Trying to build/get a USB charger that can handle phones/tablets...

Nazo

2[H]4U
Joined
Apr 2, 2002
Messages
3,613
Ok, so I've recently started getting the hang of using lithium ion batteries for everything and have a bunch of fairly good quality 18650s by Panasonic and Samsung (which are two of the best, so I'm not using cheap UltraFires or whatever.) These batteries are capable of quite a lot of performance, so should be quite sufficient to most tasks, including this one. I've been connecting them to DC-DC converters set to output 5V and they seem to work great for most things. Unfortunately, with things like my Samsung Galaxy S3 smartphone and my Google/Asus Nexus 7 tablet, they worry me. On chargers I've bought in stores (which also used 18650s, but usually cheap unbranded ones) they get so hot that insulation on the wires melt or that sort of thing. So I took one of these DC-DC converters and set it up in a metal tin with a heatsink on each side and some thermal epoxy not just holding the heatsinks on, but connecting them to the tin such that the whole tin becomes one giant heatsink to some extent. When I connected my phone though, it started making a horrible buzzing that worried me and also got extremely hot, so I didn't leave it connected for long. As I recall I had similar issues with a different one in the past.

As far as I can tell, this should be WAY within specs. The power supplies that come with these devices aren't rated for any higher than that converter is made to handle without any extra cooling (though in retrospect the ultra-tiny converters they use on those devices they sell in the stores probably can't handle more than 500mA or so though they claim 1A usually.) In fact, I think they aren't supposed to actually be pulling more than 1A, right? Or do they just pull whatever the device will allow or something crazy like that? One thing I've been wondering is if I'm going to have to do some sort of current limiting. If it's just pulling as much as it possibly can without the device actually limiting it itself, maybe it's going up to ridiculous amounts when the device doesn't actually limit it to 500mA or whatever (though it seems like with cooling it should be able to handle 3+ amps and surely it's not drawing more than that?) Or is it something else? It's hard to believe it could possibly be pulling anywhere near to as much as 2A even, much less anything more than that even if it's trying to do an ultra-fast charge. Could the problem be something else? The fact that it made even official chargers struggle so badly too makes me wonder though.

EDIT: I forgot to ask, but I do want to also ask about those DC-DC converters with a "constant current" control. Does this actually try to force even a minimum current? I know they tend to be used for things like LED driving and battery charging (so most are down stepping anyway) so I'm wondering if they literally try to make it constant to the point that they might try to force a minimum? If so, I have no idea what would happen, but I would be afraid to do that to any USB device. If they only limit the maximum obviously that would work fine, so I'm kind of hoping that's all it is.
 
Last edited:

CaptNumbNutz

Bulls[H]it Master
Joined
Apr 11, 2007
Messages
21,342
Couple things I noticed...

1. What's the source voltage you are using? I don't see that mentioned anywhere.
2. I hope you realize that DC-DC converter is a voltage booster, meaning it will pump out more volts than you give it at the sacrifice of total amps. Have you measured the voltage that is being outputted? If its too high that would be the reason your phone is getting hot. If the goal was to create some sort of quick charger using a couple low voltage alkalines (1.5v each, 2 batteries providing less than 3.5v this device is rated for), then this is the device to use.
3. Many older touchscreen smartphones can use 5v 500mah to charge with, however most newer phones and tablets need 1a or more to charge in any decent amount of time. Most phones now can slow charge at 500mah, but not all can and most tablets can't either. Even a Samsung Galaxy Note 2/3 will use a 2a charger. If you plan on charging tablets, you will want to provide nearly 2 amps. Keep in mind the device will pull as much as it needs up to a certain limit, but for fastest charging the charger needs to provide the same or more than the device wants.
 

Nazo

2[H]4U
Joined
Apr 2, 2002
Messages
3,613
1. What's the source voltage you are using? I don't see that mentioned anywhere.
Like I said, lithium ion batteries. 18650s to be specific. 4.2V down to potentially as low as about 3.0 or in the case of the newer Panasonics, 2.5V to achieve a higher total capacity -- which isn't nearly as useful as it seems since nothing can go that low -- though the converters crap out about 3.3V or so and are only rated down to 3.5V. It was a fully charged battery (at least 4.1V) providing the input. As far as I can tell, these converters have about a 0.5V minimum step up, so as long as it's below 4.5V I think it's ok even with the 4.35V batteries (not sure why they won't clearly state that, but based on their minimum input versus their minimum output I have to assume it's a 0.5V minimum step.) I haven't tried the 4.35V ones though, just to be on the safe side.

