Battery Life compromise.


^ I used to have a boombox back in the late 80's that used something like 10 D-cell batteries...and that thing was portable.  Sort of.
Ooohhh - we could derail this into a conversation of the least portable 'portable' devices we ever owned...

I had an IBM PS/2 model P-75 with a ~10" orange screen that I'd hot rodded into a 486/133, 16MB RAM, a pair of 1GB SCSI drives, a sound card, 10Mbit ethernet AND a token ring card.  It weighed @30lbs, but was the perfect dimensions to barely fit under the seat in front of you on a jet - hope you don't like your toes.  http://thinkwiki.de/Modell_P75
 
^ Wow that is an ORANGE screen.   :eek:   ..looks almost red in that picture.  I'm not sure if I could've stared at that screen for too long - messes with my rods & cones just thinkin' about it.

...okay, back on topic - I think that the OP Team should keep things simple and and less costly for the P2.  So, put me in the ONE battery category.  Remember, every dollar that the OP Team saves in production (no to mention R&D), is more profit - or a lesser price that they might charge for a P2...which would help move more units.  Keep it simple.   ;)
 
 
Powering each unit off two batteries could be the difference between ordering 1500 and 3000, possibly having a positive effect of prices due to the larger order quantity.

I'd also say that its more important to follow standards that to use the cheapest parts, although they may equate to the same thing because accessories will be made by third parties in very large numbers.
 
if you think about it, though, by using a common cell phone's knock-off batteries, we're entering into someone else's volume stream.  So, we're not talking about ordering 10,000 batteries designed, built, packaged and shipped JUST for our one device - we're talking about ordering 10,000 out of a production stream of 500,000 - much like we do with the screen component.  Taking advantage of someone else's economy of scale.  In theory, netting out the design, production and all that - dual batteries should be cheaper than a single in this scenario.  

Exophase is correct that there would have to be two battery control systems in our custom motherboard.  That piece has to be customized anyway.

The GS4 (abusing the example) ships with one 2600mAh battery.  Use two of them = 5200mAh, which is about right - we'd need to know the discharge rate/etc...  BUT from 5000 feet a pair of them looks like a good option to me.

It's a thought - at this point it's all about dreams.
 
As long as it is possible for batteries to be ordered specially (say near the end of the P2's commercial life when third parties are stopping making the chosen battery size), that sounds like a good idea.

I also like the idea of being able to run on a half-capacity battery.
 
The software and hardware mods needed to make a dual battery system work well is probably too much to be worth it.

Also, don't forget that two batteries will use more volume and weight than one with the same total capacity.

Relying on an existing battery could still be a good idea, but on the other hand: the battery is really not a very expensive part, so I doubt if what you can save by having somewhat better economies of scale is really worth it (and even for that it is to be seen if it ends up cheaper, because if you're not producing them yourselves, someone else has to sell them to you at some profit).

For the Pandora 1, as far as I understand, basically everything was designed, and THEN they looked at how much space was left for a battery and produced a battery that would fill that space. If you use an existing battery, you basically have to design everything around it. It makes sense to design around a screen (because it's such a critical and expensive part), but it doesn't really make sense to design around a battery.
 
if you think about it, though, by using a common cell phone's knock-off batteries, we're entering into someone else's volume stream.  So, we're not talking about ordering 10,000 batteries designed, built, packaged and shipped JUST for our one device - we're talking about ordering 10,000 out of a production stream of 500,000 - much like we do with the screen component.  Taking advantage of someone else's economy of scale.  In theory, netting out the design, production and all that - dual batteries should be cheaper than a single in this scenario.
This is assuming that the cell phone manufacturer is willing to sell them to you at a price comparable to what they buy them at. The battery manufacturers doesn't necessarily have to sell them to you directly, the cell manufacturer could easily have a lock on that volume. I'm sure Samsung will sell GS4 batteries to aftermarket repair shops, stores, etc. But since these come at a huge mark-up to the buyer who knows how much of the mark-up is included up-front in Samsung's price.

Last I was aware Pandora's battery in volumes of low thousands costs about 6 EUR. So you'd have to get a GS4 battery for ~3 EUR to match it. For comparison, I can't find a loose one for under $30. I doubt these stores have 1000% margins, the market is way too competitive to sustain that.

Exophase is correct that there would have to be two battery control systems in our custom motherboard.  That piece has to be customized anyway.
Customized why? Batteries have standard charging circuitry, standard PMICs for mobile devices are designed to interact with them. There's no special battery charger circuitry in Pandora, the custom geometry and capacity, charge/discharge rates etc of the battery don't require custom logic, just some numbers you program into it. Those numbers are stored on some EEPROM.

