Power Comparison Tube vs. Solid State

[JimHansonDC]

Since we have a lot of very wise people with deeper understandings of electronic engineering than me (low bar). I wonder if anyone can explain what appears to be a differential in wattage ratings in solid state vs. tube amps. I get that tubes tend to be overall more expensive for a number of reasons and that generating a large output wattage takes a lot of tube and transformer action.

But are the stated output wattages really apples to apples for the two types. I have an MC-275 rated at 75w per channel that drives my Martin Logans (notorious power hogs) subjectively better than a 250w per channel solid state. I have no objective measurement capability. So I guess what I'm asking is:
If a signal at one level hits the amp and then leaves after amplification

Are those numbers actually comparable?
Do the amps respond to power requirements from the speakers differently?
Is the perceived difference potentially just that?

I am well aware of the warm vs. cool discussion about how tubes sound v. ss but I'm more interested in the actual engineering realities.

Thanks for any edification you can provide.

 

[marpow]

It's a very good question, and I understand the way you wrote it, it's more for personal information. The MC 275 is a great 2 channel amp, I would bet if you asked McIntosh direct they would have the best answer. I have asked them questions over the years and they answer very fast.

 

[chucky3042]

The Logans are electrostatic and require a low impedance (high current drive). I had a set of summits years ago and I think they were a low of around 2 Ohms. Many solid state amps are designed for 8 Ohms and their ratings are very mush a peak rating. A good guide is to put them on a scale and if they weigh less than say 10 kg they are a toy (unless a class D).

Many Valve amps have a low impedance coupling step down transformer that can handle 2 Ohm loads. In addition they tend to soft clip, which reduces your perception of overload clipping distortion, so they sound heaps loader than the apparent rating.

I hope that helps a bit

Regards

Chucky

 

[Doug G.]

If rated legitimately, the output of a solid state amp is not a peak rating. Most amps in the audiophile world are rated legitimately. In automotive and boom box type situations, not so much.

Doug

 

[LuvMyQuad]

Since we have a lot of very wise people with deeper understandings of electronic engineering than me (low bar). I wonder if anyone can explain what appears to be a differential in wattage ratings in solid state vs. tube amps. I get that tubes tend to be overall more expensive for a number of reasons and that generating a large output wattage takes a lot of tube and transformer action.

But are the stated output wattages really apples to apples for the two types. I have an MC-275 rated at 75w per channel that drives my Martin Logans (notorious power hogs) subjectively better than a 250w per channel solid state. I have no objective measurement capability. So I guess what I'm asking is:
If a signal at one level hits the amp and then leaves after amplification

Are those numbers actually comparable?
Do the amps respond to power requirements from the speakers differently?
Is the perceived difference potentially just that?

I am well aware of the warm vs. cool discussion about how tubes sound v. ss but I'm more interested in the actual engineering realities.

Thanks for any edification you can provide.

For the record, let me say I've never owned a tube power amp, though I have owned several tube pre-amps back in the day. I have read about the phenomena you describe and also of similar instances where even lower tube power ratings were used with inefficient speakers giving good results. I had heard it is attributed to the clipping characteristics of the two design types. Tube designs are more forgiving to the ear when they clip vs a solid state design. As clipping distortion sets in, a tube design may well introduce more harmonic distortion vs the IM distortion that a SS design might produce. Harmonic distortion is said to be the source of the "enhancement" in sound that tubes produce and why many prefer the sound of tubes. Objective power measurements must limit, or at lease state the distortion levels of the test, hence the dissimilarity of the power output ratings. Tube designs do not produce more accurate sound, they color the sound in a way that pleases your ear.

 

[gene_stl]

One cause of the problem you have identified (difficulty in getting"tons" of power out of a tube amp) stems from the following.

The electrons used to make the audio power in a tube amp all have to be emitted from the cathode which either IS the filament (directly heated) or right next to the filament (indirectly heated). The cathode has a chemical impregnation that when heated and subjected to a polarization voltage gives you a stream of electrons into the vacuum space next to it. Then the plate voltage attracts it.

Since most tube circuits have paired output topology you need to increase the number or the size or both of your output tubes (to make a bigger amp). Of course doing that does increase the power consumption of the filament and bias supplies . This is essentially wasted power and dumps heat into the room. Your MC275 has two KT 88 per channel and that is about as big as you can get with two. Then you need four which gets very expensive to operate and retube. (I had a pair of MC 75 back in the day and before that MC 40s and M 40s before that. The latter had 6L6 tubes which Fender used a lot of (four or two) in their amps.

But you are always limited in how many electrons come off that cathode. At AXPONA 2019 I saw one display in a large ballroom where there was a line of very large KT 88 powered tube amps which had been stuck along the back wall , all the way across the room. I had to go look at them. I think they were four KT 88 per channel and the room was much warmer back there. The system playing was probably multi million dollar.

