(untitled)

The title says it, I built myself a piratebox, desoldered some ports, built a battery containment and soldered everything together. The whole thing is glued into a carved out book. The thing is, I used to power the exact same router/usb stick conbo with a ~4Ah powerbank and it lasted a while day. The recharable cells in the picture say they each hold a charge of 900mAh, which should add up to 6.3 Ah in total. In reality, the thing goes down in about 1.5h, which is not enough of you carry this thing around while commuting etc. I expected the voltage regulator to draw some power, but not as much as it seems to do. Any ideas how to improve this setup? I'm a software engineer and don't know too much about hardware, so any help is appreciated

It failed to set the title as it seems... I wanted to set it to "Piratebox in a book"

>>662 >which should add up to 6.3 Ah in total If the batteries are in series the capacity doesn't adds up, 6x900mAh batteries in series have a voltage of 7.2V and a capacity of 900mAh. The dropout voltage of a 7805 is 2V, this means that when the battery reaches 7 volts your piratebox will stop working. You can either use a lower dropout voltage regulator or use more energy dense batteries like lithium batteries in combination with a more efficient boost converter to output 5 volts (like any powerbank does). If you use a linear regulator, like a 1117 you'll still be losing more energy in form of heat, the voltage drop will be lower at 1.2 volts, but the piratebox will stop working when the cells reach 1 volt, so that's a lot of wasted energy that could've been used. Buck-bust converters are the way to go.

"We don't know what parallel circuit is" - /diy/

Why would you even use linear regulator for anything battery powered? Did you even check the specs?

>>662 This is very interesting to me. I have been wanting to build one for a while as I have two raspberry pis, however I have not had the knowlege or the money for parts. I will be getting paid soon, however, please share how you did this.

How much range does that get you? How far away does a person have to be to connect?

>>666 >>667 I'm sorry satan. I'm a software engineer and tinker for fun on stuff like this, but never had the chance to learn any theory about it. I posted on diy because I thought it was the right place to ask. The batteries are lithium-polymer ones. Would it work if I put the batteries in parallel with the boost converter? >>676 It's not a raspberry pi, but one of the supported routers. Now that I've got the experience with this, I'd recommend to either get a bigger book or to replace the USB stick with a microsd card. The whole thing is really tight and you can see the status led of the router when it's in a dark environment. You should also get a book with a rubber band, since it will hold everything together nice and stealthy. >>679 The range is wide enough. I often sit down in the mcdonald's of a nearby train station (without eating there, heh) and watch as people connect. I'd say 15 meters between person and router is ok. The connection speed is horrible though

>>697 You might enjoy these books: https://www.allaboutcircuits.com/textbook/ http://www.learningelectronics.net/ It's the same one. AllAboutCircuits just has this silly registration thing so you can instead download them as PDFs.

>>697 How long does it last with whatever power source you have there?

>>700 I never tried how long I could go with my 4Ah powerbank, so it should exceed ~8 hours. With the voltage regulator in the box it runs half an hour max, which is not acceptable for me, hence this thread >>698 I'll read into it. Cost usually isn't a problem when it comes to books

Out of all the things I've seen on /diy/, this is my favorite.

>>704 8 hours for a Wifi broadcasting web server is pretty good!

Ok, I got myself a little voltage booster that said it's compatible with arduinos and will boost anything that's 0.9V<. It didn't say what current it can do max though, and here's the thing: The router loses power when powering on and reboots all over again, leaving it in a bootloop. I monitored the power supply with a multimeter and the voltage drops off, so I guess it can't supply the current needed for the router to broadcast the wifi signal. Note: I tried to do it with a laboratory power source tuned to 2.4V and it worked just fine, so it's definitely not enough current. In my current setup the batteries are in pairs of 3, with six batteries in total. Is there anything I can do but buy another voltage booster and try again? The batteries should be able to supply enough current in theory, but they don't for some reason...

>>775 Yeah, power is always lost through conversion and can't be increased, so by increasing voltage you're losing current (P = IV where P is constant). What board are you using? Might be able to give you some better advice if you've got the specs. >The batteries should be able to supply enough current in theory, but they don't for some reason... Out of curiosity, where are you getting that they will provide enough current? Battery cells have a property called internal resistance, such that drawing current causes a voltage drop. IR increases with age, too - if they're old cells it's possible they'll not be able to provide enough current at the required voltage.

>>775 >>777 After looking at a few pictures and assuming you're using the TL-MR3020 board, https://wiki.openwrt.org/toh/tp-link/tl-mr3020 gives a value of 255mA peak draw at 5V, but interestingly notes that the regulator works down to 3.3V, so you can get away with using a single Li-ion cell - in which case I'd recommend grabbing a couple of 18650 cells off eBay (or gutting a modern USB battery pack, or salvaging from old laptop batteries) and just running it off them directly. If you can track down some actual NCR18650Bs then with a 1s4p configuration you'll have a nice 13400mAh to play with. One of the little TP4056 modules will let you charge off a USB source.

Bump.

>>775 It may not be easy to debug without an O-scope. I have built several SMPS and I can tell you that you may not be able to identify the problem without being able to see what is happening. A multimeter and measuring steady state power expectations is don't give you any information about startup transients. There are a couple different things you could try. One is simply to really go through the data sheet and compare the worst case performance of your PSU to expected startup transients. Another is to add a crapload of capacitance to the power supply output. Ultimately, you may have to see the transient (and the transience of other lines) to determine what it is doing. So to summarize: the random guess of a drunk internet stranger is that your little boost module does not provide enough current for the startup transient of your router. Voltage boost modules modules are often really simple little chargepumps that can't output much current. Data sheet, maybe?

>>880 I don't have a datasheet unfortunately, but I can try to find one when I'm back from work. As to the capacitors, I really might be able to grab some high caps from the ee department without anyone noticing. The problem I see is that there'll probably not be enough space in the little book, so I might have to change shells at some point. >>859 Thank you, it's nice to see that there's still people interest in this