2
\$\begingroup\$

I just got my custom PCB in the post from PCBWAY.

Plugged in small battery I got with adafruit only 2000mah at 3.7V (PKCELL LP803860) and showed the LED light to show it was being powered so I thought great its working.

However upon then connecting a battery I ordered off of alibaba. (EWT LP10048166) same voltage at 3.7V and larger size at 20AH (20000mah)

My JST 2-PH connector starts smoking.

Anyone know why this is happening as I would of thought this of worked out fine with same voltage and larger battery size.

The first part of the board which is dealing with the battery is in simple terms adafruit's 1000C Powerboost which I've added into my design. So its designed to take in 3.7V and then convert that to 5V which is the majority voltage that the components on my board require. The purpose of the PCB in conjunction with a Pi camera v2 module, Compute module 3, 3 ultrasonic sensors & a cree LED is to act as a visual awareness system for the visually impaired so the reason I'm using a 20Ah battery is I require the system have enough power to maintain up to 10 hours of use per day.

Didn't realise the JST connector is rated up to 2 Amps perhaps that might be the reason for some smoking if a 20 amp battery is connected directly to it. Was speaking to a friend of mine and she suggested it could be some of the components on the board might have their resistance change due to the higher amp current. Didn't need to worry about + & - as the connector only allows one way for the pins to meet.

Here's a picture of the larger battery

enter image description here

In Response to Daniel Tork here is the schematic for the affected section of the board.

enter image description here

And here is from board view in Eagle enter image description here

Followed exactly what MartinF posted in his new edit and that fixed the problem. Turns out the polarity was reversed on the battery's JST connector. With care managed to pop them out and switch them around. Inserted into the board and the LED on the board went on and no smoke.

\$\endgroup\$
  • 5
    \$\begingroup\$ Switched + and -? \$\endgroup\$ – Alexander von Wernherr Aug 6 '18 at 10:34
  • \$\begingroup\$ They likely designed the circuit taking the impedance of the battery into account, such that it acts as a series resistor, limiting the current. Changing the impedance of the battery lowers this resistor such your circuit is able to take more current (If it so desires) which in this case I would assume so. I think you may have been nuked by non-limited in-rush current or your LEDs sinking too much current. \$\endgroup\$ – user160063 Aug 6 '18 at 11:48
  • \$\begingroup\$ Okay cheers for this information. I know that the board still works as after pulling out the battery causing the issue I inserted a previous 2ah battery and that worked a charm. \$\endgroup\$ – Dragonfly3r Aug 6 '18 at 11:53
  • \$\begingroup\$ @sidA30: That doesn't sound plausible to me. Inrush current is usually only a spike, especially since there are no motors or that sort of stuff. Smoke only comes after a while, not because of spikes. Also the Adafruit website suggests that the 1000C board is able to handle bigger batteries. \$\endgroup\$ – MartinF Aug 6 '18 at 11:59
  • 1
    \$\begingroup\$ Would be better if you could actually edit the post instead of tacking on numerous "Edit 1:, Edit 2:" etc. \$\endgroup\$ – pipe Aug 6 '18 at 13:37
2
\$\begingroup\$

A few thoughts:

  • If your board was designed to work with 3.7V in the first place, then connecting the larger capacity battery should not be a problem since both batteries are 3,7V. Was your board designed for 3,7V?
  • As @Alexander already commented, check if you accidently switched the + and -
  • The 'JST' connector is rated upto 2Amp, so quite some current is flowing to make it smoke. Is there a fuse on the board? Is there a reverse protection diode on the board? Did you try to power up the board again with the smaller capacity battery?
  • Since specs on websites like Alibaba are not the most reliable, you could check the voltage coming out of your battery, and check if it's really (close to) 3,7V. I couldn't find the battery number you posted. There are lots of batteries called 'AWT', but then still no results for that number. (20AH is huge btw.)

Edit: Proof that you have the polarity reversed!

So I took a close look at your battery, and at the 1000C module and at the JST 2-PH connector.

JST 2-PH

This is a closeup of a generic JST 2-PH connector. Don't mind the colors of the wires as this is just a picture from the Internet. What's important is: You can clearly see that when the 'springs' that hold the contacts in the connector are on top, the '|' and two 'bumps' that hold the connector in the receptacle are on the bottom.

Adafruit 1000C

This is a closeup of the Adafruit 1000C module. The '|' and the bumps are on top, so the 'springs' must be on the bottom. With the '|' on top, the + is on the right. Put the '|' on the bottom, and thus the 'springs' on top, the + would be on the LEFT.

Large battery connector

Now here's a closeup of the picture you posted in your question. The 'springs' are clearly on top, and you can see on the bottom there is one of the 'bumps', that are on the same side of the '|'. So the '|' is on the bottom. 'Springs' on top, + should be on the LEFT, but is on the RIGHT!

So your polarity is indeed reversed! You can carefully lift the 'springs' and take the contacts out. Then you can put them back in the right place. Be careful not to bend the 'springs' too far, as they are only plastic, and easly break off. Also make sure you don't accidentally short-circuit the battery when both contacts are out.

Hope this helps!

Update: I just saw they actually have a warning on their website for third party batteries with the polarity reversed!

How do you intend to charge the battery? Via the 1000C? Because that module can only charge with 1000mA max., so charging your battery (when completely discharged) will take more than 24 hours. I don't even know if you can charge a Li-Ion/Li-Po battery with less than 0.5C. But that's a completely different topic anyway...

\$\endgroup\$
  • \$\begingroup\$ It also depends on the application. A 20Ah battery can provide a lot more power than a tiny 2Ah cell. If you switch a motor or smth. you can easily reach some kA as SCD current. \$\endgroup\$ – Alexander von Wernherr Aug 6 '18 at 11:04
  • \$\begingroup\$ You're right. Since OP didn't mention anything but a simple LED, I didn't assume big consumers. I wonder what his board is supposed to do with 20AH of battery capacity... What do you mean with SCD current? Do you mean inrush current? \$\endgroup\$ – MartinF Aug 6 '18 at 11:17
  • \$\begingroup\$ Yeah, inrush current. Im stuck with TI terminology and there it's called short circuit discharge current \$\endgroup\$ – Alexander von Wernherr Aug 6 '18 at 11:18
  • 1
    \$\begingroup\$ @MarcBrooks Please add this info about your application to your question \$\endgroup\$ – Alexander von Wernherr Aug 6 '18 at 11:26
  • 1
    \$\begingroup\$ @MartinF It was exactly as you said the polarity of the cables was the opposite way around. With a little care managed to pop out the connectors and switch opposite way and that worked. Cheers for sticking it out to the end with me over this. \$\endgroup\$ – Dragonfly3r Aug 6 '18 at 20:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.