Output Watts?

[B. Bop]

I was wondering how amp manufactures arrive at their output watt specs. I have been measuring my amps or amps that I work on by putting a resistor of the correct impedance across the speaker output. I connect a scope and a true RMS voltmeter across the resistor. I connect a signal generator set for 400 Hz at about .5V to the input and set the master volume and tone controls to 10. I increase the output of the amp with the volume control until the output waveform just hits maximum amplitude. I calculate watts using E squared / R. Is this the accepted method or should I take the measurement with the amp at maximun output. Also, is there a standard input frequency and voltage used for output watts measurements.

[Rosewood]

The problem with the number of watts is at what percent distortion. I would say most tube amps are at 5 to 10 percent where as a solid state amp would be 1 percent and less.

[PRNDL]

I have come to believe that wattage is (was, or has become) an advertising hype.
There are several completely different methods, and it certainly is not based on how loud the amp is.

[donnyjaguar]

That's the way I do it too except I generally use 1,000Hz. You may want to do spot tones at 66Hz to see what its serving up at the bottom end as many tube amplifiers fall down here.

[Rosewood]

Yea you have some 30 watt amps splitting atoms and some 50 watters laying there like a beached whale. I just look at the transformer sizes and power tube configuration for an indication of watts, but I'm a simple guy. When I see a 2 6l6 amp push a clean sine wave at 50 I call it a good day.

[B. Bop]

I asked the question becasue I am building a KT88 SE amp that puts out about 12 watts clean and 17 watts max. I have seen other KT88 SE amps reated at 20 watts with pretty much the same plate voltage. So I was wondering if the way that I measured watts was different from the way other people do it. I am not really concerned with percent of distortion. From looking at the waveform, even at low output it is not a good representation of the input sinewave. It's a guitar amp not a Hi Fi amp so I am mainly concerned with the sound as it relates to guitar tone. The amp sounds pretty good. I will use the usual verbal cliche's and describe the cleans as being shimmering and bell like and the overdrive as being warm and harmonically rich (LOL). I am not building the amp for sale, I just wanted to build a nice single ended practice amp with enough volume to play with a 15 year old drummer (my nephew). If the single tube version works out I may try to build a gigable version with parallel KT88's.

[B. Bop]

That's the way I do it too except I generally use 1,000Hz. You may want to do spot tones at 66Hz to see what its serving up at the bottom end as many tube amplifiers fall down here.

Bottom end does not seem to be a problem with KT88's. I had to tweek out some of the low frequencies for the amp to sound right and not get muddy when distorted. I use 400 Hz because I think that I read somewere that this is the frequency where speakers develope their correct impedance. I usually check the output from about 2 Khz down to 75 Hz to see how much the mids are scooped and find out where the center point is. On this amp the bass seems to keep climbing as I lower the input frequency below 75Hz.

[Old Tele man]

...take a look at the POWER specifications FMIC puts on their schematics: 1000Hz (guitar) or 400Hz (bass) sinewave signal into a resistive load (not speakers) equal to the impedance value, with all the knobs at full ON position, except VIB and REVERB which are at OFF or ZERO.

P.S.--it's "average" WATTS not RMS watts...yes, the sinusoidal voltage (V.rms) and current (I.rms) will have RMS values, but their resultant product is "average watts"...there ain't no such thing as RMS watts!

[John Phillips]

I take a sinusoidal input signal at around 330Hz (top E string on a guitar - I'll try several frequencies if I'm being fussy) and run the amp into a speaker-emulating dummy load/attenuator until I can hear clipping, then back off until it just cleans up, then measure the AC voltage across the load, square it and divide by the nominal impedance.

Technically accurate? No, not even close, there are too many errors and assumptions. Useful for determining a real-world 'output power' for an amp that lets me tell whether it's working right? Definitely. And since I always use the same load and method for different amps, it's valid to compare relative powers.

[hasserl]

Originally Posted by B. Bop

I was wondering how amp manufactures arrive at their output watt specs. I have been measuring my amps or amps that I work on by putting a resistor of the correct impedance across the speaker output. I connect a scope and a true RMS voltmeter across the resistor. I connect a signal generator set for 400 Hz at about .5V to the input and set the master volume and tone controls to 10. I increase the output of the amp with the volume control until the output waveform just hits maximum amplitude. I calculate watts using E squared / R. Is this the accepted method or should I take the measurement with the amp at maximun output. Also, is there a standard input frequency and voltage used for output watts measurements.

I do it similarly, but use 1K @ .250v (at the generator) and measure the AC across the resistive load when the output wave form just starts to distort. I'll also check maximum output and record it. I like the idea of checking multiple frequencies to plot the output.

[Old Tele man]

...I used to use a sweep oscillator to get a look at the BW "shape" but found it only (a) "pretty" to look at but (b) rarely "useful" (finding "spurious" oscillations.

...now, I just stick with the quasi industry-standard power tests of 1KHz (gtr) or 400Hz(bass) and be done with it...at least then I KNOW all the amps I've measured were done to the "same" spec's...consistency still is valid virture.

[mark norwine]

Originally Posted by Old Tele man

..the sinusoidal voltage (V.rms) and current (I.rms) will have RMS values, but their resultant product is "average watts"...there ain't no such thing as RMS watts!

Thank you! You beat me to it!

[donnyjaguar]

I'm glad OTM brought up RMS watts. That's always been a pet peeve of mine. If you DON'T use the RMS voltage value of an AC waveform the calculations for power are entirely meaningless.

[reaiken]

Originally Posted by Old Tele man

...there ain't no such thing as RMS watts!

Ah, but there you are wrong! I have discovered an ancient secret only previously known by Nikola Tesla and one of the aliens who died in the Roswell crash of 1947, so I can truly rate my amplifiers in watts RMS.

By comparison, average watts are, well...average. RMS watts are exceptional.

Randall Aiken

[mark norwine]

Originally Posted by reaiken

Ah, but there you are wrong! I have discovered an ancient secret only previously known by Nikola Tesla and one of the aliens who died in the Roswell crash of 1947, so I can truly rate my amplifiers in watts RMS.

By comparison, average watts are, well...average. RMS watts are exceptional.

Randall Aiken

And purple tolex makes amps sound better, right?

You told me that once...I do believe