One of the ways that tubes get "weak" is that their emitters get weak and don't furnish enough electrons. And this rarely happens symmetrically. So max power output drops and distortion goes up. (a process which begins the first time you turn the tube on and one of the reasons why I don't like them. The distortion and limiting sneaks in gradually so you don't notice it until suddenly you pull it out and test it and find something horribly wrong. Solid state is much more reliable and long lived)

Transistors are such a mature technology and used for other more important things than audio so output devices are available in any size and power you want and you don't have to waste power to squeeze electrons out of an emitter. The power supply furnishes the electrons and the output devices regulate them. Modern solid state output devices can be tough enough or paralleled enough to drive "almost short circuits".

And since they can drive the speaker load directly you don't need transformers which are expensive to make (try buying a McIntosh trifilar wound transformer when one shorts out) and have to be sized according to the power range. Heavy and spensive. They do contribute to soft clipping (along with emitters running out of electrons) By saturating their cores but if you were building one to beat the performance of a solid state amp , you would "oversize" it.

All of the foregoing STILL does not justify the pricing of some tube amps. Of course there are also lots of unconscionably price solid state and even class D amps, too.

I am going to bite my tongue and say that one advantage that may be with tube amps is that if you are multiamping, there are more small wattage amps with tube amps. And since you are getting rid of very inefficient crossovers you don't need big amps. Also a tube amp is less likely to thump a directly connected driver. (until you know how a transistor amp reacts at turn on and turn off you probably should under fuse and maybe capacitively couple your multiamped drivers) Tube amps are the original "soft start"devices. As you know I don't much care for tube amps. BUT a person could do WAY worse than to have a stack of four dyna Stereo 70s driving a pair of Kappas with a nice SS beast driving the wooofers. Pro amps seem to start at 300 watts per channel these days.

 

[JimHansonDC]

One cause of the problem you have identified (difficulty in getting"tons" of power out of a tube amp) stems from the following.

The electrons used to make the audio power in a tube amp all have to be emitted from the cathode which either IS the filament (directly heated) or right next to the filament (indirectly heated). The cathode has a chemical impregnation that when heated and subjected to a polarization voltage gives you a stream of electrons into the vacuum space next to it. Then the plate voltage attracts it.

Since most tube circuits have paired output topology you need to increase the number or the size or both of your output tubes (to make a bigger amp). Of course doing that does increase the power consumption of the filament and bias supplies . This is essentially wasted power and dumps heat into the room. Your MC275 has two KT 88 per channel and that is about as big as you can get with two. Then you need four which gets very expensive to operate and retube. (I had a pair of MC 75 back in the day and before that MC 40s and M 40s before that. The latter had 6L6 tubes which Fender used a lot of (four or two) in their amps.

But you are always limited in how many electrons come off that cathode. At AXPONA 2019 I saw one display in a large ballroom where there was a line of very large KT 88 powered tube amps which had been stuck along the back wall , all the way across the room. I had to go look at them. I think they were four KT 88 per channel and the room was much warmer back there. The system playing was probably multi million dollar.

One of the ways that tubes get "weak" is that their emitters get weak and don't furnish enough electrons. And this rarely happens symmetrically. So max power output drops and distortion goes up. (a process which begins the first time you turn the tube on and one of the reasons why I don't like them. The distortion and limiting sneaks in gradually so you don't notice it until suddenly you pull it out and test it and find something horribly wrong. Solid state is much more reliable and long lived)

Transistors are such a mature technology and used for other more important things than audio so output devices are available in any size and power you want and you don't have to waste power to squeeze electrons out of an emitter. The power supply furnishes the electrons and the output devices regulate them. Modern solid state output devices can be tough enough or paralleled enough to drive "almost short circuits".

And since they can drive the speaker load directly you don't need transformers which are expensive to make (try buying a McIntosh trifilar wound transformer when one shorts out) and have to be sized according to the power range. Heavy and spensive. They do contribute to soft clipping (along with emitters running out of electrons) By saturating their cores but if you were building one to beat the performance of a solid state amp , you would "oversize" it.

All of the foregoing STILL does not justify the pricing of some tube amps. Of course there are also lots of unconscionably price solid state and even class D amps, too.

I am going to bite my tongue and say that one advantage that may be with tube amps is that if you are multiamping, there are more small wattage amps with tube amps. And since you are getting rid of very inefficient crossovers you don't need big amps. Also a tube amp is less likely to thump a directly connected driver. (until you know how a transistor amp reacts at turn on and turn off you probably should under fuse and maybe capacitively couple your multiamped drivers) Tube amps are the original "soft start"devices. As you know I don't much care for tube amps. BUT a person could do WAY worse than to have a stack of four dyna Stereo 70s driving a pair of Kappas with a nice SS beast driving the wooofers. Pro amps seem to start at 300 watts per channel these days.

