Creating single cell chargers

swbluto

10 TW
Joined
May 30, 2008
Messages
9,430
Hello. I'm looking at my options for single cell chargers(I'm going to design my own LVC for the cells), and it seems it's either a case of spending 80 dollars(Including shipping) for a balancer from hobby-city(which also needs a powersupply) or I can build my own!

Ideally, I'd need to charge roughly 20 cell groups "simultaneously"(can be approximated as further explained below).

If 2A is fine, then it seems the L200 chip can be easily adapted to create a 2A CC-CV charger with minimal parts - The projected cost per single cell charger is around $3. But, it'll need a power supply. If I were to use my 12V power supply from my dead computer's PSU, then it seems 24 A is available and so I could allot upto 12 2A chargers but I'd worry about the heat dissipation issues from 12V to ~3V on the chips and I'm not entirely sure what kind of heat sink I'd need and whether I have said heat sink available. 9 volt drop*2 amps ~= 20 watts each which I've heard heat sinking for above 10W is cost-wise impractical, but maybe I have something that'd work? I picked up two somewhat large heat sinks from the robotics meeting, so I think that might work. The other possibility is to pick up a 2 lb. 7 volt switching PSU that's adjustable between 6 to 9 volts and is good for upto around 60 amps, so that'd minimize heating issues while allowing for portability for opportunity charges, but that costs about 53 dollars. So, it's either the computer PSU + 12ish simultaneous charges for 40ish dollars or the 2 lb PSU + 20 simultaneous charges for 130ish dollars.

But, then there's more! I could possibly use a buck converter regulator as charger! The good thing about this is that I could use the power from my computer's power supply more efficiently. The bad news? It seems more expensive. The chip is 3 dollars, a 330 uH inductor seems to be around 3 to 4 dollars(ouch), and the electrolytic and other capacitors, resistors, diodes, mosfets/etc. probably adds another $2 per unit implying a cost of $9 per charger. But, the up side is that I could charge at upto 10 amps, I wouldn't have to worry as much about heating issues, and I could use my 12V computer PSU without heating worries(Although it's limited to ~250 watts, so I couldn't use 20 simultaneous chargers at 10 amps each but I could probably use 10). There's also the option of getting the $50 power supply so I can get 350W charging on the fly. But, it might be feasible to minimize the number of chargers by "switching" between cells, while still getting high charging currents. I.e., a charger feeds a cell group for 30 seconds, it then feeds another cell group for 30 seconds, it switches back to the first and the cycle continues. So, it seems I could build upto 10 of these chargers to satisfy my needs, and it'd come to 90ish dollars and the 2 lb PSU would be another 50 dollars.

So, in essence, it seems I can have slow charging 10 cells at a time for 40ish dollars(Or extremely slow charging for 20 cells use the switching idea for 50 dollars), slow charging for 20 cells for 80 dollars + a 50 dollar psu(so 130 dollars), quick charging for $90 and if opportunity charging is desired, 50 dollars extra for the PSU(140).

I'm not sure if the balancers being sold on hobbycity.com can run off my computer's 12V PSU, so a 200 watt balancer might require that $50 PSU(At a higher voltage) so I'm probably looking at a necessary $50 PSU with that, so the hobbycity one is at like a $140 minimum and they also charge at 200W, I believe.

Hmmmm... I'm liking the quick chargers which matters more for opportunity charging, so I'm trying to think of ways to minimize that cost. Maybe I could only build 5 and then I could use the "switching" idea to distribute the current as needed? Then the cost might be $50+15(switchers) = $65 which could run from my computer's PSU.

The other method is get 11 2Amp single-cell chargers that plug directly into the wall from voltphreaks for 120 dollars or so, and while that minimizes the effort needed for building and design, it's kind of slow and it's not as "compact" as my own custom design could be.

So, if you didn't mind building anything, what would you do for the single cell chargers?
 
