So we all know that Push/pull cancels out even harmonics, while Single ended creates both even and odd order harmonics.
I love to use musical instruments to compare to "hifi". Hifi is extremely subjective, and there is a lot of myth and $nake oil along with the subject.
Anyway, we can all agree that obviously, all musical instruments sound different creating the same fundamental frequency. In the instrument world this is called Timbre. Timbre is the difference in harmonic overtones (harmonic multiples) created.
Reed instruments create vastly different timbre's dependent on if they have conical (cone) vs cylindrical bores. Conical instruments (Saxophone for example) are like single ended amplifiers, they create both even and odd order harmonics. Cylindrical instruments (such as a Clarinet) are like push/pull amplifiers, only create odd order harmonics.Source.
This YouTube video is a great example of both instruments performing the same song.
To me, this is a perfect explanation of the differences in sound signature with SET amplifiers vs Push/Pull amplifiers. To me, the saxaphone does sound like it has more "soul" and "magic", explaining why SET amplifiers usually are given synonymous descriptive words. On the other hand, the the Clarinet does sound more "pure"; again, going hand in hand with push/pull clarity. This is all subjective of course, and its up to you which sound you like better.
I hope this helps people in the future decide what type of sound they like, and which type of amplifier to build or design.
Cheers! 😀
I love to use musical instruments to compare to "hifi". Hifi is extremely subjective, and there is a lot of myth and $nake oil along with the subject.
Anyway, we can all agree that obviously, all musical instruments sound different creating the same fundamental frequency. In the instrument world this is called Timbre. Timbre is the difference in harmonic overtones (harmonic multiples) created.
Reed instruments create vastly different timbre's dependent on if they have conical (cone) vs cylindrical bores. Conical instruments (Saxophone for example) are like single ended amplifiers, they create both even and odd order harmonics. Cylindrical instruments (such as a Clarinet) are like push/pull amplifiers, only create odd order harmonics.Source.
This YouTube video is a great example of both instruments performing the same song.
To me, this is a perfect explanation of the differences in sound signature with SET amplifiers vs Push/Pull amplifiers. To me, the saxaphone does sound like it has more "soul" and "magic", explaining why SET amplifiers usually are given synonymous descriptive words. On the other hand, the the Clarinet does sound more "pure"; again, going hand in hand with push/pull clarity. This is all subjective of course, and its up to you which sound you like better.
I hope this helps people in the future decide what type of sound they like, and which type of amplifier to build or design.
Cheers! 😀
some of us are building high fidelity, others musical instruments or effects boxes. It's a choice.
When you use push pull, the even harmonics cancel out, and the third harmonic becomes most prominent. There are some circuit techniques that then cancel out the third, or atleast partly cancel it out. With tube based Class A, what distortion remains is then pretty small.
There's plenty of $nake oil and myth in musical instruments too. At least in the electric guitar business. However, if some wierd, and on the face of it, obnoxious, form of distortion is produced by a certain amplifier, sooner or later some musician somewhere will find a way to use it creatively and it will sound good in that song/tune. Nothing wrong with that!
There's plenty of $nake oil and myth in musical instruments too. At least in the electric guitar business. However, if some wierd, and on the face of it, obnoxious, form of distortion is produced by a certain amplifier, sooner or later some musician somewhere will find a way to use it creatively and it will sound good in that song/tune. Nothing wrong with that!
Last edited:
Operated well below their max power range both types of amp sound almost exactly the same. The secret to a "clean" amp is to overdesign its power capability beyond its anticipated usage.
For me its a bad design if the distortion profile is effecting the performance to the extent that you can hear the difference. There are many bad designs out there.
Shoog
For me its a bad design if the distortion profile is effecting the performance to the extent that you can hear the difference. There are many bad designs out there.
Shoog
A similar analogy was used by Scott Frankland in his articles for Stereophile way back in the late '90's - suggesting perhaps saxophone was not the best way to go...😉
No. Audio is (apparently) extremely subjective. Hi-fi, in contrast, is simply about reproducing a sound so is only subjective to the extent that we need to know how accurate it needs to be in order to provide good reproduction; whether that sound is pleasant or not is a matter for the musicians, not the system.Xaborus said:
If only it was that simple. If only.
