Charging in parallel

[Firecop]

As I pack up my "fast" Li-ion charger from KD, I start thinking about my WF-139 sitting on my desk...

Could I attach both feeds from each channel (charging at 450 ma) to one "D" Li-ion? If I understand the CC/CV correctly, the amps would only drop as the cell gets close to 4.2v.

BTW, how concerned should I be if that same WF-139 finishes charging a cell which immediately reads 4.23v on my Craftsman MM?

 

[LuxLuthor]

I wouldn't, but I'm not 100% sure of my answer.

First of all I put the WF-139 into the "cheap charger" category. I don't know how the sensing from CC to CV would work if the design was built around its own 450mA output, but now you added another in parallel.

It wouldn't surprise me if the QA tolerances regarding termination of the 139 was "somewhere in the ballpark of 4.2V," and varying from unit to unit. On the other hand, I don't know how precise/calibrated your Craftsman DMM is either.

 

[Benson]

Yes, hooking both channels up that way should work.

About the 4.23V, how concerned you should be depends how confident you are in the WF-139 and/or your MM. If you don't trust the WF-139 not to overcharge, and you don't trust the MM to be within 20mV, then your cell may be over 4.25, which is not good. If it is actually terminating at 4.23V, then all's well. If you have access to any other DMM, you could doublecheck the voltage, and verify that you're safe.

Edit: I'll elaborate on the first part, since I see LuxLuthor already weighed in...

I'd expect it to work; the two channels probably have slightly different voltage thresholds, so one of them will go to CV first, while the other one will still be in CC; not a problem. The channel with the lower threshold will wind up dumping less current in, and eventually drop below it's current threshold and stop charging; by this time, the other channel will be in CV, and now will be the only thing pushing current in. It'll finally drop to it's threshold, and stop charging. The cell will be charged to the higher termination voltage of the channels, but at no point will it be subjected to more than 900mA (fine for a D-cell), and the overall charge will be a 900mA CC and a ~4.2V CV (governed by the higher channel), connected by a short curve with increasing voltage and decreasing current while one channel is in each mode.

I've done this with my TR-001 on 25500s, and it seems to work fine. I don't have a WF-139, so I can't really say for sure that it works, but I think it should.

 

[Mr Happy]

I don't think I would either (connect two different Li-ion chargers in parallel, that is). Each charger by itself is trying to be in control of the charging process. If you try joining them together they will just get confused. Considering the risks of Li-ion cells going incendiary it seems like a very unwise thing even to give a passing thought to.

 

[VidPro]

cross check it, put ya meter from side to side to side, and see if there is any multiplying. example: some ni-mh chargers use a voltage divider sorta thing, where if you read slot 2 postive and slot 3 negative you would get 3v.
so one thing to check is that they are actually 2 seperated 4v channels, and not a (sorta) split regulated 8v source, with common connections, another clue is the 5-6v dc power supply vrses a 10-12V one.

then turn it OFF and check again, the conductivity between the connections you DONT want to be common, and see if they are or not again.

then you could (with nothing to lose) parellel it, test it again, see if you get an actual current increase, or it was just shared current anyways.
set it up in a "safe" area in the bottom of a metal trash can on concrete with no flamables around, and put some smaller cell in and pulverise it, then let it STAY on the charger to its max voltage end (in that safe zone) to find out the max parameters , when its functioning properly.

if it stays fully within the 4.20 +- .05, +- the .03 the meter is good for :-) then you put on your higher priced cell and away you go.
a D cell might be just happy between 4.10 - 4.20 and not any higher anyways, but its still "within spec" meaning it wont ruin it completly.
by testing it total possible MAX at least while its working proper, if you forget it, it is still within spec.

 

[mdocod]

The WF-139 does not even have a CV stage. It runs at *roughly* the same current almost all the way through the charge. (as the voltage rises, the current will drop some through the charge as the power supply in these things are severely undersized. )...

The WF-139 has a charge method that works as follows:

Charge at 450mA for a few seconds, stop charging for a fraction of a second and take a voltage reading, resume charging for a few seconds, stop charging for a fraction of a second and take a voltage reading, etc etc etc. The actual charge rate tends to be under 400mA on most of the. The "cycling" between charging and "testing" can be seen in the form of the LEDs flickering every few seconds from red to green and back. Both channels appear to run these cycles in a synchronized manner, so that's good....

The cell voltage during this type of charging will actually approach 4.3V on most Li-Ion cells, exceeding 4.3V on smaller size cells. It's trying to get the cell voltage to read 4.2V during those little "flickers" where it tests with the charge current removed.

Theoretically, pairing the channels in parallel to increase the charge speed would work on this charger, however there is a problem. Since the charger is not using a CV stage, moving up to a higher charge rate will increase the cell voltage during charging before termination is reached.

I don't like the WF-139s and other chargers that use this same charging practice.

-Eric

 

[VidPro]

if the thing actually goes beyond specs, then shouldnt it be thrown in the garbage?


so he would be hitting the 5Amp cell at .2c on termination, (it's supposed to slow down at the end for a proper charge).
there is ZERO ramping from the voltage differential dropping?

and on a single crappy 650ma cell the thing hits at ~.8C on termination? that would be way out of specs.

doesnt make sence when for $5-10 you can make a MrAl voltage max regulator charger that does a 1 amp max at startup, and terminates slowly and within limits.

 

[LuxLuthor]

Benson may be correct, but when I see "should work" in his note then certainty is lacking. :ooo:

Now comes mdocod with much more certainty and mechanism of action explained. (I was never impressed enough with these cheap chargers to even find out how it worked--so kudos to mdocod).

