazing Breakthrough”.
Hey, whatever it takes to get the Walmart rubes to fork over the dough. It’s Capitalism, facilitated by vulture capitalists and exploited by our communist Ferengi friends across the Pacific. Embrace it while it lasts — it’s all going to come crashing down in a few years. And then it won’t look so dumb to roll your own power system and hide out in the wilds of Arizona.
Sorry, I hate it too, but the whole industry has decided to call these things “generators”, because to their biggest market, RV owners, “generator” is something that puts out 120VAC, while “power pack” is something you use to charge your cell phone.
Give it a couple of years, and if you say “generator”, the kids will think you’re talking about something like this. They won’t even know there’s such a thing as an engine-driven generator. Language evolves, in sometimes messy ways.
Nice looking CG image there.
I’m sorry this will sound harsh but, man, it’s really tough to trust the description in the project log. Happily mixing “V” and “v”, or “kW”, “KW”, and “Kw”, but apparently consistently using “kW” (or Kw or KW) when it looks like it’s supposed to be “kWh”.
Errors like that imply limited understanding of basic engineering. It seriously makes one question the quality of the build.
Thank you Paul, I’ve tried to improve my photorealistic renders over the years. Photos of the physical build are included in the project page, aside from leveraging plywood for the prototype inner panel I feel the final build is pretty close to the design.
Question the quality of the build all you wish, that is the point of sharing projects. There are others out in the community building a range of equipment and the best method for improving is having someone point out potential defects in your designs.
No matter what projects I’ve seen posted to HaD over the past 13-14 years there has been something to learn from each.
I wasn’t criticizing the build itself. I was pointing out that if a description is riddled with incorrect use of units, then it makes the reader think the author either does not understand what they are doing, or don’t care. Either way, that makes the reader wonder if the quality of the build itself exhibits the same shoddiness.
It’s a valid critique and I appreciate you pointing it out; I absolutely should pay closer attention when typing up my projects for accuracy.
I tend to work on the physical aspect of the project and post photos/updates after the fact, I will set more time aside for the documentation moving forward.
This is close to something I’ve been pondering, wishing there was a commercial option. There are lot of new “generators” around now that are pricey, with internal batteries. What I wish is that there was something available that could take a 52v input. Why 52v? That’s the voltage on my three ebike batteries. Would be nice during a power outage to use the lightweight, high-output batteries to power everything from lights to phones to refrigerators. In a long duration power outage I could even ride the bike somewhere, charge the batteries and return.
Maybe the way some trucks have built-in inverters to use in the field or for emergencies, ebike makers will sell add-on boxes that let you use the bike’s batteries in an emergency.
Well, heck, the simplest thing to do is just put the three batteries in series to get around 160V and feed that directly into the AC input of any standard computer supply or brick: Most will happily accept DC, and will in fact be a bit more efficient doing it.
Or grab a RD6006 power supply or its ilk: they tolerate up to 70V input and can supply 6 amps.
Would work better if the three batteries were identical. They are not. And for a “universal” device, depending on three sets of batteries wouldn’t be great.
There are inverters made for solar panels that will take ~ 60v and give 120v AC out at 1500W. That’s where I’d start. Not all the batteries will do that much current, so final watts out would depend on the battery. I have the pelican case and have looked at the inverters. The 12v out and the various USB variations could be from a DC/DC converter or and AC adapter.
Like I’d mentioned, would be nice to buy this already packaged.