Really cheap true cc/cv usb charging board

[czAtlantis]

Hi, I tried to draw two different ways to connect your Ammeter. The first one is like I said - every A-metter must have internal resistance (in this case 0,2ohm) According to ohm's law it means therewill be voltage drop. In our case at 1A it will be 0,2V.
When you connect it between USB and input there will be only 4,8V V on the board. But it is OK because this board is stabilizer - keeps 4,2V on output (when the input voltage is in acceptable range like 4.7-6V).

On the other hand the second schematic shows connecting A-metter between board and battery. There is the same voltage drop but in this case it is a problem. The board doesn't know there is any voltage drop and stabilises still to 4,2V. But the A-mettr adds again 0,2V drop but there is no way to compensate ->On battery is only 4,0V which will prolong charging time and you battery will not be fully charged.

PS: there is little parasitic drain of the circuit itself - In the picture I showed it as 1A input and 0,99A output but this is really not significant. Just your A-metter connected between power source and board won't show 0,00mA even with fully charged battery. I just measured it is just around 5mA

[RoGuE_StreaK]

The issue is probably where to shoe-horn in the ammeter on a board using the inbuilt USB socket; if you look at the pic posted by AmperSand on the first page (I don't have my board with me to look at) I would say to remove the "R400" resistor and connect the ammeter inline there, obviously somehow re-integrating the R400. Or you could maybe cut the thick track between the USB and R400, scrape back the soldermask, and solder the ammeter between this newly created pad and the R400 input pad.

Possibly you could "tombstone" R400, and use it's exposed end as a pad?

[ToddC]

Lolaralph : What do you plan on using as the ammeter? Maybe do what RoGuE_StreaK suggested but install a current sense resistor and attach a volt meter across that?


BTW - The voltage drop of the current measurement that czAtlantis is explaining is called "burden voltage."

[Lolaralph]

czAtlantis,
thank you very much for the time and effort of the schematics, it clears things up. I probably really don't need to use the ammeter but I want to check the changes the replacement resistors have made, and be sure I'm only sending 100 ma to my 10440's.

ToddC,

I actually purchased one of the "slimline ammeters" that e-worldseller sells, orderd it with my second batch of boards.

[march.brown]

When the CC stage changes to the CV stage , I assume that the current drops to maybe 100mA or less (?) ... If the Ammeter is in series with the battery , then the voltage drop across the ammeter will be one tenth of when it was one amp ... According to previous calculations that would only be 0.02V drop across the ammeter ... I am assuming that the resistance of the ammeter remains constant irrespective of the ampere readings ... The battery would be charged up to within 0.02 volts of the correct terminating voltage ... So the battery voltage would be 20mV lower than the optimum 4.2V ... There will be some variation in terminating voltage of these little boards , so ideally we should be looking for a board that is terminating at 20mV over the 4.2V ... This would compensate for the inclusion of the ammeter into the output circuit.
.

[bshanahan14rulz]

can you just measure the voltage dropped by the known resistor that is already inline with the USB power line? Then you don't have to guess or find out the resistance of your ammeter, but also then it depends on the tolerance of the particular resistor. After that, V/R=I, and 1A=1000mA.

[NoixPecan]

cottonpickers uses the same IC in his USB chargers, and they work fantastically.

Not quite. Looking for an answer to the question in my previous post, I have read Cottonpickers' first thread about his USB chargers. In fact, his first design was based on the TP4054, but after selling a few chargers, he realized that the IC poorly implements (if at all) the CC/CV algorithm (despite what is specified in the datasheet). He ended up upgrading his design, by unsoldering the TP4054 from the charger board, and replacing it by the LTC4054. This IC from Linear (which the Chinese TP4054 tries to imitate) follows the CC/CV algorithm almost perfectly.

Since the charging board we are now discussing is based on the TP4056, a question rises. Does the TP4056 follow the CC/CV algorithm any better than the TP4054? Obviously, we cannot trust the datasheet. Should one of us own a logging multimeter, it would be a good idea to check how the board actually behaves.

