USB Battery Charging

Go To Last Post
24 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So I have a nice little project underway that is pretty much sorted from a mechanical point of view. Fits inside a nice little case with integral 2xAA battery holder and I wanted to be able to charge it from USB rather than have to remove the batteries every time.

Problem is, my searching has yeilded a large range of USB chargers/power managers specifically aimed at Li+/LiPO batteries. While I could get hold of some iPod style batteries and just run with it, I'd really prefer not to.

Has anyone found a nice USB NiCAD/NiMH charger solution out there that they would care to point me towards? Or is this becoming a thing of the past already?

Cheers,
Steve

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Did you want to build a usb NiMH charger, or buy one?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yeah sorry, I guess I wasnt too clear. Im after a chip solution to add to my board design. Something along the lines of a MAX1555 only not Li+.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

At least from a mechanical point of view converting to LiPo is simple as there are LiPo cells (14500) of the same size of AA cells.

I've started to use them in a couple of applications where I had unsatisfactory results with NiMh, mostly due to the high self-discharge of NiMh cells. Often these buggers are half dead after a couple of week on the shelf and charging them every time before use gets old fast.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Now that's interesting. I had come across AA-sized Lithium primary cells a number of years ago but they were 3.7V. I'll have to have a bit of a search.
Are the ones you have in mind 1.5V?

Hmm... I guess it probably doesnt matter. I could always just use only 1 of the 2 battery positions. The other could hold a spare - Ha!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

To confuse things, there are nickel zinc batteries at 1.6V http://www.powergenix.com/
Would a single lithium cell work? What is your voltage range?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
So I have a nice little project underway that is pretty much sorted from a mechanical point of view. Fits inside a nice little case with integral 2xAA battery holder and I wanted to be able to charge it from USB rather than have to remove the batteries every time.

Problem is, my searching has yeilded a large range of USB chargers/power managers specifically aimed at Li+/LiPO batteries. While I could get hold of some iPod style batteries and just run with it, I'd really prefer not to.

Has anyone found a nice USB NiCAD/NiMH charger solution out there that they would care to point me towards? Or is this becoming a thing of the past already?

Cheers,
Steve

Its a bit over your current voltage (3.2v vs. 2.4v) and a form factor adjustment but how about one LiFePO4 AA cell with this charger:
http://www.linear.com/pc/productDetail.jsp?navId=H0,C1,C1003,C1037,C1774,P112209
Better than NiMH except when below freezing.

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yep. That LTC4098 device is just the sort of thing I'm looking for only for NiCd/NiMH.
Its only for personal applications so I'd rather not get too exotic. I've not seen any flash LiFePO4 batteries readily available round here.
Still if I can get hold of some, its probably ok.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You probably don't want to go above the guaranteed 100mA minimum available current on the USB.

The LTC4060 uses a voltage based charge termination. If the charge current is not high enough and does not cause the expected voltage rise when the battery is fully charged, the charger may not cut off.

Also, any glitches on the battery caused by loads on the battery by your project might affect the charge cycle. The LTC4060 data sheet has some info on charging cycles and glitch control.
LTC4060 - http://www.linear.com/pc/product...

The LTC4059 will charge a NiMH battery, but you need to monitor it with a mcu / adc to terminate the charging. Make sure to get the LTC4059, not the LTC4059-A This can be put into a constant current mode to charge NiMH. A microprocessor needs to monitor the charging current (and battery voltage?) and disable the chip when done.

LTC4059 - http://www.linear.com/pc/product...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
Now that's interesting. I had come across AA-sized Lithium primary cells a number of years ago but they were 3.7V. I'll have to have a bit of a search.
Are the ones you have in mind 1.5V?

Hmm... I guess it probably doesnt matter. I could always just use only 1 of the 2 battery positions. The other could hold a spare - Ha!


The cells are 3.7V. So if you replace normal AAs one to one you add up with two to three times the voltage and will fry most devices.

In many cases you can replace three AA cells with one LiPo. A good indication is if the devices works well with NiMHs. Three NiMHs have a nominal voltage of 3.6V, so a single 3.7V LiPo is a good match.

Two examples which work well for me: My son's Lego set takes 6 AAs, two LiPos and a bit of tin-foil to the same job. I have some of those LED flashlights who take three AAAs, I now use a LiPo with some cardboard and tinfoil to make it bigger.

