Battery Powered Circuit / Charger

[jnnewton]

I'm trying to put together a project that will run on a battery. Thus far the circuit should draw < 50mA intermittently (when being used), much lower when in standby. I expect a 10% or less usage rate (active vs standby).

I am having a seriously difficult time figuring out how to do the battery / charger portion.
I have a few criteria, which don't seem all that difficult, but putting them together has been tricky.

1. Battery must be easily field replaceable (ie hard case like a cell phone battery, not a soft lipo with a cable / cord).
2. Device must be able to work in a wide range of temperatures (nothing artificial, but it will be used outside, in any climate), and not self destruct if used outside of whatever the design range is (idiot proof)
3. Device must charge from usb port or dedicated charger.
4. No BGA or leadless parts
5. Whole thing is going to be thin, akin to modern cell phones, so the battery must be thin as well.

The biggest issue is that I just randomly selected a battery without any further information (can't seem to find much, and the vendor of this battery has given me nothing but the run-around (either the people i'm talking to don't know, or aren't allowed to help me due to the fact that i'm not dropping it directly into a cell phone they know is working.

Does anyone have a suggested battery that is well documented, self protected (over charge, undervoltage discharge, current input / current draw, whatever is needed), that would be good for this application, or where I can go to get direction on this?

I'll post my current circuit (which may or may not be suitable) as soon as I can.

Thanks.

 

[cowboybob]

We'll wait for your current circuit design.

 

[alec_t]

What battery voltage?

 

[crutschow]

Not knowing your voltage requirements, I would think a flat Lithium type battery, similar to that used in a cell phone, would be what you want. For proper (and safe) charging it needs a charger specifically designed for the battery you use.

 

[Joe G]

also where will it be located at....Australia, California, Alaska, Antarctica. the temperature makes a difference if extremely cold or hot... too cold and the batterys will go dead and or freeze I ended up going with a small acid battery do to being below zero for half the year....

 

[jnnewton]

cowboybob: it's up now

alec-t: Looks like the batteries I have are 3.7V. Not sure if this means charge to 4.1 or 4.2 (I think the latter, but can't find datasheets on these cell phone type batteries). I've contacted samsung, nokia and rayovac for info, nothing back yet. I think my system voltage is going to be set at 2.8V, that keeps my battery voltage above 2.7 (at which point some die permanently, or so i've read)

crutschow: that's where I'm leaning as well, but I am having a rediculously hard time finding datasheets, and good sources for the specific batteries. I've added my initial circuit. Modifying it for one battery or another seems beyond me. The lack of information is 1st, but beyond that, all I could change is the charge rates.

Joe G: it's not just 1, this is something I hope to build or have build lots of. I obviously don't want to alienate potential customers, but I also want to stick with a standard battery. It is a handheld device, so I'm not sure how well an acid battery would work for that (I've only been looking at NiMh, NiCd, and Li-Ion / Li-Po.

 

[NorthGuy]

Why the stand-by current is so high? It'll drain the 900mA battery in 6 days!

 

[jnnewton]

Northguy,
It will be as low as possible, I'm still working out sleep modes and looking for lower power components. I'm not concerned about the draw, i can work that out for longer battery life as I go, but I want to get the battery / power portion figured out first, as it is the part I'm most concerned and least familiar with. I'll remove that from the original post, as it doesn't add to the information needed for my issue and I don't want this to get sidetracked.

 

[alec_t]

I'm not concerned about the draw, i can work that out for longer battery life as I go, but I want to get the battery / power portion figured out first

It's a chicken and egg situation. Choice of battery will dictate suitable components if you want to maximise battery life. Or choice of ICs will dictate which battery type/capacity you need.

 

[jnnewton]

It's more of a I'll take whatever I can get situation. Lets put a wide range on it, say 700mAh to 2000mAh. That help?

 

[crutschow]

It's more of a I'll take whatever I can get situation. Lets put a wide range on it, say 700mAh to 2000mAh. That help?

Not much. Such a large range will not lead to a good design. You need to determine exactly how much battery energy you need and then select the battery accordingly, since batteries are expensive and bulky, and you don't want to over-specify.