[RoGuE_StreaK]

unsoldering the TP4054 from the charger board, and replacing it by the LTC4054. This IC from Linear (which the Chinese TP4054 tries to imitate)

Out of curiousity, and following that logic, I looked up "LTC" 4056, and yes that one also does exist (probably no surprise)
Pricing-wise, a brief search gave me about $2.20 for the LTC4054 (1-99 orders), as opposed to maybe $0.20 for the TP4054. LTC4056 is about $2.15, couldn't find reliable pricing for the TP4056, cheapest was about $0.50 each.
Just in case people want to make their own!

[malow]

DealExtreme are selling now a 1A TP4056 based charging board. mini-usb, 2 status leds. $4.80

[NoixPecan]

@malow. Indeed, DX sells the TP4056 board this thread is all about. It is currently much cheaper on eBay, though (only $2).

[Norm]

Anyone familiar with TP4057, I found a site that has boards, 4.2V lithium battery charge control board lights double TP4057 lithium battery 1A

These boards uses two chips.
There is also a cheaper version using one chip for 500mA
Not sure if the current is variable.

Quantity (lot) Price (Per lot) Processing Time
1 - 3 ---------- $3.46 ------------4 Days
4 - 6 ---------- $2.01----------- 4 Days
7 - 10 --------- $1.80----------- 4 Days
11 - 15--------- $1.70----------- 4 Days



Data

Norm

[malow]

TP4056 have 4v-8v "absolute maximum" input voltage, while TP4057 is 4.5v-15v. also, theres a 8.4v TP4057, for 2 li-ion in series charging.

[bstrickler]

Well, I just bought 10 of these! (I need 7 for my 64656 build, running off of 7 26650's. 3 others just for spares/other stuff). Also bought 7 cables, and a powered 7-port USB hub (3 amp power supply), so I can charge the cells overnight, rather than having to charge one at a time with my hobby charger (positive node isn't magnetic, so I can't just use mangets to attach the leads). Total cost: ~$50 (not including figuring out how to make cradles)

Buying a 5k roll of 3k resistors (otherwise it becomes too much of a waste of $), in case anyone wants some. I'll hack 10 off and mail it to you for 75 cents PPG (45 cents to cover stamp, 30 to cover envelope and time).

[mvyrmnd]

A simpler design would be to build a 7 bay cradle, run 3 of the chargers at the full 1A each with the outputs wired in parallel across all 7 bays.

You'll waste less input power, and charge just as quickly - kinda like a giant version of my dodgy 2-up charger:

[bstrickler]

A simpler design would be to build a 7 bay cradle, run 3 of the chargers at the full 1A each with the outputs wired in parallel across all 7 bays.

You'll waste less input power, and charge just as quickly - kinda like a giant version of my dodgy 2-up charger:

Only problem is the USB hub is only designed for 500ma/channel. Fortunately, I can still cancel the order.

With your idea, I'll probably just get 2 4A wall wart chargers, and make a 7-cell 26650 bay like yours (how did you do that, by the way?)

Think it would be beneficial if I thermal epoxied a small slug of aluminum to the IC?

[mvyrmnd]

That's a 4D carrier with the guts stripped out, a piece of MDF for the top, a DC jack at the top running to both boards in parallel - I run it off a 5V 2A wall wart - then the two outputs are paralleled to give a 2A output and will balance charge the two 32650's that I stick in there.

In a pinch I can still plug into one of the USB ports and slow charge.

Oh, and a metric buttload of hot glue

For yours, since all 7 cells will be used in series it will be beneficial to charge them in a big parallel batch, to make sure all the voltages are equalized.

I'd wire a 4A brick in parallel to 3 charging boards each, then all 6 outputs linked in parallel. That way you have 6A across 7 cells, and two mains inputs with a little bit of overhead. It also will give you a lot of output flexibility. You can put all 7 cells in and charge at 420 or 840mA per cell, or even load two cells and charge them at 1.5A each (with only one mains input turned on). If you set it up so that 18650's will sit comfortably, then you'll have a very powerful and flexible charger.

The IC's will get hot if there's a log of voltage being shed ie. your batteries ate at 3V and the input is 5V - they'll get toasty. As the battery voltage increases they'll cool off.