Also, charging LiPos is much easier than NiMhs you can use a simple voltage-cut off at 4.2V. You don't need the complex di/dv detection you need for NiMh.

The only problem is, if you mistreat LiPo they may blow/burn up violently. The AA cells I buy have internal protection circuits to alleviate the problem, though.

I got my LiFo batteries & charger from http://www.dealextreme.com, search for 14500

A good read: http://www.rcgroups.com/forums/showthread.php?t=209187

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jwinkler wrote:
The LTC4060 uses a voltage based charge termination. If the charge current is not high enough and does not cause the expected voltage rise when the battery is fully charged, the charger may not cut off.

With NiMh you have essentially two charge choices:

- Fast charge: For a 2000mAh cell you feed it 500mA+ and measure the voltage rise. When the voltage stops to rise the cell is full. The problem is that your charger may not detect the stop condition, so often you complement it with a over-temperature stop.
The voltage drop is a side-effect of the cell temperature rising due to the cell being full and the incoming electricity getting converted to heat instead of chemical charge. For the effect to be measurable you need a minimum charge current.

- Slow charge: You feed the cell 200mA for 13 hours. The problem is that if the cell was not empty you overcharge it, which damages the cell over time.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

schtevo wrote:
I've not seen any flash LiFePO4 batteries readily available round here.
Still if I can get hold of some, its probably ok.
Its a bit larger than an AA but here's one from an Australia vendor:
http://www.evworks.com.au/index.php?product=BAT-PE-1150
As a comparison, a AA LiFePO4:
http://www.batteryspace.com/lifepo414505cell32v600mah06arate222whbuttontopstandardaasizeullisted.aspx
An A123 Systems 18650 from China (can order with tabs):
http://cgi.ebay.com.au/4PCS-A123-NEW-original-18650-Lifepo4-battery-element-/190444087565?pt=US_Batteries&hash=item2c575a290d

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

markus_b wrote:
At least from a mechanical point of view converting to LiPo is simple as there are LiPo cells (14500) of the same size of AA cells.

As somebody who currently works for a leading Li-ion battery manufacturer... I can't let this slide.

LiPos are never, ever in an AA form factor (if somebody can prove me wrong I'd be very interested). Li-ions are typically cylindrical form factors, but there is some flexibility there.

markus_b wrote:
Also, charging LiPos is much easier than NiMhs you can use a simple voltage-cut off at 4.2V. You don't need the complex di/dv detection you need for NiMh.

Again, this isn't quite right. Standard charging procedure for a Li-Ion or a LiPo is called CCCV. Meaning you charge at a constant current till the cell voltage hits a certain voltage, then switch to a constant voltage charge till the current hits a certain limit. If you continue to charge past that limit you'll damage the cell, even if the voltage doesn't go over the cut off voltage.

Also - the 4.2V number depends on the exact cell. minimum voltage 3.0V and max voltage 4.2V are fairly good guidelines for LiPos and LiCoO2 li-ions (most Li-ions are LiCoO2s) cells. But you'll wreck a LiFePO4 (you'll overcharge it and then cut it off too early). And even among LiCoO2 cells those voltages vary a bit so it's best to get a manufacturer's recommendation.

markus_b wrote:

The only problem is, if you mistreat LiPo they may blow/burn up violently. The AA cells I buy have internal protection circuits to alleviate the problem, though.

Strongly agree!!! Anybody that plans on working with lithium secondary cells should spend some time on Youtube watching them burn.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

nleahcim wrote:
LiPos are never, ever in an AA form factor (if somebody can prove me wrong I'd be very interested). Li-ions are typically cylindrical form factors, but there is some flexibility there.

For my practical purposes my '14500' LiPos are sufficiently close to the AA form factor to fit in all devices designed for AA I tried. Sometimes it is a bit tight. On of the bigger practical risks I've identified is my wife taking them for AAs and to blow of a device who can not take the 2-3 times higher voltage. A poor bulb of a flashlight already suffered that fate, was very bright for a very short time...

nleahcim wrote:
Again, this isn't quite right. Standard charging procedure for a Li-Ion or a LiPo is called CCCV. Meaning you charge at a constant current till the cell voltage hits a certain voltage, then switch to a constant voltage charge till the current hits a certain limit. If you continue to charge past that limit you'll damage the cell, even if the voltage doesn't go over the cut off voltage.

