Charging lithium ion battery

[GeorgeTsak]

Hi! I saw a video Dave posted about charging lithium ion batteries with a power supply and I was wondering if this procedure is safe for the power supply. I am asking this because I have seen a couple videos people posted about their supply got ruined. (I think that they applied less voltage than that of the battery resulting in current flowing from the battery to the power supply, eg they applied 2V in a 3.7V cell, I am not completely sure though)

[RGB255_0_0]

Get a TP4056 module from eBay or Ali and don't worry about it. Sure if you're testing absolute capacity what Dave did will give better results but for the man on the street  the aforementioned is perfectly fine. Or a dedicated charging station etc...

[GeorgeTsak]

Get a TP4056 module from eBay or Ali and don't worry about it. Sure if you're testing absolute capacity what Dave did will give better results but for the man on the street  the aforementioned is perfectly fine. Or a dedicated charging station etc...

Yes I am aware of this module and I have bought some (Thank you for mentioning it though). However my question is; Is charging a battery with a power supply dangerous for the power supply? I just want to know that.

[RGB255_0_0]

https://www.eevblog.com/forum/blog/eevblog-919-how-to-charge-li-ionlipo-batteries-with-a-power-supply/

Read that thread then.

[Rerouter]

 it comes down to how the power supply is made internally, and how it holds up to an external power source on its output when its turned off or set to a lower voltage.

Personally i have used this guy to charge all sorts of batteries
http://www.ebay.com.au/itm/30V-5A-Portable-Switching-DC-Power-Supply-Test-Adjustable-Variable-LED-Digital-/142029184752?hash=item2111995ef0:g:HFYAAOSw-W5Uvf-A

You can generally find them a bit cheaper, and the 10A versions only come with a single turn current set pot, while the 5A comes with a 10 turn
Now they will happily survive loosing input power, but i would say if you ever plan to walk away from it, use a diode inline, as it does have a load resistor on the output for regulation.

For most battery types it will need to have a constant current mode, especially lithium. and for the lithium types, don't get impatient, charge at 1C (the amp hour rating of the battery) or lower.

in general if it has balance leads, check the state of all the cells, to see that there pretty much even, if not you may need to charge cells in the pack to a matching voltage and monitor them carefully to see that 1 cells isnt chearging drastically different from the rest.

If the battery doesnt have balance leads, then only charge on a non flammable surface, if the cells are rated to 4.2V, only charge him to 4.1 per cell (that way if your supply is off a little you have a safety margin, and you dont really loose much capacity)

If you want something a little more plug and play and already have a power supply without a current limit, then i suggest something like this guy
https://hobbyking.com/en_us/turnigy-accucel-6-80w-10a-balancer-charger-lihv-capable.html

[Seekonk]

Love those TP4056 modules. I just hook them up with a cell phone wall wart.  Note in the spec it says to use a .5 ohm resistor in series.  You don't have to, but better to put the heat in a resistor than an IC.  This is a linear regulator.  I use up to a couple ohms to limit charging rate on iffy used cells.  It lowers maximum charge current but not end current.  These shut off charge in the end.  A power supply won't do that.  Under charged cells last longer and only a few percent of capacity is lost under charging.

[Rick Law]

If you want to limit (lower) the charge current, the TP4056 can adjust charge current from 130mA to 1A.

Change R_prog (pin 2) to different values for different charge current.

I = 1200/R_prog

So, an R_prog of 1.2K give you exactly 1Amp charge current.

Look up the specs and you can find all sorts of other goodies there.

[GeorgeTsak]

https://www.eevblog.com/forum/blog/eevblog-919-how-to-charge-li-ionlipo-batteries-with-a-power-supply/

Read that thread then.

Thanks for the info!

it comes down to how the power supply is made internally, and how it holds up to an external power source on its output when its turned off or set to a lower voltage.

Personally i have used this guy to charge all sorts of batteries
http://www.ebay.com.au/itm/30V-5A-Portable-Switching-DC-Power-Supply-Test-Adjustable-Variable-LED-Digital-/142029184752?hash=item2111995ef0:g:HFYAAOSw-W5Uvf-A

You can generally find them a bit cheaper, and the 10A versions only come with a single turn current set pot, while the 5A comes with a 10 turn
Now they will happily survive loosing input power, but i would say if you ever plan to walk away from it, use a diode inline, as it does have a load resistor on the output for regulation.

