• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Feasibility of getting 15W out of a SET tube design?

It’s amazing how little sonic difference there is between 10 and 13 watts or 12 and 15 watts. The amount of work and expense to get just that tiny bit more power is really a huge jump. As the kids say, is the juice worth the squeeze? If your speakers are on the higher efficiency side, above 94db you can hit over 104Db with a 10 watt amp, that’s loud!
With 100dB speakers, one watt will run you out of the room.
 
  • Like
Reactions: 3 users
Mhhh, good high efficiency speakers cost $$$$ as well and are big. Without a proper large room they will never perform at their best. This is a point to take into account for sure.
15W amp with the 845 at 750-800V/80-85mA is not that expensive if using good and selected Chines tubes. 5-6K OPT pretty standard too. With 750V supply a disposal there are endless options for the driver. A few years back I would have preferred the PSE route but nowadays high quality, high voltage DC-Link caps are affordable. Antek has some really good and affordable power xformers for this application. Not much difference respect to a 8-9W 300B SE really, except for the higher voltage.
Instead there are some really good bargains among loudspeakers (from monitors with 83 dB/1W to 2-way towers with 87-88dB/1W) from a few years ago that match really well with tube amps. Mostly British stuff.....
 
Last edited:
  • Like
Reactions: 1 user
For around $500, there are new bookshelf size loudspeakers that sound very good that are 93db and above. You can also just as easily buy higher efficiency speakers used as those 83db models. It makes no sense to me to try to force some older low efficiency speakers designed for high power SS amps into service with a single ended tube amp. If someone insists on using low efficiency speakers, build a PP amp. As someone else noted, the difference between 10W and 15W isn't going to get you there.
 
  • Like
Reactions: 3 users
If you are talking about speakers with cheap horn tweeZers, I cannot agree. Quality high efficiency is not cheap.
Let's put things into perspective, a 100 dB speaker is low efficiency too from a general perspective. Need to get close to 110 dB to call it high efficiency. 100 dB/1W means 2.5% efficiency. The standard speakers I talk about cannot stand high power, they were just designed for home use at civilized level. They use high quality (true) Vifa or Audax drivers (which today would cost quite some money) and have properly designed and uncomplicated cross-overs. They are just good as they are but once upgraded with modern high quality caps they get to even better level. With a bit more money one can also find stuff using ScanSpaek drivers....
Listening level depends on personal habits but it's undeniable that beyond 80 dB average level our earing system goes into protection mode. Listening at higher average level is like running with the brake on. The sense of using 100dB speaker with 1W is not the max SPL but the lower amount of tradeoffs to accept for 1W amp respect to a 15W amp. But the speakers need to be really good and need a lot of space.
The difference between a properly made 845 amp and 300B amp goes beyond output power. The big transmitting tubes are in a different league (when properly used). I jump directly from the 45 to the 211.
 
Last edited:
For around $500, there are new bookshelf size loudspeakers that sound very good that are 93db and above.
93dB/1W in a bookshelf speaker (likely using a relatively small driver is physically impossible), unless the lower frequency they need to reproduce is 100-150Hz. But In latter case the the driver is not a woofer and not even a mid-woofer. It's a midrange. Efficiency and low end response improve at expense of size. There is no way around it and 93 dB true efficiency with 50Hz @-3dB can only be a large speaker!
I suspect you are talking about basic Klipsh speakers? In that case they are miles off claimed performance. Here is how a "high efficiency" bookshelf looks like:
https://www.stereophile.com/content/klipsch-reference-premiere-rp-600m-loudspeaker-measurements
They are not even tube friendly with the impedance dipping below 4R and generally looking more like a roller-coaster. Note that the sensitivity (i.e. with 2.83V applied) is 89.6dB but the real efficiency will be in the region of 86 dB/1W as it's normal for this kind of speaker.
 
Perhaps that's true for people born after the notion of "bookshelf" shrunk by approximately 60%... Wayne at Pi has been selling 93+ speakers & kits for about 40 years, now. Now, no one has specified what room this will be in to achieve what target levels, but I'm with @stephe--esp if the box gets some boundary support from a, say, wall bookshelf etc. The main point to the OP is that the other option for amp power is sensitive speakers.
 
  • Like
Reactions: 1 user
Generalizations:

For Single Ended, Parallel Single Ended, Push Pull, and Balanced Amplifiers:

Pentode and Beam Power: Highest power out; requires global negative feedback
This could be looked at Ultra Linear, but with a 0% tap.

Ultra Linear Triode wired Pentode and Triode wired Beam Power: Medium power out; depending on the loudspeakers and the tap % (percentage), may, or may not, need global negative feedback. A tap range of 20% to 80%.

Triode wired Pentode and Triode wired Beam Power: Lowest power out; no global negative feedback needed.
This could be looked at as Ultra Linear, with a 100% tap.

DHT and Indirect heated Triodes that have low plate impedance, rp, do Not need global negative feedback.

Your Generalization(s) might have exceptions to the above.
 
  • Like
Reactions: 1 users
Perhaps that's true for people born after the notion of "bookshelf" shrunk by approximately 60%... Wayne at Pi has been selling 93+ speakers & kits for about 40 years, now. Now, no one has specified what room this will be in to achieve what target levels, but I'm with @stephe--esp if the box gets some boundary support from a, say, wall bookshelf etc. The main point to the OP is that the other option for amp power is sensitive speakers.
From bad to worse. I only see a BIG compromise in your solution.
The name says it all, a bookshelf is a small speaker you can put in a bookshelf but normally you don't do that because the result is compromised. Only put it in the bookshelf if there is no other solution.

The main point is that OP asked about a 15W SET.
 
