[Shike]

Quote:

Tubes work best where devices or speakers that are used with them are designed with tubes in mind. If that condition is met tubes offer performance in real audible terms that will leave almost all transistor gear sounding wanting.

First sentence is mostly correct, there are ways to make tubes and attached devices behave but for the most part today they aren't used in reproduction for obvious reasons.
 
The second is coming off as myth propagation again.

 

[firev1]

Quote:

First sentence is mostly correct, there are ways to make tubes and attached devices behave but for the most part today they aren't used in reproduction for obvious reasons.
 
The second is coming off as myth propagation again.

Totally agree, transistor gear was built with replacing tubes in mind. Even the best tubes degrade in performance with time and require constant replacement to maintain low distortion characteristics. Tubes do come of readily as warm and lively due to the nature of their distortion characteristics, I would not be surprised if neutral to you is actually mids-emphasis if tubes where your reference.

 

[xnor]

I wonder why any audiophile would choose horn speakers for faithful reproduction. Afaik they were invented to achieve usable sound levels - a compromise, trading in fidelity for efficiency.

 

[Mauricio]

Do you think NASA, military equipment manufacturers, telecommunication companies use tubes in applications and equipment that require precision and accuracy?  I mean, are those drones flying the tribal areas of Pakistan packed with tubes to the gills?  How about the equipment that the Iranians are using to manage and control their nuclear program?  Tube technology?

 

[Tilpo]

Do you think NASA, military equipment manufacturers, telecommunication companies use tubes in applications and equipment that require precision and accuracy?  I mean, are those drones flying the tribal areas of Pakistan packed with tubes to the gills?  How about the equipment that the Iranians are using to manage and control their nuclear program?  Tube technology?

I don't think this is relevant to the discussion. Everyone knows that transistors provide higher linearity than tubes. In addition to that they are far more efficient in terms of cost, size and energy usage.
The question at hand is whether for audio purposes they provide a subjectively better sound quality, and for what reasons this would be the case.

 

[RexAeterna]

Quote:


Totally agree, transistor gear was built with replacing tubes in mind. Even the best tubes degrade in performance with time and require constant replacement to maintain low distortion characteristics. Tubes do come of readily as warm and lively due to the nature of their distortion characteristics, I would not be surprised if neutral to you is actually mids-emphasis if tubes where your reference.

true but the actual replacement for tube guys were known as mosfet transistors. mosfets were basically improved tubes(much lower distortion, 0 ohm outputs, ability to soft-clip like tubes and ability to deliver high current loads with well designed mosfets). if someone wants something to have the ''tube'' sound mosfets are one of their choices. mosfets were famous in sony,H/K,Adcom,B/K,hitachi,Ashly,Perreaux, Nikko and even high-end yamaha power amps were known to use mosfets in their output stages. nowadays mostly car audio amps use mosfets due to their soft clipping ability.

 

[Chris J]

Quote:

 
Note that almost all modern speaker designs have inadaquate air loads. Golden era speakers had very good to excellent air loads & all impedances were mached as closely as possible back then to get the most out efficiency & best sound possible, especially in the studio environment. Horns were used back then as the horn is actually an acoustic impedance matching device. These were used with drivers & amps with matched 1 to 1 impedance thus having a damping factor of 1. These speakers had tight clean rock hard bass with the matching impedance of the amp & drivers. Crossovers were designed to take into account the source impedance that matched that of the drivers & as such could drive the speakers to very high volumes with very little wattage 5 watts was almost enough to drive you out of the room. People point to the bad specs of zero feedback tubes when run to their power limit but they never point out that in many cases turn it down to a reasonable volume which in most cases is still plenty loud & the tube amp will, if you have one that is properly designed, have distortion that is quite low, Definitely inaudable by any stretch. My friend's amp that he designed was one of these amps that at typical listening level without feedback was still only .02% distortion. Yes push the volume & you will get 2-3% distortion but nothing really noticable due to their low order nature.
 

