GP2X Leading By Inappropriate Examples?


Squidge posted on Mar 2 2007 at 07:39 PM said:
If you just want to multiply the capacity and not the voltage, then yeah, just connect them together in parallel. Ie. +ve to +ve, -ve to -ve. Make sure they are all charged first though. If some are charged more than others, then the batteries could eventually be damaged as they would be taken past there "recharge" point.

That is the problem. It is almost impossible to have 2 batteries with the EXACT same charge. Things like contact pressure, oxidation, manufacture variations will surely mean that one battery would be slightly more charged than the other if only by a very slight amount. I don't know if that could be a problem.

Can the GP2X handle 3.7V? Or would the extra amount reduce the life of the unit?
 
Last edited by a moderator:
There's always very slightly differences, yes, but that slight different isn't worth worrying about. Its if you put something like a full one in and a 3/4 full one in that you could damage the 3/4 full battery, and only then after a number of cycles which could be months.

As for the voltage, it seems to prefer 3.7V. If you provide voltages between 1.9V and 4.5V to the battery compartment via equipment that can increase that voltage in 0.1V steps, and then measure the load, it seems the load gradually decreases very slightly as you increase to 3.7V and then tails off. Before you get confused, don't mix up current and load. The current will obviously decrease constantly with voltage increase, but the load is calculated via ohm's law.
 
DaveC posted on Mar 2 2007 at 06:35 PM said:
Well not really. I wanted to make a pack with 2 of these in parallel:

http://www.all-battery.com/index.asp?PageA...amp;ProdID=1600
Yeah, I know what you wanted to do. :D I think everyone's tried it at some time.

It will work. You can stick two batteries of more or less equal charge state together and they will power the device.

It wont last twice as long though, and if you keep doing it one of the batteries will progressively become weaker until it dies completely or drags the other one down so much you end up with less performance than a single battery. It's just the way it is. :mellow:

Those cells you link to are Li-Ion, which are prone to catching fire if mistreated. Personally I wouldn't fancy pushing my luck with that.

How about a different approach? Take something like this which only requires a couple of external components, and run it from a pair of those Li-Ions in series. It may even be hackable into the existing battery compartment. Should supply something like 3x normal battery life.
 
Last edited by a moderator:
Laser posted on Mar 2 2007 at 09:18 PM said:
DaveC posted on Mar 2 2007 at 06:35 PM said:
Well not really. I wanted to make a pack with 2 of these in parallel:

http://www.all-battery.com/index.asp?PageA...amp;ProdID=1600
Yeah, I know what you wanted to do. :D I think everyone's tried it at some time.

It will work. You can stick two batteries of more or less equal charge state together and they will power the device.

It wont last twice as long though, and if you keep doing it one of the batteries will progressively become weaker until it dies completely or drags the other one down so much you end up with less performance than a single battery. It's just the way it is. :mellow:

Those cells you link to are Li-Ion, which are prone to catching fire if mistreated. Personally I wouldn't fancy pushing my luck with that.

How about a different approach? Take something like this which only requires a couple of external components, and run it from a pair of those Li-Ions in series. It may even be hackable into the existing battery compartment. Should supply something like 3x normal battery life.


So you increase the voltage, then use a device to drop it back down? Would that give more run time or just waste more energy as heat through that regulator?

That is why I thought you could use some sort of blocking diode that isolates each battery in parallel. With such a diode one battery won't be effecting the other.
 
Last edited by a moderator:
Yay, I remember this discussion like it was 8 months ago! (probably because it was 8 months ago, and I have a wacky photographic memory, so I remember better than if it was yesterday ;) ).
 
Well, my laptop battery was fubar a couple of weeks ago, and I made another by just combining 8 batteries together in parallel. Works great, and even recharges with the original lead.

So I'd suggest you try that (hey, there's no learning without experimentation :) ) and see if it blows up in your case.

(Yes, the batteries I took out of the original battery were all linked in parallel too, but they were unknown kind with no writing on them)

"Suck it and see" :)

The regulator thing will work, but it will obviously require a decent heatsink for powering the gp2x, along with most likely some stabilization caps, and then the first thing the gp2x will do is run that through a DCDC convertor :D
 
DaveC posted on Mar 2 2007 at 11:21 PM said:
So you increase the voltage, then use a device to drop it back down? Would that give more run time or just waste more energy as heat through that regulator?
Normally it would just waste heat and melt your GP2X! :lol: But the device I linked to is a switching regulator (aka a DC-DC converter) which draws a given power to match it's output power. By upping the voltage supplied to it (up to 40V in the case of that particular one) you can reduce the current drawn from the input.

