Serial battery charging with in series TP4056 modules? Sort of YES! :-D

34 posts / 0 new
Last post

Pages

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain
Serial battery charging with in series TP4056 modules? Sort of YES! :-D

Well, similar threads on this subject can be found around. We know the TP4056 module shares ground with its input and thus it cannot be used to charge cells in series unless each module's supply is fully isolated. However, what if I we feed them in series with n times the input voltage, where n is the number of cells (and modules) in series?

I am planning in a simple 2S setup fed with 10V, voltage and current I can obtain with a couple MT3608 modules connected to a standard 5V 2A supply. The MT3608 modules' 10V output would be connected to two in series TP4056 boards feeding in series 2S cells. This sounds like it should work without problems but, what do you think? 

 

Cheers 

Edited by: Barkuti on 10/26/2017 - 11:24
Jensen567
Offline
Last seen: 5 months 2 weeks ago
Joined: 04/10/2017 - 16:59
Posts: 491
Location: MD/DC Area USA

I’ve thought about trying this, but never actually have. I suspect it won’t work right as getting them to share voltage evenly might not work. They’re cheap enough that you could give it a shot and not be out much even if you blow them up. I didn’t have any spares on hand when I considered it.

ImA4Wheelr
Offline
Last seen: 1 day 19 hours ago
Joined: 02/03/2013 - 14:51
Posts: 7916
Location: SC

Seems like it would work when both chargers are pulling about the same amount of current, but what happens when one starts pulling less current.  Wouldn't the other be hit with more than 5v?

I quickly glances at a picture of the TP4056 board.  It appears that the cell negative connects straight to ground (I could be wrong as I don't have time to study the circuit).  If that is the case, couldn't the inputs to the chargers be paralleled to 5v source and then the Batt - trace to ground on the second charger be cut?  This would float the output, but the output would be grounded to the positive of the previous cell.  So it wouldn't be floating willy nilly.

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

0K, since 5V SPDT power relays are cheap as chips I've cooked up this idea:

Suffice to say the 5V input powers up the relay coils, I just didn't wanted to mess up the drawing any more. This should work… provided that the TP4056 chips behave without bursting into flames with 2S cell voltage on their BAT+ and BAT− terminals while the relay coils are disengaged.   

Cells would be protected. Innocent

What do you think?

 

Cheers Smile

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

Update, final one I believe. Gonna go the above-mentioned route, though not with bare charging modules like the ones above. I had a spare TP4056 module with battery protection (DW01 + 8205A), and of course I've tested the stuff by attaching 2S cells to its P/B+ and B− terminals. Of course it works right, the DW01 shuts off the circuit when 2S voltage is on the terminals, the TP4056 leds remain off even if power is applied to the board. Removal of the overvoltage condition (relays change cell arrangement) makes it work again. This, Sir, will kill. Smile

 

Cheers 

HKJ
HKJ's picture
Offline
Last seen: 52 min 55 sec ago
Joined: 05/24/2011 - 12:23
Posts: 6980
Location: Copenhagen, Denmark

Why not use something like this:
https://www.ebay.com/itm/USB-2S-7-4V-8-4V-Lithium-Lipo-Li-ion-18650-Batt...

It will not balance the batteries, you may need to do that manually once a year.

With two batteries in series I will also recommend a BMS board:
https://www.ebay.com/itm/Module-Automatic-Charging-18650-Charger-Battery...

Note: I have not tested the above product and do not know how well they work. Evil

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

Thanks for pointing out these boost charging modules, HKJ. I had not even bothered looking up for that sort of stuff.

The cells I'm gonna use in this setup come protected (built in BMS), and the load won't pull above 1.5A from the 2S battery in any case (no more than 8W into the driver). Should work rather well. Upon 5V PSU connection a momentary short-circuit may happen on the cells (relay coil timing), a no issue since they're protected.

We'll see how it behaves. 

 

Cheers Party

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

Hi!

I have the feeling this topic has been beaten to death but still i would like to have your guys opinion.
I also want to charge two lipo’s the cheapest way possible but also do balanced charging.
Looking at all the information online i have come up with the following (cheap) solution trying to use only ONE relay for two batt’s.
Or would the two way relay be better?
Please let me know what you guys think or are there better (small+cheap) solutions

i think this should work..

For anyone reading this and thinks this are tested circuits.
THERE NOT!
If you build them, please do, and let me know if they work Wink
IF i build them i’ll let you know here what my results are….

