Charging 18650 Li-ion w/Lab Power Supply

spectral532

Newly Enlightened
Joined
Jan 15, 2010
Messages
8
Would it be possible to charge a 18650 Li-ion with a Lab (bench-top) power supply? The power supply I have access to is voltage regulated with an adjustable current limit. I was thinking of setting the voltage at 4.2v and current limit at 0.5C (~1A). That way the current limit would hold the charge current at 1A until the current draw decreases. Then the power supply would hold steady at 4.2v for the remainder of the charge. I could then manually cutoff power when charge current drops below 0.1C.

I'm still rather new to Li-ion batteries though, and I'm not sure if this would be a safe way to charge the batteries. Any advice would be appreciated.
 

T24

Newly Enlightened
Joined
Jan 2, 2010
Messages
35
Location
Germany
when the power supply would hold steady at 4.2V --> safe

if not --> maybe :poof:
 

65535

Flashlight Enthusiast
Joined
Dec 13, 2006
Messages
3,320
Location
*Out There* (Irvine, CA)
Just make sure you have an expensive power supply, a lot of them have poor meters and less than perfect regulation. If it will hold .01V you should be fine. Keep an external VM on the cell the first time or two as a double check.

Otherwise the idea is completely sound, and ideal.
 

TranceAddict

Newly Enlightened
Joined
May 28, 2008
Messages
189
Location
Bristol, United Kingdom
i was using lab CC/CV power supply as li-ion charger as well before i built this https://www.candlepowerforums.com/threads/233932 which worked pretty fine and its algorithm far surpass ultrafire charger and even Pila IBC.


as long as you set voltage to 4.2v it will be fine even if you leave them on forever, when supply is 4.2v and cell is also 4.2 there will be no potential as so you will get 0amp if you measure it. this solution has only one drawback, which is when cell voltage raise to around 4v, it seem to take infinite time to reach 4.2v due to too small current flowing pushed by the 0.2v potential difference. same rules apply when you connect many cells in parallel to the power supply, they will balance each others and charge perfectly to 4.2v.
 

45/70

Flashlight Enthusiast
Joined
Oct 9, 2005
Messages
2,800
Location
Rural Ohio
Just make sure you have an expensive power supply, a lot of them have poor meters and less than perfect regulation. If it will hold .01V you should be fine. Keep an external VM on the cell the first time or two as a double check.

I've used my Metex MS-9140's PS for charging Li-Ion cells. It is very important to watch the current. :caution:

The DMM in the 9140 has always checked out to be better than +/- 1% @ 3.3, 5, and 12 Volts, but the meter on the PS is more like +/- 3%. What you can do, if you don't want to monitor the voltage/current with a separate DMM, is set your PS voltage to 4.10 Volts. Provided the readings on your PS aren't totally off, this will charge the cell(s) to 90% within a reasonable safety margin.

as long as you set voltage to 4.2v it will be fine even if you leave them on forever, when supply is 4.2v and cell is also 4.2 there will be no potential as so you will get 0amp if you measure it.

As I understand it, this is not a good idea. Edit: From B.U.

"No trickle charge is applied because lithium-ion is unable to absorb overcharge. A continuous trickle charge above 4.05V/cell would causes plating of metallic lithium that could lead to instabilities and compromise safety."
The CV stage should cutoff at a current no lower than 0.03C, or you run the risk of metallic lithium plating out onto the anode. A lot of hobby chargers cut the CV stage at 0.05-0.1C. This would add an acceptable safety margin without sacrificing much capacity, as well as probably prolonging cell life.

In reality, it's better to just use a good charger. As has been pointed out, using a bench supply takes a long time and you have to watch the current closely. :caution: The only time I ever use the Metex, is when I'm determining cell condition, and want an actual 4.20 Volts "off the charger". As I have chargers for most of my LiCo cells that do charge to 4.20 Volts, I rarely use the Metex anymore.