I'm thinking about the only time I'll want fast charging is when I'm opportunity charging. If I want opportunity charging, I'll need to buy a new power supply no matter what so I can toss that out to simplify comparisons for "opportunity charging"(And I think I so desire it). If I'm using opportunity charging, I probably won't need that much battery for my electric scooter as I'll be opportunity charging on the way. So, I'll probably be carrying around 10 cell groups on hand(10 A123 cell groups) meaning I could use 10 chargers or 5 chargers in conjunction with the "switching distribution" idea.

So, it'd be around 65 dollars for my own home-made balancer plus 50 dollars for the power supply for optimized opportunity charging. Does hobbycity.com have a balancer with similar charging speeds and similar cost? If they did have something that could charge at 10 amps for each channel and I could "switch" between cell groups every 30 seconds or so, I could get away with one I think. The a123s would be about 3.5 AH depleted average, 6 at max, so splitting up the charging would mean finished charging in less than an hour. Or, instead of "switching", I could just use a switch to convert the 10s3p pack into 5s6p for charging, but I'd have to figure out how to do that...

So maybe I could build my own 5-channel balancer for around $50(i.e., 5 10A single cell chargers)?
 
Ok, I just went to price 10A IC buck converters from digikey(the kind you only have to add components to) and it seems it'd cost about 10 dollars per IC. So the actual cost for the homemade single cell fast charger would be around $16, so 5 of them would cost around $80, roughly the cost of a 5-6S balancer from hobbycity.com.

So, I guess I'm looking at buying a balancer if I want fast charging. If I wanted slow-charging, it seems like my own DIY would've been cheaper, but for the higher power/quicker charging, it seems economics has me beat.

Edit: Ok, I just went to price the balancers on hobby city and it seems the "$80" models were ones that had a max charge rate of 5-7A(So about 40 watts) and the true 10A ones were 140+(includes shipping) but here's one for around $100 http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=7386&Product_Name=GT_A-6-10_200W_Balance_charger_&_discharger. Does anyone have an opinion on the quality of this charger?
 
Remember, you can't use a buck regulator on a series string of batteries unless each one has an isolated supply.

I was looking into cheap 5v power supplies that I could easily modify to run 3.65v. This seems like it might be a lot less work and possibly cheaper. Something that resembles a cell phone charger is pretty cheap these days. Changing the output voltage should be a matter of changing a resistor somewhere, depending on the design.
 
fechter said:
Remember, you can't use a buck regulator on a series string of batteries unless each one has an isolated supply.

I'll try to keep that in mind. I was thinking that each "single cell charger" would be a buck converter, so there wouldn't be a "series string of batteries" in that case... I think. I might be over-looking something, though.

Edit: Hmmm, I think I was overlooking something. It seems the output of the buck converter would need to be isolated so that it could reference a given cell group's ground as otherwise, it would reference the power supply's ground but you can't just have all different cell group's ground reference a common ground when they're in series! I'm sensing a short! Dag nabbit. A fatal flaw to ruin my idea! That balancer is looking increasingly attractive.

I was looking into cheap 5v power supplies that I could easily modify to run 3.65v. This seems like it might be a lot less work and possibly cheaper. Something that resembles a cell phone charger is pretty cheap these days. Changing the output voltage should be a matter of changing a resistor somewhere, depending on the design.

Yeah, that definitely sounds like a good idea. I'm tempted to break into my computer's PSU to see if I can change the 3.3V output voltage or 5 volt output voltage somehow. But... on that notion, may I direct your attention here:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=220388174840&ssPageName=ADME:X:RTQ:US:1123

Looking at his list of power supplies, it seems the lowest voltage he has is 5V but that can be lowerable to 4.5V. If there was someway to cheaply drop that by, say, .8V and then current regulate it(Perhaps a simple mosfet / current sense resistor + op-amp?), that might be a possible way for "direct" charging? Except.... hmmmm... would putting those current regulators in parallel with the power supply create problems? Maybe something a diode could solve?

Edit: I think that might have the same problem as the above edit noted. Darn it.
 
