True power output?

[Arjank]

Always nice to read these "amplifier" threads, I won't comment here since my field of "knowledge" is loudspeakers.
I wouldn't bring anything useful to the discussion.

 

[cfsporn]

QSC does list thorough, meaningful specs on their stuff, they are to be commended.

Yes, but dont they have to?

 

[ThisBass]

Thank's Andy for your great post.

There are a few manufaturer like QSC who provide numbers for current draw at the line socket and thermal charge of the amplifier in relation to a realistic (typical) average output.
IIRC these numbers are taken with pink noise at 1/8 and 1/4 (or 1/3) of sine output 1%THD.
Although crest numbers for the pink noise are not stated, it belongs then to the user to estimate appropriate crest number for the program material that shall be amplified by "full use of headroom" over time.
I try to avoid the term "full power" in this context because realistic signals can't rarely produce "full power".
There are only a few which really could do that! For example MI signal with lots of dirt.

Indeed if every manufacturer would follow FTC principles then everybody could easily compare apples to apples, but IMO THAT could end up with very fatal results. In this case some pro audio manufacturer had to reduce power ratings drastically downwards in a way that would make it impossible to match loudspeaker systems to amplifiers.
Some pro class-D amplifier provide four times of power burst capability versus countinous power.
If burst rating was not provided then it's impossible to estimate how much noise a amplifier may produce in combination with a distinct program material. The main problem would be it was absolutelly impossible to select speaker systems with sufficiant power handling.

IME a amplifier with well balanced burst versus contionous makes same noise and average power like a amplifier that equals a "balanced" burst number with continous rating. A lot of folks do careless underrate the potential of burst. They do underrate it careless because the nature of audio program is to often poorly understood. It's not only dynamic and transient that benefits from burst, it may also be a steady signal with high crest number that can benefit from burst.

I claim if two different amplifiers were strained with a 100Hz sine signal that in addition contains some even order Harmonics to obtain a crest number of 2 or 3 then there was (nearly) no difference in performance.
That's what I'd like to see in addition to "conventional" ratings. Burst with sine wave AND burst with signals of higher crest (than sine) to "emulate" a more realistic "worst case" scenario regarding potential of average output power.

Of course every manufacturer handles the burst versus continous proportion different. It's a trade off. Some give more weight to power whereas others give somewhat more respect to stability over time. That's no cause to blame a manufacturer X to gain unwarrented more power then manufacturer Y with less burst power but longer time constant for the burst. As music is so very various and individual and very complex regarding it's nature/physics so it's no surprise that "trade offs" may differ a little bit sometimes (on different products).

 

[ThisBass]

An interesting paper what might happen if FTC standard was the only one and if every manufacturer would regard the standard. IMO this paper is very informative anyway
Link Removed

 

[agedhorse]

Seamonkey - Lab Gruupen doesn't rate their larger amps at any continuous rating as you might think. I suggest that you read up on the foldback power limiting technology that they use, as well as every single manufacturer of power amps in this category.

They do not deliver rated power continuously, and IMO it makes no real world difference at all. What it does do is reduce size, weight, power draw, as well as greatly increase the reliability of both the amp and speakers.

Also, every single pro level powered speaker manufacturer does the same thing.

Continuous power ratings are going the way of the Dodo bird, because it's not meaningful in today's world. The question is what to replace it with?

 

[drpepper]

An interesting paper what might happen if FTC standard was the only one and if every manufacturer would regard the standard. IMO this paper is very informative anyway
Link Removed

Thanks for posting the link. Clearly, a number of you are more technically versed than I am, but I read this with interest. What stood out for me was the statement “you will hear the difference between 400 watts to 800 watts, a 3 dB increase, but you won’t hear a difference between 400 watts and 440 watts (+0.4 dB)”

I assume this was a simplification for the purposes of brevity and clarity for the intended audience, but I see it leading straight back to one of the core issues here; this is the kind of statement that leads to the paradigmatic rut that Tom Lees was apparently trying to address…ultimately, there is no such actual thing as “watts of loudness” between two separate systems.

