Power amplifiers that excel with very low impedance - and low efficiency loads

[a.wayne]

Now if totally optimized for low z operation those Boulder 8 ohm numbers could be matched at 4&2 ohms with modifications to the voltage , feedback and bias settings. Lamm attempts this route on their Hybrids to get their best low Z operation from those amplifiers, by applying both a high Z and low Z switchable operation ..


Regards

 

[atmasphere]

You're pointing out exactly what I'm talking about.

The thing is, the industry has spent a lot of time telling audiophiles that the slight difference you see in the charts is 'negligible' and 'inaudible'. Quite simply it isn't. What is being ignored here is that the ear has over a 120dB range and uses the higher orders to sense sound pressure- and so is keenly sensitive to them, far more so than anything else. As a result we can hear these 'tiny' (because that's how they look on paper) differences- that's why audiophiles hear differences in amps, not so much due to FR variation, but due to the distortion signature.

For the application the Boulder is probably one of the better choices! But saying that its going to sound its very best is an entirely different matter- the ability to drive low impedances isn't what makes an amp sound good or bad. Distortion is what does affect that and clipping has nothing to do with it. IOW the differences we hear between amps is due to the distortion they make when they are not clipping, because the ear pays a lot of attention to the tonality created by that distortion. Again, that's why you can have two amps on the bench that both measure perfectly flat, but one is bright and the other is not. The brightness is the result of non-clipping distortion.

That BTW is why the tube industry is still very much around; for the last 60 years we've been hearing solid state amps be bright (and of course the guitar world plays an enormous role). You might want to read what Bruno Putzeys has to say about distortion (if you've not already).

 

[a.wayne]

Toobs are still around because Audio has nothing to do with good sound it has and always will be about what one likes, one man’s Audio Euphoria is and will always be another's anathema, So there’s good sound available from Toobs , SS , digital and or analog and i have never owned or used a bright or Harsh SS amp in 40 plus yrs, its an old straw man argument very similar to what pro SS people do when they say Toobs are dark and slow ...

So while some toobs maybe dark and slow and some SS harsh and bright these are characteristics of poor design choices ..

That aside ,

you will have to prove That brightness between SS amplifiers were not from non clipping or running out of class A bias for the load , we have done that rodeo too many ( Bob Cordell even went as far as to do a show setup to prove this by showing how much power was necessary to prevent clipping from only 1-2 watt rms use ) times and a non clipping properly biased for the load SS amp will never sound harsh or bright unless from a very poorly designed amp, it’s academic we are not discussing poorly designed amplifiers and if you level match many properly designed amps will sound very similar, as to being without harshness and or brightness , its mostly there clipping /recovery characteristics is what really affects what most people hear and of course their choices ..!

Maybe you can explain away these high levels of distortion from toobs Or does distortion only matters with SS amps ....?

1 Audio Research REF 160S, Triode mode, 4 ohm tap, THD+N (%) vs frequency at 6.3V into: 16 ohms (left channel green), 8 ohms (left blue, right red), 4 ohms (left cyan, right magenta) 2 ohms (left gray).

 

[atmasphere]

I can,

This is an example of how the distortion signature affects the sound you hear. A large 2nd and 3rd harmonic helps mask the presence of the higher orders. This helps keep the amp sounding smooth even though it has more higher ordered harmonic content than most solid state amps.

Put another way, the distortion signature is pretty important to how smooth the amp sounds. The trick is getting it down overall- tube amps in general are not very good at that since its hard to build them with enough feedback to do the trick (IOW phase margin is an issue).

One other thing: the distortion really should be the same at 1KHz and 10KHz as it measures at 100Hz. Many amps have insufficient gain bandwidth product, which is another way of saying that as frequency is increased, the feedback decreases, so we see distortion rising with frequency.

The problem is that feedback introduces distortion in the form of higher ordered harmonics through the process of bifurcation (see Norman Crowhurst- he was writing about this 60 years ago). In addition you can get intermodulations at the feedback node if you're not careful. Since feedback suppresses the innate distortion of the amp, the remaining distortion is often that of the feedback itself and it is unmasked, so easy to hear.

If you have enough feedback you can get around this problem. Imagine a bell curve; feedback does its worst damage between 15-25 dB (which is an amount commonly used). If you get over 35dB the amp can clean this problem up and you can have a much smoother sounding amp. Class D seems to offer a nice way to do this in self-oscillating examples.

You can get tubes to sound just fine but not every manufacturer of tube amps is successful. Plus you do want to be careful to not use them with speakers that they are simply a bad match. But I've found that latter bit true of transistors as well: not every speaker works with every amp and vice versa.

 

[a.wayne]

You have to match SS amps to the load in the same way you do for Toobs , any and all distortion is audible nothing is masked to those who can discern it , Class D or any SS has its sound same as high distortion toobs and or low distortion high feedback toobs, Only amps with bad open loop gain tend to have NFB issues ...

Again a lower THD high dynamic low Z operating loudspeaker is superior sounding to a high Z variant with less drive area , this will more than compensates for any small increase in amplifier THD ....!

A committed low Z designed SS can be of low distortion and flat bandwidth 20-20K same as 8 ohm if optimized in its design phase..