I'm looking for DoctorBass's posts about single cell chargers as I know he was advocating them a year ago or so.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=2586&p=35650&hilit=single#p35650 - The post of yore, I guess.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=1984&p=26590&hilit=single#p26590 - One source; seems to be the slow 2A variety

And there's another one involving "Artec", I think. I'll try to find it.
 
http://endless-sphere.com/forums/viewtopic.php?f=14&t=2824&hilit=+single+cell+ - The post about the artesyn charging method.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=4215&hilit=+single+cell+ - And one more of his posts.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=4196&hilit=+single+cell+ - And an eBay lead of waaaay long ago.
 
http://cgi.ebay.com/Lot-of-25-Artesyn-BXA30-48S05-F-CONV-DC-DC-DUAL-5V-OUT_W0QQitemZ130243250514QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item130243250514&_trksid=p3286.m20.l1116

This looks like an interesting lead. However, I'm not sure if they're adjustable like Doc seemed to suggest the artesyn's were. Looking at the data sheet, it had an output voltage of 5 volts and then an "adjustability" of 10%.

http://cgi.ebay.com/Artesyn-BXB100-48S0 ... .m20.l1116

Here's a 20 amp 5V artesyn module for the price of $20(including shipping). If there exists a voltage adjustable 20 amp module like this implies(or even more!), maybe a 10s3p could be turned into a 2s15p pack and two artesyn modules can be purchased instead of ten! Now the problem is finding said voltage adjustable artesyn module and figuring out how the switch would be set up(That's like, what, a 5-pole 5-throw switch? 5-pole 2-throw switch? ). Or, maybe, just maybe, it could be transformed into ... 1s30p. :shock:
 
I just sent an order to voltphreaks for 10 of their older 2A single cell chargers(At 65 dollars total, that's 6.50 each. Not bad, methinks.). Slow, sure, but I think I'm fine with 2-3 hour opportunity charges on the A123s I'm thinking about getting and... it's easy to setup and (I hope) light! The only place I can think of getting an opportunity charge is at school and at the robotics club meeting, and I'm sitting down in one place for at least 2 consecutive hours at each.
 
almost all of the single output switching supplies are CV only and use a simple zener or XX431 voltage reference to control the secondary voltage.

if they use a zener just replace it with a zener of the right voltage. if they use a XXX431 (TL431, AM431, FDC431 etc.) it is easier still. the voltage is set by a voltage divider and easy enough to set.

alternatively if you want to use a CC/CV supply you can use a chip like the TSM101 (about $1 in single units) which is meant as a output controller for SMPS powersupplies. jut needs the right 4 or 5 resistors and a shunt to convert one of those power supplies to CC/CV.

i'm using one of those in what started life as a Soneil 2409SR charger. i just hacked out all of the stuff they had in there on the secondary for the feedback loof and added a little board with the TSM101 and 7 resistors. i'm using the original shunt from the charger. i've only used it a couple of times so far but it does deliver 4.5A CC until the pack hits 31.2V maintains the 31V until the current drops to 400mA and then turns off.

rick
 
Hi,

swbluto said:
I just sent an order to voltphreaks for 10 of their older 2A single cell chargers(At 65 dollars total, that's 6.50 each. Not bad, methinks.). Slow, sure, but I think I'm fine with 2-3 hour opportunity charges on the A123s I'm thinking about getting and... it's easy to setup and (I hope) light! The only place I can think of getting an opportunity charge is at school and at the robotics club meeting, and I'm sitting down in one place for at least 2 consecutive hours at each.

Ypedal posted that the voltphreaks are prone to overheating. I think he recommended plugging them into the power strip, facing in opposite directions to avoid that.

I got the following in an email from Voltphreaks, so if it turns out 2 amps isn't enough, you can always buy another 10 or 20:
you can double or triple up on the Voltphreaks chargers to reduce charge time, for example, hook up
two chargers per parallel group. The chargers are completely isolated from each other so you can safely do this.
 