For most battery types it will need to have a constant current mode, especially lithium. and for the lithium types, don't get impatient, charge at 1C (the amp hour rating of the battery) or lower.

in general if it has balance leads, check the state of all the cells, to see that there pretty much even, if not you may need to charge cells in the pack to a matching voltage and monitor them carefully to see that 1 cells isnt chearging drastically different from the rest.

If the battery doesnt have balance leads, then only charge on a non flammable surface, if the cells are rated to 4.2V, only charge him to 4.1 per cell (that way if your supply is off a little you have a safety margin, and you dont really loose much capacity)

If you want something a little more plug and play and already have a power supply without a current limit, then i suggest something like this guy
https://hobbyking.com/en_us/turnigy-accucel-6-80w-10a-balancer-charger-lihv-capable.html

hmm... Yes I should have thought to use a diode in series and using a multimeter to increase the voltage to compromise the voltage drop. Also I agree the using 4.1 volts is the safest thing to do... (Unless you have an expensive power supply or an accurate multimeter)

Love those TP4056 modules. I just hook them up with a cell phone wall wart.  Note in the spec it says to use a .5 ohm resistor in series.  You don't have to, but better to put the heat in a resistor than an IC.  This is a linear regulator.  I use up to a couple ohms to limit charging rate on iffy used cells.  It lowers maximum charge current but not end current.  These shut off charge in the end.  A power supply won't do that.  Under charged cells last longer and only a few percent of capacity is lost under charging.

Doesn't the TP4056 module limit the current by itself? Owning this module I have never used a resistor in series... As far as I know the maximum current it can provide is 1A.

If you want to limit (lower) the charge current, the TP4056 can adjust charge current from 130mA to 1A.

Change R_prog (pin 2) to different values for different charge current.

I = 1200/R_prog

So, an R_prog of 1.2K give you exactly 1Amp charge current.

Look up the specs and you can find all sorts of other goodies there.

Thanks for the info! 130ma? Is there a 18650 cell with such a low capacity? Unless you want to slowly charge a low capacity cell e.g. a 260ma at 0.5C

[Seekonk]

"Doesn't the TP4056 module limit the current by itself? Owning this module I have never used a resistor in series... As far as I know the maximum current it can provide is 1A."

Yes it does and it turns that into heat.  ! amp and 1V means you are dumping 1W into the IC.  It can take that, but wouldn't you rather put that heat somewhere else.  Check the spec sheet, they have the resistor there and it has a spot on the board that is just jumpered out.  The boards are shipped 1A and then taper to 0.1A till they shut off at 4.2V.  I grab a lot of recycled laptop power packs and they can be risky if left unattended. I prefer not to charge an old cell at the maximum. I use phone wall warts.  Many are self limiting as far as current, others are rated at just about a half amp.  The resistor easily limits maximum current, but not the end current which is fixed at 1/10 the max charge rate determined by that little snot of a resistor.  I wish you luck changing that.  The foil is so thin on these boards, think they just rubbed a penny on them, it is likely to lift off changing that resistor.  If you are that experienced with surface mount I would hardly guess you need to come on a forum and ask questions.  You can go to MUCH lower currents than the 130ma specified in the chart.  Some use this board to charge coin cells (the rechargeable ones that is).

[GeorgeTsak]

"Doesn't the TP4056 module limit the current by itself? Owning this module I have never used a resistor in series... As far as I know the maximum current it can provide is 1A."

Yes it does and it turns that into heat.  ! amp and 1V means you are dumping 1W into the IC.  It can take that, but wouldn't you rather put that heat somewhere else.  Check the spec sheet, they have the resistor there and it has a spot on the board that is just jumpered out.  The boards are shipped 1A and then taper to 0.1A till they shut off at 4.2V.  I grab a lot of recycled laptop power packs and they can be risky if left unattended. I prefer not to charge an old cell at the maximum. I use phone wall warts.  Many are self limiting as far as current, others are rated at just about a half amp.  The resistor easily limits maximum current, but not the end current which is fixed at 1/10 the max charge rate determined by that little snot of a resistor.  I wish you luck changing that.  The foil is so thin on these boards, think they just rubbed a penny on them, it is likely to lift off changing that resistor.  If you are that experienced with surface mount I would hardly guess you need to come on a forum and ask questions.  You can go to MUCH lower currents than the 130ma specified in the chart.  Some use this board to charge coin cells (the rechargeable ones that is).

I see your point and you are right. Thanks again for getting into the trouble to explain all this to me Have a nice day!

