Can you make a really good hifi amp by copying a late 70s schematic + good parts

[CJay]

     Make your PCB with 2oz of copper at least.  Old 70s design PCBs had no soldermask and usually flooded the traces with solder to thicken them up to handle higher current where needed if not everywhere.

Partly, yes to the 'heavy' copper, it definitely won't hurt and is worth doing but the tinning was, AFAIK, also to prevent corrosion, not just for current handling.

I'm talking about 70s 50-100watt designs with that huge 1/4inch thick layer of solder blobbed all over the traces around the power transistors.

Hence 'not just'

[Gyro]

If you want low distortion, Class AB, easy to build, just use this IC from TI: http://www.ti.com/product/lm3886

Use 1 per channel, or 2 per channel if you want a bridged output.
Make a stiff power supply near the IC's limit voltage, though it doesn't need to be regulated.
Get a big ass heat sink and you're away.

A different low distortion option would be to use 2 high voltage, medium current opamp from TI, each feed the bases of the transistors I mentioned earlier and with a 2x 0.2 ohm series resistors at the emitter with the center tied to your speaker output.  A little more experimenting and adjusting here, but, you can easily make a 100w / channel amp closer to the specs of a high performance, high current output amp like the pro-ones listed above, though, the distortion will be closer to 0.2%, not 0.03% like the TI LM3886 all in one IC.

The LM3886 does have a pretty good reputation in the 'Chipamp' comunity. Here's a useful site on it... https://www.neurochrome.com/taming-the-lm3886-chip-amplifier/

[BrianHG]

If you want low distortion, Class AB, easy to build, just use this IC from TI: http://www.ti.com/product/lm3886

Use 1 per channel, or 2 per channel if you want a bridged output.
Make a stiff power supply near the IC's limit voltage, though it doesn't need to be regulated.
Get a big ass heat sink and you're away.

A different low distortion option would be to use 2 high voltage, medium current opamp from TI, each feed the bases of the transistors I mentioned earlier and with a 2x 0.2 ohm series resistors at the emitter with the center tied to your speaker output.  A little more experimenting and adjusting here, but, you can easily make a 100w / channel amp closer to the specs of a high performance, high current output amp like the pro-ones listed above, though, the distortion will be closer to 0.2%, not 0.03% like the TI LM3886 all in one IC.

The LM3886 does have a pretty good reputation in the 'Chipamp' comunity. Here's a useful site on it... https://www.neurochrome.com/taming-the-lm3886-chip-amplifier/

@Beamin, just read the neurochrome website.  The TI LM3886 done right, feeding an 8 ohms speaker at 50w will give you a distortion below 0.01%.  Closer to 0.015% if you count the peak 20Khz at full power at that frequency...  That's 3x better than your current amp.

[Zero999]

I know this: Sony and most amps are around 0.8 THD my amp is 0.03 THD and sounds a whole lot better with my speakers. I haven't tried to see if cheaper speakers sound better. I don't think I, nor my speakers would notice any sound that's better and so I am shooting for the same levels at that watt range.

Nobody can dispute that.  However pinning "sound better" solely on distortion measurement is probably missing the point. There are probably at least half a dozen factors contributing to "sound better" that may or may not have anything to do with measured distortion performance.

Can anyone cite a proper test (ABX or double-blind) of 70s era BJT vs. 80s FET vs. 2010 era Class-D?  Or is this the same kind of audiophoolery as  "magic cables" and green marker on CD disks and wooden knobs? Count me skeptical.

Not class D vs class AB, but here's an interesting double-blind study done comparing several amplifier designs, back in the 70s, which showed no one can tell the difference, with everything else being equal.
http://www.keith-snook.info/wireless-world-magazine/Wireless-World-1978/Valves%20versus%20Transistors%20DCD.pdf

No doubt if the same test were conducted today, comparing amodern class D design with an older solid state design and an ancient valve design, the results would be the same.

[David Hess]

Or follow the instructions in the two books linked earlier for a completely discrete low distortion "blameless" design.  Even the simplified versions are in the 0.01% range.

Another option is the LT1166 Power Output Stage Automatic Bias System which can be used by itself but achieves much lower distortion within a feedback loop.  This might be combined with a "blameless" design to avoid the thermal compensation of a class-AB output stage.

[BrianHG]

Or follow the instructions in the two books linked earlier for a completely discrete low distortion "blameless" design.  Even the simplified versions are in the 0.01% range.

Another option is the LT1166 Power Output Stage Automatic Bias System which can be used by itself but achieves much lower distortion within a feedback loop.  This might be combined with a "blameless" design to avoid the thermal compensation of a class-AB output stage.

