Understanding the WF-139 Deficiencies Better

[andrewnewman]

Greetings:

I know that this topic has been covered in great detail in other posts over the years but as a bit of a beginner I am still a little confused. I have read (in summary) that this charger is deficient for essentially two reasons:

1. The charger only does the 'CC' part of a 'CCCV' charge to termination.
2. For smaller batteries (e.g. RCR123 batteries) the internal resistance of the battery toward the end of charge causes the charge voltage to reach undesirably high values.

I have one of the newer ones (< 5V open voltage) and decided to run a test myself. I put an *unprotected* RCR123a battery in the charger with a starting voltage of 3.89V. I measured the voltage as the battery charged. It typically ran 0.1V higher than the right-off-the-charger battery voltage until the battery voltage reached 4.05V. At this point the spread increased until the charger terminated with a charge voltage of 4.30V and a battery voltage of 4.15V. It had been asserted that this voltage at termination was due to protection circuits terminating the charge. Clearly not the case in my experiment.

I have another charger that is also regarded as non-optimal. It's one of the Nanos. This charger brings the voltage close to termination but never actually terminates the charge. Instead it just "cooks" the battery on a trickle charge for many hours. I'm told this will cause metallic lithium to plate out on the anode and lower the life of the battery.

I'm told the high charge voltage of the WF-139 will lower the life of the battery. It strikes me that the WF-139 is merely seeing the growing resistance of the cell as it approaches full charge and trying to maintain a constant current by increasing the voltage. If it didn't behave this way, but maintained a lower voltage in the face of the higher resistance then the battery's charge current would decrease and the battery would "cook" as in the second example.

Since neither is apparently optimal, I'm curious what the behavior of an "optimal" charger might be.

 

[march.brown]

Greetings:

I know that this topic has been covered in great detail in other posts over the years but as a bit of a beginner I am still a little confused. I have read (in summary) that this charger is deficient for essentially two reasons:

1. The charger only does the 'CC' part of a 'CCCV' charge to termination.
2. For smaller batteries (e.g. RCR123 batteries) the internal resistance of the battery toward the end of charge causes the charge voltage to reach undesirably high values.

I have one of the newer ones (< 5V open voltage) and decided to run a test myself. I put an *unprotected* RCR123a battery in the charger with a starting voltage of 3.89V. I measured the voltage as the battery charged. It typically ran 0.1V higher than the right-off-the-charger battery voltage until the battery voltage reached 4.05V. At this point the spread increased until the charger terminated with a charge voltage of 4.30V and a battery voltage of 4.15V. It had been asserted that this voltage at termination was due to protection circuits terminating the charge. Clearly not the case in my experiment.

I have another charger that is also regarded as non-optimal. It's one of the Nanos. This charger brings the voltage close to termination but never actually terminates the charge. Instead it just "cooks" the battery on a trickle charge for many hours. I'm told this will cause metallic lithium to plate out on the anode and lower the life of the battery.

I'm told the high charge voltage of the WF-139 will lower the life of the battery. It strikes me that the WF-139 is merely seeing the growing resistance of the cell as it approaches full charge and trying to maintain a constant current by increasing the voltage. If it didn't behave this way, but maintained a lower voltage in the face of the higher resistance then the battery's charge current would decrease and the battery would "cook" as in the second example.

Since neither is apparently optimal, I'm curious what the behavior of an "optimal" charger might be.

My understanding is that the ideal charger would bring the Li-Ion cell up to 4.2 Volts and then completely stop the charging process ... The final part of the charge should be at a constant voltage and as the cell voltage comes up to the desired voltage, the current drops to zero and ideally the charger totally disconnects from the cell.

I don't know how many of the cheaper chargers do this.

My Trustfire TR-001 has an open circuit voltage of 4.24 volts and takes ages to bring the cell up to its final voltage of 4.21 volts at which point the LED goes from red to green ... As soon as I notice that the LED is green, I take the cell off charge ... I have not checked to see if there is a trickle charge after the LED goes green ... Must measure it sometime though.

With the Charger at 4.24 volts and the cell at 4.21 volts, there will not be very much current flow though ... The difference is only three hundredths of a volt ... I still whip the cell off at this point just to be safe.

I would be interested to know if there are any cheap chargers that do this, particularly if available in the UK.
.

