1
\$\begingroup\$

In an assembly, I'm using spring loaded pins to connect a battery to a circuit board. It's possible from dropping the assembly that power may become disconnected for milliseconds at a time from the shock of dropping it. I have a capacitor bank that holds a charge for 30+ms when the positive node of the battery is disconnected. The problem is I don't think the capacitors will have any affect if the negative battery terminal is disconnected, or both at the same time because there will not be a complete circuit. Is my logic flawed? Is there a better way to do this than using a capacitor bank?

\$\endgroup\$
3
  • 5
    \$\begingroup\$ As long as the capacitor bank has it's ground still connected to the circuit when the battery's negative terminal is disconnected it should work the same. \$\endgroup\$
    – ACD
    Aug 12, 2014 at 15:53
  • 4
    \$\begingroup\$ A diagram showing how the capacitors and battery connect to the circuit board would be very helpful in answering this question. \$\endgroup\$
    – Adam Head
    Aug 12, 2014 at 15:56
  • \$\begingroup\$ The cap bank is connected in parallel with the battery contacts. Battery is connected to a daughter board with spring pins that touch the main board where the capacitor bank is. \$\endgroup\$ Aug 14, 2014 at 15:08

3 Answers 3

1
\$\begingroup\$

The battery and the capacitor should be connected in parallel. It is better to solder the capacitor to the main circuit board if possible. Considering the capacitor connection is always okay it will work just fine.

The beauty of parallel connection is that a damaged or disconnected or open branch does not affect other branches in parallel of the network.

\$\endgroup\$
0
\$\begingroup\$

You might want to look at this IC. http://www.onsemi.com/PowerSolutions/product.do?id=CAT6500 in case you can afford another small battery source. You can set preference of the source regardless of their voltage in comparison to the other one.

\$\endgroup\$
1
  • \$\begingroup\$ Interesting but not quite applicable here \$\endgroup\$ Aug 14, 2014 at 15:07
0
\$\begingroup\$

The capacitor bank will work fine if both positive & negative pins are on the circuit board side of the connector, so that it doesn't get disconnected when the battery does. If the capacitor bank is on the same side of the connector as the battery, then it will be disconnected at the same time and won't help.

A capacitor bank will draw a large inrush current when first connected to the battery, so you might want some kind of inrush limiter, like a small value power resistor (on the order 0.010 to 0.100 ohms), a ferrite, or both. This protects both the battery and capacitors from currents which are too large, and keeps the input fuse from blowing unnecessarily. The capacitors will charge a little more slowly. The longer time the capacitors can smooth out battery disconnect, the longer it takes to charge them up in the first place. Take a look to see if power input ramping up slowly doesn't have a detrimental effect on what's being powered.

Really this may be more of a mechanical engineering problem. Whatever is holding the battery in place needs to keep it pressed against the spring contacts enough that electrical contact isn't broken once made. Take a look inside a digital camera or cellphone with removable battery and see how they do it. Usually sliding into place and then click to hold it there, with a button press to release. If there's too little play it might be hard to slide into place, if there's too much play the battery contacts may bounce.

The benefit of getting the mechanical design right is that the large capacitor bank can be designed out, saving cost, and the product will run fine with the only large capacitors being the ones already in the voltage regulators (assuming you have some in the design).

\$\endgroup\$
2
  • \$\begingroup\$ i think you're right on the mechanical side. I think the solution we are leaning towards is getting pogo pins with more travel to allow more room for error in mechanical tolerances. \$\endgroup\$ Aug 14, 2014 at 15:09
  • \$\begingroup\$ If pogo pins are still too bouncy, you can try other connector types like blade-into-slot or pin-into-hole, like the old Sony camcorder pack or Thinkpad laptop. The mechanical issue may be that the battery is heavy and when the device is dropped it can push on things a lot harder than you might expect, causing springs to bend a lot, plastic to deform or break, etc. If you're building a "rugged" device you may have to defend the product from its own battery with extra sturdy battery compartment or rubber gasket etc. Or let the battery fly off so it doesn't break the rest. \$\endgroup\$
    – Matt B.
    Aug 14, 2014 at 17:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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