Most modern speaker designs have inadequate air loads? Based on what? Please support this.
True, you cannot use a driver designed for use in a horn in a non-horn loaded enclosure. It would not be damped properly.
Horn loaded systems are very high efficiency, so do not require a lot of power due to the horn loading.
But how does that relate to a modern loudspeaker system design?

To get maximum efficiency from a voltage source power amp, output impedance must approach zero.
If you mean an amp with an output impedance of 16 ohms and a speaker with 16 ohm nominal impedance, then the electrical efficiency would be very poor.
In this example, amplifier efficiency can be doubled by decreasing the amp output impedance from 16 ohms to virtually zero.
 
How do you impedance match an amp to an impedance which varies with frequency?  
 
BTW, for the most part, in spite of all the disagreements I have with this thread, I really do like your thought provoking thread.
How about posting the meat of it in the "Articles" section?
   
 

 

[Chris J]

 
Quote:

true but the actual replacement for tube guys were known as mosfet transistors. mosfets were basically improved tubes(much lower distortion, 0 ohm outputs, ability to soft-clip like tubes and ability to deliver high current loads with well designed mosfets). if someone wants something to have the ''tube'' sound mosfets are one of their choices. mosfets were famous in sony,H/K,Adcom,B/K,hitachi,Ashly,Perreaux, Nikko and even high-end yamaha power amps were known to use mosfets in their output stages. nowadays mostly car audio amps use mosfets due to their soft clipping ability.

Kinda, sorta almost true.
Hard to really compare a MOSFET power amp (or a power amp with a MOSFET output stage) with a Vacuum Tube power amp.
 
Tubes are large and need a heater filament, need a power suppy for the amp circuit and a second one the for the heater filaments.
Transistors are much smaller.   This is all rather obvious, so what is my point?
 
A typical tube power may may have 4 triodes, 2 pentodes and an output transformer. Due to the size and expense of the components simplicity is enforced upon the designer.
 
A typical solid state power amp will have AT LEAST twice as many devices, 16 transistors or more is not unusual.  The small signal transistors are cheap and small, so why not use lots of them? Due to the absense of an output transformer, it is fairly easy to make output impedance relatively lower in a solid state amp.  I can't think of a single solid state amp with a high output impedance other than a French design with variable output impedance. The name slips my mind.
 
BTW, I have nothing against the use of tubes or discrete MOSFETs, JFETs or BJTs or Op Amps in audio circuits when applied properly.
 
 

 

[germanium]

Quote:

Most modern speaker designs have inadequate air loads? Based on what? Please support this.
True, you cannot use a driver designed for use in a horn in a non-horn loaded enclosure. It would not be damped properly.
Horn loaded systems are very high efficiency, so do not require a lot of power due to the horn loading.
But how does that relate to a modern loudspeaker system design?

To get maximum efficiency from a voltage source power amp, output impedance must approach zero.
If you mean an amp with an output impedance of 16 ohms and a speaker with 16 ohm nominal impedance, then the electrical efficiency would be very poor.
In this example, amplifier efficiency can be doubled by decreasing the amp output impedance from 16 ohms to virtually zero.
 
How do you impedance match an amp to an impedance which varies with frequency?  
 
BTW, for the most part, in spite of all the disagreements I have with this thread, I really do like your thought provoking thread.
How about posting the meat of it in the "Articles" section?
   
 

Electrical efficiency of a tube amp is best when the impedances match With transistors this is obviously not true. Best power transfer can never be achieved with transitors due to not being able to impedance match due to the very nature of transistor amps being a voltage source. transistors loose 3/4 of thier power if you try to impedance match them externally. Tube do not require external impedance matching due to thier very nature & actually generate thier best power when impedance matched. As such tubes are by nature (without feedback or external impedance matching neworks) impedance matched in order to deliver best power. That is why thier output transformers have different impedance taps. so the tube is always running in its best power transfer mode which yields best power & lowest distortion. Transformers are very efficient by the way at tranfering power & 95-98% of the power is tranfered to the load. With transistors extermal matching involves resistor networks that absorb 75% of the power so only 25% makes it to the load. This impedance matching is what works best with most horn speakers if you intend to dive them with transisors.
 