 

[45/70]

My understanding is that the ideal charger would bring the Li-Ion cell up to 4.2 Volts and then completely stop the charging process ... The final part of the charge should be at a constant voltage and as the cell voltage comes up to the desired voltage, the current drops to zero and ideally the charger totally disconnects from the cell.

This is not entirely correct, but close.

A charger using the proper algorithm (from B.U.) will charge the cell at a Constant Current (CC) untill the voltage reaches 4.20 Volts. At this point, the charger will start what is referred to as "stage 2", or the Constant Voltage stage (CV). In this stage, the voltage remains at 4.20 Volts and the current applied starts to drop, as the cell approaches full charge. When the current level drops to a point no lower than 0.03C (where "C" = the charge rate in mA), the charger shuts off.

The purpose of the CV stage is to bring the cell to a full charge without subjecting the cell to a voltage higher than 4.20 Volts, and also prevent trickle charging the cell beyond the 0.03C limit, which can cause plating of metallic lithium to the anode.

You're not likely to see this in a $10 charger. Some chargers come close to approximating this algorithm at this price point, but the WF-139 is not one of them.

Dave

 

[TriChrome]

I think this is all a little overthought to be honest. I think batteries are a lot more resillient than people think. I'm using the WF-139 charger going on about 3 years now, and just did a test with my oldest 18650 cells in my JetBeam M1X and my runtime was within 7 minutes of SelfBuilt's review on this forum. I've also not noticed any other detrimental effects on any of my other cells.

I'm sure it's not quite as good to use this charger than something like the Pila, but I doubt that any casual user will see a noticeable difference.

 

[kramer5150]

I think this is all a little overthought to be honest. I think batteries are a lot more resillient than people think. I'm using the WF-139 charger going on about 3 years now, and just did a test with my oldest 18650 cells in my JetBeam M1X and my runtime was within 7 minutes of SelfBuilt's review on this forum. I've also not noticed any other detrimental effects on any of my other cells.

I'm sure it's not quite as good to use this charger than something like the Pila, but I doubt that any casual user will see a noticeable difference.

It should be pointed out that there are 3 different versions of WF-139 (identified by the behavior of the LED indicators). Not all 3 charge the same. IIRC the more recent models charge beyond 4.2V. Mine is ~2 years old and does not charge my cells past 4.21. Cell voltage hits 4.20 and cuts shortly afterwards. Seconds off the charger the cells measure 4.17

 

[old4570]

There are two cheaper chargers out there that terminate on reaching the full charge [ subjective ] .

Anyways , the WF-139 has been talked to death ...
It is what it is , and we know to use the charger only when supervised .
I am not aware of any actual disaster stories with this charger , nor the Trustfire charger ..

But if you need or want to leave the charger running unsupervised , buy a charger that terminates on completion ...

Soshine
SheKor

I own the SheKor , and it does terminate on completion , but @ 4.13v , so its a little shy of a full charge .
And the Soshine is on my shopping list .

 

[march.brown]

This is not entirely correct, but close.

A charger using the proper algorithm (from B.U.) will charge the cell at a Constant Current (CC) untill the voltage reaches 4.20 Volts. At this point, the charger will start what is referred to as "stage 2", or the Constant Voltage stage (CV). In this stage, the voltage remains at 4.20 Volts and the current applied starts to drop, as the cell approaches full charge. When the current level drops to a point no lower than 0.03C (where "C" = the charge rate in mA), the charger shuts off.

The purpose of the CV stage is to bring the cell to a full charge without subjecting the cell to a voltage higher than 4.20 Volts, and also prevent trickle charging the cell beyond the 0.03C limit, which can cause plating of metallic lithium to the anode.

You're not likely to see this in a $10 charger. Some chargers come close to approximating this algorithm at this price point, but the WF-139 is not one of them.

Dave

.
Hi Dave, it is always a pleasure to read your comments, particularly as I do not know much about the internals or the chemistry of Li-Ion cells ... I think it is important to keep the old grey matter working, even at my age (old).

My Trustfire TR-001 only has an open circuit voltage of 4.24 Volts ... The green LED comes on at 4.21 Volts ... I have not left a 18650 on charge for more than a few minutes after the green LED comes on ... If I did leave the battery on for hours after the green LED is lit, it could not go above the chargers open circuit voltage of 4.24V ... Would these extra hours have much effect on the life of the cell or on the the plating that could occur on the anode ? ... How much effect would the extra 0.03 Volts make ?