Dave
 
Last edited:

rmteo

Flashlight Enthusiast
Joined
Feb 11, 2009
Messages
1,071
Location
Colorado, USA
I have charged Lithium cells using a bench power supply but no longer do it. My PSU is a 1200W (0-30V, 0-40A) switching Xantrex unit. It is just too easy to accidentally nudge the controls (or to make an incorrect setting setting due to loss of concentration) to make the process potentially dangerous to your health.

With 6-cell, 5A balancing hobby chargers starting at under $50, there really is no need to look at alternative charging methods. Also, these hobby chargers have several safety mechanisms built in such as cell count determination, low voltage, auto cut-off, etc. which you do not have with a bench supply.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello Spectral532,

Welcome to CPF.

As others have mentioned this is a great way to charge Li-Ion cells.

However there is one "issue." Li-Ion chargers will terminate the charge when the charge current drops below 50 - 100 mA. The reason they do this is because Li-Ion chemistry does not handle "trickle charging" very well.

In this case, we are defining "trickle charging" as the charge that occurs when the cell is at 4.2 volts and the current is in the range of 50 - 0 mA.

If you let your power supply go, it will continue to charge until the current drops to 0 mA. The "trickle charging" phase tends to oxidize the electrolyte inside the cell, and is detrimental to long term life. It also takes a long time to complete this final stage, so charging time is extended. This is especially true when the cell has developed some internal resistance and is no longer holding 4.2 volts rested. With a cell that has been in service for awhile and is at around 90% of its original capacity, the amount of time spent "trickle charging" can be equal to or exceed the total charge time if the charge was terminated when the current dropped to around 100 mA. As the cell ages further, you get into the situation where the "trickle charge" never ends.

There are a couple of ways to work around this.

One way is to be very vigilant and stop the charge when the current falls below a certain amount. This is a very good way to go, but sometimes we get distracted and let things go a little longer than we had planned.

A better way is to simply reduce the voltage you are charging to.

Cycle life is related to the maximum voltage you charge the cell to. If you charge to 4.2 volts, you should get somewhere around 500 charge/discharge cycles. However, if you limit the charge to 4.1 volts, the cycle life extends to somewhere around 1500 charge/discharge cycles.

With this information in hand, you can now adjust the settings to avoid any problems with "trickle charging." You simply set the voltage to 4.15 or 4.18 volts. With cells in good condition I usually use 4.18 volts. As the cells age and develop some internal resistance, I drop that to 4.15 volts. Many people simply use 4.15 volts all the time.

With this method you still have to watch the charge process, but you don't have to be glued to it. Since our definition of "trickle charge" is very specific, at a lower voltage we are not "trickle charging." Charging using a power supply is safest using this method.

There is one other method that can be used. Normal charge rates are in the 0.5C - 1.0C range. An optimum rate is about 0.7C. The idea is to balance the CC phase and the CV phase of the charge. The faster you charge, the more time you spend in the CV phase, and the greater the possibility of "trickle charging." The ideal rate (from the cells perspective) is 0.7C, but everyone wants faster charging and I am not sure there is much of a difference in cell life between charging at 0.7C or 1.0C.

At any rate, if you charge at a low charge rate, it seems that you can eliminate the CV stage completely. I have not tried this, but have read that if your charge current is about 0.18C, or lower, you can simply terminate the charge when the voltage reaches 4.2 volts.

Once again you have to watch the end of the charge closely, but it is another option. The general rule that the termination current is usually about 10% of the initial charge current does not apply in this case, but you simply terminate upon the voltage reaching 4.2 volts.

One final caution... from experience.

If you have curious cats, watch them carefully while charging. I had a cat become interested in the changing displays and was pawing at the control nobs... :)

Tom
 

BVH

Flashaholic
Joined
Sep 25, 2004
Messages
7,023
Location
CentCalCoast
Tom, I've read where a Schotkey diode should be installed in the + lead of the PS to prevent backflo of power. I did not have one in my 1st Mastec 30V-50A and I believe I had some backflo when I happen to carelessly make a final connection to energize an HID ballast. With a bare wire, I ended up making and breaking hot contact probably 3 or four times in a fraction of a second. I smoked the supply. I now have a Schotkey in-line and have had no issues, although I am much more careful about my connections and have a toggle switch in-line to make the final connection. Your thoughts?
 