Woofers on almost all modern speaker  designs are in fact inadaquately air loaded. That is why they need so much compliance in thier suspension. Because they have to move a lot to get a little sound whereas the older high efficiency speakers could deliver lots of bass power & not have to move much more than the thickness of it's own driver cone thickness. That is what having an adaquate air load does for you.You will hardly see any driver movement in such a speaker even when generating bass that will rattle everything in the house. Properly designed horn speaker can in fact be very high fidelity & have much lower distortion than any direct radiator. Great care must of coarse be taken in thier design to get this performance. Horn speakers were used in studios for many years but as control rooms got smaller the horn speakers became impractical as it takes some distance fot the sound of the woofer & the tweeter in such a system to combine in a way that sounds proper  even when care is taken to time align the drivers as the more recent (less than 60 year old) designs.

 

[xnor]

germanium, could you post a link to a few of those (non-DIY please) horn speakers you talk about? Btw, a properly designed horn loudspeaker needs a few meters large horn for the bass woofer...

 

[Chris J]

Quote:

Electrical efficiency of a tube amp is best when the impedances match With transistors this is obviously not true. Best power transfer can never be achieved with transitors due to not being able to impedance match due to the very nature of transistor amps being a voltage source. transistors loose 3/4 of thier power if you try to impedance match them externally. Tube do not require external impedance matching due to thier very nature & actually generate thier best power when impedance matched. As such tubes are by nature (without feedback or external impedance matching neworks) impedance matched in order to deliver best power. That is why thier output transformers have different impedance taps. so the tube is always running in its best power transfer mode which yields best power & lowest distortion. Transformers are very efficient by the way at tranfering power & 95-98% of the power is tranfered to the load. With transistors extermal matching involves resistor networks that absorb 75% of the power so only 25% makes it to the load. This impedance matching is what works best with most horn speakers if you intend to dive them with transisors.
 
Woofers on almost all modern speaker  designs are in fact inadaquately air loaded. That is why they need so much compliance in thier suspension. Because they have to move a lot to get a little sound whereas the older high efficiency speakers could deliver lots of bass power & not have to move much more than the thickness of it's own driver cone thickness. That is what having an adaquate air load does for you.You will hardly see any driver movement in such a speaker even when generating bass that will rattle everything in the house. Properly designed horn speaker can in fact be very high fidelity & have much lower distortion than any direct radiator. Great care must of coarse be taken in thier design to get this performance. Horn speakers were used in studios for many years but as control rooms got smaller the horn speakers became impractical as it takes some distance fot the sound of the woofer & the tweeter in such a system to combine in a way that sounds proper  even when care is taken to time align the drivers as the more recent (less than 60 year old) designs.

I may be way off base here, but here goes:
 
http://en.wikipedia.org/wiki/Impedance_matching
 
Don't confuse "Impedance matching" with using a transformer in a vacuum tube power to step the output voltage down and step the output current up.
It is an impedance changing transformer, not an impedance matching transformer.
To get maximum power efficiency out of a vacuum tube power amp you still want the output impedance to be as low as possible.
As you said, transformers are very efficient. In a vacuum tube power amp, the output impedance of the output transformer will be very low, lower than the load impedance it is driving.
The output taps on an output transformer are so the voltage and current clip at the same level into a 4, 8 or 16 ohm load.
 
Keep in mind that both vacuum tube power amps and SS power amps are normally voltage amplifiers (or voltage sources).
In addition, they are usually both feedback amplifiers.
 
Best power transfer (or most efficient power transfer) in a solid state amp occurs when the outpout impedance is as low as possible.
 
Don't confuse Maximum Power Transfer with Maximum efficency.
In Maximum Power Transfer efficiency can never be higher than 50%.  Maximum Power Transfer:  when output impedance  = load impedance.
 