What an interesting little chip.  I had no clue something like that existed.  One of my early home made bipolar amps replicated it's function by using 2 normal, medium output current op-amps, one driving the NPN's base and the other driving the PNP's base in a push-pull output emitter follower.  The 2 series resistors at the output emitters and the fixed voltage offset at each op-amp's inputs created my regulated bias current, correcting for the thermal warming Vbe offset characteristics of the power transistors.  Worked like a charm.  It had an excellent 75 watt performance with 100w peak.  I biased the thing to around 50 watts idle.  Boy did it get toasty, but, it had such a warm filled rich sound.

[NiHaoMike]

Not class D vs class AB, but here's an interesting double-blind study done comparing several amplifier designs, back in the 70s, which showed no one can tell the difference, with everything else being equal.
http://www.keith-snook.info/wireless-world-magazine/Wireless-World-1978/Valves%20versus%20Transistors%20DCD.pdf

No doubt if the same test were conducted today, comparing amodern class D design with an older solid state design and an ancient valve design, the results would be the same.

Isn't the whole point of a tube amp the specific kind of distortion they make that alters the sound in a good way?

[David Hess]

Or follow the instructions in the two books linked earlier for a completely discrete low distortion "blameless" design.  Even the simplified versions are in the 0.01% range.

Another option is the LT1166 Power Output Stage Automatic Bias System which can be used by itself but achieves much lower distortion within a feedback loop.  This might be combined with a "blameless" design to avoid the thermal compensation of a class-AB output stage.

What an interesting little chip.  I had no clue something like that existed.  One of my early home made bipolar amps replicated it's function by using 2 normal, medium output current op-amps, one driving the NPN's base and the other driving the PNP's base in a push-pull output emitter follower.  The 2 series resistors at the output emitters and the fixed voltage offset at each op-amp's inputs created my regulated bias current, correcting for the thermal warming Vbe offset characteristics of the power transistors.  Worked like a charm.  It had an excellent 75 watt performance with 100w peak.  I biased the thing to around 50 watts idle.  Boy did it get toasty, but, it had such a warm filled rich sound.

The design I came up with long ago used a complementary common base input stage to drive current mirrors and produce an output.  The supply current to the common collector output stage was monitored on alternate half cycles from one collector when it should be idle and fed back to the biasing of the complementary common base input stage to trim the idle for every stage.  Full power bandwidth with 2MHz output transistors was about 500kHz, power supply rejection was surprisingly high, and stability and transient response were ideal.  The first version used a stack of diodes for thermal compensation and it ran away on a hot day; a Vbe multiplier might have worked better.

The next design I might try is a power version the one shown below which could be considered a voltage feedback operational amplifier driving a current feedback amplifier which neatly avoids the problems of achieving high full power bandwidth at the expense of input noise.

[Alex Nikitin]

Isn't the whole point of a tube amp the specific kind of distortion they make that alters the sound in a good way?

No - unless it is a guitar amp or another "effect processor" .

Cheers

Alex

[Zero999]

Not class D vs class AB, but here's an interesting double-blind study done comparing several amplifier designs, back in the 70s, which showed no one can tell the difference, with everything else being equal.
http://www.keith-snook.info/wireless-world-magazine/Wireless-World-1978/Valves%20versus%20Transistors%20DCD.pdf

No doubt if the same test were conducted today, comparing amodern class D design with an older solid state design and an ancient valve design, the results would be the same.

Isn't the whole point of a tube amp the specific kind of distortion they make that alters the sound in a good way?

That might have not been the case when the study was done. Back then (I can't be sure, as I wasn't around) I think audiophiles were still convinced that old glassware sounded better than solid state. In any case it proves that if a decent amplifier is not over-driven, it will not cause any audible artefacts, irrespective of whether it's solid state or thermionic. 

[dzseki]

Hi-Fi (let alone High-End) is a though topic, there is no ultimate solution, but opinions and preferences...

Relying only on THD measures may be misleading, as these parameters are born with some sort of static tests. To my understanding in the 70’s-80’s enthusiast engineers found out what TIM (Transient InterModulation) is. Until that amplifier designers focused on high open loop gain, and relied on the global feedback loop to decrease THD. Indeed this way it is possible to get very good THD measures for the borchure, but the trade off here is that the open loop bandwidth is low, which implies that the feedback loop „speed” is also bad. High TIM is the result of this suffering, and is often even noticeable when listening to music on such amplifiers.
Making a Low-TIM amplifier is a design concept, which relies highly on symmetrical circuit layout, and on more local feeedback loops rather than big global feedback.

There is plenty of information on the net over this topic, IMHO, this is the real game changer in the amplifier construction.
But this is only one aspect of the things, because a lot depends on actually what music do you listen usualy. Electrical music (Yello, J.M. Jarre, etc.) may sound very impressive on virtually any sound system, which may not be the case with a brass big band style music...

