Watts?

[diagrammatiks]

i believe what you are saying would disprove feedback theory.

[GregS]

I found a thread on here already discussing solid state vs tube watts. So there is technically speaking, no difference between the two. Watts is watts. Without opening that whole can of worms and having a redundant post I have a couple q's. I recently fixed an SVT Pro 7. It is a great sounding amp (class D). However, it certainly did not sound like it was 1000 watts when I plugged my guitar into it. I was using a 4ohm speaker too. Why would this be? Fyi I typically find bass amps like GKs and Hartke to also seem like they are NOT as loud as their stated wattage.

So I ask this question because I'd like to build a hybrid bass amp. Tube pre, ss PA. My bassist is saying he needs 300watts at 4ohms for the gigs we do. (we're a loud blues/rock/r+b band). He is obviously saying this based on the fact that his Hartke head is 500 watts and works well on our gigs. If I were using LM3886 PAs (ha ha!) I would need 6 of 'em driving 6 separate 4ohms speakers!!! Maybe this is incorrect and someone will enlighten me on commercial ss bass amps' wattage ratings. This seems like a ridiculous project so if need be I will build a discrete 300watt PA.

I suspect the high wattage amps aren't putting out nearly their rated wattage with normal signal levels input. The gain within the amp may not be enough to get anywhere near the limits of the power amp headroom with a normal level of input signal. However, if a very large spike in input came through, it would more likely pass through the power amp completely undistorted with the headroom of a high wattage amp.

I think to bass players (who have some understanding of this stuff) high wattage means assurance that they will always have a firm, tight response and not encounter mush. Obviously, the louder the band the more headroom (wattage) required to avoid mush.

Also, I suspect the really high wattage ratings are more of a selling point rather than a practical necessity. With solid state power amps, this can be done at a reasonable price and weight.

Greg

[Jazz P Bass]

The Ampeg SVT-7 is "rated" at 1000 watts into 4 ohms.
600 watts into 8 ohms.
My question to Lowell would be :
"When you final tested the amp, what is the maximum unclipped RMS Vac signal that was observed at the output, assuming you used a dummy load with a 1000hz signal.?"
65 Vac at 4 ohms is roughly 1000 watts.
70 Vac into 8 ohms is roughly 600 watts.
(which begs the question" what did you use for a 4 ohm, 1000 watt dummy load?")
(I use Watlow industrial heater strips)

[diagrammatiks]

i'm sure teemuk is going to disagree...but the amount of available power after spec sheet maximum is for a transistor is nearly 0. Using that same amp the loudness available at 1000watts will only be twice as loud as 100 watts.

You probably hit 100 watts at around 2 on the volume of the amp.

[teemuk]

"When you final tested the amp, what is the maximum unclipped RMS Vac signal that was observed at the output, assuming you used a dummy load at with a 1000Khz signal.?

Pretty much this. The amp runs two class-D switching amps in bridge, each powered by pair of 55V rails. Technically, from that respect there should be no issue why the amp couldn't produce 1000 watts of output power.

I'm still wondering though how one can "hear" 1000 watts since power never has been a measurement of loudness. The difference to 100 watt amp isn't astounding and mostly shows itself as an increased clarity of sustained low notes or transients. Not so much as sheer loudness. Overall, a clipping or compressed 100-watt amp may easily sound louder if the average signal levels are higher.

Also, I suspect the really high wattage ratings are more of a selling point rather than a practical necessity.

No. They actually do make sense in applications such as PA or bass that tend to have a need for clean amplification, often extending to quite low frequencies and having a large varience in signal dynamics. High output power isn't as much a neccessity with things like guitar amps operating on a narrow band of frequencies and often deliberately overdriven to distort.

Power demand raises exponentially. Hence jumps from eg. 10W to 100W and from 100W to 1000W are not extremely astounding loudness-wise, but keeping the signal intact and clean requires huge amount of power from the amp.

A cranked 30W guitar amp can easily hang with a 300W bass amp if the intention is to play with more or less distorted tones. On the other hand, that same 30 watts for bass would be pretty much unuseful.

[diagrammatiks]

well it would matter where and how they calculated the wattage right?

If it was true wattage 100 v. 1000 sustained output of a given frequency there should be a volume difference...

but actually playing...i'd imagine that it would be more of a increased clarity and speed of response to low frequency transients.

[kg]

Yes and no. Depends on how you define "fall apart". The amp will try to use some or all of its open loop gain to correct the difference it sees between input and negative feedback input. In other words, the feedback loop is still doing its exact job but you usually run into problem of combination of high open loop gain and requirement for high magnitude error signal pushing things into clipping as well.

during a hard OUTPUT STAGE clip, (ie, contained withing the global loop) i can assure you that the NFB is doing absolutely nothing to decrease output impedance.

[lowell]

Can 'o worms. It was unavoidable I guess. Seriously thanks to everyone for the input. Most if not all points are good ones. Especially the point about speaker systems. A 1000w amp is obviously gonna sound louder because of the need for more speakers. I should have done what JPB suggested which was to measure the output(and input) voltage. I thought about all this after the customer took back the head. A few more q's.

Teemuk if the rails are 55v does that not mean that the maximum output voltage is 55v? In my LM3886 amp the rails are 30v. Does this mean that the maxium output (and input) voltage is 30v? Obviously I'd want to stay clear of this as to avoid clipping. Do solid state emitter follower PAs have gain? I'm pretty sure they do, and it comes from the VAS. I don't think that the LM3886 has voltage gain though, just current gain. ??