That all makes perfect sense and I know rationally that given my desire is to be as close to sitting in the master room as the engineer plays the final mix for the band, I should buy solid state. But, I'll readily admit I just love tube amps.

I have a lot of guitar heads where it makes sense for tubes as they function as a sound effect generator rather than a high fidelity audio reproduction tool. But the look and feel of a tube amp provides a combo of engineering and art that makes me willing to compromise on the very real points you make. I just like to see them glow and know that is where my juice comes from.

I think the placebo effect is in full effect on a lot of high end gear and accessories. I've had many electrically wise folks say that a lamp cord serves just as good a purpose for speaker wire as a pricy one. But I think Kimber Kables (the less expensive ones) look cool and that is worth the cost to me. When I see a $1500 power cord I just laugh. But that guy is making the same argument I am, just burning significantly more $100 bills. I find reading reviews of power cables for amps similar to wine reviews, lots of hint of cassis berry and note of unicorn farts. But I simply do not believe anyone can A/B test power cords and tell the difference consistently.

I have a bit more insight on the tube wattage game and I appreciate it. I like to know as much as I can to guide my decisions, but in the end. If I like it, it jumps the line.

 

[marpow]

This has been a good thread. I have learned a lot.
Like you Jim, I am a bit of a gear head and sometimes I think I get more enjoyment out of playing with gear than actually listening.
I love listening to stereo and was thinking of getting a tube preamp from McIntosh, the C2700 that would be what is called a pass though relationship. Just thinking about it at the moment.

 

[gene_stl]

Heavy gauge (10 12 or 14 ) are worth having, depending on run length. If you look in the owners manuals of pro power amps you find that the requirements aren't too much. Twisted pair or twisted quad is good. I like Canare pro cable. I see people talking about not using plated speaker cable. I doubt you can hear a difference. I used to use surplus aircraft 10 gauge twisted pair. You want to try and preserve damping factor. Plastic jacketing and trestles of rare woods do not improve sound. I buy my stereo more than enough presents as it is, I try to avoid ones that are completely unnecessary.

At AXPONA the power cable companies made cables available to people showing gear for the price of a little sign "power cables by whomever" Inside the wall are miles of 12 gauge solid Romex or equivalent. To think the last five or ten feet makes a difference is well , stupid. Power conditioners are also pretty dumb unless you know your power is bad. (like you live way out in the country, or in a third world area, or something) I do have a lot of MOVs and Arc surge suppressors hung on my system.

Expectation bias and placebo effect are why you need to test things blinded. Or have no expectation or prejudice. I used to do some tests on my system such as changing the crossover points and slopes to see if anything sounded better. The tests were not blinded but I had an open mind and did not really expect much difference. And what happened was that things sounded so alike that I knew if I was testing blind it would be the proverbial coin flip. I found out that the much vaunted Linkwitz Riley crossover arrangement, which I was a great follower of, is not actually audible. There are a lot of other audiophile tropes and memes that everyone "knows" are true that actually if you set up a blind test you most likely would not hear a difference or if you did hear a difference you could not decide which was which or which was "more accurate"

The amount of stuff you CAN'T hear or distinguish or decide which is really better/more accurate would probably amaze most people. The auditory mechanism of accommodation makes everything sound fairly good after a while. The audiophile's ear is not a good test instrument. Even when you "live with a piece of equipment for a period of time" you don't really know that much.

 

[ar surround]

The Logans are electrostatic and require a low impedance (high current drive). I had a set of summits years ago and I think they were a low of around 2 Ohms. Many solid state amps are designed for 8 Ohms and their ratings are very mush a peak rating. A good guide is to put them on a scale and if they weigh less than say 10 kg they are a toy (unless a class D).

Many Valve amps have a low impedance coupling step down transformer that can handle 2 Ohm loads. In addition they tend to soft clip, which reduces your perception of overload clipping distortion, so they sound heaps loader than the apparent rating.

I hope that helps a bit

Regards

Chucky

That's why I like my 2006 vintage Lexicon amps. Their rated output is down to 2 ohms with all seven channels driven. It's one of the few beasts that can handle my seven low impedance speakers, five of which are notorious for having troughs down to 3.2 ohms.

 

[Sonik Wiz]

Heavy gauge (10 12 or 14 ) are worth having, depending on run length. If you look in the owners manuals of pro power amps you find that the requirements aren't too much. Twisted pair or twisted quad is good. I like Canare pro cable. I see people talking about not using plated speaker cable. I doubt you can hear a difference. I used to use surplus aircraft 10 gauge twisted pair. You want to try and preserve damping factor. Plastic jacketing and trestles of rare woods do not improve sound. I buy my stereo more than enough presents as it is, I try to avoid ones that are completely unnecessary.

