USB Battery Bank Prevent Shut Down

Status
Not open for further replies.
J

joepr

New member
Hello Everyone

It's has been a pleasure finding this forum. I have been searching on the Net for couple of weeks looking for a solution to my project. I will really appreciate if someone could point me to the right direction. Here's my scenario..

I have a GPS tracker that has a internal battery that could last for 3 days on normal operation. I will like to extend the battery life on the tracker since it will be on the field for several days. This small GPS tracker can be charged by connecting a regular 5v cell phone charger on its micro-usb in this case a Power Bank as you may see on this picture.

battery-bank.JPG

The problem I have is that as soon the internal battery its fully charged the battery bank turns it self OFF. The only way to turn the Battery Bank ON again its by removing the USB cable and plugging it back OR pressing a small button on the BBank. I have found a link on dorkbotpdx website http://www.dorkbotpdx.org/blog/paul/battery_pack_load that shows a way to prevent a battery bank to shut it self OFF when little power is consumed. I appreciate the time of Paul Stoffregen on sharing his diagram.

The model of this battery bank is Astro E1 5200mah. I wrote the manufacturer about the Auto Shut OFF and this is what they answer me

The Astro E1 you bought has minimum current requirement 30mA~70mA. That means if the output current is below this range, then, the battery won't work.

Since I'm no expert on electronics I will like to ask for advice. Does placing this circuit between the BBank and GPS Tracker will prevent the BBank to shut it self OFF? I know there will be power loss in the long run but at least the GPS tracker will have more days of work rather than using the internal battery.

I really appreciate your time of reading this thread.
 
Not All battery banks do that and they have different thresholds - I have some of each. Sounds like you need a dummy load to draw power to keep it alive. How long does it live before shutting down - like 10 seconds? You might get away with a dummy that pulses 70mA just short of that period for a second. You'd have to 'make' that - you could do it with a Teensy 3.1 or an LC in some fashion if that is what you were aiming for. The Teensy could stay alive and watch the clock and light some LEDS to get the current - or maybe warm some resistors?
 
Generally what you have to do with these battery packs is draw a pulse of current every so often. I don't remember the parameters, but perhaps draw current for 0.1 s every 5 s would work.

The Teensy itself can't easily do this alone -- taking 50 mA and 3.6 V as an example, you would need to connect a resistance of 3.6/0.05 = 72 ohm across the cell. You could use a (easy to get) transistor - e.g. 2N2222 and some resistors. You can't get 72 ohm resistors -- use 68 ohm or 62 ohm. Connect the 62 ohm from the power supply to the NPN's collector. Connect the emitter to GND, and connect the base via 1 kohm to a Teensy I/O port. Then drive the port as an output and low, with pulses high of 0.1 s every 5s.

Alternatively, you could just connect a 62 ohm between the Teensy's 3.3 V pin and a number of I/O pins together. Each pin is only rated at 25 mA (I think) -- so 3 together would be safe (actually you'll need 3.3 V/R=50 mA; ==> R = 66 ohm; use 62 ohm or lower). Connect those 3 I/Os together to the other end of the resistor, and drive the pins the same way (0.1 s on, 5 s off).

Note that if the I/O was on (low) 100 % of the time, you'd need a 1 W resistor, but since you are only applying small pulses, a normal 1/4 W should be OK. If you are concerned about this, instead of a single 62 ohm, use 4 240 ohm 1/4 W in parallel.

You'll need to check what parameters your battery pack really needs -- some experimentation may be needed.

Note that the
 
Pulsing a load onto the power bank would be the simple solution, but an alternative if you have more time to tweak would be to have a circuit wired to the 'activate output' button on the battery bank and turn that on say every 4 hours. GPS will draw enough on it's own while charging and everything will auto shut off when GPS is done. Done right this should be more efficient since the battery bank step up circuitry is only running while under a reasonable load. Critical parts are closing the button contacts via Teensy/Arduino/micro of your choice, and keeping the power draw of whatever you use low enough to actually come out ahead, since it will be drawing power from the GPS battery.

Do note that this 'charge battery bank battery to 3.7V, convert to 5V, GPS converts to it's own battery' is not going to be particularly efficient so the optimal solution is to upgrade the GPS battery, though this may require more engineering than you want in changes to the charge circuit for the larger capacity.

If you consider both devices 'sacrificial' and are in a hurry it is quite possible you can remove the battery from the GPS and connect the internal power bank battery in it's place (the 3.7V battery out, not the 'too clever' 5V step up converter), assuming they are both using the same battery chemistry. Doing so you are following advice from some amateur on the internet, and result may be destruction of both devices and a fire that gets worse when you put water on it so some reading first would be mandatory.
 
Thanks all! Its overwhelming getting so many points of views. I test this battery bank and saw it turn it self Off after 15 sec of inactivity. I also got today on my mail this little usb meter. My idea was to connect the "charger doctor device" and it will basically could tell the battery something was connected even if the GPS in the other end was fully charged. At least the battery will stay ON. Seems it do not consume the minimum of 30ma that the manufacturer recommends to leave the BBank ON. Do you think there is a way to modify this little gadget? It only cost me $1.50 us

charger-doctor1.JPGcharger-doctor2.JPGcharger-doctor3.JPG
 
Paul S investigated a USB battery pack with this issue a few years ago. Evidently the Teensy 3.0 was too efficient :). Paul posted a small circuit that every so often wakes up and does enough charge to keep the battery running: http://www.dorkbotpdx.org/blog/paul/battery_pack_load.