DavidEF
DavidEF's picture
Offline
Last seen: 9 hours 3 min ago
Joined: 06/05/2014 - 06:00
Posts: 7678
Location: Salisbury, North Carolina, USA

I happen to have some cheap TP4056 charger modules on the way to me from eBay. But I’m not smart enough to do any of that stuff. Sorry. Facepalm

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

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

No worries.
Any feedback is good Wink
If anyone can help out please do! ;) ;)
(especially on the single relay circuit)

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

For anyone reading this.
I have ditched the idea for 2s because i could not find a 2s BMS (CHARGER).
So i’m switching to 3s as there seems to be a good one for that!

Described by this guy:
https://www.youtube.com/watch?v=ar6efZS0fdw

search on ebay for:

3s PCB BMS Protection Board For 3 Packs 18650

NOTE! you have to see/get the one from he video as it has a proper charger IC on it.

Lexel
Lexel's picture
Offline
Last seen: 55 min 14 sec ago
Joined: 11/01/2016 - 08:00
Posts: 5487
Location: Germany

honestly I would search for a 2S BMS instead this solution, cant be hard to find a proper one

Jerommel
Jerommel's picture
Offline
Last seen: 5 months 2 weeks ago
Joined: 01/04/2014 - 13:18
Posts: 5721
Location: the Hague, Netherlands

Something like this:

https://www.ebay.com/itm/2S-20A-7-4V-8-4V-Li-ion-Lithium-Battery-18650-C...PCB-BMS-Protection-Board/202358675893

But this is not really balance charging, it’s just individual termination and bypass when one of the cells is fully charged (isn’t it?).
So you have to continue charging until both cells are fully charged.

I think you also need a current limited input.
It’s probably not advised to let the BMS limit the charging current.
That’s only a protection feature, not intended to regulate your charging current.

So, what you would need with this kind of BSM is some boost circuit (5 Volts to 8.5 Volts) with a constant current output.
I guess you could use a LED boost driver for that?

There’s always very little information in those listings…

2Q19

Jerommel
Jerommel's picture
Offline
Last seen: 5 months 2 weeks ago
Joined: 01/04/2014 - 13:18
Posts: 5721
Location: the Hague, Netherlands

Oh, another thing, regarding those USB charging boards with protection circuit.
I don’t understand this:
It seems to me that the outputs are mislabeled.
B+ and B- seem to be OUT+ and OUT- in reality.
And so OUT+ and OUT- seem to be B+ and B- in reality.
Because the OUT connections come straight from the TP4056, and usually go straight to the battery (on versions without protection).
And then comes the protection circuit to prevent discharge-over-current and over-discharging based on voltage.

I tried both options (according to the labels and according to my (possibly faulty) logic).
I found that it worked better according to my (possibly faulty) logic.
Unfortunately i can’t remember exactly what the problem was when you go by the labels, but i think i lost some voltage to the cell or it would take much longer to terminate charging.
Because the TP4056 stops charging when it sees 4.2 Volts over the output, but it doesn’t reach 4.2 Volts when there’s that protection circuit in between.
Something like that..

2Q19

Lexel
Lexel's picture
Offline
Last seen: 55 min 14 sec ago
Joined: 11/01/2016 - 08:00
Posts: 5487
Location: Germany

the main problem is that the 2S charge circuits with 5V boost to 2S simply charge the battery
and depend on the balancing BMS board of the battery pack

an example
https://www.maximintegrated.com/en/products/power/battery-management/MAX...

Jerommel
Jerommel's picture
Offline
Last seen: 5 months 2 weeks ago
Joined: 01/04/2014 - 13:18
Posts: 5721
Location: the Hague, Netherlands

Lexel wrote:
the main problem is that the 2S charge circuits with 5V boost to 2S simply charge the battery
and depend on the balancing BMS board of the battery pack

an example
https://www.maximintegrated.com/en/products/power/battery-management/MAX...

So you need both.

But then still, it’s only balanced when both cells are fully charged.
You often see this on cordless drills too.
They strongly advise you to charge the battery pack until it’s fully charged.

2Q19

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

There are plenty 2S BMS models: https://www.aliexpress.com/w/wholesale-2S-BMS.html?SortType=price_asc&SearchText=2S+BMS

540 hits I am seeing.

To charge a 2S battery, use a CC/CV DC/DC converter. They're cheap as chips (at least ≈6.5V of input required for constant current regulation):

Also, do not expect balancing boards to do jackshit below well above 4.2V/cell. 

 

Cheers Smile 

Enderman
Enderman's picture
Offline
Last seen: 1 day 6 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3965
Location: Vancouver, Canada

What does “terminal of the charger using high voltage device” even mean…

If those BMSs don’t shut off power to the battery when charging is completed then the DC DC converter will continuously ‘trickle charge’ the lithium cell at CV which is horrible.