Here's what I'm doing:

16-Cell%20Individual%203.7V%20Charger%20Setup.jpg



Up until recently, Bill Zelman (Powermed here...) was selling Vicor 48V/3.7V 20A DC-DC convertors for $10 each, and that was shipped, so I bought a bunch (24...). He was also offering a nifty 48V/4.2A supply, so I bought some of them as well. In the box above, I've mounted 16 of the 3.7V supplies on a 2-1/2" x 1/4" x 12" piece of aluminum, which is then bolted to the bottom of U-shaped box. I mounted the two 48V supplies, which are amazingly small, to the sides. This is still a work-in-progress, like most of my projects ( :roll: ), but eventually I need to add the wires and multipin plug, like I had in this setup:

16-Cell%20CMS-06.jpg



This plug will mate up with the LVC board-mounted connector on the front of my 16s5p a123 packs:

a123-16s5p-v3-09.jpg



Using these two 48V supplies won't be enough to run these at 20A, but these connectors are only good for about 8-9A anyway, which is about what I will get per cell. That's still enough to do a completely dead 16s5p pack in under an hour and a half.

I am doing a special PCB that will mount over the output pins on the convertors and that will have one two-color LED per channel. For cells that are dead, the LED will start out red. It will be fully green when each convertor is in the CV mode, holding the voltage at 3.7V. In between it will transition from red to orangish-yellow to yellowish green and finally to fully green. The LEDs will poke through holes in the box cover, which will also have two small fans on one end.

-- Gary
 
Gary, you can safely do 20A with this arrangement. Just add separate heavy duty cable to the charger connecting first and last DC converter to the pack deans (in addition to the LVC connector). Most of the current will flow through the loop formed by series connected DCDC converters and the pack through the deans. Only a small fraction of the current will flow through LVC connector (unless the pack is completely out of balance).

That high power outer loop does not destroy the concept of individual charging. It just creates a "fast lane" to the common path.
 
MitchJi said:
Hi,


Ypedal posted that the voltphreaks are prone to overheating. I think he recommended plugging them into the power strip, facing in opposite directions to avoid that.

Thanks for mentioning this. I was merely thinking about taking the chargers apart and hooking them up to a more compact circuit arrangement, so it's good to know I should check heating issues as well. So, now I'm convinced I should take them apart instead of merely thinking of it! :lol:
 
Hi,

Would it be hard set up a few Voltphreaks Chargers so I have the option of 500mA charging? If so how would I do that?

I'm supposed to start receiving aproximately 2 spent Dewalt packs per week and I would like to try this with the low cells:

Doctorbass said:
There is something i discovered that impressed me a lot about these low cells that we easily declare: dead .... I had many of them that i kept for me instead of selling them that i tested again.

I was surprized to see that many cells close to zero V was still good with capacity and RI !! :shock:

I mean... these famous 1 or 2 "bad"... or even low cells that had 0.9V or 0.2V still gaved me 2150mAh and 10 mohm !! ..

and same results with 5 cycles after that!!

I also had a pack that i kept for me that measured 0.8V with every cells between 0.1 and 0.2V that i successfully charged
using individual charger/cell and that gaved me 2100mAh ! with Ri between 10 and 13mohm /cell!!

since the last full charge of that pack i let ist to sit for 2 month on my desk and it still measure 34.75V of SOC!

That's why i say these cells are like bulletproof and so incredible!

i'm not saying that everybody should keep these to match with their ebike batt project.. but for something like boost pack they are excellent!

If you try to recharge these lower cells, you will see quickly if they survived or not... the really BAD/DEAD cells will just heat up and will not keep their voltage when you remove the current

The best solution to revive low cells is to begin the charging process at 500mA and when they reach 3.0V you can increase the current.

The DEAD cells will never reach more than 1.5V during the 500mA charge process.. the rest should succeed.

I'm not sure what I'll do with the resuscitated cells but I hate to recycle something that can be reused.
 
I would imagine the chargers use a shunt for current sensing, so basically you'd need to replace the shunt with something that has 4 times as much resistance and your current would be limited to 4 times as much(2A/4 = 500 mA). There shouldn't be any problems with this as decreasing current has the effect of decreasing heat.

However, "where" this shunt is and how it can be replaced is beyond me at the moment. I'll let you know when I start taking apart about the chargers.
 
swbluto said:
I would imagine the chargers use a shunt for current sensing, so basically you'd need to replace the shunt with something that has 4 times as much resistance and your current would be limited to 4 times as much(2A/4 = 500 mA). There shouldn't be any problems with this as decreasing current has the effect of decreasing heat.