BTW, just to be clear, these batteries are able to handle at least 3C as far as I know (I forget exact numbers and I have different ones, but the point is, massive current output is possible without danger to the battery, albeit at the cost of the battery's total capacity and lifetime. This shouldn't be using anywhere near to 1C though from the battery.)

2. I hope you realize that DC-DC converter is a voltage booster, meaning it will pump out more volts than you give it at the sacrifice of total amps.
That is the point given that it's to power USB devices which require 5.0V +/- ~10% from a source that isn't 5.0V.

Have you measured the voltage that is being outputted?
Very little variance. It's hard to get it exactly on 5.00 since you have to turn a little pot, but the minimum is 4.98 and the maximum about 5.01 or so.

your phone is getting hot.
I didn't say it was the phone getting hot. Yikes! If it were I would unplug it very very quickly and never touch that thing again! I don't want to tear it up... It's the DC-DC converter that gets exceptionally hot despite the added heatsinks and makes the buzzing noise. But then for all I know that could still tear up the phone, so I want to be very very careful still.

3. Many older touchscreen smartphones can use 5v 500mah to charge with, however most newer phones and tablets need 1a or more to charge in any decent amount of time. Most phones now can slow charge at 500mah, but not all can and most tablets can't either.
I'm aware of this. And presumably they won't even try to limit to 500mA unless the input itself is being limited externally rather than by them. Well, I actually don't mind a slow charge, but this is part of wondering if I'm just going to have to give up and put a current regulator on there.

Actually, I meant to ask something important but forgot. Some of these DC-DC converters (albeit usually the down step ones instead of up) have "constant current" controls so they can do things like handle LEDs and charging (though I think it's a bad idea to use one for charging lithium batteries since you're supposed to use current on a curve instead of constant.) Does this "constant current" literally try to force a minimum current, or does it simply limit the maximum? If it tries to force a minimum it could be a problem and probably I'd just have to set it to something low like 500mA and use it only for this stuff, but if I could use it to set a maximum then I could continue to keep these converters as adaptable devices that I can use for multiple things (for instance, I've powered 12V fans via these before.) I'd rather keep it a relatively universal tool if possible.
 
Last edited:

CaptNumbNutz

Bulls[H]it Master
Joined
Apr 11, 2007
Messages
21,342
Like I said, lithium ion batteries. 18650s to be specific. 4.2V down to potentially as low as about 3.0 or in the case of the newer Panasonics, 2.5V to achieve a higher total capacity -- which isn't nearly as useful as it seems since nothing can go that low -- though the converters crap out about 3.3V or so and are only rated down to 3.5V. It was a fully charged battery (at least 4.1V) providing the input. As far as I can tell, these converters have about a 0.5V minimum step up, so as long as it's below 4.5V I think it's ok even with the 4.35V batteries (not sure why they won't clearly state that, but based on their minimum input versus their minimum output I have to assume it's a 0.5V minimum step.) I haven't tried the 4.35V ones though, just to be on the safe side.

BTW, just to be clear, these batteries are able to handle at least 3C as far as I know (I forget exact numbers and I have different ones, but the point is, massive current output is possible without danger to the battery, albeit at the cost of the battery's total capacity and lifetime. This shouldn't be using anywhere near to 1C though from the battery.)


That is the point given that it's to power USB devices which require 5.0V +/- ~10% from a source that isn't 5.0V.


Very little variance. It's hard to get it exactly on 5.00 since you have to turn a little pot, but the minimum is 4.98 and the maximum about 5.01 or so.


I didn't say it was the phone getting hot. Yikes! If it were I would unplug it very very quickly and never touch that thing again! I don't want to tear it up... It's the DC-DC converter that gets exceptionally hot despite the added heatsinks and makes the buzzing noise. But then for all I know that could still tear up the phone, so I want to be very very careful still.


I'm aware of this. And presumably they won't even try to limit to 500mA unless the input itself is being limited externally rather than by them. Well, I actually don't mind a slow charge, but this is part of wondering if I'm just going to have to give up and put a current regulator on there.