[IanB]

Hi! I saw a video Dave posted about charging lithium ion batteries with a power supply and I was wondering if this procedure is safe for the power supply. I am asking this because I have seen a couple videos people posted about their supply got ruined. (I think that they applied less voltage than that of the battery resulting in current flowing from the battery to the power supply, eg they applied 2V in a 3.7V cell, I am not completely sure though)

If you can find detailed operating specifications for the power supply (good power supplies will have this information), it will tell you what the power supply can safely handle in terms of reverse voltage on the output terminals.

There is no general answer to the question of "is it safe?". It depends on the specific supply you are using-- although most good adjustable laboratory power supplies will be able to charge lithium ion batteries without harm.

[Rick Law]

"Doesn't the TP4056 module limit the current by itself? Owning this module I have never used a resistor in series... As far as I know the maximum current it can provide is 1A."

Yes it does and it turns that into heat.  ! amp and 1V means you are dumping 1W into the IC.  It can take that, but wouldn't you rather put that heat somewhere else.  Check the spec sheet, they have the resistor there and it has a spot on the board that is just jumpered out.  The boards are shipped 1A and then taper to 0.1A till they shut off at 4.2V.  I grab a lot of recycled laptop power packs and they can be risky if left unattended. I prefer not to charge an old cell at the maximum. I use phone wall warts.  Many are self limiting as far as current, others are rated at just about a half amp.  The resistor easily limits maximum current, but not the end current which is fixed at 1/10 the max charge rate determined by that little snot of a resistor.  I wish you luck changing that.  The foil is so thin on these boards, think they just rubbed a penny on them, it is likely to lift off changing that resistor.  If you are that experienced with surface mount I would hardly guess you need to come on a forum and ask questions.  You can go to MUCH lower currents than the 130ma specified in the chart.  Some use this board to charge coin cells (the rechargeable o4nes that is).

For current limiting, the TP4056 does not dump that much into heat.  It lowers the charge voltage to keep the CC (constant current) selected by R_prog.

So the waste heat does not equal to V_supplied*(1A-I_desired), rather,
the waste heat is the I_desired*V_drop.
If the I_desired is lowered, your waste heat is also lowered.
[ see EDIT below ]

You can see that for yourself and connect an Amp-meter before the TP4056 and after the TP4056.  The current coming out of it is the same as the current going in.  V_drop is the difference in voltage going into TP4056 and the voltage coming out of it.

Personally, I think letting TP4056 manage the current is just fine as long as your power supply is around 5V.  I am however more concern about the (per-spec) 4.2V +/- 1.5% in charge voltage.  On the real thing, 4.263V output is a bit more than I would like.  If you get a fake-TP4056, it may be worst.

I have a cut-tape of 20xTP4056es.  I actually paid more from what I thought would be a reputable dealer.  I tested two of them and they were in the 4.4V to 4.5V range.  Since it was to repair a board, it could be other things on my board that caused the 4.5V out, but I now adopt the "Trust but verify before use" mantra before I "deploy" the darn thing.

Once tested, they do work like a charm.  Love it.

[EDIT: forgot to mention:]
If the I_desired is lowered, it is done by increasing V_drop.  So it may seem to be just a wash with I_desired*V_drop.  Experience with charging shows you don't need to drop the voltage by 1/2 just to drop the current by 1/2.  So, V_drop requires a small increase to cause a much bigger decrease in I_desired.  End result is, I_desired*V_drop got smaller.

[Seekonk]

"For current limiting, the TP4056 does not dump that much into heat.  It lowers the charge voltage to keep the CC (constant current) selected by Rprog.

So the waste heat does not equal to Vsupplied*(1A-Idesired), rather,
the waste heat is the Idesired*Vdrop.
If the Idesired is lowered, your waste heat is also lowered."

It's the internet.  If I say it is a nice day, someone will come along and say there is a little cloud over there.  That isn't at all what I said.  Typical supply voltage is about 5v so you very easily can get 1W of dissipation with about 1V of drop.  I still think that is a lot of dissipation in a chip that you don't know exactly who made it on a board with no dissipating copper.  Don't think the OP was confused about that power loss at all.

[Rick Law]

...
It's the internet.  If I say it is a nice day, someone will come along and say there is a little cloud over there.  That isn't at all what I said.  Typical supply voltage is about 5v so you very easily can get 1W of dissipation with about 1V of drop.
...

I do get too argumentative.  If that offended you, sorry!  Didn't mean to be offensive...