redundant batterys

[mid-air]

how would i set up a power isolator to make a true redundant bat. setup, so if one bat. shorts it will disconect and not affect the other or the rx. or servos. would it be as simple as wireing in a dioad, hope you guys can help!
AL

[Dan Baldwin]

This is a hotly contested topic here on RCGroups. Many insist that there is no need to isolate the two battery packs, just hook them in parallel. Others (myself included) think that it would be possible to have a short in the wiring on one pack that could draw both packs down, so it would be better to isolate the packs with diodes. There would be a voltage drop with diodes, but you could use 5 cell packs, and still be above the normal voltage for a 4 cell pack after the diode (diode drop would be about .7 volts).

Dan

[village_idiot]

To do any good as a redundant system, they must be isolated from each other

[Dan Baldwin]

According to Red Scholefield (of RC Battery Clinic and contributor to Model Aviation magazine) (http://www.rcbatteryclinic.com/parallel.html) it is not necessary to isolate the packs with diodes.

Dan

[EloyM]

Companies such as Sanyo, Panasonic, etc recommend against using batteries in parallel,, in spite of the "experts" we have amongst us. There are a number of true battery backers on the market - see cermark.com for its CEMBK at $20.

[MatC]

mid-air: you need the diodes if you want to protect against a battery shorting. I would guess with lipoly's this is a possibility. Anything that you put between your battery and your circuit has to be more reliable than a battery pack, or you lose out anyway - so anything more than a diode probably isn't worth it.
If you do decide to go the diode route, all you do is add a diode (high power one!) to each battery + and then put the two diodes together to the + of the receiver. Get the diodes the right way round, and test that the system works with either battery.
Silicon diodes will lose you 0.6v, germanium (if you can find high powered ones) 0.2v, look at schottky diodes too.

[tb47]

Companies such as Sanyo, Panasonic, etc recommend against using batteries in parallel
I understand well why I shouldn't connect Nixx batteries in parallel during charging. I am curious however why I shouldn't parallel them for discharge. A link or reference to statements/explanations from the mentioned manufactureres would be appreciated.

I found this from panasonic, but it doesn't mention discharghing in parallel:
http://www.panasonic.com/industrial/battery/oem/images/pdf/Panasonic_NIMH_Precautions.pdf

This one advocates the use of parallel discharging for RC models:
http://hangtimes.com/rcbattery_faq.html

[tb47]

Anything that you put between your battery and your circuit has to be more reliable than a battery pack, or you lose out anyway - so anything more than a diode probably isn't worth it.
That's a good point. But what is the risk that a charged battery will short compared to the risk of failure in diodes, switches and (not the least) connectors?

[village_idiot]

Failure of a discreet diode would be extremely rare if you size them correctly. And normally when they do fail the open, so you would be running on the other pack with a failure, which is the whole point behind redunant batteries.

[MatC]

> That's a good point. But what is the risk that a charged battery will short compared to the risk of failure in diodes, switches and (not the least) connectors?

If you put your switches and connectors close to the battery side, then you are paralleling those too. That puts your single point of failure down to your receiver and your wiring.

[Acetronics]

Hi, MatC

I've built such a coupler/decoupler for High value scale planes ... I use two pairs of back-to-back Power Mosfets as switches ... ( see Micrel Handbooks for details )

that permits to have a security switch ( open circuit is Power-on ).

Now, you're right: the more components and wires, the more failure reasons ...

and all the plane wirings have to be done according to that ...

Twin Power schottky diodes are perfect for their Simplicity/Service ratio ... as some have a .2v maxi Voltage loss !!! ;)

Alain