I could see that a large trickle charge current would make a difference, but this would surely need a greater voltage differential than the existing 0.03 Volts that exists on my system ... Do you know if anyone has experimented with the damaging effects of smaller trickle charge currents on Li-Ion cells ?

My individual 18650s will probably not be charged more than a handful of times a year (maximum), so I would think that my cells will die of old age rather than any other effects ... I don't know this for sure, as it is an assumption that I have rightly or wrongly made ... What do you reckon ? ... The 18650s will probably be charged whether they need it or not at a low voltage of about 4.0V ... If they get below about 4.0 Volts, it would be under exceptional circumstances ... Do you think that it would be better to leave the charging till the voltage gets well below this point ? ... This would mean that the cells would then probably only be charged maybe three or four times a year.

I have one spare 18650 protected (Ultrafire grey) cell, plus two unprotected (Ultrafire blue) that are spare spares ... I won't buy any more cells till my existing cells start to die, which brings me to another question ... With my useage pattern, how many years can I expect to get out of an average 18650 ?

What would be the value in Milliamps of 0.03C for a 2400mAh 18650 cell at which the ideal charger should cut off ? ... Would that be 72 milliamps ? ... The charger volts would have to be higher than my 4.24 volts to produce this current.

I have just checked what current is flowing from the charger when the green LED is on ... The protected 18650 is being trickle charged at 3.1 mA and the unprotected is measuring 2.9mA ... The protected cell is 4.20 volts and the unprotected cells are both 4.19 volts ... So this Trustfire doesn't switch off when the green LED comes on ... I assume that the unprotected cells are slightly higher internal impedance than the protected one, based on the fact that the current is lower and the voltage difference (charger & cell) is higher on the unprotected cells.

If we don't ask questions, we don't learn !

Anyway, thanks again Dave, I really appreciate your comments and those of other knowledgeable posters.

By the way, are there any detailed articles on Li-Ion cells that would perhaps help me understand this technology better than my present minimal level ?
.

 

[march.brown]

I own the SheKor , and it does terminate on completion , but @ 4.13v , so its a little shy of a full charge .
And the Soshine is on my shopping list .

.
There are several different Soshine Li-Ion chargers ... I have sent for this one as it only has two battery slots ... The advert says that it does switch off after charging and if that is correct, it seems a good price for this facility ... Just waiting now for it to arrive.


http://www.bestofferbuy.com/soshine...-for-186501767017650-110v240v12v-p-51564.html
.

 

[old4570]

.
There are several different Soshine Li-Ion chargers ... I have sent for this one as it only has two battery slots ... The advert says that it does switch off after charging and if that is correct, it seems a good price for this facility ... Just waiting now for it to arrive.


http://www.bestofferbuy.com/soshine...-for-186501767017650-110v240v12v-p-51564.html
.

Its also advertised as a fast charger ! I will order one when the Chinese new year is over and things return to normal .. I expect there will be some what of a backlog of orders like last year .. So I will wait till March .
$15 = Worth a look see !

 

[andrewnewman]

Ugh. As the OP I have to concur with the observation that this general topic has been talked to death. My original question was neither, "why can't I leave this charger unattended", nor was it, "what's wrong with this charger I've been using for years", nor was it even, "what's the best charger". I've read many many explanations of all of these.

My original question was, "What is the behavior of an optimal Li-Ion charger"? In particular, how would it differ during the CC phase from a WF-139 if the cell's resistance increases as the cell approaches full charge?

Anyone?

 

[SilverFox]

Hello Andrewnewman,

Dave in post #3 gave a link to the proper Li-Ion charging algorithm. A charger that utilizes that algorithm is doing a proper job of charging Li-Ion cells.

One thing you have to keep in mind is that there isn't a lot of internal resistance change in Li-Ion chemistry as the cells approach full charge. This is why it is possible to easily overcharge the cells. This is also why the cells remain cool during charging and why charging efficiency is so high.

With use the internal resistance will increase. This is why the voltage drops below 4.2 volts when you remove the cell from the charger and let it rest. When the internal resistance increases to the point where the cells voltage after charging and at rest is below 4.0 volts, it is time to recycle the cell.