However, "where" this shunt is and how it can be replaced is beyond me at the moment. I'll let you know when I start taking apart about the chargers.
I believe it's the one next to the transformer and one of the output leads. It's a 2W, 2 ohm, 10% resistor, mounted vertically.

MitchJi, you should be able to cut the lead that's bent over the top of the resistor and install a 6 ohm resistor. That will bring the total resistance up to 8 ohms, 4x the original value. Then see what happens. :mrgreen:
 
That doesn't seem like a shunt I've encountered before. 2 ohms? I take it all the current that goes to the battery doesn't go through the shunt... otherwise, it sounds like it'd have a power dissipation 2^2*2 = 8W. I thought they'd use something like copper "mini-pipes" or something with an incredibly low resistance.
 
Good point! I didn't do the math for the power level. :oops:

I checked the resistor and it's the only one that's even a remotely low value on the secondary side of the transformer. There is a bare wire along the top of the board, labeled J2, but it looks like it is just a jumper.

If there's shunt in there, I don't know where. :mrgreen:
 
Interesting.

So, it has a transformer? Can you briefly explain how it works? Does the transformer convert it from 110V AC to whatever-volt AC and then it has a bridge rectifier, smoother cap, etc. and then use a buck converter? Or does it have a "fly-back" like topology where it switches on and off directly off the transformer?

The images kind of made the unit look pretty small and (assumably) lightweight, so it seems surprising they have a transformer in there to drop from 110V to 3-5ish volts.

I suspect they're hitching a "ride" off another component which has a known VI curve that is reasonably temperature stable(or they've compensated for temperature changes), in which case changing the current value might not be that easy. It is possible to add external circuitry to limit the current, but that can get kind of complicated.
 
The charger uses an OB2354 PWM controller to create a flyback converter.

Checking page 1 of the datasheet (http://www.micro-bridge.com/data/on-bright/OB2354_P1.pdf), all I could easily find, the converter has cycle-by-cycle over-current protection and no shunt resistor on the secondary (only voltage feedback going to the optocoupler?). According to the "Typical Application" schematic, there does seem to be a current sense resistor on the primary side, but I couldn't find it when looking at the PCB,
 
Hi SW,

swbluto said:
Edit: Ok, I just went to price the balancers on hobby city and it seems the "$80" models were ones that had a max charge rate of 5-7A(So about 40 watts) and the true 10A ones were 140+(includes shipping) but here's one for around $100 http://www.hobbycity.com/hobbycity/...e=GT_A-6-10_200W_Balance_charger_&_discharger discharger. Does anyone have an opinion on the quality of this charger?

Luke really likes this one:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=9194#p142743
I have 4 of these
http://www.hobbycity.com/hobbycity/store/uh_viewitem.asp?idproduct=6609
iCharger 1010B+ 300W 10s Balance/Charger
I have a fluke 189 meter, and this charger drops them off at EXACTLY the voltage I set it to charge each cell. Works fantastic, never had a problem with any of them. You do need a 30amp supply for each charger to run it at full potential. If you don't have a 30amp supply, it runs fine off even a little 5amp supply, but you obviously have to lower the charging current settings. It does 1-10s cells, and as long as you run each one from it's own 12v source, you can easily run a pair of them for charging an 11-20s pack. They never get above luke warm during charging, and they are very quick, painless, user friendly, plug-it-in-and-walk-away. If they detect any errors or funny stuff, they will shut down the charging.

This model is probably just as good as the model Luke likes, but its about $40 more than the one you linked to:
http://www.hobbycity.com/hobbycity/...me=iCharger_106B-plus_250W_6s_Balance/Charger
iCharger 106B-plus 250W 6s Balance/Charger
Capable of handling up to 250W or 10A charge rates. Lipoly, LiFe and Lilo compatible as well as able to charge up to 6 cell packs.
PRICE: $119.95
 
Back
Top