So, for example, if you gave it 2x3.7V = 7.4V, and it has a 3.3V output, you'd be drawing 3.3/7.4 = 45% of the current from the batteries, less some efficiency figure. (Might end up nearer 60% of the output current.)

This is exactly the same as what Squidge was saying about the GP2X having some kind of DC-DC converter inside so that supplying it with 3.7V draws less current (same power) as 2.4V. If we knew for sure that the internal regulator could handle 7.4V, you could dispense with this external one I was proposing and just use your 2x3.7V cells directly in the battery compartment. :eek:
 
Last edited by a moderator:
Laser posted on Mar 3 2007 at 04:00 AM said:
This is exactly the same as what Squidge was saying about the GP2X having some kind of DC-DC converter inside so that supplying it with 3.7V draws less current (same power) as 2.4V. If we knew for sure that the internal regulator could handle 7.4V, you could dispense with this external one I was proposing and just use your 2x3.7V cells directly in the battery compartment. :eek:
The spec sheet on the DC-DC converter is around on the forums, I may have a copy on my HDD at home.

If you can change a bit of the circuitry on the input side it would probably take a lot of voltage, I was thinking of using 4x AA NiMH yesterday. The real bonus is I believe it will make a few custom voltages at 100hz cycle (AKA pure clean juice), if 5volts is one of those voltages and we use 4x 2500mAh batteries we can power WiFi no problem, and probably a 1.8" or 2.5" HDD too.

If we can take a look at the circuitry and see what it is doing we can probably make it play nice with more voltage, probably not 40v, but maybe ~12v-20v.
 
Last edited by a moderator:
nubie posted on Mar 3 2007 at 04:24 PM said:
The real bonus is I believe it will make a few custom voltages at 100hz cycle (AKA pure clean juice),
Sorry, not sure what you mean by that?

If we can take a look at the circuitry and see what it is doing we can probably make it play nice with more voltage, probably not 40v, but maybe ~12v-20v.
It's likely it will tolerate at least 5V, but be careful because there are some devices that wont.

I'm more worried that there may be other things inside the GP2X that are not connected through the DC-DC converter, so they would see the actual input voltage.
 
Last edited by a moderator:
Squidge posted on Mar 3 2007 at 10:22 AM said:
Well, my laptop battery was fubar a couple of weeks ago, and I made another by just combining 8 batteries together in parallel. Works great, and even recharges with the original lead.

So I'd suggest you try that (hey, there's no learning without experimentation :) ) and see if it blows up in your case.

(Yes, the batteries I took out of the original battery were all linked in parallel too, but they were unknown kind with no writing on them)

"Suck it and see" :)

The regulator thing will work, but it will obviously require a decent heatsink for powering the gp2x, along with most likely some stabilization caps, and then the first thing the gp2x will do is run that through a DCDC convertor :D

So there was no blocking diode isolating those batteries? Would something like below work?

battery.png


What were the dimensions of those batteries? were they the same size as these 16850 ones?

http://www.all-battery.com/index.asp?PageA...&ProdID=471

Those are pretty standard and alot of laptop batteries use these. I have one of these batteries I use for a portable laser and I looked at how they could fit into the battery compartment. They are bigger than AAs but not much bigger. The length is still a bit under the length of the battery bulge now. I would have to cut out the current compartment and make a slightly bigger one. Then I could make a caddy to fit in the normal AAs as well so I would have dual power. Would it be better to wire them to the AC port or use the AA compartment contacts (re-arranged)?
 
Last edited by a moderator:
The batteries you link to seem like AA-type to me, whilst the batteries in the laptop were more like C-type.

stripping all from there plastic, the batteries were stacked together like so:

--------------------- +
||||||||||||||||||||
--------------------- -

So basically, there was a strip of metal connecting all the tops, and another connecting all the bottoms. At the end was two wires connecting the metal strips to a circuit board (Which I assume was the charging circuit). There also seemed to be a kind of thermistor in there, so obviously the battery case temperature was loosely checked.

It seems if you use the DC input, you bypass the DCDC convertor, as if you don't provide adequate voltage to that input, all kinds of strange things happen, but you can easily provide 1.9V to the battery connectors and it still works well. I'd use the battery compartment for neatness and the acceptable variety of input voltages.
 
Back
Top