Going tubes...

[BobW]

The issue was cross-over distortion

I think an even bigger problem was transient IM distortion. The early designers realized that with low signal level transistors being small and cheap, they could put a lot of them into the circuit, to give phenomenal voltage gain, and then add gobs of negative feedback in a single feedback loop from the output back to the input, to reduce the harmonic distortion practically to zero. However, the slow NFB loop couldn't respond to transients, and gave horrible IM distortion on transients.

 

[Nigel Goodwin]

TID wasn't even 'discovered' until many years later, and transistors (while small) certainly weren't cheap in the early days, and circuits used relatively few of them.

The 'TID' situation was relatively small compared with cross-over distortion, and once that was overcome there were many truly amazing sounding amplifiers from companies like Quad and Leak, who's products from back then are still revered and sought after even now. I did have an excellent Leak Stereo 70, I only replaced it as it wouldn't 'play nicely' with a three head cassette deck I bought, and I would have kept it, but someone I knew begged me to sell it to them (and I can't even remember who it was now?). But what I always wanted was a Quad 33/303, but I never managed to come across one for a decent price.

 

[oemcar]

Thanks to all for your thoughtful responses-
Not sure if those that paid close to $20K USD for tube designs are smart or crazy per link below...

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 - The Absolute Sound

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 -

www.theabsolutesound.com

What I have gleaned from online research is double blind studies concluded that 2nd and 4th harmonic noise of tube amps were more pleasing (perhaps acceptable?) to the ear than 1st and 3rd from solid state- even though solid state did have, on average, much less % THD than tube variants.

From this, it seems some opinions here are undoubtedly good points...
So yes, Gophert, I will build said tube amp for my own evaluation... not to make a profit at someone else's expense.
And I'll give you my word that I'll report on sound quality, best as these 55yr old ears can detect- lol. If it's a turd, then so be it- you were right.

Thank you electro-tech..
Jim

 

[Ylli]

Another issue with the tube amp is that the output impedance is much higher than the output impedance of a transistor amp. The higher output impedance gives a much lower damping factor, allowing the speaker cone to overshoot at lower frequencies. This produces a more 'basey ' sound.

 

[Nigel Goodwin]

Another issue with the tube amp is that the output impedance is much higher than the output impedance of a transistor amp. The higher output impedance gives a much lower damping factor, allowing the speaker cone to overshoot at lower frequencies. This produces a more 'basey ' sound.

Valve amps use impedance matching, hence MUCH lower efficiency, and as you say, much lower damping. However, it wasn't a problem as the speakers were designed for that type of use, and you shouldn't really use modern 'transistor' speakers on valve amps - or the other way round.

 

[Nigel Goodwin]

Thanks to all for your thoughtful responses-
Not sure if those that paid close to $20K USD for tube designs are smart or crazy per link below...

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 - The Absolute Sound

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 -

www.theabsolutesound.com

What I have gleaned from online research is double blind studies concluded that 2nd and 4th harmonic noise of tube amps were more pleasing (perhaps acceptable?) to the ear than 1st and 3rd from solid state- even though solid state did have, on average, much less % THD than tube variants.

Quite true, odd harmonics are FAR more displeasing than even ones, which is why valve amps are often used by guitarists for their specific distortion.

From this, it seems some opinions here are undoubtedly good points...
So yes, Gophert, I will build said tube amp for my own evaluation... not to make a profit at someone else's expense.
And I'll give you my word that I'll report on sound quality, best as these 55yr old ears can detect- lol. If it's a turd, then so be it- you were right.

By all means build one, it should be fun, and valve amps sound 'OK', at one time they were all you could get - just that the specs are far inferior to even cheap transistor designs. Please post pictures of your design as you go along.

If you want any schematics drop me a PM, as I have a number of old valve amplifier design books.

 

[gophert]

So yes, Gophert, I will build said tube amp for my own evaluation... not to make a profit at someone else's expense.