At least a 15W 845 amp is real. 93 dB/1w bookshelf is simply a myth or commercial nonsense.
I am actually surprised you believed it as you normally measure everything....
I have measured the in room response using these speakers with 5-10W amplifiers, you can easily get to in room SPL louder than most people are interested in (mid to upper 80's), with very decent bass response down to 50-60hz or so, if you are smart about speaker placement. If you truly need high db down to 20hz, add a small powered sub. They work well in smaller rooms that aren't suited to the large 100db size speakers. It's why I recommended the OP consider this rather than trying to get a few more db out of the amplifier.

I've personally used these speakers in several folks homes along with a 7W SEUL amp and they sound great when setup right. By your post here, I'm not sure you have any personal experience with these and are just repeating what you read online somewhere as "experience"? I honestly was done with this thread until you decided to toss out this personal attack. Not sure why you repeatedly attacked me on this forum, but here we are once again.
 
  • Like
Reactions: 1 user
As system designers (which we all are), we'd be in a better head space by measuring amplifier output in dBW (Watts as refererred to 1 Watt = 0dB) rather than our conventional linear Watts scale. We build and pay for amplifier peak output on a linear scale, but we hear (and use) amplifiers on a log scale. dB is the log of a power ratio, and is the way we specify all other acoustic measurements (and a lot of others, too, for example RF).

Thinking about amplifiers on a linear scale misleads us into believing that the difference between a 5 Watt amplifier and a 10 Watt amplifier is that one can play twice as loud as the other, or at least much louder. Actually 3dB difference in peak output ability is much less important than good overload recovery, in the real world, even in respect to how loud an amplifier can be played.

A 5 Watt peak output per channel amplifier makes +7dBW peaks; 10 Watt makes +10dBW peaks. Nearfield (1M), equidistant from two 90dB SPL/1W/1M speakers with uncorrelated (music) source, which adds as +3dB, the first can make peaks of 100 dB SPL and the latter 103dB SPL. Room path length losses are of course the same for either.

The first step in system design should be to properly specify a real acoustic goal, rather than a number of amplifier Watts. I wish we could get beyond this antique thinking, but it'll probably never happen.

All good fortune,
Chris
 
  • Like
Reactions: 1 users
I have measured the in room response using these speakers with 5-10W amplifiers, you can easily get to in room SPL louder than most people are interested in (mid to upper 80's), with very decent bass response down to 50-60hz or so, if you are smart about speaker placement. If you truly need high db down to 20hz, add a small powered sub. They work well in smaller rooms that aren't suited to the large 100db size speakers. It's why I recommended the OP consider this rather than trying to get a few more db out of the amplifier.

I've personally used these speakers in several folks homes along with a 7W SEUL amp and they sound great when setup right. By your post here, I'm not sure you have any personal experience with these and are just repeating what you read online somewhere as "experience"? I honestly was done with this thread until you decided to toss out this personal attack. Not sure why you repeatedly attacked me on this forum, but here we are once again.
Quite a few times I have made the example of your 300B when seeing people spending a lot of money to get worse performance. Read my posts on 300B amps....
About measurements I was referring to the efficiency/sensitivity.
I only disagreed. If you see it like an attack I am sorry, I will ignore you from now on. But....do you really think that if I had something against you, and attacked you repeatedly (Where????) I would care to reply to your mail a couple of weeks ago and give you the schematics of my PCL82 amp with all the info on how I did it and how it works????
Should I agree on a 93 dB bookshelf that does not exist? Can I have my preference about the sound? I don't like affordable horn tweeZers and the 300B is not my cup of tea. I can't see an attack. Other than that I think the only other time I remember I disagreed was the case of the distorting preamp to "fix" a garbage SS amp when one could buy a proper SS amp with the same money. And it was just a disagreement, not attack. Do you need help? Maybe you have problem and see ghosts where there aren't....
You have no idea of what I know and what I have experienced.
 
Last edited:
As system designers (which we all are), we'd be in a better head space by measuring amplifier output in dBW (Watts as refererred to 1 Watt = 0dB) rather than our conventional linear Watts scale. We build and pay for amplifier peak output on a linear scale, but we hear (and use) amplifiers on a log scale. dB is the log of a power ratio, and is the way we specify all other acoustic measurements (and a lot of others, too, for example RF).

Thinking about amplifiers on a linear scale misleads us into believing that the difference between a 5 Watt amplifier and a 10 Watt amplifier is that one can play twice as loud as the other, or at least much louder. Actually 3dB difference in peak output ability is much less important than good overload recovery, in the real world, even in respect to how loud an amplifier can be played.

A 5 Watt peak output per channel amplifier makes +7dBW peaks; 10 Watt makes +10dBW peaks. Nearfield (1M), equidistant from two 90dB SPL/1W/1M speakers with uncorrelated (music) source, which adds as +3dB, the first can make peaks of 100 dB SPL and the latter 103dB SPL. Room path length losses are of course the same for either.

The first step in system design should be to properly specify a real acoustic goal, rather than a number of amplifier Watts. I wish we could get beyond this antique thinking, but it'll probably never happen.

All good fortune,
Chris
My is experience is that SPL numbers do not have all the importance normally people give to them. The intensity of the sonic/musical feeling is more related to actual dynamics and behaviour in the time domain. Room acoustics and the quality of the software are the first two and most important things. These are the components that compromise most on dynamics. Then the source and speakers and last the amplifier simply because it is technically the strong link. Often, when one feels the need of turning up the volume to higher levels that is a bad sign. Without a well behaved room and a good source it's just a perennial game without a solution.
 
Last edited:
  • Like
Reactions: 1 user