There is another aspect that occurred to me in picturing these being used in parallel after reading mdocod's post.

There is going to be input of voltage and charging current from one charger into another in circuit, in addition to sensing of changing circuits being done by one unit while the other is doing the same thing, but not synchronized. How would the changing resistance and voltage/current flowing through two parallel PCB's affect each other that was not intended...and perhaps leading to more problems since you are starting with cheaper (less QA/consistency in tolerances of) components.

 

[VidPro]

and if its that crappy, there would be one more thing to check, put a overcharged cell in , that the "brain" didnt see get to the =4.2v and see if it still keeps charging it, via some stupid microprocessor that expects to see a specific voltage location as: 4.2=stop that routine, and not >4.2=stop the routine. wherin the one channel could cause the other channel to skip termination. also for checking, when topping off a cell that was put higher on another charger.

 

[mdocod]

did this a long time ago:
http://www.youtube.com/watch?v=w05TzgYHayY

protected RCR123s actually terminate via PCB rather than the charger itself.

We had a long thread about it a while back.

-Eric

 

[VidPro]

well you know what we say about that.

BAN IT FROM THE FORUM lovecpf
and have Gretta make it a stickey, and talk about responcibility and how others would like to be treated. :-)

the charger took the voltage so high that it was the cells internal protection circuit that terminated the charge.
and arent todays internal protections now set at 4.25, not 4.35? so even though that isnt the purpose of the protections, there is one thing left stopping it, luckily.

this is what i dont understand tho, and i said it before, why say Ohh take it off when the light turns green, and not , the charger is one of the few ways that li-ion becomes completly UNSAFE, and this one is, DONT USE IT. It is completly unsafe and "people sometimes make misteaks" so it should never be in anyones house.

if the chargers all do that given the TIME, then they should all be put on notice not coddled.

 

[LuxLuthor]

No, no no, you misunderstood. Mdocod was demonstrating a substandard DMM in that video.

Ban the charger...OMG, I almost got keelhauled, lynched, and boiled in oil when I said that about this GB Li-Ion charger, after the first one shorted internally and melted. I keep this one for posterity.

 

[VidPro]

No, no no, you misunderstood. Mdocod was demonstrating a substandard DMM in that video.

:crackup:


well its a USER forum, if the dealers and manufactures want to come in and hack up the users, that is thier problem :-)
if the users are saying it isnt so, then mabey thier model doesnt do that , . . YET, , mabey the parameters were changed? like with versions of those smart chargers?

Firecop probably knows what is going on by now, after all he was already charging with it and the D cell before. now he can test it safely, and determine. that is if he doesnt think we all have battery acid on the brain :) or he can just upgrade , knowing the money cant be worth having to deal with us

this couldnt happen in california , because there is 1 lawyer for every 10 people , and the first one that burnt down a hose, would result in a stoning in the public square :naughty:
we have to go to china to get this kind of service.

 

[Firecop]

Thank you all. I think it's time to spend yet more money on another piece of flashlight equipment.

Eric, thanks for the specific description of the charge method.

VidPro, I don't have battery acid on the brain, but I have been exposed to lots of smoke. (and not the good kind, either)

 

[mdocod]

Wow, I think I may have inadvertently started a grassroots militia against the WF-139, lol....

Lets be clear though. In order to justify a complete and outright banning of the wf-139, you would, in order to apply equal justice in all directions, have to ban all wolf-eyes chargers, all old [read pre-IBC] pila chargers, and probably tons of others on the market in devices we aren't even aware of. The Wolf-Eyes chargers and old Pila chargers use the PCB to terminate ALL the time on any size cell! The WF-139 does actually stop the charge voltage at around 4.28V on larger cells like an 18650, and on a D size Li-Ion, the cell charging voltage may actually never exceed 4.25V, which is perfectly safe. If you could find cells that actually did have a PCB that was set to 4.25V, then the wf-139 would be an ideal charger as the PCB would almost always kick in and terminate at a healthy stop point, resulting in OCV at the cell of ~4.15V give or take for most cells after the charge.

In defense of the 139, (not that I like to come to the defense of lousy crap, but I also don't want to just conveniently leave out the facts): They did change the design at some point in it's life, the ones sold in the last ~year and newer all appear to have a different internal design, and can be identified as the newer version via a simple open circuit voltage test. (old ones were ~11V OC, new ones are ~5V OC). The new version uses the same charge algorithm, however, it no longer trickle charges after "termination"... Now, with that in mind. It's near-impossible for a cell who's charge was terminated via PCB to (RCR123 size almost always will) to be trickled charged anyways, as the PCB will hold the circuit open for as long as the open-circuit voltage of the charger is applied. So the old version used to charge smaller cells will not trickle...

The CC/CV charge method is not always adhered to in the industry as tightly as we might expect. Many cell phones and other devices uses a pretty straight forward CC only terminated at 4.25V and call it good charge method.

I'm finding out that there are probably an enormous number of laptops out there that used a charge method similar to the wolf-eyes tailcap chargers. There's a global PCB that links to each cell, and when one cell (or grouping of parallel cells) reaches a voltage high enough to trip the PCB, the charge is terminated. There's no balance or CV stage going on. Just the first cell to reach trip point terminates for the whole pack.

It's a weak argument for the wf-139, sort of like the argument "well all politicians lie, so in this politicians defense, it's not like he's acting out of the norm here or anything..."

-Eric

 

[LuxLuthor]

Great and additionally informative post, Eric.