I didn't mean to imply that you were going to make a profit. I had assumed you were going to buy one and pump $/€/£/¥ or some permutation thereof into the economy. If you are just buying pieces, I'm disappointed in myself for not catching that part of your question.

Also, don't give your silicon-amp to your son until you can do a side-by-side comparison.

 

[schmitt trigger]

The key to a good sounding tube amp are its output transformers.

Good output transformers are massive, with a price to match.

The golden standard for many years were the McIntosh output transformers.

 

[gophert]

The key to a good sounding tube amp are its output transformers.

Good output transformers are massive, with a price to match.

The golden standard for many years were the McIntosh output transformers.

The other problem of DIY amps are that people try to pack everything in a small space but then they get cross-talk of 50/60Hz power supply humm into the speaker output transformers. Orientation + Distance can overcome the problem easily but once the board is cut and populated, most people don't want to start over.

 

[crutschow]

My main dislike of tube amps was that they are high-impedance, low-current devices, so their output signal has to go through a less-than-ideal, impedance-changing transformer before going to the speakers.
It costs a small fortune to make a good output transformer to minimize the transformer deficiencies over the full audio range, especially at power outputs over a few tens of watts..
When transistor amps first appeared, the fact that they could direct couple to the speaker without the transformer seemed the ideal way to make a speaker amp.
I have seen a tube design with dozens of tubes to get a low enough output impedance to directly drive the speaker, which a solid-state amp can do with a pair of transistors.

 

[Nigel Goodwin]

The other problem of DIY amps are that people try to pack everything in a small space but then they get cross-talk of 50/60Hz power supply humm into the speaker output transformers. Orientation + Distance can overcome the problem easily but once the board is cut and populated, most people don't want to start over.

BOARD! - BOARD! - you don't use 'boards'

Just wire the components to the bottom of the valve holders and large capacitors, or use tag strips if and where required - use twisted wires for the heater connections. You can even use tag boards, but there seems little point other than in a production environment, where the tag boards could be pre-assembled.

KB (Kolster Brands) even sold valve colour TV's hand wired with no PCB's - now those had some serious wiring.

Most of the designs in my valve books give actual wiring instructions, or at least a photo of a completed amplifier.

If you're using PCB's you may as well make a transistor amp.

What's the Point of Point-to-Point?

This is a Guest Post by Bruce Clement of BC Audio I am often asked why I use what I refer to as true point-to-point wiring in my amps. Why go through the extra work? Does it really make a difference in the sound? While point-to-point wiring is certainly not the only path to great tone, I'd like...

www.guitartonetalk.com

 

[gophert]

BOARD! - BOARD! - you don't use 'boards'

Just wire the components to the bottom of the valve holders and large capacitors, or use tag strips if and where required - use twisted wires for the heater connections. You can even use tag boards, but there seems little point other than in a production environment, where the tag boards could be pre-assembled.

KB (Kolster Brands) even sold valve colour TV's hand wired with no PCB's - now those had some serious wiring.

Most of the designs in my valve books give actual wiring instructions, or at least a photo of a completed amplifier.

If you're using PCB's you may as well make a transistor amp.

What's the Point of Point-to-Point?

This is a Guest Post by Bruce Clement of BC Audio I am often asked why I use what I refer to as true point-to-point wiring in my amps. Why go through the extra work? Does it really make a difference in the sound? While point-to-point wiring is certainly not the only path to great tone, I'd like...

www.guitartonetalk.com

I'm sorry you assume my use of Board with a Printed Circuit board.

Once the mounting FACEPLATE and Top BOARD are cut, it is very demotivating and challenging to convince a. First-time amp builder to cut all of those transformer wires and pilot lite wires, some POT wires (maybe), lengthen them as needed and design & build a new Chassis.

