Multiple Battery Setup

This came from a separate thread "Typo in Battery Article in Soaring"

On Feb 2, 10:11 am, Karl Kunz wrote:
John, also enjoyed your article and will try your technique on a couple of batteries I am unsure of. Also, since we have an EE on the line I wanted to ask about battery setups. My partner and I have installed some new toys in our ASW20 (transponder, etc) and are wondering what is the best way add more power. We are currently running a single 12v 9amp battery. Would it be better to add another 9amp battery or to go to a single higher capacity battery. If using two batteries, what is the best way to tie them together.

-karl

I suggest multiple batteries for a couple of reasons;

Redundancy - you can have a failure either in the air or on the ground
so with two batteries you have a spare.
Expense - I replace one of my batteries (I own 4 and fly with 2) every
year so over the course of four years I replace them all (the older
units are given to the club). While two 9ah batteries are more
expensive than a single 12ah battery, my yearly replacement cost is
less of a single time bite to the budget.
Size - You are probably set up for the 9ah size and a larger 12ah may
not fit. However, someone mentioned a 12ah to me that has the same
footprint as the 9ah only taller. Anyone seen this?
Weight - Is 2x9ah heavier than 1x12ah? Is that important? Hmmmm.
Power - 2x9ah has more energy capability than 1x12ah

Now, how to connect them? What I recomment is three runs of 14-16
gauge Tefzel wire between the batteries and the panel. This gauge is
overkill but I don't want any voltage drop with a smaller gauge. 3
runs = a common ground, battery 1 and battery 2. I use red (battery)
and black (ground) heat shrink to mark each run. I use PowerPole
connectors - very robust and secure.

At the panel I use three switches to run the batteries in parallel.
Battery 1 on/off, battery 2 on/off and then the output of both to a
master switch and then to the avionics. The thinking of separate
battery switches is that you can disconnect a "bad" battery. I always
have all three switches turned on during flight - the need to be able
to turn off a bad battery hasn't arisen for me and I am unsure how I
would even know (everything dies? smoke?). I do, however, have a
voltmeter, so I turn on/off each battery to see how they are doing (on
the ground typically).

Some people wonder that with both batteries connected to one another
in parallel, is there an issue with one battery charging the other?
Certainly cross charging happens (no two batteries are identical) but
unless one battery is significantly discharged from another, this
cross charging would be slight. I always swap out both batteries
right off the chargers at the same time to minimize this. Some
solutions suggest installing diodes in series to prevent this cross
charging but diodes "steal" 0.5v-0.8v of your voltage and every volt
counts in a glider. So I don't use diodes. Does anyone else?

Finally - ABSOLUTELY CRITICAL - install a fuse on each battery right
at the terminals. Not a breaker, a fuse. Fuses are faster than
breakers. This fuse is more important than fuses/breakers in the
panel.

My $0.02. Let me know how your testing turns out.

- John DeRosa
http://aviation.derosaweb.net

John, you addressed all the issues we where concerned about. Thanks for some great info.

-karl

On 3 helmi, 18:10, JohnDeRosa wrote:

At the panel I use three switches to run the batteries in parallel.
Battery 1 on/off, battery 2 on/off and then the output of both to a
master switch and then to the avionics. *The thinking of separate
battery switches is that you can disconnect a "bad" battery. *I always
have all three switches turned on during flight - the need to be able
to turn off a bad battery hasn't arisen for me and I am unsure how I
would even know (everything dies? *smoke?).

You would know that by total power loss of all your batteries. You
have essentially parallel connection with all your batteries during
flight. When one fails, it drains your good batteries empty. When your
radio or computer starts blinking it's too late. Best option is to use
simple selector switch to choose one battery at time, and keep other
batteries disconnected.

rk

On Feb 5, 7:43*am, rk wrote:
On 3 helmi, 18:10, JohnDeRosa wrote:

At the panel I use three switches to run the batteries in parallel.
Battery 1 on/off, battery 2 on/off and then the output of both to a
master switch and then to the avionics. *The thinking of separate
battery switches is that you can disconnect a "bad" battery. *I always
have all three switches turned on during flight - the need to be able
to turn off a bad battery hasn't arisen for me and I am unsure how I
would even know (everything dies? *smoke?).

