Dual battery PSU with low voltage changeover

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi folks, I have a few ideas to float and some help required.

I am currently building a DAC based on the DDDAC schematic, and I am stuck trying to decide how to power it.

I love the idea of having it battery powered, but because the DAC will be used with the optical out from my cable TV, the battery would need to be huge (expensive) to give long enough playing times.

I had thought about is having 2 smaller (cheaper) SLA batteries, with a switch on the front panel that changes over the battery used to power the DAC and also charges the battery not in use using a cheap charger.

This would work provided I or the other half remember to turn the thing off to prevent one of the batteries being completely discharged, this kind of usage will kill the battery very quickly.

So, I would like to make the change over automatic using a pair of bistable/changover relays to change over the charging and supply batteries.

The relay arrangement is simple, the hard bit (for me anyway) is working out how to trigger the switchover.

I would like the changeover to occur when the output of the source battery drops to below 11V, this means I need to generate a pulse to the relay coils at this point.

I want this process to carry on indefinatley so the system swaps battery every time the source drops below 11V.

Any ideas how I could do this? Perhaps comparing the output to the output of a 11V regulator? I a really in the dark here.

There are a few other "open items"

-It is likley that at the changeover point there would be a dropout in the DAC which could produce undesirable/damaging noise. A cap in the supply might prevent this? or is ita matter of briefly putting both batteries in parallel before completing the changover.

-Are relays the way to go, or should I be thinking about a solid state solution?

-Need to think about what to do when its powered off, either keep charging the last used battery, or try to charge both.

-Charge rate needs to be higher than drain rate for indefinite running otherwise the changeover from battery A to B back to A will get faster and faster until the system is powered down .

I would really appreciate you feedback and any ideas you have.

Cheers
 
So you ned a circuit basically that turns a relay powering transistor on or off when the threshold voltage reaches 11V

Basically yes, thats the part I am unable to do myself, the rest of it is just arranging the relays correctly.

There are other cuircuits around on the web that do similar things, but these tend to compare the two batteries, which creates a link back to the charging battery and hence could introduce noise. (also a charging battery will show good voltage so it wouldnt work anyway)

11V was chosen because thats about as low as I'd like the DAC to run from, but you are right it is probably too low for the battery cyclic life. I am assuming that whatever circuit eventually emerges will be tunable.
 
However the difficult part is that you don´t want the relay switched again when the voltage goes over 11v.You want the charger on for a specific time i.e. 6hrs to ensure a full charge.So you would need a timer circuit as well or a sensor that switches the relay out when the charging current reaches a minimum indicating a fully charged battery.
That probably means programming a PIC microcontroller.It is doable but a little complicated.

I would suggest first try a very good superegulated supply based on alw supreregulators.In many ways they sound better than batteries.
 
dwitt4 said:
...
-It is likley that at the changeover point there would be a dropout in the DAC which could produce undesirable/damaging noise. A cap in the supply might prevent this? or is ita matter of briefly putting both batteries in parallel before completing the changover.
...
a simple gyrator/capacitor-multiplier in series with the supply (i.e. after the battery changeover) will «exterminate» the «problem», just use a long enough time-constant ...
...
-Are relays the way to go, or should I be thinking about a solid state solution?
...
Relays usually consume a little more power, and uses more time to switch. Solid state may use less power, takes less time to switch, and is possibly easier to integrate with switching logic, control logic for the charging, voltage sensing etc.
...
-Need to think about what to do when its powered off, either keep charging the last used battery, or try to charge both.
...
For charging both batteries, you should use two chargers. I believe I would either use two chargers, or design the charging logic such that the unused battery is charging; and when fully charged, either manually switching out (and start charging) the used battery, or automatically switching when the unit is not in use ... something like that.
[/quote]...
-Charge rate needs to be higher than drain rate for indefinite running otherwise the changeover from battery A to B back to A will get faster and faster until the system is powered down .
...
Not necessarily, you need control of thresholds and states (in use vs. not in use, and charging states
...
I would really appreciate you feedback and any ideas you have.

Cheers [/B]
That’s my two cents.

Regards
KJ
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.