0
\$\begingroup\$

I am in no means a electrical engineer or student but i dabble.

I have a project to make a BIG smartphone charger using multiple 1800mAH batteries.

you can choose which cell to use and charge via a three way rotary switch.

My question is do i need a ground somewhere? is there something i am doing wrong?

schematics

\$\endgroup\$
1

3 Answers 3

2
\$\begingroup\$

No, your design seems fine in theory. All grounds should be connected at a single point on your board. And make sure you properly size the wires or pcb traces for the current you are expecting. As well as making sure your rotary switch can handle the current as well.

\$\endgroup\$
6
  • \$\begingroup\$ if i use really big wires there should be no problem right? \$\endgroup\$ Jul 14, 2016 at 16:21
  • \$\begingroup\$ @marcdanielt Yes. But you should know that this is a bit odd. There are setups that let you use multiple LIPO cells in parallel without needing to switch between them. Many charger ICs handle this without any issue. Just don't mix and match different cells. youtube.com/watch?v=Z1KBccrNYHE \$\endgroup\$
    – Passerby
    Jul 14, 2016 at 16:56
  • \$\begingroup\$ i read somewhere that connecting them in series was a little dangerous as one could get a screwed up charge and BOOM! It was my first idead to connect them in series and if you suggest its safe, i might just do that. \$\endgroup\$ Jul 14, 2016 at 20:14
  • \$\begingroup\$ You can connect lipo in series or parallel. You need a charger designed for either setup. And you should never mix old and new cells, or different brands, etc. \$\endgroup\$
    – Passerby
    Jul 14, 2016 at 20:17
  • \$\begingroup\$ im modifying a charger that only as 1 LIPO connected to it. but all the three chargers are the same. You seem to be quite more knowlegeable than me. What would you do? \$\endgroup\$ Jul 14, 2016 at 20:50
1
\$\begingroup\$

If you want to modify an exsiting circuit, things will get a bit more complicated. In that case, your first idea - using a switch to basically swap out the LiPo cell - is probably best.

The reason is simple: a battery charger is designed for the cell type it is charging. and for the voltage of the entire battery. LiPo are a bit tricky because they have this tendency to combust; the youtube videos of LiPos bursting into flames are real!

When modding and exsisting charger with a switch, make sure the switch is a break-before-make type. This means that when rotating, there is a point where all contacts are open before any contact is closed. This will avoid having two cells connected in parallel.

If you connect 2 cells in paralel, one charged, one discharged, the discharged cell will draw current from the full cell. This seems no big deal, but the current flowing into the discharged cell is not regulated or limited - and that can harm the cell: LiPo charging must be done with a limited current. This would happen every time you switch over with a make-before-break type switch.

If you use a cell type different from the original cell type in the power bank, make sure it is a cell with larger capacity. LiPo chargers limit the current they feed to the cell. The current limit is defined by the cell capacity. "too little" is harmless while "too much" damages the cell; so the current limit designed for a say 1aH cell is usually harmless for a 1.8aH cell. "usually" not "always" because there are cells designed to allow for high charge current and others not so much; so the limit for a 1Ah quickcharge cell could be damaging for a 2aH "standard" type.

So your original schematic was very much on the right track: all negative terminals wired together and to the ground of the whole contraption, the positive terminals of the cells feed a selector switch and that feeds the positive input of the charger board.

If all cells are of the type the charger is designed for, this will work; from the chargers point of view, you are simply swapping out the battery cell.

To increase the life of your cells, recharge them soon after discharge: the cells self-discharge then has little chance to deep-discharge the cell (which damages the cell).

Read up on LiPo batteries on the web. For example: http://linrunner.de/en/tlp/docs/tlp-faq.html#battery (considering laptop batteries but follow the lead from there). If you still dare :-) go for the switch.

In a recent project (an art installation) I needed some battery: the power was fed to a small vehicle hanging from a rail; this used a sliding contact and those graphite-on-brass contatcs get wonky over time (the brass oxidizes). So I wanted a battery in the vehicle to avoid power interruptions to the onboard computer. I chose lead, one of the oldest battery chemistries out there :-) (actually in the more modern form with an absorband glass mat to bind the electrolyte). Lead does not have the best energy-to-weight or energy-to-volume ratio (actually: it has the worst). But they are really safe...

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

LiPo cells can indeed "go boom" - well, start to burn or even explode. Especially when charged over their maximum voltage, they may exothermally decompose.

Let's assume a cell has a maximum charge voltage of 4.2V. Now you connect 3 of those in series and limit your charging voltage to 12.6V precisely; everything should be fine, right? But it isn't: each cell is slightly different in capacity (and temperature and wear state). So each has a slightly different voltage. Therefore, one will reach the 4.2V before the others and the series is still below the 12.6V applied to the stack. This one will then be overcharged while the others are still fine.

This is the reason that all LiPo cell packs have a balancer connector. This connector simply has a wire for each cell, os the charger can measure each cells voltage independently from the other cells. And also limit the charging voltage for each cell to a safe level, again individually.

When connecting cells in parallel, the overcharging problem does not exist: all cells are charged to the exact same voltage every time. But you will face the "uneven wear" and possibly "deep discharge" problems instead.

Again, imagine 3 cells in parallel, each slightly different in capacity. This is always the case, even two absolutely identical cells will differ after a few charge cycles. Now the freshly charged pack is discharged - that is its purpose after all.

One cell will have the lowest capacity of the triple. This one will drop in voltage faster than the others. Actually, the voltage will be the same - they are wired in parallel - but the current distribution within the pack will be uneven. And this uneven load will cause uneven wear and uneven self-heating of the cells. Worst case, the weakest cell will be completely depleted /deep discharged before the discharging process is stopped. In this case, the cell will be damaged, i.e. loose even more capacity or even become a short circuit (and in turn kill the other cells in the pack).

Your design avoids all that trouble - which is good. But your solution has a cost: only one cell is active at an time and you manually have to switch the charge current as well as the discharge current from cell to cell.

I think, the best solution for you might be - a charger with integrated balancer, - a cell pack with balancer connection and - a stepdown converter able to handle >13V input and deliver the 5V/2A output you desire. The cell packs and balancer-chargers are pretty much standard in the RC model scene. You only need to make sure the battery does not get discharged below the safe voltage per cell; simple stepdown (aka Buck) regulators will usually deplete your battery and kill it.

Hope this helps hase

\$\endgroup\$
5
  • \$\begingroup\$ Please answer all of the users questions. This is not so much a forum for discussion but has a Q&A format. The OP asked a question about the ground. Please see: electronics.stackexchange.com/help/how-to-answer \$\endgroup\$
    – Voltage Spike
    Jul 14, 2016 at 22:25
  • \$\begingroup\$ Thank you for your input. Would it look a little like this? \$\endgroup\$ Jul 14, 2016 at 22:49
  • \$\begingroup\$ @laptop2d there is nothing wrong with this answer, it adds significant information for OPS real purpose... \$\endgroup\$
    – Passerby
    Jul 15, 2016 at 19:15
  • \$\begingroup\$ @Passerby But they still didn't answer the OP's question about the ground. \$\endgroup\$
    – Voltage Spike
    Jul 15, 2016 at 19:35
  • 1
    \$\begingroup\$ Hello Hartmut. Could you please refer to my last comment and see if im right. The ground question got answered but i like your input. \$\endgroup\$ Jul 16, 2016 at 13:45

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.