Battery backup & Supercaps how to connect ??

Go To Last Post
11 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Guyzz

I am planning on doing battery backup on a system containing of a GSM Module controlled by a Mega8 board , and a Mega32 board.
I need the system to send me an SMS when the AC power is lost.

I expect to feed it 24v via PDS-cable , and then use a National Switcher (those TO-220 ones) on each board , to regulate down to 5v.

I would then use the 12v battery as backup power , and have 2 Schottkey diodes to do the input selection , stolen from the "Butterfly" schematics

 

                  | Switcher Vcc Input                     
                  |
                  |
 24v --------|>---O---<|------- 12v Bat

I have 4 questions ...

1: Would the diode trick work here ??

2: In order to have power in the time where the 24v is lost , and until the 12v battery kicks in , i was thinking about a Panasonic supercap on the 5V side of the regulator , is that an overkill ?? , would a normal 470uF be sufficient on the M32 board.

I allready have 2x 470uF on the M8 board as the GSM module draw current in pulses.

3: When i switch on the system would the Supercap kill my switcher (max 1.5A i think) or is the internal resistance high enough to prevent this.

4: The National Switcher would like low ESR caps what would happen if i put a Supercap on the 5v side , along with the 470uF low esr.

Well a lot of questions , but you guys usually have some useful guidelines.

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi

1. Yes, it will work fine. No need for schottky's, ordinary 1N400x will suffice. In case of the Butterfly, schottky's are needed because the diodes are on the 3.3V
2. Assuming that the 12V is always present, there is no need for a supercap. If you have a 470 or 1000 uF on the input of the switcher, that will (by far) be enough
3. IIRC these supercaps have a pretty high impedance, but nevertheless I think the recommendation is to limit the rush-in current. But as said, there is no need for a supercap
4. Low ESR caps are sufficient

Nard

A GIF is worth a thousend words   She is called Rosa, lives at Mint17.3 https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Can't really help out, but I'll try.

1) What's the reverse breakdown for the diodes? If it's around 12V, then you'll run into some problems. Also check the max current to ensure you don't blow up the diodes when the switcher is under full load.

2 & 3) I'd say you'd need a charging resistor, as I think they'll cause one heck of a current spike (charges in 5 timer periods, t=RC, almost 0 ohm R value). You want to charge up the capacitor over a short period of time, to prevent a huge current draw from the switcher.

EDIT: Seems Nard's corrected me before I even posted. Ignore me while I go study first year Electronics ;).

- Dean :twisted:

Make Atmel Studio better with my free extensions. Open source and feedback welcome!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If one is worried about saving every watt, a schottkey dissipates less than a silicon diode... .5v vs .7V times the amps... at 10 amps, this is 5 watts vs 7 watts (hot diode!). A big mosfet with real low ohms is better than this watts wise. In a 100W system, that goes from 95% to 93% effciency (for example).

I have one of those Maxwell boostcaps that is rated 350 farads at 2.7V. It takes a few MINUTES to charge to 2.5V on a 2A lab supply set to 2.5V. My question to my fellow EEs here is: if I hooked my discharged boostcap to a voltmeter then connected a 12V battery, and disconnected the batt before the V got > 2.5V, would it be damaged? In other words, there is a microsecond of 12V before the IR drop sucks the V down to .1V as the cap charges. Does that zap the cap?? Damage the dielectric?

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Bob, I checked the datasheet of the goldcap I have .... all it says is 5.5V as abs.max. If I would face that issue, I wouldn't try it: goldcaps are too expensive to destroy.

And as for Bingo's application: I don't expect currents like 10 amps ... given the Mega8, Mega32 and GSM-module, so I think that my recommendation is ok

Nard

A GIF is worth a thousend words   She is called Rosa, lives at Mint17.3 https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanx guyzz.

I have some Panasonic 1F 5.5v goldcaps , i can't see if they are F or NF models.
But they have either less than 30 or 50 ohm resistance at 1Khz

To minimize inrush current , one would normally use a resistor.
But that would also limit Vout , but maybe not much at moderate loads.

I have done some searching , and it seems that most Goldcaps are used for memory or clock's. And then they use a diode and a resistor to Vcc
The diode is for blocking the whole circuit from drawing power from the Cap , and the resistor would then be for limiting the inrush i guess.

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Btw: Does anyone have a simple charger circuit for a 12v 7.2AH LED-GEL Battery (Eagle-Picher).
I can buy a charger for 16 Eur , but if it's "Ultra simple" to do then i might build one.

I was thinking like a Tiny13 & a FET

/Bingo

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

reading the google I'm starting to get the feeling, that this gel cell battery is like sealed lead acid battery. Same charging principle. But I'm not sure about that... this is just google we are talking about.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Does anyone have a simple charger circuit for a 12v 7.2AH LED-GEL Battery (Eagle-Picher).
A nice project in itself :) but I think it will take more than just a Tiny and a FET ....

Problem with standard low-cost chargers is the cheap design :)

They are pretty well okay for just the chargingjob, but are not designed to maintain the battery (permanent connection). F.i., temperature has to be taken into account.

I don't have a design at hand, but in one of the latest Elektor-magazines there was a design

Nard

A GIF is worth a thousend words   She is called Rosa, lives at Mint17.3 https://www.linuxmint.com/

Dragon broken ? http://aplomb.nl/TechStuff/Dragon/Dragon.html for how-to-fix tips

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Gel cells need a constant voltage, and quick charge or constant trickle need slightly different voltages. Here's a link to an application note.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here is the battery DS

How quick should one react on a charger ??

I mean a TinyXX , and the AD connected to the battery poles , and maybe another AD via a NTC glued to the Battery. Then PWM the 24v until the charge voltage is obtained. I'd prob just need trickle charge , as the 220v is up most of the time.

/Bingo

Attachment(s):