With average sensitivity modern speakers and an average sized and furnished room, for a comfortable listening level, the average amplifier output is around 0.5 to 1 watt. (1960's speakers were rather more sensitive, but they had noticeably more distortion and colouration.)
Peaks routinely occur reaching to 10 to 20 watts. More rarely, if playing modern recordings on CD's, peaks reach to 50 to 100 watts. If you like bass, and most of us males do, you'll turn the bass up a bit. That will make the peaks 100 to 300 watts.
Almost all domestic audio gear reprodicing from digital media clips now and then with sound volumes perceived as quite ordinary! Try listening with an oscilloscope on the amplifier output - you may be surprised how often clipping occurs.
Natural music would have even rarer peaks even louder. But recording engineers know domestic equipment couldn't handle it (and the neighbours would complain), so they limit the dymamic range.
It all means pretty much nobody has equipment meeting your recommendation. It would cost far too much (remember, it's no good having an amplifier with more power output than the speaker can handle in its linear range.) and you couln't use it.
Loudspeakers generate intermodulation distortion from doppler effect. Consider a woofer reproducing 70 Hz (from one instrument) and a smaller amplitude 190 Hz (from another instrument). The cone moves back and forth at 70 Hz, effectively shifting the 190 Hz source toward and away from you, causing it to be doppler distorted. So you get to hear not just 70 and 190 Hx, you hear 70, 190, 120, and 260 Hz, and tiny level of other frequencies.
The only way to fix this is to increase the cone area, so the amount of travel back and forth doesn't have to be as much for the same sound power. With a woofer around 8 inch diameter (200 mm), you are limitted to about 10 watts for 1% distortion. To repoduce 40 watts with only 1% distortion, you need a 15 inch woofer.
People who visit me comment on how clean my system sounds. But I have dual 15 inch woofers in each channel, in big boxes. The lady of the house HATES how big the boxes are. But she always knew I was a hi-fi entusiast. Bass reflex design helps efficiency and thus keeps distortion down. But that makes the boxes really big and there are other problems with the response.
100W against a 1W RMS is a 20dB span which I see on classical, but rarely elsewhere and I don't have a big enough house to hit 100dB peaks without it hurting. But for those lucky enough to have big enough listening rooms to be out the near field you need more power.
For me a system that sounds good at average levels around 75dB is far more satisfying than one you have to crank up before it gets realistic.
Not just on classical.
Actually most classical recordings around are faily old and are compressed quite a bit.
The most difficult CD I have, and what I use to demostrate to people what in terms of distortion, clipping , etc can be tolerated is The Buddy Holly Story London Cast Recording Exallshow D31292 (1994). Certain tracks cannot be reproduced cleanly at conversation level with anything less than 100 watts per channel, even with the bass control set to flat.
Movie sound tracks can be very peaky and difficult. I have the U574 DVD. If you set the volume so that the dialog is the same level that people actually talk at, the depth charge sound effects just about lift the roof off the house. There is a point in this movie where some off-camera pipe gets banged. On a good system its hard not to think that some dangerous bang occurred just above your head!. But on a fairly standard store bought 5+1 sound system the effect is completely lost - its just another bang on the soundtrack. I also have the Titanic DVD (the one featuring DiCaprio and Winslet). If the dialog is at normal conversation level, the mayhem noise during the scene where it upends and sinks is enormous!
In one point in Titanic I had to laugh. When the ship up-ends and there's all this mayhem noise, somguy falls off the guard rail at the stern and falls into the ocean 200 feet down. On the way down he bangs his head on a propellor blade and starts spinning. In all the very loud mayhem roaring noise, a tiny little crack is clearly audible when he bangs his head. On a friend's store bought system (probably about 30 watts per channel), the crack is lost in all the other racket.