I enclose a mid 80’s amplifier schematic, which uses classic semiconductors but feature the „novel” circuit arrangement already, this amplifier was a part of Hungarian made active studio monitors, each amplifier drove a single speaker after active crossover network. It is a rather complicated circuit but if one takes the time to analyze can see each component have a good reason to be there...

[CatalinaWOW]

Hi-Fi (let alone High-End) is a though topic, there is no ultimate solution, but opinions and preferences...

Relying only on THD measures may be misleading, as these parameters are born with some sort of static tests. To my understanding in the 70’s-80’s enthusiast engineers found out what TIM (Transient InterModulation) is. Until that amplifier designers focused on high open loop gain, and relied on the global feedback loop to decrease THD. Indeed this way it is possible to get very good THD measures for the borchure, but the trade off here is that the open loop bandwidth is low, which implies that the feedback loop „speed” is also bad. High TIM is the result of this suffering, and is often even noticeable when listening to music on such amplifiers.
Making a Low-TIM amplifier is a design concept, which relies highly on symmetrical circuit layout, and on more local feeedback loops rather than big global feedback.

There is plenty of information on the net over this topic, IMHO, this is the real game changer in the amplifier construction.
But this is only one aspect of the things, because a lot depends on actually what music do you listen usualy. Electrical music (Yello, J.M. Jarre, etc.) may sound very impressive on virtually any sound system, which may not be the case with a brass big band style music...

I enclose a mid 80’s amplifier schematic, which uses classic semiconductors but feature the „novel” circuit arrangement already, this amplifier was a part of Hungarian made active studio monitors, each amplifier drove a single speaker after active crossover network. It is a rather complicated circuit but if one takes the time to analyze can see each component have a good reason to be there...

This post is a good example of why audio discussion often ends in hurt feelings and fist fights.

It starts with good factual information.  TIM is a real thing and very credibly one of the sources of "poor" listening experiences on some styles of music.

Then goes on with irrelevant or partially relevant statements.  Symmetrical layout has little, if anything to do with TIM, though it does have some other benefits and definitely helps the aesthetics.   Local feedback is one approach, but not "the" answer to reducing TIM (or to getting low THD while minimizing the use of feedback loops).

Even if the original designer and/or this poster understands the details of a particular approach there will be a lot of folks misunderstanding, quoting and fighting over these particular implementations of a design and their efficacy in generating a good listening experience.

[Zero999]

Hi-Fi (let alone High-End) is a though topic, there is no ultimate solution, but opinions and preferences...

Relying only on THD measures may be misleading, as these parameters are born with some sort of static tests. To my understanding in the 70’s-80’s enthusiast engineers found out what TIM (Transient InterModulation) is. Until that amplifier designers focused on high open loop gain, and relied on the global feedback loop to decrease THD. Indeed this way it is possible to get very good THD measures for the borchure, but the trade off here is that the open loop bandwidth is low, which implies that the feedback loop „speed” is also bad. High TIM is the result of this suffering, and is often even noticeable when listening to music on such amplifiers.
Making a Low-TIM amplifier is a design concept, which relies highly on symmetrical circuit layout, and on more local feeedback loops rather than big global feedback.

There is plenty of information on the net over this topic, IMHO, this is the real game changer in the amplifier construction.
But this is only one aspect of the things, because a lot depends on actually what music do you listen usualy. Electrical music (Yello, J.M. Jarre, etc.) may sound very impressive on virtually any sound system, which may not be the case with a brass big band style music...

I enclose a mid 80’s amplifier schematic, which uses classic semiconductors but feature the „novel” circuit arrangement already, this amplifier was a part of Hungarian made active studio monitors, each amplifier drove a single speaker after active crossover network. It is a rather complicated circuit but if one takes the time to analyze can see each component have a good reason to be there...

This post is a good example of why audio discussion often ends in hurt feelings and fist fights.

It starts with good factual information.  TIM is a real thing and very credibly one of the sources of "poor" listening experiences on some styles of music.

Then goes on with irrelevant or partially relevant statements.  Symmetrical layout has little, if anything to do with TIM, though it does have some other benefits and definitely helps the aesthetics.   Local feedback is one approach, but not "the" answer to reducing TIM (or to getting low THD while minimizing the use of feedback loops).

Even if the original designer and/or this poster understands the details of a particular approach there will be a lot of folks misunderstanding, quoting and fighting over these particular implementations of a design and their efficacy in generating a good listening experience.