[bstrickler]

Thanks for the info

I decided to get a 380 watt MicroATX power supply instead of 2 4a wall warts, since it's going to run the same price after shipping (~$25), and I'll be able to get the full capabilities out of my B6, while still charging the 26650's via the other method (would've used the B6 for everything, but it only charges up to 6S Li-Ion, and I wasn't gonna spend a fortune on a charger).

~Brian

[march.brown]

Just been looking on Ebay at 5V switch mode power supplies ... There are 2 amp (10 watts) ones for only $9-99 (£6-39) including postage ... There are some very good ones up to 5 amps and even higher for about £10 or so ... I might consider something like this.

It would be possible to feed several of these chargers from one PSU.
.

[march.brown]

Just been looking on Ebay at 5V switch mode power supplies ... There are 2 amp (10 watts) ones for only $9-99 (£6-39) including postage ... There are some very good ones up to 5 amps and even higher for about £10 or so ... I might consider something like this.

It would be possible to feed several of these chargers from one PSU.

I have just sent for a switched-mode power supply off ebay ... It is 5V 4A (20W) ... Input voltage 100 to 264 Vac and the size is only 96mm X 58mm X 25mm ... It says that ripple and noise is less than 80 millivolts peak to peak ... I don't know how this compares with other 5V supplies ... It was from marcmart and it was only £5-19 including postage ... The intention is to use this to power two of the charging boards.

I have got to figure out the best way to connect the PSU to the charging boards ... I might possibly make up two mini USB cords and simply attatch them to the screw terminals on the PSU ... On the other hand I might add a couple of USB sockets to the PSU ... As they are coming from the Far East to the UK , I have about a month to think about it ... If needs be , I can use one of my wall-chargers to power a single board if the PSU comes before I have sussed out the 5V distribution ... I don't suppose that there is a set of ganged USB sockets available which could be hard-wired to the PSU ... This would be neater ... Decisions , decisions.
.

[wshyang]

Not quite. Looking for an answer to the question in my previous post, I have read Cottonpickers' first thread about his USB chargers. In fact, his first design was based on the TP4054, but after selling a few chargers, he realized that the IC poorly implements (if at all) the CC/CV algorithm (despite what is specified in the datasheet). He ended up upgrading his design, by unsoldering the TP4054 from the charger board, and replacing it by the LTC4054. This IC from Linear (which the Chinese TP4054 tries to imitate) follows the CC/CV algorithm almost perfectly.

Since the charging board we are now discussing is based on the TP4056, a question rises. Does the TP4056 follow the CC/CV algorithm any better than the TP4054? Obviously, we cannot trust the datasheet. Should one of us own a logging multimeter, it would be a good idea to check how the board actually behaves.

Are the 2 ICs pin compatible?

[RoGuE_StreaK]

Are the 2 ICs pin compatible?

From what I could see from the spec sheets, yes in both cases (4054 and 4056). Will double-check exact dimensions later today.

[Lolaralph]

From what I could see from the spec sheets, yes in both cases (4054 and 4056). Will double-check exact dimensions later today.

I have some of each of the chargers discussed, the original board uses the 4056 which has 8 pins, the smaller board uses the 4054 and it only has 5 pins, hope this helps

[march.brown]

Do you connect the supply first then put the battery in OR put the battery in first then put the supply in ?

My 1000mA HTC wall charger is inputting 5.11V on a thickish cable ... The 850mA wall charger is inputting 4.97V on a thin cable , so I've sent for a couple of the thicker cables and another HTC 1000mA charger ... Didn't try swopping the cables ... These new cables and charger will do till my 4A 5V switched mode PSU arrives from the Far East.

I'm not certain how long the charge should take using these modules on a 2400mAh cell.

I have been thinking of getting some more (better) 18650 batteries , possibly Senybor or Yezl or Xtar ... The price is right ... I can now probably justify getting some new batteries as my Wife has just bought four new dresses today that she didn't really need (birthday is later this month) ... I probably need four new (protected) 18650 batteries more than she needed the dresses (personally speaking) ... And my purchase would be a lot cheaper than hers !

