Xref: utzoo rec.autos:10339 sci.electronics:3238 rec.audio:7066
Path: utzoo!utgpu!water!watmath!clyde!att!ihnp4!ihlpf!chrz
From: chrz@ihlpf.ATT.COM (Chrzanowski)
Newsgroups: rec.autos,sci.electronics,rec.audio
Subject: Car stereo RMS power
Keywords: Bridged amps, DC-to-DC converters
Message-ID: <5244@ihlpf.ATT.COM>
Date: 8 Jul 88 16:26:18 GMT
Organization: AT&T Bell Laboratories - Naperville, Illinois
Lines: 61

>>>>
>>> ...12 volts driving a 4 ohm load yields 36 watts RMS...
>>> ...The unit bench tested at 35 watts RMS which is about what I was
>>> hoping for.
>> 
>> wouldn't that be closer to 25W:  12**2/4 = 36W peak, 36 * 2**(-0.5) ~= 25W.
>> i'd expect a sharp change in the price/power curve where the power hits
>> (an honest) 25W, because of the need for a voltage-boosting gizmo.
>> 

>Try dividing by 2 and then divide by the square root of 2 for RMS.

	[Correct]

>Sine wave is + and - so that gives you + and - 6 volts peak to peak.

	[No.  It can vary from zero to +12, that's 12 volts peak TO PEAK]

>6 / 1.41 = 4.24 V RMS which yields ~4.5 Watts into 4 ohms.
>Since cars are usually above  13 Volts you get slightly more than
>5 watts  per channel.  If you want more than 5 watts you need to

To convert volts peak to peak (VPP) to RMS (assuming a sinewave)
>Try dividing by 2 and then divide by the square root of 2 
Soo, If we've got 12 VPP into 4 ohms then RMS power is

	(12volts * 12volts) / ((2*(2**.5) * (4ohms)) = 12.7 watts.

Of course, to actually deliver 12 volts to the load with a twelve volt
supply the amp's output impedance would have to be zero.  In fact, with
a nominal 13.6 volt source this (~12 watts) is about the max you will
get from such an amp WITH mucho distortion, as the output transistors
will be at or near saturation (don't forget that a current limiting/sensing
resistor is in series with the transistors).

Actually, many modern car stereos use "bridged" amps.  This means there
are actually TWO amps driving each speaker: the amps are (180 degrees) 
out of phase; each is connected to one terminal on the speaker.  Remember
that warning not to connect ANY speaker wires to chassis ground?

With a bridged amp, you can get (theoretically -- assuming the amp's
output impedance is zero) FOUR times the power into the load: with
12VPP out of each amp, that's 50.9 watts into 4 ohms.

To visualize how this happens, let's call the "+" speaker terminal V2
and the "-" speaker terminal V1.  An oscilloscope connected from EITHER
V2 or V1 to GROUND will see 12VPP -- but a 'scope connected BETWEEN
V2 and V1 will see 24VPP!  

Try drawing it on paper.  V2 is a sine wave starting at T=0, V1 is out of
phase.  At t=1/4 wave, V2 = +12, V1 = 0 and (V2-V1) = +12.  
        At t=3/4 wave, V2 = 0, V1 = +12 and (V2-V1) = -12.

BTW, with a bridged amp driving a 4 ohm load, EACH amp in the bridge will
"see" a 2 ohm load.

One problem with DC-to-DC converters (other than cost and size) is
that they are power pigs: efficiency goes down in a hurry when 
not delivering max power, as is normally the case with a stereo.

Remember when the U.S.A. had a consumer electronics industry?  Oh, well ...