45/70

Flashlight Enthusiast
Joined
Oct 9, 2005
Messages
2,800
Location
Rural Ohio
As others have mentioned this is a great way to charge Li-Ion cells.

One way is to be very vigilant and stop the charge when the current falls below a certain amount. This is a very good way to go, but sometimes we get distracted and let things go a little longer than we had planned.......

With this method you still have to watch the charge process, but you don't have to be glued to it.......

Once again you have to watch the end of the charge closely, but it is another option.......

One final caution... from experience.

If you have curious cats, watch them carefully while charging. I had a cat become interested in the changing displays and was pawing at the control nobs... :)


What's really neat, is charging single NiMH or NiCd cells with a bench PS. Your eyes are so "glued" to the display, watching for that 0.005 Volt -dV that the cat isn't a problem! :crackup:

Before anyone asks, no, I haven't done this. :)

Dave
 

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
I've read where a Schotkey diode should be installed in the + lead of the PS to prevent backflo of power. I did not have one in my 1st Mastec 30V-50A and I believe I had some backflo when I happen to carelessly make a final connection to energize an HID ballast. With a bare wire, I ended up making and breaking hot contact probably 3 or four times in a fraction of a second. I smoked the supply. I now have a Schotkey in-line and have had no issues, although I am much more careful about my connections and have a toggle switch in-line to make the final connection.
I know this was addressed to Tom, but how were you using the power supply?

Here is what I think is a typical safe way to use a lab power supply:


  1. Set the power supply to the minimum voltage
  2. Set the current limit just above the highest expected load value
  3. Discharge the power supply terminals with a small load (bulb or medium value resistor) and remove
  4. Discharge any capacitors in the equipment to be tested with a resistor
  5. Connect the equipment to be tested to the power supply
  6. Slowly increase the power supply voltage to the required value while watching for any unexpected events in the equipment

This should save you from untoward happenings, protecting both the power supply and the equipment being tested.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello BVH,

I have also heard that using a diode is a good thing to do, but I have never used one. While I frequently charge Li-Ion batteries with my power supply, that is not its main use. If I ever get to the point of dedicating a power supply to charging, I will look further into this.

Tom
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
Hello BVH,

I have also heard that using a diode is a good thing to do, but I have never used one. While I frequently charge Li-Ion batteries with my power supply, that is not its main use. If I ever get to the point of dedicating a power supply to charging, I will look further into this.

Tom

There are both positive and negative aspects to using a diode for charging on a power supply.

The plus side is that the batteries will be not discharged if the power supply is turned off, while connected to the batteries.
A badly designed power supply might also be damaged, if turned off while the batteries are connected.

The minus side to diodes, it the voltage drop, the diode will drop some voltage (between 0.5 to 1 volt depending on current), making it a bit difficult to adjust the power supply to exactly the required voltage.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello Mr Happy,



I know this was addressed to Tom, but how were you using the power supply?

Here is what I think is a typical safe way to use a lab power supply:


  1. Set the power supply to the minimum voltage
  2. Set the current limit just above the highest expected load value
  3. Discharge the power supply terminals with a small load (bulb or medium value resistor) and remove
  4. Discharge any capacitors in the equipment to be tested with a resistor
  5. Connect the equipment to be tested to the power supply
  6. Slowly increase the power supply voltage to the required value while watching for any unexpected events in the equipment

This should save you from untoward happenings, protecting both the power supply and the equipment being tested.



I set things up a little differently for Li-Ion battery charging...

I turn the current down to near 0 mA, then adjust the voltage to 4.15 volts. I then hook up my battery and adjust the charging current. When the charge is complete, I zero the charging current and disconnect.

Tom
 

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
I set things up a little differently for Li-Ion battery charging...

I turn the current down to near 0 mA, then adjust the voltage to 4.15 volts. I then hook up my battery and adjust the charging current. When the charge is complete, I zero the charging current and disconnect.
I agree. The procedure is a little different as you describe when connecting to an energized device like a battery (which is a slightly unusual use for a lab power supply).

I believe BVH was connecting to an HID ballast which is more of a passive load.