[teemuk]

if the rails are 55v does that not mean that the maximum output voltage is 55v

No. Circuit's limitations (such as voltage drops across transistors) prevent the amplifier from swinging to full rail potential. My point was merely that those 55V rail potentials are high enough for a bridged amp to produce 1000W (and more) output power to 4-ohm load, naturally assuming the power supply can sustain those potentials.

In my LM3886 amp the rails are 30v. Does this mean that the maxium output (and input) voltage is 30v?

No. Consult the datasheet, there are graphs for clipping voltage versus supply voltage, for both 8-ohm and four 4-ohm loads. The chip clips approximately 2.5 - 4V below the rail potentials, depending on loading.

Do solid state emitter follower PAs have gain? I'm pretty sure they do, and it comes from the VAS. I don't think that the LM3886 has voltage gain though, just current gain. ??

What's an emitter follower PA? LM3886 is a high power op amp, it has both enormous (open loop) voltage gain (approximately 115 dB) and a high current gain.

[jmaf]

Teemuk if the rails are 55v does that not mean that the maximum output voltage is 55v? In my LM3886 amp the rails are 30v. Does this mean that the maxium output (and input) voltage is 30v?

In linear amplifiers it is. I don't know about class D and other pwm or digital technology.

In car audio for example 4 ohms is the standard so with the 12v supply from the battery the max you can achieve is 36 watts. The big Rockford Fosgate amps need to oscilate the 12 v with a high frequency, step it up to a higher voltage via a transformer and then rectify that and work with the higher voltage to be able to push more than 36 watts out.

[GregS]

Also, I suspect the really high wattage ratings are more of a selling point rather than a practical necessity.

No. They actually do make sense in applications such as PA or bass that tend to have a need for clean amplification, often extending to quite low frequencies and having a large varience in signal dynamics. High output power isn't as much a neccessity with things like guitar amps operating on a narrow band of frequencies and often deliberately overdriven to distort.

Power demand raises exponentially. Hence jumps from eg. 10W to 100W and from 100W to 1000W are not extremely astounding loudness-wise, but keeping the signal intact and clean requires huge amount of power from the amp.

A cranked 30W guitar amp can easily hang with a 300W bass amp if the intention is to play with more or less distorted tones. On the other hand, that same 30 watts for bass would be pretty much unuseful.

Yes, I understand these concepts. No argument here. What I meant by the high wattage being more of a selling point rather than a practical necessity is most bass players are playing small clubs and venues where 200W-300W is more than adequate headroom to remain clean - especially if they have a feed into the PA as well.

For players who understand the concept, more wattage is still better as it is extra insurance they won't run out of headroom in any situation. Others that don't get it are still attracted to it even if they don't need it as they believe somehow it will make them sound bigger or something like that.

[NorCalTuna]

Ok, my .02. There may be a fundamental confusion going on here between watts, db, phons and sones. Yes, there has been a fifty+ year controversy about the perceived ability of a tube watt vs a silicon watt to push paper, but that's one of two things going on.

Somebody said that nobody has ever heard a watt, which is spot on. It's literally an expression of the electromotive force on the voice coil of the driver. It has almost nothing to do with the output, as the drivers eff will be that determining factor. That gets us from watts to sil, or db, but not to our consciousness yet, which is what were looking for (unless there are robots among us). DBS follow the fletcher munson curves, so that 60db at 1khz makes 60 phons. Now, sound is so relative, that some smart guy (Stanley smith Stevens) in the 30s decided to try to describe it better, and he proposed the sone. You may have seen it on the box of your bathroom fan. Now, not counting for critical bands in your cochlea, the sone sums up things pretty well, so that at 1khz, 40 db is 40 phons is 1 sone. My 60 db up there is going to be 4 phons.

That's 1khz, while the 12 fret e on the guitar is 629 hz. However, ALL this notes harmonics, which in an overdrive amp is the whole point, are well above that. Bass, though, is very different. Taking your 110hz a string now, let's assume the amp and speaker are linear, and will make the same voltage and db at 110hz than 1k. It's a miracle! And makes 60db again. What is that to your consciousness though? It's about 50 phons, or 2 sones. It's half as loud! The decibel is nothing but the measure of energy in the air, related to w/m^2 ( db= 10 log sil/10^-12) and for perception purposes incomplete without frequency. The watt is perhaps, to the ear, absurd.

[printer2]

I'm still wondering though how one can "hear" 1000 watts since power never has been a measurement of loudness.

Actually the proper measurement is Watts per Meter Squared, W/m2.

Sound Intensity and Decibels

The dB, and the loudness level we associate with the number, is only a ratio of power compared to the threshold of our hearing. That threshold is approximately 1*10-12 W/m2.



I know, others hate me too.

[NorCalTuna]

To unpack that one step furthur, that's watts of sound, actual oscillating air. 1 w = 120 db (10 log 1/10^-12) but speakers aren't gonna allow 100% transduction, more like .6-1% Take the ev 12l. It makes 100 db with a watt, so we find the ratio of elec. energy in and acou. Energy out. 120 db is 100x more intensity (sil) than 100, giving us roughly 1% eff. That's what arrives at your ear, what happens next is even worse!

[printer2]

To unpack that one step furthur, that's watts of sound, actual oscillating air. 1 w = 120 db (10 log 1/10^-12) but speakers aren't gonna allow 100% transduction, more like .6-1% Take the ev 12l. It makes 100 db with a watt, so we find the ratio of elec. energy in and acou. Energy out. 120 db is 100x more intensity (sil) than 100, giving us roughly 1% eff. That's what arrives at your ear, what happens next is even worse!

They hat you too, huh?