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

Well i’m ditching the 3s idea also….
It’s just all really confusing.
Anyway i’m probably stick with the PT4065 and a disconnect circuit, not sure on the relais though…

I made a overview of the type of board that are out there.
This is partially guesses so not sure on everything Wink

PDF with overview of LIPO chargers:
http://realretro.nl/SharedPics/LIPO%20charging%20types%20of%20boards.pdf

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

Enderman wrote:

If those BMSs don't shut off power to the battery when charging is completed then the DC DC converter will continuously 'trickle charge' the lithium cell at CV which is horrible.

A BMS shuts off charging power to the battery if overvoltage condition is reached, this won't happen with a DC/DC module with carefully adjusted voltage output.

What do you mean with “trickle charge”?

A DC/DC converter tunes a fixed output voltage, technically this is not trickle charging as it can't overcharge a battery unless a too high output voltage is tuned. Battery voltage cannot raise above output voltage and current flow tapers exponentially to ridiculous levels after a while in CV mode. You're supposed to remove the battery from the charger once charging is finished anyway.

 

Cheers Smile 

HKJ
HKJ's picture
Offline
Last seen: 52 min 55 sec ago
Joined: 05/24/2011 - 12:23
Posts: 6980
Location: Copenhagen, Denmark
Barkuti wrote:

A DC/DC converter tunes a fixed output voltage, technically this is not trickle charging as it can’t overcharge a battery unless a too high output voltage is tuned. Battery voltage cannot raise above output voltage and current flow tapers exponentially to ridiculous levels after a while in CV mode. You’re supposed to remove the battery from the charger once charging is finished anyway.

This is definitely a trickle charge, a battery needs to settle a bit after being charged.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

I no offence guys but in the batteryuniversity this is stated:

“The absence of trickle charge further simplifies the charger. “
In below piece of text

Summary
Charging lithium-ion batteries is simpler than nickel-based systems. The charge circuit is straight forward; voltage and current limitations are easier to accommodate than analyzing complex voltage signatures, which change as the battery ages. The charge process can be intermittent, and Li-ion does not need saturation as is the case with lead acid. This offers a major advantage for renewable energy storage such as a solar panel and wind turbine, which cannot always fully charge the battery. The absence of trickle charge further simplifies the charger. Equalizing charger, as is required with lead acid, is not necessary with Li-ion.

Consumer and most industrial Li-ion chargers charge the battery fully. They do not offer adjustable end-of-charge voltages that would prolong the service life of Li-ion by lowering the end charge voltage and accepting a shorter runtime. Device manufacturers fear that such an option would complicate the charger. Exceptions are electric vehicles and satellites that avoid full charge to achieve long service life.

This is a copy from this website:
https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

Thanks for all your help and comments!
Cheers

HKJ
HKJ's picture
Offline
Last seen: 52 min 55 sec ago
Joined: 05/24/2011 - 12:23
Posts: 6980
Location: Copenhagen, Denmark

I do not hope that was a comment on my post.
With LiIon you charge to a specified voltage (Usual 4.20V) then you continue to charge until the current is dropped below a specified level (Termination current). When this happens you stop charging and let the voltage drop a bit.

The termination current it usual stated in the datasheet for batteries, continuing to charge will increase the wear on the battery. The problem is not using a different termination current that stated (100mA or 50mA is a minor detail), but not terminating at all.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

DavidEF
DavidEF's picture
Offline
Last seen: 9 hours 3 min ago
Joined: 06/05/2014 - 06:00
Posts: 7678
Location: Salisbury, North Carolina, USA

Do the TP4056 modules not terminate when they reach the end of the charge cycle?

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

HKJ
HKJ's picture
Offline
Last seen: 52 min 55 sec ago
Joined: 05/24/2011 - 12:23
Posts: 6980
Location: Copenhagen, Denmark
DavidEF wrote:
Do the TP4056 modules not terminate when they reach the end of the charge cycle?

The TP4056 is a well behaved LiIon charger and will terminate.
When used in this module it may not because the output power is drawn from a combination of battery and TP4056. It requires something more advanced it you want a good control in these circumstances.

