Pandora 2: Battery

[WizardStan]

Exactly. Making it so you can stack two batteries is theoretically possible, but the complexity is sufficiently high and easy to get wrong that I would actually prefer a device that didn't have that feature.

 

[Binky]

I was suggesting that the P2 could accept one 3Ah battery, or one 6Ah battery. This would only require one set of charging circuitry.

I'd imagine that this would only work if large amounts of control were done inside the battery itself, to ensure that the small battery was not overcharged on the assumption that it held 6Ah.

For example, the battery could be held (semi)-permanently in a removable base section of the case, along with a charge controller.

As a rough suggestion (don't know if this would be at all practical), there could be three lines out of the battery unit:

1. earth/ground/0V
2. +ve (5V or 3.7V depending on other considerations)
3. serial data out

The charge controller would 'know' the capacity of its battery, preventing mistakes and dangerous software features - and would report capacity and approximate charge left periodically to its host.

Another possibility would be have a charging port on the battery (as with Lego NXTs), to allow charging multiple batteries simultaneously, even in the absence of a P2

 

[EssoAir]

I was suggesting that the P2 could accept one 3Ah battery, or one 6Ah battery. This would only require one set of charging circuitry.
I'd imagine that this would only work if large amounts of control were done inside the battery itself, to ensure that the small battery was not overcharged on the assumption that it held 6Ah.

For example, the battery could be held (semi)-permanently in a removable base section of the case, along with a charge controller.

As a rough suggestion (don't know if this would be at all practical), there could be three lines out of the battery unit:

1. earth/ground/0V
2. +ve (5V or 3.7V depending on other considerations)
3. serial data out

The charge controller would 'know' the capacity of its battery, preventing mistakes and dangerous software features - and would report capacity and approximate charge left periodically to its host.

Another possibility would be have a charging port on the battery (as with Lego NXTs), to allow charging multiple batteries simultaneously, even in the absence of a P2

And you plan on designing the board that regulates this? Seriously! Do some research! The charging regulators have to be analog. Digital cannot account for the analog charging patterns of DC power. (yes, there is a digital controller, I know) You would have to design the board by yourself. (I personally know the men that design the chargers for 70% of the world's electronics - Google "Texas Instruments Knoxville Tennessee") The people that design those boards spend months on a individual design and they all have masters in electrical engineering. Correct me if I'm wrong, but no one on the Open Pandora team has any experience in designing and manufacturing analog board from scratch. 

So this whole concept may sound easy, but it's not. Unless you wants to spend millions of dollars paying someone else to do it, it isn't happening.

 

 

[Binky]

As I said, the above was a random sketch. The below is the sort of thing which is actually available:

This looks somewhat suitable, does it not?

http://www.atmel.com/devices/ATMEGA8HVA.aspx?tab=documents

(a programmable microcontroller with advanced battery management features)

 

[EssoAir]

imagining

There's the problem

 

[Binky]

imagining

There's the problem

See massively edited post - I did some research and think I have found an answer

 

[EssoAir]

Cool! What did you find?

 

[Binky]

http://www.atmel.com/devices/ATMEGA8HVA.aspx?tab=documents
(a programmable microcontroller with advanced battery management features)

These things called Integrated Circuits (google it) are great for simplifying PCB designs, aren't they?


I apologise for being sarcastic, but did you really think that your friends at TI were designing PCBs with only discrete components? The stuff they build is designed to be incorporated into other people's systems.

Same goes for the people over at Atmel.

 

[EssoAir]

Ofcourse I didn't think they were descrete. They made me sit through a 2 hour presentation on EXACTLY what they do (I wish I had those two hours back)

 

[WizardStan]

Two batteries of differing capacities is an easy thing to charge. I was concerned with the other idea someone had about stackable batteries: there's a lot of complexity in that and even if someone said they got it working I wouldn't have 100% trust that MWeston could implement it as such. Theoretically he could implement it perfectly, but batteries are weird sometimes.

 

[Binky]

Ofcourse I didn't think they were descrete. They made me sit through a 2 hour presentation on EXACTLY what they do (I wish I had those two hours back)

OK. Sorry for any misunderstanding there.


My surmise was than an electronic system could be made so that it charged batteries correctly and reported remaining charge to a host system. I believe this to be correct. I also believe that such a system, incorporating commercially available ICs, would be within the abilities of OPt

However the details I gave were not, and did not attempt to be accurate, although I think they were plausible. I was providing them to clarify the broad concept of what I was suggesting.

I don't understand battery battery management, but I know you can buy chips from people who do.

The example up there ^^^ somewhere relies on plenty of wizardry from your friends at TI (or their contemporaries), but still seems possible.

 

[EssoAir]

Actually, next simester, my school gave me permission to do another internship with them and have it count as credit. I'm going to be working with wireless chargers. I think it would be cool if we could integrate wireless charging into the Pandora

 

[Binky]

Do you know how efficient wireless charging is?

I'm guessing that it would be quite in efficient and only work if the P2 were seated in exactly the right place on a charging base.

 

[EssoAir]

Its just as efficient as a micro usb cord. Check out this http://www.ti.com/ww/en/analog/wireless_power_solutions/index.shtml?DCMP=hpa_pmp_wireless11_en&HQS=pmp-wireless11-bpowerportal-en

 

[3_bit]

If we could get a better processor and graphics, then make it thinner, but I think 5 hours is a good target, and with ARM it's hard to undershoot that. I would love to see at the very least 32 GB internal storage and 1GB RAM, possibly upgradable?

 

[Binky]

If we could get a better processor and graphics, then make it thinner, but I think 15 hours is a good target, and with ARM it's hard to undershoot that. I would love to see at the very least 32 GB internal storage and 1GB RAM, possibly upgradable?

I think you missed out a '1' there

 

[monstercameron]

the battery wont be removable so we dont go through the whole pandora battery thing again and lose all the money to piracy

seriously, no one got this joke?

 

[thatgui]

the battery wont be removable so we dont go through the whole pandora battery thing again and lose all the money to piracy

seriously, no one got this joke?

maybe too few psp owners around, or it was just too obvious, and nobody dared to say it

 

[laurencevde]

Moving the USB-port won't help, as you still have the shoulderbuttons.

If you move the stylus-storage to the screen, you can get 33% additional space for the battery (though currently the headphone-connector and volume-wheel are a bit in the way). If you do that, you can make (well, almost must, as there's not much case left) the entire bottom-side of the Pandora, except for the shoulderbuttons and connectors, one big flat surface onto which you click the battery. You could then also offer a bigger battery that also adds two additional shoulderbuttons.