The interesting thing is, the mayhem noise in Titanic is somewhat distressing its so loud. The peaks in The Buddy Holly story cause considerable clipping and crumbling of the sound on 20 to 25 watt systems, but on a 100 watt per chan system don't seem all that loud. More punchy than loud if you can get what I mean.
Actually most classical recordings around are faily old and are compressed quite a bit.
The most difficult CD I have, and what I use to demostrate to people what in terms of distortion, clipping , etc can be tolerated is The Buddy Holly Story London Cast Recording Exallshow D31292 (1994). Certain tracks cannot be reproduced cleanly at conversation level with anything less than 100 watts per channel, even with the bass control set to flat.
Movie sound tracks can be very peaky and difficult. I have the U574 DVD. If you set the volume so that the dialog is the same level that people actually talk at, the depth charge sound effects just about lift the roof off the house. There is a point in this movie where some off-camera pipe gets banged. On a good system its hard not to think that some dangerous bang occurred just above your head!. But on a fairly standard store bought 5+1 sound system the effect is completely lost - its just another bang on the soundtrack. I also have the Titanic DVD (the one featuring DiCaprio and Winslet). If the dialog is at normal conversation level, the mayhem noise during the scene where it upends and sinks is enormous!
In one point in Titanic I had to laugh. When the ship up-ends and there's all this mayhem noise, somguy falls off the guard rail at the stern and falls into the ocean 200 feet down. On the way down he bangs his head on a propellor blade and starts spinning. In all the very loud mayhem roaring noise, a tiny little crack is clearly audible when he bangs his head. On a friend's store bought system (probably about 30 watts per channel), the crack is lost in all the other racket.
The interesting thing is, the mayhem noise in Titanic is somewhat distressing its so loud. The peaks in The Buddy Holly story cause considerable clipping and crumbling of the sound on 20 to 25 watt systems, but on a 100 watt per chan system don't seem all that loud. More punchy than loud if you can get what I mean.
Last edited:
You can always go to a line array where you have 20 or more drivers per side reproducing the sound. Not only do you get increased cone area but you also get multiplied motor capability per unit cone area, thus more control over the increased cone area.
If you stay in the nearfield and space the drivers closely enough together the comb effects from the multiple drivers can be minimized.
I intend to check out (build) a line array or two when time permits; although I love my current setup there is always that little voice whispering "it could get better"...
True. One reason why those multi-speaker Bose loudspeakers that came out in the 1970's sounded so good.
There's three ways to reduce the comb effect.
a) Deliverably throw sound around in all different directions, to bounce off walls etc as Bose did;
b) Put all the drivers in a vertical line, with the centre of the array the same height as your ears;
c) Introduce electronic delay between speakers. The centre speaker should have no delay, and the other drivers get greater delay in accordance with their distance away from the centre one. The amount of delay should be calculated so that the composite wavefront is the same as it would be from a point source.
The Quad Series 2 electrostatics employed method (c).
Method (c) in theory requires an infinite number of very small drivers. Since you can only in practice have a limitted number of drivers, there are remaining small comb effects, and to fix this the amplitude fed to each speaker should also be taperred off as the driver position moves away from the centre.
I have a lengthy paper on the design approach for this written by a Philips speaker engineer. It was never commercially viable for obvious reasons. It's been on my to-do list to try for years. I get that little voice too.
Last edited:
You can also curve the line to physically compensate for the distance differences. Makes for a small sweet spot, though.
I'm growing impatient for someone to invent that mythical pulsating sphere already!
So 2 films (including one so carp I never want to watch it) and one original cast recording. Hardly an endemic problem for power handling then? plus the kids complain if I try to reach cinema SPLs!
OK I admit it. I cheat as below 120Hz I have a powered sub.
OK I admit it. I cheat as below 120Hz I have a powered sub.
I just gave two examples of movies that highlight deficiencies in sound systems. Lots of modern movies have high dynamic range sound tracks.
If your going to indulge a passion for valve amps then you need to consider the need for high sensitivity speakers to match and to allow operation at low amplifier distortion.