This increase in internal resistance has nothing to do charging or terminating a charge.

Tom

 

[andrewnewman]

Thank you Silverfox! I re-examined the graph referenced in Dave's post. If I read it correctly, you can apparently apply a 4.0V charge at an appropriate current (~ 0.7C I think I once read) and eventually cause a Li-Ion battery to absorb sufficient energy so that it presents a battery voltage in excess of 4.0V (4.2V in the graph). This surprised me.

Thanks again.

 

[mfm]

I buy Soshine battery carriers and spacers, but would never touch their chargers.

Don't buy Soshine chargers thinking they actually do what the specifications say, they don't:

Soshine SC-S1 (MIX)

http://www.candlepowerforums.com/vb/showpost.php?p=2930726&postcount=1

"This charger has a counterfeit UL listing on the back of it"

Claims to charge by CC/CV but as you can see from the above post, it is bullshit.

Soshine SC-S1(MAX)

http://www.dealextreme.com/details.dx/sku.15608

"Burned down a friends bedroom.. Only use this charger in in a fireproof canister or in a house you are prepared to loose. "

"I agree, this is not a safe product. Mine began to smell like magic smoke and I noticed it getting hot but was able to shut it off on the first attempt at charging."

 

[45/70]

My Trustfire TR-001 only has an open circuit voltage of 4.24 Volts ... The green LED comes on at 4.21 Volts ... I have not left a 18650 on charge for more than a few minutes after the green LED comes on ... If I did leave the battery on for hours after the green LED is lit, it could not go above the chargers open circuit voltage of 4.24V ... Would these extra hours have much effect on the life of the cell or on the the plating that could occur on the anode ? ... How much effect would the extra 0.03 Volts make ?

In short form, your TR-001 having a voltage of 4.24 Volts will have less ill effect than a charger with a higher voltage. You are correct that the voltage cannot go above the voltage of the charger. Still, any charging taking place once the cell (note: not the charger) has reached 4.20 Volts, is trickle charging. I suppose you could say, less of a bad thing is better than more of a bad thing, but it's still not a recommended approach to charging Li-Ion cells. It's not so much whether the trickle charge is large, or small, it's the fact that a trickle charge is taking place at all. No chemistry of Li-Ion cells should be trickle charged past their full charge state.

I think your cells would last longer and probably have a better safety margin if you used a charger that didn't trickle charge your cells, but yes, many people are using the TR-001, it will work. I would definitely recommend not leaving the cells on the charger after they have completed charging, however. If nothing else, it's just a bad habit to develop when charging Li-Ion cells.

... The 18650s will probably be charged whether they need it or not at a low voltage of about 4.0V ... If they get below about 4.0 Volts, it would be under exceptional circumstances ... Do you think that it would be better to leave the charging till the voltage gets well below this point ? ...

The best state of charge (SOC) for Li-Ion cells is somewhere between the limits of, fully charged (4.20 Volts OC for LiCo and LiMn), and discharged (3.60 Volts OC). As SilverFox has mentioned several times on the Forum, at either extreme, oxidation of the electrodes occurs faster. The most "neutral" SOC is around 40%, or ~3.80 Volts OC.

As far as how much to discharge your cells, this is probably somewhat debatable, but the general rule is to stop the discharge of cells at about 20% remaining capacity (~3.7 Volts OC) to prolong cell life. On the other end, charging LiCo cells to 4.10 Volts has been shown to increase the cycle life of the cells 100%, or more. Combining these two extremes, would most likely ensure the longest life of your cells, both in number of cycles, and calander life as well.

Since it's not that easy for everyone to charge their cells to 4.10 Volts, unless you have a hobby charger, what I do is store my cells at ~40%. When I am going to use them, right away, I charge them up to 4.20 Volts, figuring that they will be discharged to well below 4.20 Volts soon enough, so as not to really have that much impact. I then go ahead and use them until I think they are near the 40% level, and then store them. If I discharge the cells below the 40% level, I simply charge them up to the 40% level, and then store, or rotate them. On another note, charging Li-Ion cells frequently has no ill effect, as opposed to attempting to run them through a larger percentage of discharge each time. The only difference may be, how long and frequently, the cells are at, or near the 4.20 Volt level. In any event, properly cared for LiCo Li-Ion cells of recent manufacture, seem to have about a 5 year life, provided they don't reach the cycle limit first.