You have a valid point with horn loaded loudspeakers, because of the horn loading, the driver doesn't need a large excursion to output a lot of SPL.
The problem is, low end bandwidth is limited by the size of the woofer horn.
Modern loudspeaker are often designed for a large excursion and designed to have a high power capacity compared to older loudspekers. So you can use a long throw, high power capacity 8" woofer in a sealed enclosure in place of a short throw, low power capacity 15" woofer in an Altec Voice Of The Theatre loudspeaker.
A bit of an oversimplification, I know!

You could argue that a moden approach is to use a fairly low efficiency, long throw speaker with a high powered amp to get the same volume as a horn loaded louspeaker and a very low powered single ended vacuum tube power amp.
 
I wouldn't mind hearing an Altec Model 19 with a high quality vacuum tube power amp! Last time I heard one was with a crappy solid state amp. maybe that's why it sounded so bloody awful.
 

 

[germanium]

Quote:

I may be way off base here, but here goes:
 
http://en.wikipedia.org/wiki/Impedance_matching
 
Don't confuse "Impedance matching" with using a transformer in a vacuum tube power to step the output voltage down and step the output current up.
It is an impedance changing transformer, not an impedance matching transformer.
To get maximum power efficiency out of a vacuum tube power amp you still want the output impedance to be as low as possible.
As you said, transformers are very efficient. In a vacuum tube power amp, the output impedance of the output transformer will be very low, lower than the load impedance it is driving.
The output taps on an output transformer are so the voltage and current clip at the same level into a 4, 8 or 16 ohm load.
 
Keep in mind that both vacuum tube power amps and SS power amps are normally voltage amplifiers (or voltage sources).
In addition, they are usually both feedback amplifiers.
 
Best power transfer (or most efficient power transfer) in a solid state amp occurs when the outpout impedance is as low as possible.
 
Don't confuse Maximum Power Transfer with Maximum efficency.
In Maximum Power Transfer efficiency can never be higher than 50%.  Maximum Power Transfer:  when output impedance  = load impedance.
 
You have a valid point with horn loaded loudspeakers, because of the horn loading, the driver doesn't need a large excursion to output a lot of SPL.
The problem is, low end bandwidth is limited by the size of the woofer horn.
Modern loudspeaker are often designed for a large excursion and designed to have a high power capacity compared to older loudspekers. So you can use a long throw, high power capacity 8" woofer in a sealed enclosure in place of a short throw, low power capacity 15" woofer in an Altec Voice Of The Theatre loudspeaker.
A bit of an oversimplification, I know!

You could argue that a moden approach is to use a fairly low efficiency, long throw speaker with a high powered amp to get the same volume as a horn loaded louspeaker and a very low powered single ended vacuum tube power amp.
 
I wouldn't mind hearing an Altec Model 19 with a high quality vacuum tube power amp! Last time I heard one was with a crappy solid state amp. maybe that's why it sounded so bloody awful.
 

Output impedance of a tube amp is determined by the tubes not the transformer & the transformer is what is used to allow the tubes to operate in thier most efficient manner with the best power & lowest distortion. Without feedback the most efficient operation occures when the impedances are matched as you will get the highest power & lowest distortion. The tranformer is there to get the best impedance match for the tubes & therefore best performance. If you add feedback of coarse the output impedance will drop as will distortion but that doesnot mean nessessarily best audio performance as the feedback will make amps that are capable of running without feedback lifeless & drab sounding like bad solid state. In fact under that cercumstance solid state may even sound less drab than the tube amp with even modest feedback. So much for higher distortion as causing the more open sound of tubes as the lower distortion modern tranny amp will often sound more open than the tube amp with feedback as the tube amp even with feedback will still have more distortion hence by the reasoning of some should sound less drab than the solid state
 

 

[Chris J]

Quote:

Output impedance of a tube amp is determined by the tubes not the transformer & the transformer is what is used to allow the tubes to operate in thier most efficient manner with the best power & lowest distortion. Without feedback the most efficient operation occures when the impedances are matched as you will get the highest power & lowest distortion. The tranformer is there to get the best impedance match for the tubes & therefore best performance. If you add feedback of coarse the output impedance will drop as will distortion but that does not mean nessessarily best audio performance as the feedback will make amps that are capable of running without feedback lifeless & drab sounding like bad solid state.