 

[jnnewton]

That sounds nice, from an academic exercise point of view, but when you're actually trying to build something and a single part (the battery) is the most difficult and undocumented part to find and the most expensive, it makes more sense to start there. If i can find a well documented battery, of any size (within reason), I can make it work. If i design to a specific life with specific current usage, I believe that I will not find a battery suitable.

However, for argument's sake, I'll play ball.
Operational current: 50mA
Standby current: 2mA
Operational Duty Cycle: 20%
Required Life between Charges: 48hrs
Required circuitry voltage: 3-5V

Now that I have that, I can estimate the battery size (ignoring self-discharge and variations in draw at different voltages).
48hrs * (50mA * 20% + 2mA * 90%) = 556.8mAh, but I'm only discharging to 3V, so there will be 0.2V left, but as i understand it they drop off pretty fast after 3V, so say we lose 5% of the battery
556.8 /95% = 586mAh

So, who sells a well documented, self protected, off the shelf battery that is thin and small like a cell phone battery (please don't tell me to design the case and then pick a battery to fit in the arbitrary hole size I chose), and is not overspecified for 586mAh, and will work with my charger circuit?

If anyone can actually answer this question, I will ship you a case of beer. Otherwise, I would like some help finding a battery supplier to design my circuit around.

 

[NorthGuy]

How about AAA sized Li-ion battery? You can get 600mAh, possibly slightly more, and you get standard size - easy to buy a holder, replacement etc.

With flat phone batteries, you probably won't get any documentation because they're all proprietary, but you do not need much. These batteries usually have 3 terminals +, -, and thermistor. I guess by heating/cooling the battery you can figure out thermistor behaviour.

Or you can just ignore the thermistor. If you do not charge above 3.9V, there's very little chance of overheating. You'll loose some capacity, but you prolong battery life.

 

[NorthGuy]

Duplicate

 

[jnnewton]

Northguy,

Good points. I'd like to talk about the AAA sized Li-Ion's further. Initially I thought this was the route to go. However, there are a couple of problems that I couldn't seem to address.
1. Reverse insertion protection. It seems that the only way to guard against reverse battery insertion is mechanically, but many (all that I saw that were compact) did not implement this. I could not find a good solution to reverse protect the circuit / cells.
2. Other battery type insertions, it's bound to happen, someone is going to stick an alkaline, NiCd or NiMh battery in there in a rush to get the thing going again. Then it won't work because the voltage is too low, so they'll plug it in to charge. I don't know what would happen, but I doubt it would be good.
3. AAA are a bit thicker than I was hoping for, but it may work if #1 & 2 are solvable.

Do you have any recommendations for the two above scenario's. I could custom make the holder to mechanically limit the insertion, but the other chemistry types problem is one I don't know how to tackle. A reverse protection circuit would be a better solution for #1, as i could then use regular pcb mount battery contacts.

As for the Cell Phone battery, it seems that many don't even have a thermistor, but rather the 3rd pin is used to identify the battery capacity. I don't remember off hand, but maybe it was called BIS? Camera batteries tend to have thermistors, but they are quite a bit thicker. I went to batteries plus and poked through their selection looking for a battery marked with a T, similar to the camera batteries, but was unable to find one.

I have a couple cell phone batteries, and have tried heating / cooling without a remarkable change in the center pin (measured resistance wrt the "-" and "+" pins). I may have been unlucky in my selection, and if you think this is the case, I'll go back and buy a couple more to heat / cool for the sake of being thorough.

 

[jnnewton]

Possible answer to #1: https://www.maximintegrated.com/en/app-notes/index.mvp/id/4572

 

[audioguru]

Many fake Li-Ion batteries are sold on ebay. One video shows a battery rated at 2400mA and inside was powder and a tiny battery cell that measured only 66mAh.
There are also Li-ion battery cells on ebay with fake protection circuits. A Chinese company sells photos of a protection circuit that looks the same as a real circuit.

 

[crutschow]

Here's a simple reverse voltage protection circuit that has negligible voltage drop in the forward direction. The P-MOSFET can be a much smaller size than shown for your low current and R3 can be eliminated (it can be an N-MOSFET if it's placed in the negative line from the battery). The MOSFET must be a logic-level device that full turns on at 3V or 3.3V (not the Vgs threshold voltage).