Yes. You “will hear the difference between 400 watts to 800 watts, a 3 dB increase” in the same or identical systems. Between two different systems, this may hold true, or it may not. No. You “won’t hear a difference between 400 watts and 440 watts” in the same or identical systems, but you may or may not hear a difference between 400 watts and 440 watts between two different systems.

Using the same measurement of wattage, demonstrably, you may hear a difference between 236 watts and 440 watts, and find that the difference is that the 236 watts are louder.

 

[seamonkey]

I remember Bob Lee talking about a project for unified, meaningful spec sheets amongst all builders but I can't find back the info.

Yes, Bob posted a link to CAF Link Removed

 

[Arjank]

In the past I have used two PA amps for my live setup. One was a Bell 9024 PCX, the other a Behringer EP2500 (which got good reviews back then). Although both had approx the same power rating the Bell was way more dynamic, lively and it could get (significantly) louder then the Behringer. I wish I still had the Bell, it weighted a ton but as far as sound quality goes it was top notch. Nowadays I use a "head" , the Bugera BVP5500. On higher volumes it doesn't have the dynamics of the Bell but it definitely sounds closer to the Bell then the Behringer.

 

[alaskaleftybass]

In the past I have used two PA amps for my live setup. One was a Bell 9024 PCX, the other a Behringer EP2500 (which got good reviews back then). Although both had approx the same power rating the Bell was way more dynamic, lively and it could get (significantly) louder then the Behringer. I wish I still had the Bell, it weighted a ton but as far as sound quality goes it was top notch. Nowadays I use a "head" , the Bugera BVP5500. On higher volumes it doesn't have the dynamics of the Bell but it definitely sounds closer to the Bell then the Behringer.

What do you think of the BVP5500? Sound? Reliability? etc.

 

[/\/\3phist0]

Wait a minute seamonkey, uou have to remember that the amps designed for this purpose are also rated for a power bandwidth of 800Hz - 4kHz, and I recall the THD+N at <20%. Going from memory, as I don't have a subscription to this particular UL standard, there is also a duty cycle that's included in the test that is consistent with the voice evac. standards. I also recall that the amp must perform at 24 volts DC (or the inverter equiv. which must also be listed) and deliver this spec as well... So the standard is REALLY unsuited for a bass guitar amplifier.

You can't arbitrarily apply a spec. without understanding the details behind the spec. and this spec is for a really unrelated application.

Dam. Andy you are one smart cookie I learn something from every one of your posts. Thank you for sharing these Gems. Seriously. .thanks

 

[agedhorse]

I have spent most of my professional life (>35 years) designing products that focus on real world performance rather than leaning on boiler plate standards as a cop out.

If you are going to suggest and quote standards then you better darn well understand exactly what you are proposing, and if the standard doesn't apply to the product (as in this case) then it further supports my previous comments that providing detailed (and accurate) information just confuses the average amp purchaser.

Another spec. That I think is very descriptive of real world performance but is rarely discussed (nor is it included in the CAF) is the ability of an amp to source and sink current into a sub-nominal reactive load (read as a REAL WORLD LOAD). If this current capability is inadequate, the amp will (at best) prematurely voltage limit on current sense, a limit with particularly nasty phase related artifacts. At worst, the amp will release it's magic smoke.

Incidentally, this is one of several key factors I have focused on in some new designs I am working on.

 

[agedhorse]

Ok, here's another thought related to pigeonholing to inappropriate standards.

Let's say, for example, that I was modeling (in either DSP or analog domain) a particularly famous and desirable mildly overdriven tone for a new product. Let's also say that I have several of you on the forum serving as my test players and together we fine tune the "perfect" tone. We them find that upon measuring that the distortion (THD) always exceeds say 5% because of the way the preamp and power amp is tuned. How do we specify power if the power must be specified by a standard at less than 1% (or 0.1%) when you test players "specified" the higher THD as ideal?

For some amp approaches and specific models/applications, this is exactly what might happen. So how should an inappropriate standard apply in this case?