What would be the value in Milliamps of 0.03C for a 2400mAh 18650 cell at which the ideal charger should cut off ? ... Would that be 72 milliamps ? ... The charger volts would have to be higher than my 4.24 volts to produce this current.

This all is dependent on the charge rate you are using, as "C" in this use, represents the charge current. So, if you were charging your cells at 600mA, 0.03C would be 18mA. If you charge your 2500mAh 18650's at 0.5C ("C" here, represents the capacity of the cell, and is not related to charge current) as I do, that means the CC stage will be at 1250mA, and 0.03C (again "C" here is representative of the charge current) will be 37.5mA.

By the way, are there any detailed articles on Li-Ion cells that would perhaps help me understand this technology better than my present minimal level ?

Battery University is a good place to start, and often referenced here in the Forum. [FONT=Verdana, Arial, Helvetica, sans-serif]Isidor Buchmann is with Cadex, which manufactures battery analyzers and chargers, so has a vested interest in taking care of batteries.[/FONT]

......you can apparently apply a 4.0V charge at an appropriate current (~ 0.7C I think I once read) and eventually cause a Li-Ion battery to absorb sufficient energy so that it presents a battery voltage in excess of 4.0V (4.2V in the graph).......

Umm, that ain't gonna happen. You can't get more Volts out, than you're putting in.

Dave

 

[SilverFox]

Hello Andrewnewman,

I think you better take another look at the graph. You are not reading it correctly.

Tom

 

[old4570]

I buy Soshine battery carriers and spacers, but would never touch their chargers.

Don't buy Soshine chargers thinking they actually do what the specifications say, they don't:

Soshine SC-S1 (MIX)

http://www.candlepowerforums.com/vb/showpost.php?p=2930726&postcount=1

"This charger has a counterfeit UL listing on the back of it"

Claims to charge by CC/CV but as you can see from the above post, it is bullshit.

Soshine SC-S1(MAX)

http://www.dealextreme.com/details.dx/sku.15608

"Burned down a friends bedroom.. Only use this charger in in a fireproof canister or in a house you are prepared to loose. "

"I agree, this is not a safe product. Mine began to smell like magic smoke and I noticed it getting hot but was able to shut it off on the first attempt at charging."

Yes ! Thats why I have resisted buying the 4 bay charger ..
The one Im talking about is the new 2 bay 18650 charger ...
I am not aware of any tests for it ... dx/sku.31897

 

[andrewnewman]

Hello Andrewnewman,

I think you better take another look at the graph. You are not reading it correctly.

Tom

Indeed I'm sure I am. My 30 year old high school physics suggests that for electrons to flow into a storage cell the voltage potential of the charging circuit must be higher than that of the cell w/o a load. If we presume that a Li-Ion cell must be charged no higher than 4.2V, then as the CC portion of the charge reaches it's logical end, then the charging voltage will be higher than 4.2V. If we presume that we want to maintain, say 450mA of constant current then the charger voltage will always float above the current voltage reading of the cell being charged. The internal resistance of the cell will determine how much higher the charging voltage needs to be. A cell with a *very* high internal resistance can't be safely charged and, if I understand correctly, is nearing end of life.

So. In my experiment, my WF-139 charger reached 4.30V before the red light turned green. This was with an unprotected Li-Ion RCR123 cell. I measured the battery voltage within a second of the light turning green and it was 4.15V. While I realized that the charger should have entered a CV stage at this point and WF-139 chargers don't "do" CV stages,

1. Is 0.15V above the battery voltage for the charging voltage reasonable (assuming this charger is trying to maintain CC at 450mA)?
2. In what ways (other than entering a CV stage) would an ideal charger differ from this behavior?

I'm really not trying to be intentionally dense here. Just trying to learn.

 

[SilverFox]

Hello Andrewnewman,

30 years ago the physics books didn't address Li-Ion chemistry...

What you learned is true for NiCd and NiMh chemistry, but it doesn't apply to Li-Ion chemistry.