In every case, it is always a good plan to short the power supply terminals with a dummy load after setting the voltage and before connecting the real device. This will discharge any output capacitors in the power supply and prevent any unexpectedly high voltage spikes being fed into delicate equipment.

When working with electronic circuits one should always be careful of residual charge on capacitors. In some cases they can kill.
 

45/70

Flashlight Enthusiast
Joined
Oct 9, 2005
Messages
2,800
Location
Rural Ohio
I turn the current down to near 0 mA, then adjust the voltage to 4.15 volts. I then hook up my battery and adjust the charging current. When the charge is complete, I zero the charging current and disconnect.

That's the way I've always done it for Li-Ion cells, except I adjust the current to zero, initially. When I use the PS for forming NiMH or NiCd packs, I turn the voltage up to 30V, and then once the pack is connected, adjust the current up (from zero) to 0.1C.

The 0-30V section of my PS is only 3A max, but I guess that really wouldn't make any difference, in regards to the intensity of any spikes, or would it?

Also, I added a quote and link from Battery University to post #6

Dave
 

spectral532

Newly Enlightened
Joined
Jan 15, 2010
Messages
8
Thanks for the replies everyone. I couldn't have asked for better answers.

In response to the concerns of a few members, I am using a digital power supply with a built in volt/ammeter. It has an accuracy of about ±4mV, so it should be fine for this purpose. I plan on manually monitoring the current, and will stop the charge when current drops to ~.1C.

@Silverfox

Why is it that "trickle charging" at 4.15V is fine, while doing so at 4.2V isn't? I dont think I completely understand your explanation. Additionally, why is 0.7C the optimal charging current? Would it be any better for the batteries if I charged them at less than 0.7C? Say, if I decided to charge the batteries at 0.1C all the way, would that in any way effect the cycle life?

Again, thanks for the help.
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello Spectral532,

If you review the definition, you will remember that "trickle charging" occurs when the cell is at 4.2 volts and the current drops very low. Slow charging at 4.15 volts is simply slow charging and not "trickle charging."

I have not had the opportunity to explore, study, test, or experience plating first hand, so I am just providing information from others who have run into this. It seems that the conditions for plating are most favorable when the cell is at full charge, or is overcharged. This is one of the main reasons for recycling cells when they fall below 80% of their initial capacity.

The 0.7C charge rate comes from the battery manufacturers. They feel that this is the optimum rate to balance cell life and charge time.

Tom
 

45/70

Flashlight Enthusiast
Joined
Oct 9, 2005
Messages
2,800
Location
Rural Ohio
My interpretation of "trickle charging" with regard to Li-Ion cells, is any additional charge beyond the predetermined current cutoff point (usually 0.03-0.1C).

If your charging your cells to 4.15 Volts, as SilverFox suggests (and I normally do, as that's about where my "el cheapo" hobby charger stops (actually, 4.16V) when set to "LiPo" anyway), that's OK, provided the cell doesn't continue to receive current past the predetermined cutoff charge current level, of the CV stage. If the cell does continue to charge past this point, then that constitutes a trickle charge.

According to multiple sources, "trickle charging" can happen at any point from 4.05 Volts, or above. Below this voltage, I guess it's just considerd a really slow charge. :)

Dave
 

SilverFox

Flashaholic
Joined
Jan 19, 2003
Messages
12,449
Location
Bellingham WA
Hello Dave,

The early Li-Ion cells, and many of the current Saft cells, require limiting the charge to 4.1 volts. With these cells, plating can begin at a lower voltage. I don't think the current mix of chemicals that allow for 4.2 volt charging plate out at that low a voltage. Unless the cell has deteriorated to below 80% of its initial capacity.

The point is that Li-Ion cells can not tolerate overcharging. When the cell reaches full charge, additional charging is harmful. If you are charging to 4.2 volts, and the cell settled down to 4.16 volts, you would be better off using a higher cut off current. On the other hand, if you have a cell that is designed to be charged to 4.2 volts, and you limit the charge to 4.15 volts, you never achieve full charge, and the risk of plating is greatly reduced.

Tom
 
Top