True enough, for now I used chargers, chips built for the purpose, like the OP mentioned. Might have been too sloppy here.

nleahcim wrote:
Strongly agree!!! Anybody that plans on working with lithium secondary cells should spend some time on Youtube watching them burn.

Plenty of fun searching for 'lipo fire'.

Back to the OPs problem. A charger for NiMh from USB should probably be a switching converter to be able to drive the cells with enough power. I even have one, sold by Sanyo for their Eneloop range of NiMh calls (and it contains an AVR), but I don't have the schematics.

Otherwise most gadgets these days come with LiPo. Biting the bullet might be the best option.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

markus_b wrote:
nleahcim wrote:
LiPos are never, ever in an AA form factor (if somebody can prove me wrong I'd be very interested). Li-ions are typically cylindrical form factors, but there is some flexibility there.

For my practical purposes my '14500' LiPos are sufficiently close to the AA form factor to fit in all devices designed for AA I tried. Sometimes it is a bit tight. On of the bigger practical risks I've identified is my wife taking them for AAs and to blow of a device who can not take the 2-3 times higher voltage. A poor bulb of a flashlight already suffered that fate, was very bright for a very short time...

The 14500 cells on DealExtreme are Li-ions, not LiPos. I have never seen a cylindrical LiPo. Sorry I was not more clear about that.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks for the suggestions so far poeple, and rest assured I'm very much aware of the potential dangers of lithium based batteries. One of the reasons I wanted to use NiMH/NiCd in the 1st place.

At this point I was planning a dedicated charger/power path manager chip solution but, since there's a 1281 on board that's very under utlilised, I might consider a software based charger with NiCds. (particularly as I haven't yet found a chip to do the job)
Only problem I see its getting the micro to fire up 1st so it can start running the switcher.

On the other hand, I'm also keen to "bite the bullet"
and give Li a shot. It's got to happen eventually.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You might consider nickel zinc batteries. They can be charged with a constant current to a (temperature dependent) voltage. Constant voltage charge mode can be skipped for a loss of 5% capacity. The 1.6+V would allow a blocking diode for the USB / battery and enable processor startup from USB supply only. They are more sensitive to overcharging then NiCd though.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

BTW: Have you seen the Battery Management Section app note on Atmel.com? There is a "AVR463: Charging Nickel-Metal Hydride Batteries with ATAVRBC100" http://www.atmel.com/dyn/product...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I did check out those NiZn batteries and they look great. Certainly if the marketing hype is to be believed.
Unfortunately they don't seem all that well proliferated in Aus. It also appears that a manual charging solution is required.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just read what Wikipedia says. NiZn charging is recommended at C1, so your USB port will have difficulties to supply enough power. Trickle charging is specifically discouraged.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The lithium chemistries that release oxygen exothermically have thermal runaway pathways, a particularly nasty one occurs after an overdischarge but is not triggered until the next charge cycle which could be days later. Hence, an onboard circuit that prevents recharge if any cell voltage falls below the minimum.

However, LiFePO4 is as safe to experiment with as any other battery type. 3.0-3.6 volts, low leakage, charge at any time and rate (less than the maximum), and can be safely tested after overdischarge (loses capacity after a few days below 2 volts).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

PowerGenix fast charger charges 4 AA NiZc in 3-5 hours: C/3 - C/5. The charge cutoff algorithm does not use thermal heating effects for cutoff, only voltage. Presumably one could use the same method (maybe with a lower cutoff voltage?) for lower charge current. 100mA would be about C/20.

The Powergenix AA spec is for C-C/2 http://www.powergenix.com/docs/p...
MSDS: http://www.powergenix.com/docs/p...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jwinkler wrote:
PowerGenix fast charger charges 4 AA NiZc in 3-5 hours: C/3 - C/5. The charge cutoff algorithm does not use thermal heating effects for cutoff, only voltage.

That looks to me to be one mayor advantage of NiZc over NiMh/NiCd. With NiZc you can charge to a certain voltage without dv/di complexities.

Of course, in the end the OP must decide on the duty cycle of the gadget. How long must it run on battery, how fast must it be charged again, etc.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you can handle low charge rates, you could just use a high temperature rated NiCD or NiMH and charge it at 0.05C using a resistor and a diode.

Good quality high temperature batteries can have a service life of around 4 years provided they get a discharge once every 6-12 months.

At least then you can avoid some of the headaches the other chemistries have.

oddbudman