This is the sort of challenge real engineers encounter.

 

[Arjank]

What do you think of the BVP5500? Sound? Reliability? etc.

I really can't complain about the BVP5500, for the price it's a steal. Soundwise, it's neutral (that's what I want) but the tone controls add more then enough flexibility to dial in anything from dub/reggea tones to flat out sizzle fusion. I did swap two of the four tubes though, installed some NOS Philips tubes, it gives a more hifi vibe to the sound detailed but still warm. I once compared it to an Eden (Traveler) pre-amp + master audio power amp, the BVP5500 held it's own against that combo but of course on extreme volumes the BVP5500 (I really pushed it to it's limits and it was plenty loud)had less control on the woofers since the master audio power amp has more then twice the rated output power.
Reliability, I own it for more then 4 years now and it's still going strong.

 

[teemuk]

Let's say, for example, that I was modeling (in either DSP or analog domain) a particularly famous and desirable mildly overdriven tone for a new product. Let's also say that I have several of you on the forum serving as my test players and together we fine tune the "perfect" tone. We them find that upon measuring that the distortion (THD) always exceeds say 5% because of the way the preamp and power amp is tuned. How do we specify power if the power must be specified by a standard at less than 1% (or 0.1%) when you test players "specified" the higher THD as ideal?

I mentioned it earlier, but many manufacturers "solve" this issue simply by rating the power amplifier only and ignoring effects of additional "signal processing" by inboard preamps, limiters, compressors, etc.

They could probably quote honest ratings (as per some particular spec.) but the numbers wouldn't be too impressive. And let's face it, it's the bigger numbers that generally tickle interest of people. 50W @ 1%THD doesn't seem nearly as impressive as 400W @ 5% THD, even if both of those ratings were derived from the very same amp.

Yep, musical instrument amps are treated more like effect processors instead like conventional amps. There's much more "subjectivism" involved: e.g. Distortion and clipping is not only preferred, it serves as a musical effect. ...But then we have subjective opinions about what kind of distortion "sounds good", and in what particular applications. Let's not even wonder to signal compression or overall EQ:ing that works with the musical instrument the amp is supposed to complement somehow. It's hard enough to derive meaningful ratings for conventional amplifiers, probably impossible for amps that actually serve as effect processors.

 

[Arjank]

I mentioned it earlier, but many manufacturers "solve" this issue simply by rating the power amplifier only and ignoring effects of additional "signal processing" by inboard preamps, limiters, compressors, etc.

That sounds logical to me.
But it indeed gets tricky when the "standard" "tone" of the amp already has 5% of harmonic distortion, it would be easier to specify with Hybrid amps (tube pre and SS or Class-D power amp section).

 

[JimmyM]

I have spent most of my professional life (>35 years) designing products that focus on real world performance rather than leaning on boiler plate standards as a cop out.

If you are going to suggest and quote standards then you better darn well understand exactly what you are proposing, and if the standard doesn't apply to the product (as in this case) then it further supports my previous comments that providing detailed (and accurate) information just confuses the average amp purchaser.

True enough. I certainly make no great claims to understanding every spec I see, and at the end of the day, I just want a rig that sounds good and can put out at the volume level that I need.

BUT...

There's got to be some sort of line in the sand, no? Like some manufacturers quote wattage specs with the level of THD they're measured at. That to me is perfectly acceptable, while claiming a certain wattage just because it "sounds like" an amp with a certain wattage isn't. But if there's one thing I've learned from discussions like this, it's that any new standards are not going to be derived from my input

 

[Chef]

Howdy Guys:
thanks for the many interesting and informative posts in here since I was hailed in last time. It's nice to see that we can, in fact, discuss some "historically difficult topics" in civil ways.

thanks.

 

[agedhorse]

Now, for Jimmy's sake (note that I am NOT picking on Jimmy here), let's take "another of his favorite amp models" with a power amp that happens to be very low feedback and cathode biased. Let's say, for discussion purposes, that the power amp can do 5 watts at 1% THD, but also does 20 watts at 4% THD, and 50 watts at 10% THD.