I mentor a robotics/mechatronics club at a high school. As soon as they realize other things around their house use electronics and they start googling & hear about former students' projects, Interested students want to build a guitar effects pedal (I advise them) and then a guitar tube amp (I insist they are on their own - until they come back to complain about 60Hz buzz). Some work from kits with pre-drilled chassis (some are well-designed, some not), some work from a purchased kit of parts and must build their own box, and some work just from a schematic as their first project. Of the ~10

 

[Nigel Goodwin]

I'm sorry you assume my use of Board with a Printed Circuit board.

As that's the only use I know of for 'board' in such a project, then obviously I did

From what you've since said presumably you're refering to the chassis?, or some such mechanical construction?.

 

[audioguru]

In about 1962 (58 years ago!) I built a Heathkit amplifier that used tubes. Then a MacIntosh clinic came to town and offered a free test. I thought it sounded great but they measured 25% distortion near full output power and sold me replacement tubes (the originals were less than 1 year old) and the distortion dropped to 0.5%. A few months later the distortion was 20%. I sold the amplifier to an old man who liked old stuff and in 1964 I bought a solid state stereo receiver that I still use today.

My first color TV used tubes and ICs. I replaced its tubes many times for it to last for many years.
Last year I bought my first large flat screen 4K TV (at a food store on Black Friday) for a very low price.

 

[Nigel Goodwin]

In about 1962 (58 years ago!) I built a Heathkit amplifier that used tubes. Then a MacIntosh clinic came to town and offered a free test. I thought it sounded great but they measured 25% distortion near full output power and sold me replacement tubes (the originals were less than 1 year old) and the distortion dropped to 0.5%. A few months later the distortion was 20%. I sold the amplifier to an old man who liked old stuff and in 1964 I bought a solid state stereo receiver that I still use today.

My first color TV used tubes and ICs.

Were you a late adopter of colour then?.

I know North America continued making valve TV's for years after Europe had gone all solid state, but I would have hoped that they would have been phased out before IC's made any kind of impact in TV's?.

Obviously colour came earlier over there, but IC's weren't used until a fair bit later - even the early solid state sets over here were before IC's was used (in TV's).

 

[audioguru]

My first job was working for Philips so my first color TV was a Philips designed in Holland but made in Canada. Its "brains" was a Philips "jungle" IC and it used an RCA CRT. I modified it to decode pay TV.

 

[gophert]

I modified it to decode pay TV.

I trust you did it to eliminate the ugly cable box but you still paid the monthly fee. (ha. ha, ha!)

 

[audioguru]

Fee? If they wanted me to pay for it but I didn't want to pay then they should have stopped sending it to me.
It allowed my "education" about *ex and about encryption. I did not record then sell the recordings.

 

[atferrari]

Thanks to all for your thoughtful responses-
Not sure if those that paid close to $20K USD for tube designs are smart or crazy per link below...

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 - The Absolute Sound

2019 High-End Audio Buyer’s Guide: Power Amplifiers $5,000 - $20,000 -

www.theabsolutesound.com

What I have gleaned from online research is double blind studies concluded that 2nd and 4th harmonic noise of tube amps were more pleasing (perhaps acceptable?) to the ear than 1st and 3rd from solid state- even though solid state did have, on average, much less % THD than tube variants.

From this, it seems some opinions here are undoubtedly good points...
So yes, Gophert, I will build said tube amp for my own evaluation... not to make a profit at someone else's expense.
And I'll give you my word that I'll report on sound quality, best as these 55yr old ears can detect- lol. If it's a turd, then so be it- you were right.

Thank you electro-tech..
Jim

Gear even costlier **broken link removed**

 

[audioguru]

A long time ago, new solid state amplifiers produced odd-numbered harmonics but tubes produced even-numbered harmonics that old people think was in the music. Now old people like the harmonic distortion produced by tubes better than having no distortion.

The buyers guide has no spec's. Maybe the output transformers in the tube amplifiers muffled high frequencies and hiss the music?

It is normal for men to have high frequency hearing loss beginning when they are only 20 years old without playing with guns and loud acid rock noises.
At 65 years old, a man does not have golden hearing, instead he can barely hear any sounds above 4kHz like in this graph: