Gp2x And Li-ion Batteries.


DaveC

Internal Development
Joined
Mar 4, 2004
Messages
9,208
Well the big quandry with batteries has been life vs availability.

AA batteries are easily replaced and pretty cheap but the bad thing is they don't last very long sometimes, especially with overclock.

Li-Ion batteries are usually propriatary so getting and replacing them can be expensive and a pain in the ass. It can be a dead end when the company making the custom packs goes under or discontinues the product. It can leave your handheld a worthless brick. This is why I have railed against them.

Here can be a solution that could be the best of both worlds. They are a bit bigger than AA batteries. They are also a kind of STANDARD. You can get them very easy from many sources, and alot of products use them. This almost insures that you will probably be able to get them for a lifetime.

I wonder how hard it would be to mod a GP32/GP2X to use these? You could put 2 in parallel for a whopping 4000 MAh. The voltage is higher too. Remember POWER= Current x Voltage. So a battery at 3.7 volts at 2000 MAh gives more power than (2) 1.2V (2.4V) @ 2000 MAh.

I think 3.7V should be within the 3V power requirements of the GP32/GP2X. I don't think it would fry it.

http://batteryjunction.com/186502200pcb.html
 
I think 3.7V should be within the 3V power requirements of the GP32/GP2X. I don't think it would fry it.

http://batteryjunction.com/186502200pcb.html
I think epicenter did bench supply tests of varying voltages and 3.7 is OK, some others may have too.

3.7 runs more stable and for longer too according to tests (or popular opinion), especially when overclocking. I think maybe one at a time would be best, and probably last as long as 2x nimh AA's.

As far as a solution with these batteries, first they need to fit, no problem it's just plastic after all.

Secondly we need an intelligent battery controller, can be as simple as a chip that cuts power if the Li-ion discharges too quickly or too low (no need for an exploding GP2X section to the forum ;)), or can have an integrated charging circuit included which makes for a more complex circuit.

Either way I am not an electrical engineer, I would have to buy a Li-ion battery controller, and they ain't cheap.

If there are any electrical engineers about, maybe they have a suggestion for a simple cheap controller?

EDIT: Me and my big mouth:

http://batteryjunction.com/li18322mahre.html

Turns out these batteries here have a controller, just cut the protection PCB off and use it with any replacement cell when these batteries die :D
 
Last edited by a moderator:
So with two of these batteries hooked in parallel (4400mAh at 3.7v) I would get 16280 watts, as opposed to my typical 7250 (2900mAh at 2.5v). That certainly looks nice.

I think the main thing I wonder, is how long would these batteries last before dropping to a voltage unusable by the GP2X. I mean, the voltage on a NiMH won't drop significantly until right at the end of it's life, but I was thinking that Li-Ion had a steady decline to it. (or am I missing something.. oddly enough I think I only ever owned one thing which used a Li-Ion battery. So I might not be the best judge of them.)

Also, there is the issue of how they fit in.. Length wise, it looks like it just might be possible, and the height looks doable as well, but it seems the compartment might have to stick farther out the back then it already does... (but hopefully not too far)

I wouldn't go with the ones with the built in PCB, they add an extra 2mm (and the extrusion on the case looks like it might only allow the 65mm batteries.) They sell PCBs on the same site for not too much so it wouldn't be too hard to build them into the GP2X (it even includes a nice little wiring diagram)

http://batteryjunction.com/prcimopfor3l.html

I wonder if there is any way this mod could be done and still look pretty. (at this moment I'm thinking, "just paint the batteries black, and live with the fact that they're showing..") I think I'll give this a try as soon as my disposable income returns to me.
 
Would it not be much easier, and possibly cheaper to use C or D batteries.

D batteries are available at over 10000 mah and could easily be hooked up to the standard battery compartment or even via a dc plug. If the dc plug is used you could just use a slightly longer lead and have the batteries in your pocket.
 
Would it not be much easier, and possibly cheaper to use C or D batteries.

D batteries are available at over 10000 mah and could easily be hooked up to the standard battery compartment or even via a dc plug. If the dc plug is used you could just use a slightly longer lead and have the batteries in your pocket.
Yeah D batteries would work well imo.

Would it not be much easier, and possibly cheaper to use C or D batteries.

D batteries are available at over 10000 mah and could easily be hooked up to the standard battery compartment or even via a dc plug. If the dc plug is used you could just use a slightly longer lead and have the batteries in your pocket.
Yeah D batteries would work well imo.
 
Last edited by a moderator:
Be careful - the batteries will probably be at a higher voltage when fully charged, and it might be higher than you'd want to connect to the voltage regulator. Kinda like NiMH are rated at 1.2V, but are typically around 1.5V briefly when fully charged. 3.7V is probably OK, but once you start getting around 4V, I'd be a little more worried. The regulator is supposed to be able to handle up to 5V, but it may be wired in a configuration that doesn't want it that high.
 
Be careful - the batteries will probably be at a higher voltage when fully charged, and it might be higher than you'd want to connect to the voltage regulator. Kinda like NiMH are rated at 1.2V, but are typically around 1.5V briefly when fully charged. 3.7V is probably OK, but once you start getting around 4V, I'd be a little more worried. The regulator is supposed to be able to handle up to 5V, but it may be wired in a configuration that doesn't want it that high.


I guess you could always throw in a zener diode to block anything over 3.7V.

Would it not be much easier, and possibly cheaper to use C or D batteries.

D batteries are available at over 10000 mah and could easily be hooked up to the standard battery compartment or even via a dc plug. If the dc plug is used you could just use a slightly longer lead and have the batteries in your pocket.

C or D batteries would be too big. You couldn't just mod the back of the GP2X to accomodate the new batteries. These li-ion cells are only a bit bigger so it would still be possible to integrate with the unit.

I already tried making a pack and plugging it in to the AC port, didn't work for some reason. It just came up with a screen full of vertical lines as if the batteries were low.
 
Last edited by a moderator:
I tried making an external battery pack that used 4 D cells (which I planned on using for camping trips) It would have worked nicely, but the plug I put in the GP2X was cut off of an IDE plug, so it doesn't allow enough amperes through.. (this plug was wired straight to the battery terminals, I've always been to scared to try plugging it into the mains port) So despite my little set back I still love external packs, however I think it would be cool to have a good internal solution. (or at least, mostly internal :p )
 
I already tried making a pack and plugging it in to the AC port, didn't work for some reason. It just came up with a screen full of vertical lines as if the batteries were low.
Apparently, the 2X draws more power from the DC port than it does from the batteries. When I was testing a power supply, I noticed that at 4.5v (The supply I use is only 850mA so 3.0v isn't enough) I'd just get vertical bars when plugged to the DC port, but it would work when plugged in where the batteries would go.
 
I already tried making a pack and plugging it in to the AC port, didn't work for some reason. It just came up with a screen full of vertical lines as if the batteries were low.
Apparently, the 2X draws more power from the DC port than it does from the batteries. When I was testing a power supply, I noticed that at 4.5v (The supply I use is only 850mA so 3.0v isn't enough) I'd just get vertical bars when plugged to the DC port, but it would work when plugged in where the batteries would go.

The fact is that from the battery terminals it goes into a voltage regulator.

If using the side port damage can result from over-voltage, or it just won't work as you discovered.

On a side-note, try measuring the voltage under load, i am sure it drops quite low.
 
Last edited by a moderator:
You guys might want to be careful about building your own adapters and battery packs. I don't know how the MAX1566 power supply chip in the GP2X is wired up, but from the datasheet I see that there are control pins that determine the input RANGE that is expected from the battery terminals.

Voltages higher than what is put out by two cells seems to require different settings than what the GP2X would be set to. Now, I only skimmed the datasheet but since this is a 2AA cell application, there is no reason GPH would wire it differently than what would be expected. As a result, powering the GP2X with more than 3.3V could damage it.

Some boost only regulators (such as the one that creates 3.3V if that is a power rail in the GP2X) can start to track the input voltage if the input starts to exceed the desired output. I can't say exactly how bad this could be, but lets say the input is 4V and the output starts boosting up to 3.8V. That would put everything out of range which could fry parts or make them warm and also could prevent proper boot up of the system. I think some people have had good luck using inputs greater than 3.3V but they might be just pushing their luck or reducing the long term life of their GP2X.

I think making a two cell external pack such as D cells should be successful but someone with more time and interest may want to dig into that datasheet more before using higher cell voltages such as 3.6V lithium packs. There might be an overvoltage tolerance range and 0.3V over could be acceptable. It's just a thought. I haven't studied it that hard and maybe a quick email to someone at GPH would be your best bet. Good luck! :)
 
This is just a heads up that I'm working on this, in case anyone is still even monitoring this thread.

I've almost completed the mod, I still want to add a toggle switch to disconnect the PCBs so that I can make a cradle for using regular AAs without the PCBs sucking all the current out of them.

I couldn't get the batteries tucked in as far as I originally planned because there was a capacitor in the way.

I've tested it before wiring it all up and the batt test gave me 9.5 hrs at 200MHz. I want to give it another test before I post the official report on this.

And here are some pictures:





My next post will most likely be a main thread announcing this concept.
 
DawnOfTheRent said:
This is just a heads up that I'm working on this, in case anyone is still even monitoring this thread.

I've almost completed the mod, I still want to add a toggle switch to disconnect the PCBs so that I can make a cradle for using regular AAs without the PCBs sucking all the current out of them.

I couldn't get the batteries tucked in as far as I originally planned because there was a capacitor in the way.

I've tested it before wiring it all up and the batt test gave me 9.5 hrs at 200MHz. I want to give it another test before I post the official report on this.
My next post will most likely be a main thread announcing this concept.



Cool that is exactly what I was going to do, you saved me time to test :p I am a bit suprised that you only got 9.5 hours at 200 MHz though. Kind of odd considering that two 1.2 V NiMH batteries which gives 2.4V at 2500 MAh can give 5 hours at 200 MHz. With your batteries it should be 3.6V at 5000 MAh which I would think would give way more than slightly less than double the time. It is more voltage and double the capacity. Do you have any kind of diode to prevent the batteries from discharging into each other?
 
Last edited by a moderator:
You should NOT use diodes to prevent one batteries discharging into each other. Why? Because you're gonna waste hell out of your batteries' power!
Each diode would consume 0.7(some just 0.5) out of 3.7v of the battery, so it's like having a 3.0V battery on gp2x... that can be the answer for those who are afraid of overloading gp2x with more than 3.7vdc, but it's doom for those who are trying to push the longest gameplay into account.

The easiest way I've found to use rechargeable batts in parallel is to let it charge and after that having a single 0.1ohm resistor between them for 10 minutes... As one can think, the bigger the resistence and the longer the time, more precise and safer the result shall be, it all depends on patience.

And talking about patience, as long as you have a higher-enough DC source, you can use it to recharge your batteries, it all depends on how much you care about the battery life span, worry about the possibility of blowing it away and how desperate you are to charge the batt, a +5vdc source is easily obtainable (from USB or a molex connector, both from compy), then use a rectifier diode just to make the voltage drop from 5 to 4.3bdc (often 4.4 or 4.5) and then a resistor to limit the current and we're set in a quite decent but slow charger (all chargers that don't have any sensor/timer/etc feedback must be named slow, but you can let higher current in circuit for the first hours, just be cautious!)

If you don't want to wait 10 minutes after recharge, you can just use 1 resistor in each battery, they will by no mean let batts blow each other since the tension difference gotta be less than 0.1vdc, so take a very, very low resistance
 
Wow,.. I wonder how defective my GP2X might be.. I get just over 4 hours with my 2900MAh AAs (and I'm also wondering if my AAs I cared so much about are actually just a gimmick of some sort..)

I also used the 2600MAh Li-Ions so my total amp hours goes to 5200MAh... I think the main culprit is the PCBs,.. I'm pretty sure (I just woke up) I read somewhere that they cut off at 2.5 volts to prevent over drain on the batts, so that could be a big killer. I did hook up each battery to it's own PCB in it's own circuit, so if there is any diode in there it would have to be on the PCB because I didn't add one of my own (and since the PCBs seem to be pulling current from the hook up for AAs (I realize this might require a little more explanation, I'll draw you a pic when I get back from work, but for now I have to go) it would seem that they would not have such a diode).
 
when you use 2x2600 batteries on the gp2x, they're placed in series connection, not parallel, so you should not claim 5200 mAh rating, instead you should say 2600mAh@2.4Vdc (which is the same power rating as 5200mAh@1.2Vdc)

4hours at 2.4v/2.6Ah is pretty much acceptatble
 
Lint said:
when you use 2x2600 batteries on the gp2x, they're placed in series connection, not parallel, so you should not claim 5200 mAh rating, instead you should say 2600mAh@2.4Vdc (which is the same power rating as 5200mAh@1.2Vdc)
Li-ions give 3.7V (or at least all those I have seen do). These would be wired in parallel.
 
Last edited by a moderator:
I don't even want to think about how dead my GP2X would be if I placed those batts together in series.. (yes, mine give around 3.7v too).

I've basically finished the mod though, I'm about ready to start writing the new thread opener, but I still want to do some more testing first.

Oh, and to answer Tripminkey's question (which I didn't even realize was a question until today); No, there isn't too much cutting done on the GP2X it's self, mainly I just widened the battery bay's opening as much as I could to fit the larger batteries. There was a lot of cutting that I done on a CD case in order to make the brackets that hold the battery terminals and stuff.

Oh no, here comes more cheap camcorder pictures:


 
Back
Top