Since I play with neopixels a lot, I can imagine every so just light up a led or two at full power will use enough power to keep the USB charger running.
 
Last edited:
That's just a meter displaying current draw, and would have been designed to draw the minimum current so not going to help overmuch keeping the battery bank active. It's not going to have any smarts in it so not much to modify, though case is a handy size if you do go for the 'pulsed draw' method.
 
MichaelMeissner said:
Paul S investigated a USB battery pack with this issue a few years ago. Evidently the Teensy 3.0 was too efficient :). Paul posted a small circuit that every so often wakes up and does enough charge to keep the battery running: http://www.dorkbotpdx.org/blog/paul/battery_pack_load.

Since I play with neopixels a lot, I can imagine every so just light up a led or two at full power will use enough power to keep the USB charger running.
Click to expand...

Thanks MichaelMeissner that same link tutorial was my first stop to get into this forum. I saw Paul blog and it seems to be the solution to my problem. I will like to ask.... If I will like to make the circuit how much approximate will cost. For this project we will start with 50 GPS units and Battery Banks.

If this helps someone else I find out a USB Battery Bank that has a "Always ON" Feature already embed . The brand is Voltaic V15 you can search on the web. My concern now is if it will be cost effective buying a Battery with this feature already or designing my Own. My concern is that the End User who will be in charge of working with the equipment will be a regular employee and most of the time not TechSavvy.
 
defragster said:
If you have a choice and want 50 - maybe get a different pack that just stays on as noted

http://www.newegg.com/Product/Product.aspx?Item=9SIA4TZ2J29346

This is same brand as my 3Ah SCUD HOBBIT version I got for <$10 at NewEgg - it shows a similar push switch that keeps mine on - with nothing or just a Teensy blinking.

It is here at Amazon for $19 - doesn't declare pushbutton as "POWER ON" like mine: http://www.amazon.com/SCUD-PC-6000E-GD-Ultra-Slim-Portable-Champagne/dp/B00R756SU2
Click to expand...

Greetings defragster I don't mind at all getting other BBank. If it has a option to always stays ON it will suit my needs. So you have one of this http://www.newegg.com/Product/Product.aspx?Item=9SIA4TZ2J29346
And it stays always ON ?
 
The two overt links are to the same 6Ah model - mine was the out of stock 3Ah. Mine has a single LED - the larger units have a similar looking pushbutton - and multi LED's so the design is changed and the button may act differently - you'd have to buy one to see. Mine is a non-latching button - but once powered it stays on until low power shutdown. If the 6Ah one works the same you'd be set.
 
Attempting to do a capacity test i ran into a "Doshin" battery bank tripping out at 1A resistive load. the power bank is doing some sort of check. Its rated at 2.1A.
adding a 220 uF capacitor across the USB connector fixed the problem. That suggests some sort of pulsing is going on.
There are a lot of decidedly over specified power banks being sold, particularly on ebay, so a discharge test is recommended if the application is in any way critical.
 
I'm dealing with the same problem, I want a battery bank to keep my Trackimo charged until the battery bank is exhausted.
With all the switchless battery banks I've tried, they start charging again if I unplug & plug back in the USB cord.

If I could find a circuit that would replicate that effect, disconnecting & reconnecting the USB every X hours, I'm thinking that would make most every auto-on battery bank work. Also, wouldn't it be more power efficient?

Anyone know if there is a circuit I could fab up or a device that I could buy that would do that?
 
OfficeGlenn said:
I'm dealing with the same problem, I want a battery bank to keep my Trackimo charged until the battery bank is exhausted.
With all the switchless battery banks I've tried, they start charging again if I unplug & plug back in the USB cord.

If I could find a circuit that would replicate that effect, disconnecting & reconnecting the USB every X hours, I'm thinking that would make most every auto-on battery bank work. Also, wouldn't it be more power efficient?

Anyone know if there is a circuit I could fab up or a device that I could buy that would do that?
Click to expand...

The link in post #12 should show some history on this - including that I came across a brand that typically stays on - but also ( on some units ) will both output power and charge up the unit without dropping the output voltage at all. YMMV may vary and I have not gotten any recently I got pairs of a couple models that are still in service. Those may not be available or the design of the packs may have changed.
 
Hello guys. I was searching for information because my power bank was closing down when I use it with a breadbord and basic circuits. And this thread is always on top. So I thought it maybe useful for people. What I did was to keep powerbank see there's a mA value bigger than 50. This is the limit. If you cant reach that it shuts itself down.

I used circuits.io to make a simple way to pass that limit with what I have in my hands and simulate its mA level. So I used 4 x 330ohm resistors to get 50+ mA. Tested it for 30 minutes and it didn't shut down. For your battery issue you may need to find a way to add these resistors at the beginning. And I'm not sure if its safe to force a battery to overload. But this is the circuit...

I don't know much about resistors. So they may heat up but It looks fine so far.

Adsız.jpg
Adsız2.jpg
untitled.png
untitled2.jpg
untitled3.jpg
 
blackleon, your circuit works, the design posted above trades increased complexity for lower wasted power by only pulsing having the resistors in circuit for short periods, given anything going into the resistors is wasted in terms of maximising battery life.
 
Status
Not open for further replies.
Back
Top