At AXPONA the power cable companies made cables available to people showing gear for the price of a little sign "power cables by whomever" Inside the wall are miles of 12 gauge solid Romex or equivalent. To think the last five or ten feet makes a difference is well , stupid. Power conditioners are also pretty dumb unless you know your power is bad. (like you live way out in the country, or in a third world area, or something) I do have a lot of MOVs and Arc surge suppressors hung on my system.

Expectation bias and placebo effect are why you need to test things blinded. Or have no expectation or prejudice. I used to do some tests on my system such as changing the crossover points and slopes to see if anything sounded better. The tests were not blinded but I had an open mind and did not really expect much difference. And what happened was that things sounded so alike that I knew if I was testing blind it would be the proverbial coin flip. I found out that the much vaunted Linkwitz Riley crossover arrangement, which I was a great follower of, is not actually audible. There are a lot of other audiophile tropes and memes that everyone "knows" are true that actually if you set up a blind test you most likely would not hear a difference or if you did hear a difference you could not decide which was which or which was "more accurate"

The amount of stuff you CAN'T hear or distinguish or decide which is really better/more accurate would probably amaze most people. The auditory mechanism of accommodation makes everything sound fairly good after a while. The audiophile's ear is not a good test instrument. Even when you "live with a piece of equipment for a period of time" you don't really know that much.

So are you saying my Tice Digital Audio Clock is... bogus?

I think it was the Absolute Sound I read an advertisement disguised as an interview in regards to line level cables. The mfg'er had noticed as the reader must have, music sounds better in the dark. He attributed this to the effect the darkness had on the inter-connects. After much research & testing it was discovered that the same beneficial effects could be achieved by just the opposite: bathing the conductors in light! So these mucho expensive light up cables came with there own power supply. However you could only see a bit light at the ends because they still had a conventional opaque covering.

 

[gene_stl]

I have picked up issues of the Absolute Sound many times and I always put them down very quickly. Even in the seventies my speaker building buddy and I called it "the Absolute Baloney". Were those cables from Lirpa Labs? (I don't know what a Tice Digital Audio Clock is.....)
That must have occured during the heyday of internally lit gaming computers.

I would guess that 99% of the forum members here operate systems that are so much better than whatever "average" is , that many multitudes of sins get covered up.

I did adore my McIntosh tube preamp (C20). Also AFTER I got my C21 which I was very happy with the aforementioned speaker buddy placed one of his SchMarantz 7C tube preamps with me. He had several. It sat on the floor for a long time because I "knew" it could not be as good , especially not as quiet as the C21. Finally he goaded me into hooking it up. I had concern that it would not be able to drive my Pioneer D23 crossover and especially I was demanding about noise. It was ,unbelievably, dead quiet. It worked and sounded great. I played with it for a long time until he sold it to another friend (who had built a pair of Audio Amateur powered electrostats: Danger Will Robinson!).

PS the loudness compensator on the C20 was the best I ever saw.

 

[Sonik Wiz]

I have picked up issues of the Absolute Sound many times and I always put them down very quickly. Even in the seventies my speaker building buddy and I called it "the Absolute Baloney". Were those cables from Lirpa Labs? (I don't know what a Tice Digital Audio Clock is.....)
That must have occured during the heyday of internally lit gaming computers.

I would guess that 99% of the forum members here operate systems that are so much better than whatever "average" is , that many multitudes of sins get covered up.

Those cables were real not Lirpa! And the Tice Clock cured the harsh sound of digital CD's by injecting opposite phase digital signals into the power line:
https://www.stereophile.com/artdudleylistening/listening_85/index.html

 

[gene_stl]

When I went to AXPONA I was walking down an aisle. Suddenly I was grabbed by a hook and a guy was talking to me about something, I needed to have. He could tell by looking at me that I had an expensive stereo. Or maybe I just looked like a sucker. I think it called ADD audio. They were gadgets of whatever size you wanted to spend on, from wall warts up to a couple of rack units. They went across the line and appeared to have inductors. "Fascinating" I said before I withdrew. For the rest of the show I walked on the opposite side of that aisle. Probably the same guy as Tice.

https://add-powr.com/products
He also sold holistic health products. through a sister company.

 

[bluelightning]

https://add-powr.com/products
He also sold holistic health products. through a sister company.

WTF!

 

[bluelightning]

I have picked up issues of the Absolute Sound many times and I always put them down very quickly. Even in the seventies my speaker building buddy and I called it "the Absolute Baloney".

Funny, you mention TAS, I have long felt the same about most of Stereophile as well.