With Li-Ion chemistry, the cell is charged at a constant current until the voltage reaches 4.2 volts. Then the voltage is clamped at 4.2 volts (not over) and the current reduced to hold the voltage at 4.2 volts. When the current drops off to a certain value, it is shut off and the cell is considered fully charged.

If you parallel a fully charged NiMh cell with a fully discharged cell, they will equalize in voltage, but not in capacity. In this case you need a higher voltage to charge the discharged cell. If you do the same thing with Li-Ion cells, they will equalize in voltage as well as capacity.

It is possible to end up with an ending voltage of 4.2 volts by running the CC phase up to 4.3 volts, and then let the cell settle down from there. The problem with this approach is that it is hard on the cells, and it increases the internal resistance of the cells. Once the cells have been damaged, they will no longer charge up to 4.2 volts.

Very simply put, the proper Li-Ion charging algorithm never subjects the cells to greater than 4.2 volts.

An alternate method for charging without the CV stage is to monitor the cell voltage. When it drops a little, the charger kicks back in and charges it back up to 4.2 volts, then shuts off. This results in a see-saw voltage pattern, but eventually the cell will become fully charged.

This is a little tricky because you would have to pick a lower voltage that would allow for used cells, and you may want to incorporate a timer. The logic would be something like this. If the voltage drops to X voltage in Y time start the charger, otherwise keep things shut off.

Tom

 

[andrewnewman]

Thank you *very much* for this cogent explanation. I may have to noodle it a bit but I think it makes sense superficially. Funny I never considered by 30 year old physic education 'dated' but then it's nice to know I can still learn new things.

Andy

 

[march.brown]

Indeed I'm sure I am. My 30 year old high school physics suggests that for electrons to flow into a storage cell the voltage potential of the charging circuit must be higher than that of the cell w/o a load. If we presume that a Li-Ion cell must be charged no higher than 4.2V, then as the CC portion of the charge reaches it's logical end, then the charging voltage will be higher than 4.2V. If we presume that we want to maintain, say 450mA of constant current then the charger voltage will always float above the current voltage reading of the cell being charged. The internal resistance of the cell will determine how much higher the charging voltage needs to be. A cell with a *very* high internal resistance can't be safely charged and, if I understand correctly, is nearing end of life.

So. In my experiment, my WF-139 charger reached 4.30V before the red light turned green. This was with an unprotected Li-Ion RCR123 cell. I measured the battery voltage within a second of the light turning green and it was 4.15V. While I realized that the charger should have entered a CV stage at this point and WF-139 chargers don't "do" CV stages,

1. Is 0.15V above the battery voltage for the charging voltage reasonable (assuming this charger is trying to maintain CC at 450mA)?
2. In what ways (other than entering a CV stage) would an ideal charger differ from this behavior?

I'm really not trying to be intentionally dense here. Just trying to learn.

.
My charger has an open circuit voltage of only 4.24 Volts and does (eventually) take the 18650 cell up to 4.21 volts ... At 4.21 Volts the green LED comes on but as the charger is a cheapie, it does not cut off the charging current ... With the green LED on, there is still 2.9mA of charge current ... So the charger voltage is greater than the battery voltage as you said ... In my case it is only a difference of 30 millivolts.

In post number 3, the link to Battery University is a great read ... It does give some internal resistance figures typical of aging batteries ... It quotes 155mOhm for a NiCad, 320mOhm for a Li-Ion and 778mOhm for a NiMh cell.

I have not measured the initial charging current as I was more interested in the final stages of charging, however next time I put a 18650 on charge I will check the initial charge current ... From what I understand, the Constant Current part of the charge cycle will not bring the cell up to exactly 4.2 volts ... It seems to take the cell fairly quickly up to almost 4.20 volts ... The charger will not bring the battery voltage to the recommended 4.2 volts without the longish Constant Voltage part of the charge cycle ... This part looks for the diminishing current and when the charge current is low, the ideal or perfect charger would then finally remove the charging current from the cell ... Looking at various reviews and posts, it seems that it takes at least 3 hours to charge a 18650 cell ... In some cases considerably longer as with my Trustfire.

I am hoping that the new Soshine charger will follow the correct algorithm ... It is not the four cell model that has had bad press, but a new two cell model ... There have not been any reviews done yet on this model so hopefully it will be OK ... If not, then I will use the Trustfire that I already have.
.