Relatively meaningless question/opinions

[Earl]

The term watts "RMS".

Valid, or slang?

[black_labb]

root mean square

its the standard way of showing the wattage of an alternating current. root mean square is how it is calculated. its nessisary as an ac voltage will flucuate periodically between +x and -x volts, where x is the peak voltage. the average voltage is 0 volts, but we both know that putting your finger into a wall socket is a bad idea as its not 0 volts. 220 volts rms is somthing like 317 volts peak (thats when its a sin wave)

[MWJB]

Hearing perception being what it is, it's just a number that gives a rough guide as to output. There are so many other factors going on (speaker efficiency, voicing, envelope) that it's only really relevant when comparing like amps with like.

I've never seen a guitar player turn down and quote that he did so because he was exeeding 5% THD.

Isn't it odd that (virtually) nobody officially makes a 13W or 17W amp?

You could just as easily assume W figures based on no. & type of tubes, rough B+ & method of bias...I'm sure plenty of builders do exactly this.

It is usual for output figures to be quoted as W RMS, rather than peak.

Diagnostically though, it can be a useful reference & indicator of a fundamental problem if voltages into a resistive load are available.

It's a bit like BHP with cars & bikes...a rough yardstick, open to some massaging as not everyone uses the same procedure to get to the magic number.

[Sock Puppet]

Root mean squared as black labb says.

You take a succession of samples across a waveform, square each one then take the square root of the arithmetic mean which for a sine wave is the peak voltage divided by the square root of 2 (.7071 x Vpk). This A.C voltage is the one that will cause the same heat as a D.C. voltage of the same value when applied across a resistive load.

If you have 10W r.m.s. then that is exactly what is dissipated in the load, resistive or otherwise whereas 10W peak or peak music power means nothing.

S.

[Earl]

My thoughts are that since voltage and current (AC) are cosine waves, the S(t) functions apply....but power is not a "wave", just a number.

Not that it makes any difference, but I would think that 'average' power would be more descriptive, than RMS.

[Bruce / Mission Amps]

My thoughts are that since voltage and current (AC) are cosine waves, the S(t) functions apply....but power is not a "wave", just a number.

Not that it makes any difference, but I would think that 'average' power would be more descriptive, than RMS.

My customers are always very disappointed when I plug their 50 watt, super-duper, belch fire, woof-woof amp it into a pure resistive dummy load, monitor the output on the scope for zero distortion, read the voltage across the resistor with my "RMS reading" voltmeter, do the math and they find out that their amp only makes 32-37 watts..."RMS" watts at 85HZ, which is about 4 times the frequency of what a Hi-Fi amp could do.
Then I show them what their fantastic sounding, "I think I could almost gig with this", Champ does... and that thing makes an amazing 1.7 watts at 85Hz!! ha ha

[Earl]

What about this -

Volts RMS X Current RMS = Watts RMS

then,

Volts RMS / Current RMS = Ohms RMS ... ?

[Old Tele man]

...the RMS term does NOT apply to WATTS

...the RMS term applies to the VOLTAGE and CURRENT values, specifically as a mathematically operation to determine 'effective' (DC-quivalent) value from the "peak" of a sinusoidal shape.

...power is AVERAGE, as in "time-averaged" over one sinusoid "cycle."

...pertinant reading: http://www.eznec.com/Amateur/RMS_Power.pdf

[loudthud]

As Sock Puppet pointed out, it's the (square) Root (of the) Mean Square. What's important about "RMS" values is that and Watt's Law (is that a real law?) still works. One volt RMS across one ohm will cause one amp RMS to flow and the resistor will dissapate one watt. The RMS value of an AC voltage will allow us to predict how much power will be dissipated by a purely resistive load.

For a DC voltage E, if we want to calculate how much power P will be dissipated by a resistor R connected to it (and we are too lazy or unable to measure the current), the formula is P = E^2 / R. If we want to go back to calculate E from P, the formula is E = square root of (P*R). Note that if E was negative to begin with, we have lost the sign. The Mean or average value of DC is just the DC value so in effect the RMS value of a DC voltage is just the DC voltage.

If you ever took calculus and couldn't figure out a practical application, here it is:

Imagine the unit sine wave. What do you get if you square the Y value of every point? Over the first 90 degrees, you get sort of a S shaped line that is always below the sine wave line. (Square a number between zero and one and you always get a number smaller than what you started with.) It meets the sine wave at X=0 and X=90 degrees. Closer examination reveals it's a cosine wave from 0 to 180 degrees! Over the next 90 degrees of the sine wave, the squared value wave makes a backwards S curve. Over the portion of the sine wave that goes negative, the squared wave stays positive completing a second cycle of a cosine wave. The Cosine wave is twice the frequency of the sine wave, it's half the peak to peak amplitude and offset by 0.500 .

Now the Mean value of the our cosine wave is 0.500 . The square root of 1/2 is 1/SQR(2) or about 0.707 .

(sin(t))^2 = (1/2) + (cos(2t))/2 or something like that I can't remember

[Old Tele man]

P.pk = (V.pk * I.pk)

..converting "pk" to "DC-effective" via RMS-sinusoid conversion:

V.eff = V.pk/SQRT[2]
I.eff = I.pk/SQRT[2]

P.avg = V.pk*I.pk/(SQRT[2]*SQRT[2])
P.avg = V.pk*I.pk / 2 = W.avg (ie: "time-averaged" power)

P.avg = V.rms * I.rms = W.avg (ie: "time-averaged" power)

...NOTE: today's microprocessor-based DVM's actually perform "sample-by-sample" ROOT-MEAN-SQUARE conversion on the fly and so do not assume a sinusoidal input...as the "rule-of-thumb" value 0.707 implies!

[Enzo]

Volts RMS / Current RMS = Ohms RMS

Nice try. But "RMS" is not a unit of measure, it is descriptive of the units of measure. In the same sense that when you say it is 10 degrees cooler, although you have specified the units of measure - degrees - you still have to specify the type of degrees - centigrade, fahrenheit, Kelvin. So if you went from 100 degrees F to 90 degrees F, you would not say there was a reduction of 10%F. 10 degrees F yes, but not 10%F.

[Steve Conner]

The term "RMS power" is really just slang shorthand. As other posters pointed out: only voltages and currents can be RMS. Power is always just average or instantaneous.

If you compute Vrms * Irms into a resistive load, you get average power. So when you say Watts RMS you're just saying that it's average power and not some cheesy PMPO figure, got by taking the maximum instantaneous power during a lightning strike to the apparatus

I rated my home-made amps by measuring how much sine wave power they would deliver into a dummy load resistor before starting to clip visibly on a scope. I wouldn't run the test longer than a few minutes, though. When you ask how long the test should last, you open a whole new can of worms.

PS: If it was 60'F yesterday and it's 30' today, does that make it twice as cold? How many percent colder if it was 20 degrees yesterday and it's -3 today? Can you see why people use the Kelvin and Rankine absolute temperature scales for stuff like this?