Actually, I meant to ask something important but forgot. Some of these DC-DC converters (albeit usually the down step ones instead of up) have "constant current" controls so they can do things like handle LEDs and charging (though I think it's a bad idea to use one for charging lithium batteries since you're supposed to use current on a curve instead of constant.) Does this "constant current" literally try to force a minimum current, or does it simply limit the maximum? If it tries to force a minimum it could be a problem and probably I'd just have to set it to something low like 500mA and use it only for this stuff, but if I could use it to set a maximum then I could continue to keep these converters as adaptable devices that I can use for multiple things (for instance, I've powered 12V fans via these before.) I'd rather keep it a relatively universal tool if possible.

Ok. Thanks for clarifying. Sounds like you know more about this than I do. Someone with more electronics expertise needs to chime in.:cool:
 

Mohonri

Supreme [H]ardness
Joined
Jul 29, 2005
Messages
5,751
If you're willing to wait a few weeks for shipping, this may be right up your alley (EDIT: you are obviously already familiar with DX, so maybe this isn't an option). It's an enclosure that takes four 18650 cells of your choice, charges them when it's plugged into a USB port, and acts as an emergency battery for a cell phone or tablet.

There are a couple things to note here:
1) when you supply 5V to a phone or tablet, a regulator inside the phone drops that down to 3.6-4.2 for the Li-ion or Lipo cell inside (and the rest of the phone's electronics). So the actual voltage going into the phone doesn't have to be perfectly at 5V. That leads to...
2) The cheap AC adapters aren't particularly well-regulated (except for Apple's, I suspect), so their output voltage is likely to sag when the device is drawing a lot of current. The voltage sag also frequently comes with a very nasty-looking waveform, which is probably what caused the noise from your phone.
3) I looked up the datasheet for the regulator on that DC-DC converter, and I'm not convinced it would be able to output a very clean waveform in the best of cases, particularly when under a heavy (1-2A) load. If you had an oscilloscope, you could test it out :) I'm guessing the device you bought wasn't designed for the higher currents that you're using.

Out of curiosity, what kind of wire are you using for this?
 

Nazo

2[H]4U
Joined
Apr 2, 2002
Messages
3,613
I used heavy wires as much as I could. The actual battery clip has some fairly short wires that are relatively thin, but I think they can still handle more than I've come remotely close to using in any of this (they've never gotten warm at all in fact.) With the exception of the battery clip itself, I used wires from a computer power supply in fact.

Anyway, what I really wanted was a DC-DC converter that's powerful enough to handle a variety of things, not just these. I particularly liked the idea of it being adaptable to different voltages in fact. I've powered things like 12V fans, used these to test effects of lower voltages on things (for instance, finding the starting voltage of some of these fans so I could run them at a lower, quieter voltage in certain things.) I want to be able to use these in emergencies and situations where I'm without power for a long time to do any of a number of things (for instance, powering a 3.5" drive long enough to pull some data from it.) So that's part of why I don't want one like that which is specifically tailored to one specific task. But I also want to be able to build something relatively small that uses only one battery rather than four...

And yeah, I know they are essentially doing their own downconverting -- whether a regulator, simple resistor, or whatever. This is technically waste, but short of directly connecting the batteries together with some sort of regulator (and again voltage is still an issue since lithium ion is charged with constant voltage and only the current varies -- doubly so since my phone uses a newer technology 4.35V battery and almost beyond a doubt the Nexus 7 would as well.) Plus it could be quite tricky to actually externally charge the tablet's battery, even if it doesn't use weird screws or anything. Basically I just have to play along with their intended methods unless I want to shut it off and use an external charger even if there is technically a lower efficiency than theoretically there could be. (At least these things generally stay really close to or above the 90% range.)


Anyway, I guess I'm asking just what DC-DC converter I need for something relatively universal (and preferably adaptable) if that one is as insufficient as you say. I'm guessing by the buzzing noise that it was indeed insufficient for the task (though I'd swear I remember this phone being no more than than 1A...) but I have no idea what to look for then. In particular though, I wan to be able to use one battery.

Actually, I'm assuming that thing listed above runs them in series and then downconverts. It's not parallel is it? That could potentially be very bad... There is a reason batteries are supposed to be carefully matched to use them in packs.
 
Last edited:
Top