My website with reviews of many chargers and batteries (More than 1000): https://lygte-info.dk/

snafky
Offline
Last seen: 10 months 2 weeks ago
Joined: 01/04/2019 - 04:06
Posts: 6

yes they do as far as i have read/seen online.
On this video is the proof ;) ;)

https://youtu.be/wfrm6lbt8Pc?t=681

Enderman
Enderman's picture
Offline
Last seen: 1 day 6 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3965
Location: Vancouver, Canada

Barkuti wrote:

A DC/DC converter tunes a fixed output voltage, technically this is not trickle charging as it can’t overcharge a battery unless a too high output voltage is tuned. Battery voltage cannot raise above output voltage and current flow tapers exponentially to ridiculous levels after a while in CV mode. You’re supposed to remove the battery from the charger once charging is finished anyway.


That is not what trickle charging means.
A lithium cell self-discharges, especially when it has just finished charging at 4.2V it will drop by several tens of millivolts.
By keeping a constant 4.2V attached to the battery, you are trickling energy into it continuously to keep it topped off at 4.2V.
This current fed into the battery to keep it ‘topped off’ is trickle charging, and this is what damages lithium cells.

All high quality chargers (even the TP4056 as mentioned in the posts above) will terminate charging, aka it send 0A and 0V to the battery after charging, it does NOT hold the battery at 4.2V.

Trickle charging the battery by holding it at a full charge voltage is common with SLA (and I think also NIMH), but should never be done to lithium cells.
You will also notice that phones and laptops will not begin charging until it is several percent below 100% charge, this is to prevent the charger from constantly feeding power into the battery, or having it cycle on and off as the battery voltage drops after charging and the charger thinking it needs to recharge again.

DavidEF
DavidEF's picture
Offline
Last seen: 9 hours 3 min ago
Joined: 06/05/2014 - 06:00
Posts: 7678
Location: Salisbury, North Carolina, USA

Just for curiosity, what about trickle charging at a lower voltage, say 4.0V or even 4.1V instead of “full charge”? Would it still be problematic? Is it the mere fact of keeping a steady voltage level that is the problem, or is there a lower level of voltage where it’s no longer an issue?

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

Enderman
Enderman's picture
Offline
Last seen: 1 day 6 hours ago
Joined: 11/03/2016 - 22:42
Posts: 3965
Location: Vancouver, Canada

DavidEF wrote:
Just for curiosity, what about trickle charging at a lower voltage, say 4.0V or even 4.1V instead of “full charge”? Would it still be problematic? Is it the mere fact of keeping a steady voltage level that is the problem, or is there a lower level of voltage where it’s no longer an issue?

I’m not sure what the self-discharge curve of a lithium cell would look like, but clearly near full charge it drops voltage much faster than at lower charge %.
There’s probably some point where the microamps being fed into the cell will cause minimal damage, but I’m not sure at what voltage or what current.
DavidEF
DavidEF's picture
Offline
Last seen: 9 hours 3 min ago
Joined: 06/05/2014 - 06:00
Posts: 7678
Location: Salisbury, North Carolina, USA

It would be interesting if somebody wanted to test it scientifically. Maybe using four identical cells, a test rig could be set up to check level of damage over time. For this test, the one “control” cell would be fully charged, then discharged to optimal storage voltage, kept for an amount of time, and the cycle repeated continuously over the duration of the test. The charge/discharge/store cycle would approximate “usage” of the cell. The three “testing” cells can be kept at a non-terminating voltage level, one cell at 4.2V, one at 4.1V, and one at 4.0V, and “vital signs” of all four cells checked regularly (maybe once a week?) over several weeks, or even months, to see what damage occurs. Anybody want to do it? Crazy

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

Barkuti
Barkuti's picture
Offline
Last seen: 2 hours 40 min ago
Joined: 02/19/2014 - 14:46
Posts: 4303
Location: Alhama de Murcia, Spain

0K people, so you are worried because of “trickle charging to whatever maximum voltage” which for li-ion happens to be 4.2/4.35V, mostly. This is a no issue for me with CC/CV modules. I've already built two dual voltage output custom chargers for power tools, consisting of XL4015 DC/DC modules with in series schottky diodes at the output for reverse polarity protection and a switch for changing the output voltage resistor divider value. For both setups I carefully measured voltage drop at the schottky for a current flow of around the balancing circuit resistor drop. With this in mind I adjusted maximum output voltage a little bit higher than Vmax plus Vschottky: 12.6V + 0.27V <= ≈12.93V for 3S. I did this because balancing circuits activate at a higher than maximum value, albeit in my experience it seems they may be doing it too high so I question their serviceability. For long lifespan Vout I tuned a variable resistor which could be engaged via switch, reducing Vout value close to 0.9V in the above example.

So trickle charge I fear not. Lots of reasons as you can infer from the above. Excluding NOR capacity tests on my Lii-500 I cannot recall when was the last time I fully charged a cell on purpose. Seriously. 

 

Cheers Party 

Pages