DIY: Generic way to lower voltage termination on li-ion chargers

7 posts / 0 new
Last post
Barkuti
Barkuti's picture
Offline
Last seen: 1 hour 6 min ago
Joined: 02/19/2014 - 14:46
Posts: 3917
Location: Alhama de Murcia, Spain
DIY: Generic way to lower voltage termination on li-ion chargers

Used this “technique” on certain custom chargers based on generic CC/CV supplies in the past. It ended up working very well.

Recently made some more voltage drop measurements on a few schottky diodes applying a low constant current across them. For example, measured ≈270mV of drop on an SK54 (1N5824 equivalent) when ≈90mA of current goes through it. A 20SQ045 is at ≈250mV with ≈100mA if I do recall right.

As you may guess, setting a diode in series with the charger's output will effectively lower the termination voltage by the diode's Vf at the charger's termination current.

This can be of service for modified gadgets with very or ultra long service life batteries in mind. And for those of you who may want to add a selectable output voltage switch to a existing chargers using SPDT switches and a few schottkies. A couple schottkies or a standard rectifier should drop the standard 4.2Vmax to LiFePO4 levels.

As an unintended bonus it will also protect TP4056 chips from reverse polarity mishaps where removable batteries are involved.

Just that. Hope it can be of DIY service.

 

Cheers Smile 

Innocent

Edited by: Barkuti on 08/03/2019 - 15:23
roguesoul
Offline
Last seen: 6 hours 24 min ago
Joined: 06/25/2019 - 15:16
Posts: 49
Location: Lat: 43.14°NLon: 89.35°WElev: 860ft.

I really like my hobby charger’s ability allowing me to select a charge voltage of 4.1v rather than 4.2v, Doubles the life of the battery by only charging to a capacity of 92%.
Voltage / capacity

Barkuti
Barkuti's picture
Offline
Last seen: 1 hour 6 min ago
Joined: 02/19/2014 - 14:46
Posts: 3917
Location: Alhama de Murcia, Spain

Now that I'm back here, I'll pinpoint the main article which inspired my li-ion loving care practices: BU-808: How to Prolong Lithium-based Batteries @ Battery University. Suffice to say I had already experienced unexpected low cycle life with some batteries (mainly in smartphones).

Batteries have evolved. Many modern quality batteries/cells display higher cycle life and higher energy density at lower charge voltages than the figures claimed in the above article. Looking at the discharge curves of many of the reviewed cells at lygte-info.dk I can clearly see most cells displaying ≈70% or more available capacity when charged to 3.92/4.2V (14/15 of maximum charge voltage). So, at least 65% of the maximum rated energy is available at that point in most cases.

Innocent

Lightbringer
Lightbringer's picture
Offline
Last seen: 3 hours 42 min ago
Joined: 08/30/2016 - 14:12
Posts: 8856
Location: nyc

I’d be happy with 4.1V or so. Better cycle-life, yet still about fully-charged.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

DavidEF
DavidEF's picture
Offline
Last seen: 3 hours 50 min ago
Joined: 06/05/2014 - 06:00
Posts: 7395
Location: Salisbury, North Carolina, USA

Lightbringer wrote:
I’d be happy with 4.1V or so. Better cycle-life, yet still about fully-charged.

If you subtract 250mV from a 4.2V charge, you should get 3.95V instead. I don’t see in this thread a way to get just a 100mV drop from termination voltage, except where roguesoul said his hobby charger is capable of doing it by selection.

The Cycle of Goodness: “No one prospers without rendering benefit to others”
- The YKK Philosophy

Lightbringer
Lightbringer's picture
Offline
Last seen: 3 hours 42 min ago
Joined: 08/30/2016 - 14:12
Posts: 8856
Location: nyc

Whereas a Si diode drops 0.7V at Vγ, a Ge diode would drop only 0.3V.

In my always-on car voltmeter, I had a big 0-1mA meter (labelled 10V-15V), so had to drop pretty much 10V exactly for an accurate read. Used a 9.1V zener diode and pair of plain ol’ 1N400x diodes to drop that ~1V, about 0.5V each at low currents. If I would’ve cherrypicked zeners, I could’ve gotten it spot-on, but didn’t care about 0.1V here’n‘there.

So even if a Ge diode drops 0.2V, it’s at least closer, maybe even closer than with a Schottky diode.

A saturated transistor could drop VCE to 0.1V or less, too.

Unno, seems smarter to have the thing just built-in, for 4.2V cells, 4.3V cells, “storage mode”, and “long-life mode”.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

Barkuti
Barkuti's picture
Offline
Last seen: 1 hour 6 min ago
Joined: 02/19/2014 - 14:46
Posts: 3917
Location: Alhama de Murcia, Spain

This is a sort of generic trick. If you really need fully custom output voltage a CC/CV power supply/module is due, if you can live without termination that is. This is no problem for me.

Using a charger with 4.35V output and a schottky should deliver something around 4.1V. The exact figure depends on a combination of termination current, diode size and type plus its operating temperature (!). Just re-measured a 20SQ045, 264mV at ≈92mA, the room was a bit cooler than the other day with the SK54. With enough diodes in parallel to share a really small current each diode we could get into ≈200mV territory, but that would look pretty (>ლ) frankenstein.

Innocent