self powered speakers

[H.Johnston]

I am trying to decide what wire size to use for 10 parallel
connections of an array of self powered speakers each with an
amplifier incorporated in it.
..
They are 5 mounted on each parallel line making a total of 50
speakers.
I measured the input resistance of the speaker and it is about 1Kohm .
Power use is 25 W and I have a 50Amp 48Vdc power source to give power
to the speakers.

Any suggestions on how to improve design , calculate wire size.

The total lengths of the wiring provoding is 70meters.

Thanks all for the help.

H.Johnston

 

[pimpom]

H.Johnston wrote:

I hope you won't take this amiss, but I thought it would be best
to first clarify what you wrote.

I am trying to decide what wire size to use for 10 parallel
connections of an array of self powered speakers each with an
amplifier incorporated in it.

Are you talking about the power connection, the audio signal or
both?

.
They are 5 mounted on each parallel line making a total of 50
speakers.
I measured the input resistance of the speaker and it is about
1Kohm .

Are the amplifier-speakers designed to be driven from a
microphone or from the line level output of a preamp/mixer? 1K is
quite low for the line-level input of a power amplifier. It's
more in the range of an input for a low impedance mike. Or did
you measure the power supply input?

Sorry if you know your stuff about these things and what you said
are what they appear to be. But I thought it would be best to
make sure.

Power use is 25 W

Is 25W the power consumed from the power supply or is it the
output power rating?

and I have a 50Amp 48Vdc power source to give power
to the speakers.

Are the amplifiers specced to be powered from an external 48Vdc
source?

Any suggestions on how to improve design , calculate wire size.

The total lengths of the wiring provoding is 70meters.

More details about how the speakers are to be located with
respect to each other will also help.

 

[H.Johnston]

H.Johnston wrote:

I hope you won't take this amiss, but I thought it would be best
to first clarify what you wrote.


Are you talking about the power connection, the audio signal or
both?
both

Are the amplifier-speakers designed to be driven from a
microphone or from the line level output of a preamp/mixer? 1K is
quite low for the line-level input of a power amplifier. It's
more in the range of an input for a low impedance mike. Or did
you measure the power supply input?

Well the 1 K is for the transformer which connects to the speaker
inputs. You can change the transformer settings though to vary the
output audio power.

Sorry if you know your stuff about these things and what you said
are what they appear to be. But I thought it would be best to
make sure.


Is 25W the power consumed from the power supply or is it the
output power rating?

Consumed by speaker and I would guess the amplifier unit though the
data sheet does give sufficient detail on that.

Are the amplifiers specced to be powered from an external 48Vdc
source?

Yes the 50A source mentioned above

More details about how the speakers are to be located with
respect to each other will also help.

Thanks.

H.Johnston

 

[Paul Keinanen]

Well the 1 K is for the transformer which connects to the speaker
inputs. You can change the transformer settings though to vary the
output audio power.

Are you sure that you are talking about active speakers and not about
remote speakers intended for 70 V or 100 V line operation ?

Your descriptions fits quite well to remote loudspeakers as found in
schools and other large buildings for PA applications.

These contain an ordinary 4-16 ohm speaker, bit if it is driven by
long lines directly, most power would be lost in the cabling due to
the high current. To avoid this, each speaker contains a transformer,
which is intended to fed from the nominally 100 V audio line. The
speaker unit usually contains a step attenuator implemented by
selecting the appropriate winding tap from the transformer.

A 1000 ohm impedance on the 100 V line would draw 0,1 A, thus the
speaker could deliver up to 10 W. Thus for 50 units, a single power
amplifier would be required with 500 W output power. If the current
really is 0,1 A for each speaker, the total current would be 5 A,
possibly distributed along several branches, so wires less than 1 mm
thick should be sufficient.

 

[H.Johnston]

Are you sure that you are talking about active speakers and not about
remote speakers intended for 70 V or 100 V line operation ?

Your descriptions fits quite well to remote loudspeakers as found in
schools and other large buildings for PA applications.

These contain an ordinary 4-16 ohm speaker, bit if it is driven by
long lines directly, most power would be lost in the cabling due to
the high current. To avoid this, each speaker contains a transformer,
which is intended to fed from the nominally 100 V audio line. The
speaker unit usually contains a step attenuator implemented by
selecting the appropriate winding tap from the transformer.

A 1000 ohm impedance on the 100 V line would draw 0,1 A, thus the
speaker could deliver up to 10 W. Thus for 50 units, a single power
amplifier would be required with 500 W output power. If the current
really is 0,1 A for each speaker, the total current would be 5 A,
possibly distributed along several branches, so wires less than 1 mm
thick should be sufficient.