You would know that by total power loss of all your batteries. You
have essentially parallel connection with all your batteries during
flight. When one fails, it drains your good batteries empty. When your
radio or computer starts blinking it's too late. Best option is to use
simple selector switch to choose one battery at time, and keep other
batteries disconnected.

rk

I've been using a somewhat different approach. Two batteries, each
powering half the avionics. So batt 1 has the radio and nav computer
bus; batt 2 has the logger/PDA and backup vario bus. Instrument
"busses" hooked up via individual 3 position switches so that either
buss can be powered by either battery or disconnected. Normally, run
the batts/busses independent, but have had occasions when one batt
went low early, and just switched the bus it was on to the other
battery for all instruments for rest of flight.

I also use the built-in voltmeters in various instruments (AR-4201,
SN10) to monitor the health of the batteries during flight.

Seems to work OK for the past 12 years...

Kirk
66

On Feb 3, 8:10*am, JohnDeRosa wrote:
This came from a separate thread "Typo in Battery Article in Soaring"

On Feb 2, 10:11 am, Karl Kunz wrote:

John, also enjoyed your article and will try your technique on a couple of batteries I am unsure of. *Also, since we have an EE on the line I wanted to ask about battery setups. My partner and I have installed some new toys in our ASW20 (transponder, etc) and are wondering what is the best way add more power. We are currently running a single 12v 9amp battery. Would it be better to add another 9amp battery or to go to a single higher capacity battery. If using two batteries, what is the best way to tie them together.

-karl

I suggest multiple batteries for a couple of reasons;

Redundancy - you can have a failure either in the air or on the ground
so with two batteries you have a spare.
Expense - I replace one of my batteries (I own 4 and fly with 2) every
year so over the course of four years I replace them all (the older
units are given to the club). *While two 9ah batteries are more
expensive than a single 12ah battery, my yearly replacement cost is
less of a single time bite to the budget.
Size - You are probably set up for the 9ah size and a larger 12ah may
not fit. *However, someone mentioned a 12ah to me that has the same
footprint as the 9ah only taller. *Anyone seen this?
Weight - Is 2x9ah heavier than 1x12ah? *Is that important? *Hmmmm.
Power - 2x9ah has more energy capability than 1x12ah

Now, how to connect them? * What I recomment is three runs of 14-16
gauge Tefzel wire between the batteries and the panel. *This gauge is
overkill but I don't want any voltage drop with a smaller gauge. *3
runs = a common ground, battery 1 and battery 2. *I use red (battery)
and black (ground) heat shrink to mark each run. *I use PowerPole
connectors - very robust and secure.

At the panel I use three switches to run the batteries in parallel.
Battery 1 on/off, battery 2 on/off and then the output of both to a
master switch and then to the avionics. *The thinking of separate
battery switches is that you can disconnect a "bad" battery. *I always
have all three switches turned on during flight - the need to be able
to turn off a bad battery hasn't arisen for me and I am unsure how I
would even know (everything dies? *smoke?). *I do, however, have a
voltmeter, so I turn on/off each battery to see how they are doing (on
the ground typically).

Some people wonder that with both batteries connected to one another
in parallel, is there an issue with one battery charging the other?
Certainly cross charging happens (no two batteries are identical) but
unless one battery is significantly discharged from another, this
cross charging would be slight. *I always swap out both batteries
right off the chargers at the same time to minimize this. *Some
solutions suggest installing diodes in series to prevent this cross
charging but diodes "steal" 0.5v-0.8v of your voltage and every volt
counts in a glider. *So I don't use diodes. *Does anyone else?

Finally - ABSOLUTELY CRITICAL - install a fuse on each battery right
at the terminals. *Not a breaker, a fuse. *Fuses are faster than
breakers. *This fuse is more important than fuses/breakers in the
panel.