Just noticed (after an hour) that one of the chargers blue LED is flicking on occasionally and the voltage is 4.19V ... That's the one with the thin cable ... The other is still charging and at 4.13V ... Blue flicking is getting quicker and the volts are flicking from 4.19V to 4.20V ... Got fed up and took the unprotected Ultrafire cells out.

Tried with my protected Ultrafires and a blue light still doesn't flicker even though the volts are up to 4.19V and 4.20V.

Can't understand why the protected cells are apparently acting better than the unprotected cells ... Perhaps it's age related ... The cells age not mine !

Gotta go now unfortunately , the dog's needs are greater than mine ... Will have another play tomorrow with more protected batteries.

Gotta get some better 18650's , I've made my mind up as long as she doesn't find out ... Wives have different priorities to men ... i've noticed this with all of my Wives ... I mean sequentially not concurrently ! ... Concurrently is illegal in the UK.
.

[march.brown]

Found a small problem with the double battery holders.

The positive and negative connections are crimped with sort of rivets and are not a very good connection ... Couldn't understand why a bit of wire was open circuit on the positive end till I checked the rivets ... Rather than hammer the rivets to make a better connection , I will do it a different way with my next tries ... I have sent for some more of the double holders and will make some better (straight through) connections next time.

When my 4 Amp 5V PSU arrives , I will fit four USB sockets to it so I can use four one amp chargers if or when necessary ... Whilst this construction job is very interesting , I could have bought another charger for what I will be eventually spending ... I don't actually need any more Li-Ion chargers as I already have two double slot for 18650's and a four slot for the shorter ones.

Just a shame that items take a month to get to the UK from the Far East ... It makes the construction job take a long , long time.

Does the Postal system in the Far East cost peanuts ? ... Everything (plus postage) is really cheap ... How can they send five double battery holders to the UK for less than £4 including postage ?
.

[Norm]

What do you guys think of this?

I bought a couple to see what they were like.

Pros : finished product, cost $8.95
Cons : my longer 18650s are a firm fit (keepower 3100mA)

Description:

Charging current is 500mA when connected to a computer USB port.

The charging current is 1000mA when it's connected through an adapter to the mains socket.

This device is portable and can be carried with ease. Provided that the charger has a 18650 or 26650 battery in the cradle, it can charge devices that can be charged by an USB connection, like cellphones, MP3 players, GPS devices and so on.

Double TP4057 IC for battery charging
Single PT1301 IC for charging devices
Charging termination 4.2V, no trickle charge.
PC-ABS flame retardant material
The output to other devices will cut-off at around 2.9-3V
Output current to other devices : from 500mA to 700mA, generally 600mA
Low voltage and over-discharge protection.
Reverse polarity protection.
LED light description:
Red light: Charging
Green light: Charging done
Low Blue light: Ready to output to other devices. No Blue light means low voltage and it will not output to other devices.
Intense Blue light: Charging other devices (For some Nokia cellphones the Blue LED will actually flash, it is normal)
Green & Red light flashing: Charger inputting current but battery not making contact or battery not detected

Additional aluminum spacers can be used to charge 14500, 18500, 17335 and 16340 cells.

Click thumbs for a larger image

[Lynx_Arc]

would be nicer if there was a cover over the battery so you could carry it around without worrying about the battery falling out

[mvyrmnd]

I think we might be teetering in the brink of an era of cheap chargers that won't risk blowing up...

[Norm]

would be nicer if there was a cover over the battery so you could carry it around without worrying about the battery falling out

A rubber band might be OK, there are no exposed contacts when the battery in inserted, the spring as shown above is compressed totally back into the body.

Norm

[NoixPecan]

Pros : finished product

Is it not a cons? I mean, this thread is mostly about DIY.
From what I have read around the web, notably on BudgetLightForum, the ML-102 charger seems praised by its owners. Good cheap stuff. It certainly deserves a dedicated thread, here on CPF. Maybe you could start one?

[Lynx_Arc]

A rubber band might be OK, there are no exposed contacts when the battery in inserted, the spring as shown above is compressed totally back into the body.

Norm

Well the way it is I wouldn't want to give one to someone as a gift to charge their cellphone if they weren't a little battery savvy for fear they would not be careful with it.