Maybe I wasn't clear enough.
No I'm familair with the units you're talking about and we've set up a
lot of those at 70V /100V and as a matter of fact for PA systems they
are the most used and the literature is full of advice on setting
those up
Self powered speakers or active units are not as common in PA , at
least for me, which is why I was asking . The attenuating unit on the
output of the inbuilt amplifier is a transformer or inductor is what
the manufacturer is using to reduce the voltage instead of a resistor.
The data sheet shows a combination of 6 dip switch settings that are
set on this unit to vary the output power on the loudspeaker from
values ranging at 0.76W to 25W. The datasheet gives the speaker
impedance as 16Ohms, I measured it during audio reproduction and it
varies from 1 Ohm to 25Ohms depending on frequency.

The loudspeaker+amp unit have practically speaking 5 wires entering
into them 2 for the 48v source which powers the internal amplifier
unit
2 for the audio signal + 1 wire for the shield .

The input impedance to the amplifier , which I think is more important
then the figure I gave earlier as the audio signal passes through
here, is 11Kohms before getting to the speakers.

Thanks again.

Johnston

 

[pimpom]

Well the 1 K is for the transformer which connects to the
speaker
inputs. You can change the transformer settings though to vary
the
output audio power.
Consumed by speaker and I would guess the amplifier unit though
the
data sheet does give sufficient detail on that.

Yes the 50A source mentioned above

Thanks.

H.Johnston

For the input signals, a shielded microphone cable to each
speaker will do.

The power supply cable is more involved as we have to work out
the best way to group, separate and run the wires from the common
power supply to the speakers. We need to know the distances
involved as this is crucial to calculating the size of wires
needed. It's vital to know how far the speakers are spaced from
each other and from the common power supply. We have to consider
voltage drops over the wires, not just the current carrying
capacity.

 

[pimpom]

Well looking at just the power your looking at fifty 50W loads
(assuming the amps are 50% efficient which is typical for
class AB) That's 2500W which is ~52A from your 48V power supply
at full out so
you don't have much margin.

http://www.powerstream.com/Wire_Size.htm

You will more design for voltage drop than anything else.
Assume a
10% voltage drop due to the resistance of both the 48V and the
return
leg.
Assume each set of to amps gets its own power cable from the
supply.
At 11 amps (500W/48V) the resistance would be 4.8V(10%
drop)/11A =
0.43 ohms. This is the total loop resistance of the wires from
the
power supply to the ten amps and back (70 there and 70 back so
140
meters total length).

Looks like you'll need 10 AWG wire to meet that specification.

You'll need 5 runs of this to accomodate your 50 speakers.

.14Km * 3.27 ohms/Km = .45 ohms

If you can allow the 48V to sag 20% say down to 40V then you
can get
away with 12 AWG wire.

All of this assumes you need max power from the speakers, if
you know
you'll need less you can get away with smaller wires.

And that is assuming all 50 speakers are bunched close together,
powered from a common supply 70 meters away. I'm not sure that's
what the OP is trying to do. The speakers or groups of speakers
may be scattered over a sizeable area.

 

[H.Johnston]

Well looking at just the power your looking at fifty 50W loads (assuming
the amps are 50% efficient which is typical for class AB) That's 2500W
which is ~52A from your 48V power supply at full out so you don't have
much margin.

http://www.powerstream.com/Wire_Size.htm

You will more design for voltage drop than anything else. Assume a 10%
voltage drop due to the resistance of both the 48V and the return leg.

Assume each set of to amps gets its own power cable from the supply.
At 11 amps (500W/48V) the resistance would be 4.8V(10% drop)/11A = 0.43
ohms. This is the total loop resistance of the wires from the power
supply to the ten amps and back (70 there and 70 back so 140 meters
total length).

Looks like you'll need 10 AWG wire to meet that specification.

You'll need 5 runs of this to accomodate your 50 speakers.

.14Km * 3.27 ohms/Km = .45 ohms

If you can allow the 48V to sag 20% say down to 40V then you can get
away with 12 AWG wire.

All of this assumes you need max power from the speakers, if you know
you'll need less you can get away with smaller wires.



As for input signals that is more tricky because noise pickup and
crosstalk are a bigger concerns than line resistance.

Thanks there for the insight on calculating the power cable size
though one thing I didn't mention was that the amplifiers were class D
which should give a higher efficiency figure . Well still the exact
power is still unknown because before the
signal goes to the amplifier it passes through a mixer and attenuator
unit (there is one in an audio rack cabinet ... whatever the power
consumption is here is unknown to me ... I'll see if I can find the
documentation somewhere. The attenuator units (who knows why its
called so ) from the documents I got are supposed to amplify the
signal to sufficient levels as it comes from a microphone so as to get
to the amplifier + loudspeaker units sufficiently intact. But
actually what goes on in these units is still unknown to me at the
moment.

I was thinking of for the sake of simplicity using a 4 wire cable (2
twisted pairs, each sized 10AWG if the 500W power consumption
assumption holds) with a shield for both power and audio.

H. Johnston