But something meant for handling two batteries would definitely require something different, who knows if you can even find a reference implementation to get ideas from. You could be carrying a lot of extra risk in addition to carrying cost overhead for the more complex design, which could be substantial if it means deviating from a normal PMIC setup.
 
Last edited by a moderator:
This dual battery discussion started because I/we were looking for a solution that could accommodate a heat spreader and some external fins with the idea of being able to handle an SoC with 2-3x the thermal dissipation needs as the current one.

OK - so dual batteries with heat dissipation fins between them is difficult.

To get heat out of the unit without fans requires surface area - some sort of finned heat exchanger likely.  That type of structure would work best where your hands -don't- grip the device.  That leaves an inverted T space - wide from the front where the SD slots are now then a spine bisecting the base of the unit to the back.

Single battery with heat fins idea.

Start with the existing Pandora layout.

Move the sd slots to the left (or right) side.

Use a ~24ga aluminum heat spreader with a bunch (12?) of folds making a heat exchanger on the front/bottom where the SD slots used to come through.  That's the spot where the hands don't grip the device - not that we're talking about all that much heat anyway if it has someplace to go (air).

Single battery inserts/slides in from the side opposite the SD slots to a set of contact springs inside the unit.  A little door/slide/lock holds it in place.

Battery sits under the heat spreader.

Motherboard & CPU sit on top of the heat spreader - in contact with it.

No doors/hatches/ports on the bottom of the unit.

So - could that solve the question of thermal dissipation for a hypothetical SoC with 2-3x the requirements of the current SoC without being a design disaster?
 
Last edited by a moderator:
^ I used to have a boombox back in the late 80's that used something like 10 D-cell batteries...and that thing was portable.  Sort of.
Ooohhh - we could derail this into a conversation of the least portable 'portable' devices we ever owned...

I had an IBM PS/2 model P-75 with a ~10" orange screen that I'd hot rodded into a 486/133, 16MB RAM, a pair of 1GB SCSI drives, a sound card, 10Mbit ethernet AND a token ring card.  It weighed @30lbs, but was the perfect dimensions to barely fit under the seat in front of you on a jet - hope you don't like your toes.  http://thinkwiki.de/Modell_P75
I love derailment in the P2 SPAM section..


I had this when I was in Junior High School..

http://www.computercloset.org/AmstradPPC640.htm

Also I still have this puppy, 386SX.. two ISA expansion slots a 16 and 8-bit slot.

http://www.sinasohn.com/cgi-bin/clascomp/bldhtm.pl?computer=t52001
 
Last edited by a moderator:
Could go for a Silicon-anode Li-Ion battery, they supposedly can carry 30% more capacity .  So a smaller battery with the same capacity can be done using these.
 
Why not use the P1 battery?
Progress! I'm sure in 2015.. there will be better tech
If the current battery gets cheaper that way and we do not need a redesign in that area,

the more reason to keep the old ones!

They are good, the design is already finished, they are existent in a certain number,

they are cheap and easy to replace (or increase the number of the ones you got with you).
 
There's a lot of other things that makes this unattractive, like needing a charge circuit that can handle multiple batteries.
 
Kind of sucks, considering we may be able to get a bigger capacity battery at the same size.
 
Kind of sucks, considering we may be able to get a bigger capacity battery at the same size.
How are you imagining this exactly? A slot that takes two half-sized batteries? Or a slot that takes two full sized batteries but defaults just with one?

For the first option, everyone would be paying the price of two batteries; while they'd individually be cheaper than the full sized battery together they'd cost more. Plus, Pyra was announced to reuse Pandora's battery so they'd need to get a new one made, adding to the development cost.

For the second option, the unit would become substantially thicker and, and those who don't want two batteries would have to deal with having a partially hollow base. It's annoying to use a current Pandora with its battery missing, doesn't feel good holding it.

If the idea is to have a slot that can take either one full sized battery or two half sized batteries, that sounds difficult to engineer in a way that can hold two batteries securely in without obstructing the full sized battery. Maybe with a divider that bends out from the base, but this too would make the unit at least a little thicker.

In any solution, the added circuitry (especially if a custom solution is needed, and this is much riskier) and

These would be big prices to pay, when the alternative is to have an easy way to suspend the Pandora to storage (a certain button combination could work, also have it happen automatically on low battery) and having to wait a little bit to pull it and out of this mode, then swap the spare battery. It'd be nice if there were a way to charge the battery without the Pyra. I wonder if some people will use their old Pandoras for this. Too bad you aren't supposed to remove or insert the battery while it's on.
 
Back
Top