PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2018-12-10, 01:59:44
News: Forum TIP:
The SHOUT BOX deletes messages after 3 hours. It is NOT meant to have lengthy conversations in. Use the Chat feature instead.

Pages: [1]
Author Topic: Powerbanks and minimum loads ?  (Read 1181 times)
Group: Moderator
Hero Member
*****

Posts: 1338
... .-.. .. -.. . .-.
A general question..looking for the best route to solve it.
USB/5V powerbanks are a great idea for charging cellphones or running medium wattage LED's while camping etc. However, there seems to be a minimum output load required, else they shut off after approx 15 seconds.

The one i'm struggling with, uses a TP4333 (functionally very similar to the common TP4056 and clones), which can charge an 18650 battery well to 4.2V and also handle up to 800mA output too. 1 chip + an inductor and a couple of LED's, caps and resistors. Elegant, in it's own Chinese way. I bought 5 'kits', with the circuit, battery contacts and plastic cases.
https://www.ebay.com/itm/H1-Portable-Power-Bank-USB-18650-External-Backup-Battery-with-Charger-Key-Chai/282434770730

But, the whole problem is that the devices i'm trying to power don't draw enough current !
If you plug say an Arduino into a chargebank, it will run for a few seconds and then everything turns off.
Minimums appear to be around 65mA, though that is just a regularly written figure on various websites and videos.
Here's the only datasheet I could find: http://www.wendangku.net/doc/f3cc639e48d7c1c709a145a4.html
It's either a lot newer than the TP4056 or just not as often used.
Have tried connecting Pin 5 (SWT) to Ground and the circuit did fire up, but went off again a few seconds later, even if the 10K resistor was still in place.
I'd perhaps be best to put a switch and a resistor across the output rails for around 50mA short circuit type of load, then the actual load would send it above its switch off minimum.
Is there a workaround ? it seems very wasteful and the batt won't run the device as long as it could.
 


---------------------------
ʎɐqǝ from pɹɐoqʎǝʞ a ʎnq ɹǝʌǝu
   
Sr. Member
****

Posts: 253
Hi Slider,

Unfortunately I have not found a data sheet on the TP4333 either.  So I 'blindly' suggest to put at least a 10 or 22 uF capacitor in paralell with the 1 uF shown as C3 between pin VDD and the negative rail, if it is indeed 1 uF only.  Maybe it influences the ON time you find to be lasting for a few seconds.  This a guess only.

Another suggestion would be: what if you made the connection of Pin 5 to Ground automated by an oscillator whose output would pull down Pin 5 periodically?  CMOS RC oscillators draw microamps for their own operation and could pull down Pin 5 either directly or via a suitable value series resistor.  74HC14 Schmidt trigger for instance needs a capacitor and a single (some MOhm) resistor to oscillate in the some Hz range, here is a pdf on such from a random search, in page 5:  http://users.ics.forth.gr/~kateveni/120/10f/extra03_osc.pdf
Hopefully the 3-4 Hz oscillations introduced to Pin 5 would not disturb the normal operation of TP4333.

Gyula
   
Group: Moderator
Hero Member
*****

Posts: 1338
... .-.. .. -.. . .-.
Thanks Gyula, that pdf was an interesting read.
About 10 years back I started building a 4 oscillator music synth and used somewhat similar techniques..didn't really know what I was doing but figured out that some logic chips would oscillate. Below is a pic of 'Big Blue' named after the large cap on the right. The 3 chips on the left were the sound source, including a damaged TC4011BP.
The project stopped when all I could make out of the thing were sounds like Scottish bagpipes  C.C

Periodic retriggering makes sense, it does seem to be a power button on the Chinese diagram.
Will report back with any findings.

Btw, these TP4333 chips are a wonderwaffen for old dead 18650's. The 3 that were charged yesterday are ex 1990's laptop cells and 2 of them started out at around 0.3V. My decent RC hobbygrade charger would beep a none connection or error about being too low to do anything with them, or would shoot to 4.2V and beep away saying they were charged after about 2 minutes. However, all 3 TP4333 chips took around 3hrs to slowly and carefully bring them to approx 4.14V. This morning, they've settled to > 4V on them and that's quite the result.



---------------------------
ʎɐqǝ from pɹɐoqʎǝʞ a ʎnq ɹǝʌǝu
   
Sr. Member
****

Posts: 253
Probably a blinking LED and a suitable value resistor would also work. The resistor would be between Pin 5 and the negative rail (gnd) and the LED would be between BAT input and Pin 5.  Whenever the LED is OFF the resistor would pull down Pin 5 and whenever the LED blinks ON the voltage on Pin 5 will be higher than zero,  this latter would probably be indifferent because there should be a certain voltage, >> 0, at Pin 5 wrt the gnd anyway, you can check this of course).  OF course the blinking LED should be able to work in the BAT voltage level range versus the gnd with the chosen series resistor in itself as a normal blinking LED.  It would not matter much if the blinking comes  say in every 1 second, the operation of the charging process on the long run would become disturbed only negligibly I think. 

EDIT:  Now that I found a data sheet, the blinking LED should be connected to the VDD  where I wrote the BAT pin above if this suggestion is apllicable at all in this circuit.

Gyula
« Last Edit: 2018-04-06, 15:38:14 by gyula »
   
Sr. Member
****

Posts: 253
Well I found the detailed data sheet here:   https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwj4zdfq5aXaAhXhQZoKHTSGDjMQFggmMAA&url=https%3A%2F%2Fumnaya-elektronika.ru%2Findex.php%3Fdispatch%3Dattachments.getfile%26attachment_id%3D28&usg=AOvVaw3ZzOfwfKI3l6XxaX4HmIAz 

and after download, the characters from the PDF file can be copied and pasted into the google translator.   :)

So here is what Pin 5 is, there are 3 different lines that include characters SWT, these are their translations: 

Flashlight output
SWT can drive LED light for flashlight illumination, maximum drive current
For 30mA, LED series resistance can be used to reduce the power of the indicator flashlight
Stream, SWT is also a button switch pin, if long press S1 button 2S, flashlight
When the canister is opened, press and hold the S1 key again. The 2S flashlight output is turned off.

Connect the button and flashlight LED, short press the button to display the power, long press the button 2S flashlight to open or close

SWT button and flashlight LED light, short press the button to display power, long press the button 2S flashlight to open or close.

All the Chinese text can be translated by copy and paste but I miss from the data sheet the minimum load current you mention as around 65 mA ? Strange it is not covered or I did not find. 

EDIT  This means my blinking LED or periodic switching on and off suggestions should be revised...   :D

Gyula
   
Group: Moderator
Hero Member
*****

Posts: 1338
... .-.. .. -.. . .-.
Oh that's cool and thanks very much for the translations  O0

In the main table, half way down the datasheet, is the following:
ITRK 100mA - which seems to be the 'keep alive' mA
*Edit - maybe a nope, translated it comes out as "The trickle-charging threshold voltage"

Unfortunately i've no blinking LED's, but there is now a temporary solution. A white LED and 150ohm resistor across the power rails. The draw is about 20mA and makes the Arduino project stay on. A nice bright power light for a finished project would make use of such waste and, I mean, that's exactly what the thing is doing and showing.


 
« Last Edit: 2018-04-06, 19:23:05 by Slider2732 »


---------------------------
ʎɐqǝ from pɹɐoqʎǝʞ a ʎnq ɹǝʌǝu
   
Pages: [1]
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2018-12-10, 01:59:44