Clean vs Distorted Wattage

[Littlewyan]

When measuring my TW Express I got 35W Clean and 75W Distorted. I haven't measured my 50W Marshall but I've heard good examples put out 50-60W Clean and close to 100W Distorted.

Reason I bring this up is I'm building an attenuator which can take up to 100W and as my Marshall may put out close to 100W I'm a bit worried. I spoke to my Grandad and he said it will be fine as the Clean measurement is RMS and the Distorted is Peak. What does this mean exactly?

[vibratoking]

How exactly did you measure/determine the clean and distorted output power?

[Littlewyan]

I sent a 1Khz Sine Wave into the front of the amp, plugged it into an 8Ohm Load, adjusted the controls until the signal coming out was clean and then measured with my multimeter. For the distorted I did exactly the same thing except I dimed it .

I can't remember what calculation I did exactly but it was one I found on here.

[vibratoking]

I sent a 1Khz Sine Wave into the front of the amp, plugged it into an 8Ohm Load, adjusted the controls until the signal coming out was clean and then measured with my multimeter. For the distorted I did exactly the same thing except I dimed it .

I can't remember what calculation I did exactly but it was one I found on here.

That's a reasonable approach with some caveats. Is your meter a true RMS meter? The RMS voltage of a sine wave is not the same as the RMS voltage of a square wave. At full power the output is probably closer to a square than a sine. Some meters don't measure this correctly. You calculate Prms= Vrms^2/Load. So Vrms must be correct to calculate the power correctly.

With that said, your measurements seem plausible.

[Firestorm]

I tested a parallel SE amp (two tubes) using the same method. I got 14W with 2x6V6, 24.5W with 2XEL34, but 38W with 2x6L6, which is theoretically impossible. The OT (actually two OTs paralleled) were set at 5K for the paired 6V6s, and 2K5 for the ELs and 6Ls. I was surprised at the 6L6 output numbers {and the fact that they didn't sound 6L6ish at all). I think maybe the impedance was wrong (but kind of by the book). Plate volts were about 390 with the 6V6s, about 375 with the others. All separately cathode biased, bypassed. Weird.

[pdf64]

With no power supply sag, a square wave will have twice the rms power as a sine wave of the same peak level.
Tube guitar amps usually have some sag, so the heavier current draw of the max square wave condition acts to pull down its peak level, so that the overdriven power is ~1.5 x the clean power.
Saggy supplies / cathode bias can lower that ratio even more.
50 watt Marshalls can be anywhere from ~30 watts to ~60, depending on whether the VB+ is up near 500 or down below 400.

[xtian]

Yes, peak wattage can be twice clean wattage--but this is for short periods of time when you're actually playing guitar, compared to the continuous duty cycle of a generated sine wave distorted into a square. Any 100-watt resistive load should have no trouble keeping up with a 50-watt amp.

[Stevem]

I have never seen any 800 or 900 Marshall reach its 50 or 100 watt RMS rating even with the best testing output tubes I can stuff in them, tubes that for instance in a Highwatt 50 will make 60 watts RMS !

[Littlewyan]

Its not a true RMS meter its a standard Beckman 3020. I did measure 1.5W out of my JTM1 running full blast which isn't far off.

Firestorm - I measured 75W with TAD EL34B-STRs in my Express and 55W with JJ EL34 IIs in it. Quite a difference! Then again as I'm not using a true RMS meter these measurements may not mean much at all! The clean rating was spot on though.

I need to build myself a load box so I can properly test my amps. Be interesting to see what the Marshall pushes out as its definitely louder than the Express.

So basically when doing number crunching for my attenuator build should I use the RMS wattage? Or the peak wattage that I measured?

[Firestorm]

Technically, there isn't such a thing as RMS watts (despite what government labelling authorities insist). What you are measuring is AC volts across your load, and for this you want the RMS voltage. The quality of the meter is critical. My True RMS Radio Shack DMMs were useless for this; only my Flukes produced usable results. If you are confident in the accuracy of your meter for Vpp results, multiply by 0.3535 for the RMS value.

[Structo]

RMS watts

Peak watts

Peak to peak watts

Music Power watts?

The audio world has always dreamed up new units to measure amplifier output.

Unless you can measure live speaker driven power, then most of these things are meaningless.

Ever heard that tube amp watts are louder than solid state watts?
Oh yeah.

How about a scale?

That's friggin' loud!

What?

[vibratoking]

RMS wattage is mathematically defined and makes good sense to me. Much more relevant than peak wattage IMO. I'm not sure why RMS wattage doesn't exist?

[Firestorm]

RMS wattage is mathematically defined and makes good sense to me. Much more relevant than peak wattage IMO. I'm not sure why RMS wattage doesn't exist?

Peak power. Continuous average power. At what frequency? The term RMS watts was dictated by the FTC to rein in power claims by amp manufacturers (Hi-Fi, not MI). But RMS is a voltage related term, not a power one.

[xtian]

Folks said the same thing when I started collecting this data--measuring output wattage is relative and mostly meaningless. However, after measuring with the same methodology a growing number of amps, it becomes a diagnostic tool, like many others. I started compiling a database of type of amp, number of power tubes, plate voltage, fixed vs cathode biased, clean output wattage. It's helpful info!

[pdf64]

Be interesting to see what the Marshall pushes out as its definitely louder than the Express.

So basically when doing number crunching for my attenuator build should I use the RMS wattage? Or the peak wattage that I measured?

At the moment you're just guessing, you really need a resistive dummy load and true rms meter to put valid numbers on it.

But if we assume that the Express really is 75 watts and the Marshall even louder, then I suggest building an attenuator rated for 200 watts.
As a key benefit of an attenuator is being able to crank the amp, but if your nose is constantly twitching to detect smoke escape whilst cranking, it won't be much fun.
OK, as xtian notes, the time averaged power most of your playing will be <1/10 of that.
But when you go for the Parisienne Walkways solo and that note sustains into feedback, the amp may kicking out peak power for a long time. For the attenuator to keep its cool, it would be beneficial for its rating to be twice that power.