My speakers are over 100db/w which fits the bill. In this situation a modest 7watt amp will stay well behaved for most of its operation and the main issue will be keeping your amps quiet enough not to be annoying when there is no signal.
Trying to match valve amps to modern speakers is something of a fools errand.
Shoog
My speakers are over 100db/w which fits the bill. In this situation a modest 7watt amp will stay well behaved for most of its operation and the main issue will be keeping your amps quiet enough not to be annoying when there is no signal.
Trying to match valve amps to modern speakers is something of a fools errand.
Shoog
Well actually you more gave a justification for a 5.1 system with separate adjustment on the dialog channel. If you listen to movies on a 2 channel system you have to accept compromises or big power handling or hearing damage! 🙂
Either that or KT88's or 6CA7/EL34 in parallel push pull.
You don't get something for nothing. There is a good reason why modern speakers are inefficient. There is a trade-off between distortion, coloration, and transient response on one hand, and sensistivity on the other hand.
Before transistors, loudspeakers had to be efficient. The compromised sound quality had to be accepted. But when transistors came along, high power was cheap, so the speaker makers could adjust the compromise to better distortion, colouration, and transient response.
No. In amplification, the signal to noise ratio is largely constant regardless of power output, unless the designer is incompetent. High power amps make more noise than low power ones, because of the greater amplification, but as they are intended for inefficeint speakers, the acoustic signal to noise ratio remains much the same.
Perhaps. There is almost always a solution to any problem - throw more money at it - and that applies with tube amplifiers.
Last edited:
It's not as simple as that.
With 5.1 systems, you (or the installer) is supposed to adjust the gain of each channel with the white noise test, so that the percieved volume where you sit is equally loud from all channels. The front channels and especially the subwoofer need more power output to reproduce the dynamic range required, but the gain of all channels shall be such that the white noise test gives equal volume from all channels.
If you don't do that, the reproduced sound tracks will not have the artistic merit the producer intended, and the percieved direction of sound effects will be wrong.
If the channel balance is correct, and actor's conversations at conversation level, aircraft engines will be loud, and explosions will be near deafening, as the producer intended.
Of course, it's your choice.
But I rather enjoyed the movie U574 in part because of its excellent sound effects - engine noises, air valves, miscelaneous structural creaks and groans from all around you in submarine control room scenes was very good. And the sounds of general hubbub and the band in the dinner venue really made you feel you were there if you watched with a 100 W multi-channel system.
Another example: In the movie, "The Right Stuff", there is a scen where a jet aircraft passes low overhead, going from front to back. In a correctly adjusted multichannel system, and only in a correctly adjusted multichannel system, it really does sound like a jet passing over you at very low altitude. On other systems it is just a whole lot of noise that builds and fades away.
Last edited:
hmm interesting you talk about how the producer intended given this thread is about SET which is giving you what you want and not necessarily what the producer wanted 🙂
I can barely fit 2 channels in my living room so a 5.1/7.1/atmos is a long long way in the future (sofa hard on rear wall, piano where centre channel could go). Doesn't stop me thinking 'damn the foley team had fun there' 🙂
I can barely fit 2 channels in my living room so a 5.1/7.1/atmos is a long long way in the future (sofa hard on rear wall, piano where centre channel could go). Doesn't stop me thinking 'damn the foley team had fun there' 🙂
Many of those old speakers (1940's to 50's era) with super high efficiency perform far better than most people would admit. There is a caveat though - almost all of them require open baffles to give of there best since any back loading at all will compromise their performance.
Where they cannot perform well is in the bass region - but it is a small matter to actively cross them over to a good SS bass amp and some modern high efficiency bass drivers (many of the car audio subwoofers make cheap but excellent candidates).
its about going the extra mile to allow the components to give of their best, and that involves both valves and vintage drivers.
Shoog
Where they cannot perform well is in the bass region - but it is a small matter to actively cross them over to a good SS bass amp and some modern high efficiency bass drivers (many of the car audio subwoofers make cheap but excellent candidates).
its about going the extra mile to allow the components to give of their best, and that involves both valves and vintage drivers.
Shoog
- Status
- Not open for further replies.