I'd head of intermodulation and knew that non-linear amplifiers can cause it, but hadn't heard of transient intermodulation distortion, which seems like a different name for what I already knew. The trouble is, when dzseki started talking about symmetrical layout, I started to smell BS, which made me question TIM and the rest of the post.

Although layout can be important, in audio amplifiers, especially for high current paths and high impedance nodes, it's not as critical as in an RF circuit (unless it's a class D amplifier). There's no need to do anything crazy such as making it look a certain way i.e. symmetrical.

The LM3886 has a more than acceptable level of intermodulation distortion, probably better than many old discrete designs from the 70s.
http://www.ti.com/lit/ds/symlink/lm3886.pdf

[dzseki]

I may have been imprecise in my previous post, sorry, but at least gave some practical aspect to it (posted a schematic that the OP may appreciate).
While I have to admit there is no hard definition for a symmetrical layout for a low-TIM power amplifier, it is just all of them appear to be fully symmetrical (that I know of).

And while CatalaniaWOW started his post with this:

This post is a good example of why audio discussion often ends in hurt feelings and fist fights.

Then he goes on with qualifying my post (without any actual evidence or alternative solution offered) which eventually lead to the hurt feelings in me.  And now this trail goes on with Hero999’s suspicion on me BS-ing, which makes me curious where is it stated the LM3886 excels in low-TIM, because Intermodulation Distortion is not the same as Transient Intermodulation Distortion.

EDIT: added an article about TIM too.

[GeorgeOfTheJungle]

The next design I might try is a power version the one shown below which could be considered a voltage feedback operational amplifier driving a current feedback amplifier which neatly avoids the problems of achieving high full power bandwidth at the expense of input noise.

"Op Amp Applications Handbook" by Walter G. Jung :-) I had all his cookboks, were the best back in the day!

[dzseki]

Not class D vs class AB, but here's an interesting double-blind study done comparing several amplifier designs, back in the 70s, which showed no one can tell the difference, with everything else being equal.
http://www.keith-snook.info/wireless-world-magazine/Wireless-World-1978/Valves%20versus%20Transistors%20DCD.pdf

No doubt if the same test were conducted today, comparing amodern class D design with an older solid state design and an ancient valve design, the results would be the same.

Isn't the whole point of a tube amp the specific kind of distortion they make that alters the sound in a good way?

That might have not been the case when the study was done. Back then (I can't be sure, as I wasn't around) I think audiophiles were still convinced that old glassware sounded better than solid state. In any case it proves that if a decent amplifier is not over-driven, it will not cause any audible artefacts, irrespective of whether it's solid state or thermionic.

Interesting find, it is often overlooked that tube amplifiers emplyos different feedback structure (current feedback vs. the voltage feedback of the solid state amplifiers) this results in (an inherent) non-linear frequency response that depends on the actual speaker's impedance (and the woofer's impedance resonant point dominates here), this behaviour is contributed as the "warm sound" of the tubes.

[Zero999]

I may have been imprecise in my previous post, sorry, but at least gave some practical aspect to it (posted a schematic that the OP may appreciate).
While I have to admit there is no hard definition for a symmetrical layout for a low-TIM power amplifier, it is just all of them appear to be fully symmetrical (that I know of).

And while CatalaniaWOW started his post with this:

This post is a good example of why audio discussion often ends in hurt feelings and fist fights.

Then he goes on with qualifying my post (without any actual evidence or alternative solution offered) which eventually lead to the hurt feelings in me.  And now this trail goes on with Hero999’s suspicion on me BS-ing, which makes me curious where is it stated the LM3886 excels in low-TIM, because Intermodulation Distortion is not the same as Transient Intermodulation Distortion.

EDIT: added an article about TIM too.

Don't take it personally. It's impossible to distinguish between an honest error and deliberate bullshit. If this was about most other subjects related to electronics, one would normally assume any inaccuracies are honest mistakes, but because audio is full of bullshit, one's judgement is biased to presuming the worst.

[CatalinaWOW]

Sorry I hurt your feeling dzseki.  But I stand by my comments.  Which included a mention that you probably knew what you are doing.

The key points - you made an assertion that symmetrical layout is needed to control TIM.  I don't believe this is true, and you have partially retracted that statement.  Persons with limited understanding will take your original comment and run with it, leading to spreading conflict and misunderstanding.  As far as I can understand the only benefit to symmetrical layout for TIM is to make sure that unintended signal paths are balanced resulting in identical TIM in each channel which is a possible listening benefit.  Symmetrical layout has other benefits unrelated to TIM which are sufficient unto themselves.  Matching TIM is a speculative benefit.

You asserted that "the" way to control TIM is the reduced and local feedback.  I agree that it is a way, and perhaps one of the best.  Again a path for the relatively uninformed to start a religious war over the "right" way to design amplifiers.