A bit of an oversimplification.
I agree, the output transformer greatly improves the efficency of the amp.
However,  output impedance of a tube amp is characterized by the output tubes, output transformer and amount of (or lack of) feedback in the circuit.
An amplifier will usually be optimized to work best with a certain feedback factor.
However, feedback does not have a large effect on efficiency.
Define "best power" please.
Define "impedance match".
 

 

[Isaansound]

From Wikipedia:"In electronics, impedance matching is the practice of designing the input impedance of an electrical load (or the output impedance of its corresponding signal source) to maximize the power transfer or minimize reflections from the load."
The issue of air loading of bass speakers was approached in the Altec A7 by combining a short bass horn with a tuned bass reflex cabinet. The Klipschorn rather brilliantly folded a long bass horn in the internalls of a cabinet, then coupled the cabinet to the room by placing it in a corner. A 1947 design which without equalization was essentially flat down to 15hz. Interesting that a pair of these is on fleabay for $12,000........
I suspect that much of the " tube hating" and the reaction to it arise as much from social disconnect as engineering. I admit there is a perceived elitism among the tube crowd, and I suppose "rebellion" against that perception is natural and human.
I find it remarkable that more acknowledgement isn't given of the two most non linear components; the lessor, the ear and that brain being infinitely non-linear.
Audiophilia originated before electronic instruments. The concept was to reproduce the live listening experience of human driven and produced instruments. Ask a violinist, it is the very harmonic character imparted by the wood, varnish, etc. that makes a great instrument, not linearity.
The usefulness of dampening is entirely dependent on the system as a whole. Good in some designs, bad in others.
My point......?  As human hearing and the brain are non linear, sound engineers skew the sound to several criteria, people like many different types of music, so engaging in these debates while interesting and hopefully educational is largely fruitless  IMO better to play your music on different gears and see what works for you and your budget. Personally I don't like hip-hop, but anticipating  my modded Altec horns in DIY boxes, tube amp, and a Sansa Fuze V2 will play Diana Krall "Live in Paris" to my extreme satisfaction. We all have too many choices in gear currently. All I can ask  is that when the music is over  "was it good for you??".
BTW anyone interested in the plans for the Klipschorn or the A7 or principles behind them as well as other DIY speaker projects should consider picking up an old copy (out of print, but on Amazon and fleabay) "How to Build Speaker Enclosures" by Alex Badmief and Don Davis.

 

[Chris J]

Quote:

From Wikipedia:"In electronics, impedance matching is the practice of designing the input impedance of an electrical load (or the output impedance of its corresponding signal source) to maximize the power transfer or minimize reflections from the load."
The issue of air loading of bass speakers was approached in the Altec A7 by combining a short bass horn with a tuned bass reflex cabinet. The Klipschorn rather brilliantly folded a long bass horn in the internalls of a cabinet, then coupled the cabinet to the room by placing it in a corner. A 1947 design which without equalization was essentially flat down to 15hz. Interesting that a pair of these is on fleabay for $12,000........
I suspect that much of the " tube hating" and the reaction to it arise as much from social disconnect as engineering. I admit there is a perceived elitism among the tube crowd, and I suppose "rebellion" against that perception is natural and human.

 
Did you read the COMPLETE Wikipedia article?
If you did then you would see that impedance matching does not normally apply to audiophile loudspeakers.
Often, both sides of the Tube argument seem to have a lot to do with contempt for Science and Engineering and a belief in Audiophile magic, myths and voodoo.