How do you rate this amp so that it is representative of the performance it delivers rather than a number that looks dismal to those who do not understand what is going on that is responsible for a particular and highly desirable tone?

Often, a tremendous effort goes into developing both a tone profile as well as feel feedback through the player's fingers. Much of the time, the results are somewhat counter-intuitive unless you understand the complete process, and how the variables inter-relate with the results.

One thing I have always done is to study historical designs and the philosophy of the designers responsible. As an engineer, I tend to resolve these approaches using numbers and formulas but I never forget the original process and always respect those designers efforts because of how they were earned.

It's somewhat ironic that am now working for Mesa, as Randy Smith is one of the designers I have studied for decades. His approach is similar, though with his experience, he relies more on the gut instincts and experience from the guitar world whereas I reinforce my gut with numbers. When I "interviewed", I probably had a better understanding of how he arrived at his results than he did, because I had analyzed the historical paths to his approaches where he just went with his gut and hit home run after home run.

 

[Martin Beer]

Now, for Jimmy's sake (note that I am NOT picking on Jimmy here), let's take "another of his favorite amp models" with a power amp that happens to be very low feedback and cathode biased. Let's say, for discussion purposes, that the power amp can do 5 watts at 1% THD, but also does 20 watts at 4% THD, and 50 watts at 10% THD.

How do you rate this amp so that it is representative of the performance it delivers rather than a number that looks dismal to those who do not understand what is going on that is responsible for a particular and highly desirable tone?

Would it be useful in that case to look at the point at which distortion begins to increase beyond a certain rate as the power increases? (Genuine question, as I don't know). I've noticed that where power amp manufacturers publish a graph of %THD against power output, it tends to increase gradually up to a certain power level, after which it tends more and more towards the vertical. So in an amp with some distortion as part of the desired tone, would they have rated it at the point where that corner occurs in the graph?

 

[JimmyM]

Now, for Jimmy's sake (note that I am NOT picking on Jimmy here), let's take "another of his favorite amp models" with a power amp that happens to be very low feedback and cathode biased. Let's say, for discussion purposes, that the power amp can do 5 watts at 1% THD, but also does 20 watts at 4% THD, and 50 watts at 10% THD.

How do you rate this amp so that it is representative of the performance it delivers rather than a number that looks dismal to those who do not understand what is going on that is responsible for a particular and highly desirable tone?

I think in the designer's case, he just went by what the RCA Tube Manual specified as he was very seat-of-the-pants, was drafted into amp design and sent to electronics night school by Everett Hull based upon a chance meeting, lifted the design straight out of the RCA manual, and was more of a gut-instinct type like Randy Smith. Interesting to note that the Heritage B-15 manual specifies its cathode bias setting as 25w RMS @ 5% THD and 30w @ 5% THD in fixed bias mode. No attempt is made to hide what role the THD plays, and while someone looking for an amp that has < .0001% THD might not be interested, I think it's a fair and honest representation that reflects the currently accepted standards.

Often, a tremendous effort goes into developing both a tone profile as well as feel feedback through the player's fingers. Much of the time, the results are somewhat counter-intuitive unless you understand the complete process, and how the variables inter-relate with the results.

One thing I have always done is to study historical designs and the philosophy of the designers responsible. As an engineer, I tend to resolve these approaches using numbers and formulas but I never forget the original process and always respect those designers efforts because of how they were earned.

It's somewhat ironic that am now working for Mesa, as Randy Smith is one of the designers I have studied for decades. His approach is similar, though with his experience, he relies more on the gut instincts and experience from the guitar world whereas I reinforce my gut with numbers. When I "interviewed", I probably had a better understanding of how he arrived at his results than he did, because I had analyzed the historical paths to his approaches where he just went with his gut and hit home run after home run.

Another very good real-world example. I agree that there are complicating factors, but I still think it can be done in a reasonable and relatively honest manner that gives people the idea of what to expect without getting busted in a lie.