My $0.02. *Let me know how your testing turns out.

- John DeRosahttp://aviation.derosaweb.net

I've been using very low loss Schottky diodes (95SQ015 - Vf stays at
about 0.2v @ 1A last I checked) to run all my batteries (sometimes 4)
in parallel. Increased AH capacity from parallel operation offsets Vf
and provides isolation of a misbehaving battery. Absolutely fuse at
the battery terminals with fuse sized appropriately for the size/
length of conductor. Comments welcome.
QT

On Feb 3, 8:10*am, JohnDeRosa wrote:
right off the chargers at the same time to minimize this. *Some
solutions suggest installing diodes in series to prevent this cross
charging but diodes "steal" 0.5v-0.8v of your voltage and every volt
counts in a glider.

Ideal diode, such as LTC4358, would eliminate the voltage drop at the
cost of higher complexity.

Bart

I've been using very low loss Schottky diodes (95SQ015 - Vf stays at
about 0.2v @ 1A last I checked)
QT

Ideal diode, such as LTC4358, would eliminate the voltage drop at the
cost of higher complexity.

Bart

First, a teaching moment for those non-techies about using diodes to
isolate two battery setups so that a bad battery won't suck the life
out of the good battery, see
http://aviation.derosaweb.net/presen...iring_2007.pdf
(page 21-25) for some explanations.

While I know about Schottky diodes and their low forward voltage drop
(0.2v which seems like the best of the diode worlds) but I had not
heard of the Ideal Diode which seems custom made for our application
of two battery sources and if one dies the other takes over. Good to
5A. What's not to like? But what is the critical forward voltage
drop? I read;

"The internal MOSFET turns on and the amplifier tries to regulate the
voltage drop across the IN and OUT connections to 25mV. If the load
current causes more than 25mV of drop, the MOSFET is driven fully on
and the voltage drop is equal to RDS(ON) x ILOAD."

25mv? GREAT NEWS!
RDS(ON) x ILOAD = 20milli-ohm x 5A = 0.1 volt. Better than a
Schottky. WHOOP!
(BTW: the small m for milli instead of big M for mega tripped me
up)

But then in the Specs Table I read;

ÄVSD Source-Drain Regulation Voltage (VIN - VOUT) 1mA IIN 100mA
l 25 mV (Typ)
ÄVSD Body Diode Forward Voltage Drop IIN = 5A, MOSFET Off l 0.8 V
(Typ)

So aren't we back to a silicon diode like forward drop? OK, when the
MOSFET is ON then we are back to the 0.1 volt. Interesting.

Lousy SM packaging for our application though. Bart - How are you
mounting it? Pictures?

On Feb 5, 8:47*pm, JohnDeRosa wrote:
But then in the Specs Table I read;

* *ÄVSD Source-Drain Regulation Voltage (VIN - VOUT) 1mA IIN 100mA
l 25 mV (Typ)
* *ÄVSD Body Diode Forward Voltage Drop IIN = 5A, MOSFET Off l 0.8 V
(Typ)

So aren't we back to a silicon diode like forward drop? *OK, when the
MOSFET is ON then we are back to the 0.1 volt. * Interesting.

The 0.8V voltage drop is for the "body diode" of the internal MOSFET:
http://en.wikipedia.org/wiki/Power_MOSFET#Body_diode
In other words, that's the drop you will get if the controller dies,
or if the input voltage drops below 9V.

Lousy SM packaging for our application though. *Bart - How are you
mounting it? *Pictures?

I have never used one in a glider. If I were to, I would put the two
chips on a small PCB and cover it with heatshrink. Something like
this: http://i00.i.aliimg.com/img/pb/905/8...834905_364.jpg

Actually, if I find time, I can make a prototype within a week or two.
Then Karl can test it if he wants to :-)

Bart

I would certainly be willing to give it a try.

I would certainly be willing to give it a try. Would 5amps be adequate for this? Are these diodes fail operational, i.e. if it fails will I continue to get power from the effected battery.