Lamm M1.2 Reference in low Z operation mode optimized , "1–6 ohms," THD+N (%) vs frequency at 8.95V into: 8 ohms (blue), 4 ohms (magenta), 2 ohms (red).

Is that flat enuff for you ...?


Regards

 

[atmasphere]

Class D or any SS has its sound same as high distortion toobs and or low distortion high feedback toobs, Only amps with bad open loop gain tend to have NFB issues ...

If I understand you correctly I partially agree. I'll make it a bit more clear: If they have the same distortion signature it won't matter what kind of amp it is, it'll sound the same. But what is meant by 'bad open loop gain'??

Again a lower THD high dynamic low Z operating loudspeaker is superior sounding to a high Z variant with less drive area , this will more than compensates for any small increase in amplifier THD ....!

It may well- but increase the Z and you have the best of both worlds.

A committed low Z designed SS can be of low distortion and flat bandwidth 20-20K same as 8 ohm if optimized in its design phase..

Of course!! Why did you bring this up?

Lamm M1.2 Reference in low Z operation mode optimized , "1–6 ohms," THD+N (%) vs frequency at 8.95V into: 8 ohms (blue), 4 ohms (magenta), 2 ohms (red).

Is that flat enuff for you ...?


Regards

Sure! But what does its distortion signature look like? I've heard that amp a few times so I think I can take a good guess that its low in 2nd and 3rd harmonic content. IOW another example of exactly what I've been talking about. You can't know how the amp sounds by looking at the THD; you'll have a greater understanding by knowing what its primary distortion product actually is. In audio amps you tend to get two types of non-linear functions- quadratic and cubic. A fully single-ended circuit will exhibit the former while a fully differential circuit will exhibit the latter (and so be lower in overall distortion). If both types are present you'll get some algebraic summing, like you see in a Dyna ST70 or the like where the 5th will have a bit more amplitude. I recommend the writings of Norman Crowhurst and Bruno Putzeys regarding this particular aspect of topology. Sorry if this seems off in the weeds but its directly related to the issue of distortion being audible.

 

[a.wayne]

I’m familiar with Crowhurst , Geddes , Putzeys, et al , its not relevant to the discussion about low Z SS amps as we are not debating distortion signature, we are debating your comment on why low z is bad in loudspeakers...



Regards

 

[TheOctopus]

Loved my Lamm M1.2 hybrid monos. Should have kept them.

 

[a.wayne]

Yea ,

unlike your Mac those can handle the Pre-conditioning test at 1/3 power ..

 

[TheOctopus]

Yea ,

unlike your Mac those can handle the Pre-conditioning test at 1/3 power ..

Never had a problem with either.

 

[atmasphere]

I’m familiar with Crowhurst , Geddes , Putzeys, et al , its not relevant to the discussion about low Z SS amps as we are not debating distortion signature, we are debating your comment on why low z is bad in loudspeakers...



Regards

Hm. If you really think that then you missed the point of my comment. It was in fact entirely about how low Z loudspeakers cause all amps to have increased distortion. And yes, the writings of those authors are quite relevant. They can be quite helpful in guiding one towards something more musical as opposed to something less musical.

And with a lot of solid state amps that have a distortion signature that allows the higher ordered harmonics to be heard as brightness and harshness, this simply means that the system will sound less 'real'. But in the OP's situation, the distortion is corralled a bit by being limited to the woofer or woofer array, thus limiting the effect somewhat (unless the woofer also has a choke in series, some of that distortion will be reproduced).

By simply employing a higher impedance (which might be as simple as changing the series/parallel arrangement if its an array) this distortion will be reduced.

BTW you didn't answer my question about what is meant by 'bad open loop gain'. I'm very curious what you meant by that.

 

[a.wayne]

Well..


You missed my point entirely Ralph , mainly how SS amps designed for low Z operation( proven and shown as you requested look at lamm ) can have low distortion as well as how low Z loudspeakers if designed correctly has less distortion than high Z , making the whole combination superior granted far more technically difficult and not as easy as say an high Z , 2 way or 3 way system with usually higher Thd

As to your umpteen time Strawman argument about distortion signature , that’s all getting into subjective audio blah blah as to who likes what and is a total side step of the point being made.

Measured thd , not subjective likes and dislikes ...!


BTW , your suggestion of simple change of series / parallel to increase impedance load is nothing more than band aid for what is really an amp mis-match situation , you should investigate how series coils affects the sound vs parallel circuits, in regards to loudspeakers ..


Regards

PS: Maybe i should have used poorly designed open loop gain, instead of bad ..



My Bad ..

 

[atmasphere]

Well..


You missed my point entirely Ralph , mainly how SS amps designed for low Z operation( proven and shown as you requested look at lamm ) can have low distortion as well as how low Z loudspeakers if designed correctly has less distortion than high Z , making the whole combination superior granted far more technically difficult and not as easy as say an high Z , 2 way or 3 way system with usually higher Thd

As to your umpteen time Strawman argument about distortion signature , that’s all getting into subjective audio blah blah as to who likes what and is a total side step of the point being made.

Measured thd , not subjective likes and dislikes ...!

This is almost entirely false. All humans use the same hearing perceptual rules- for example all humans use the higher ordered harmonics to sense sound pressure. That's not a matter of taste. The masking principle is also common to all humans. There's no strawman going on; simply the nuance of what happens when you try to use any amp to drive an impedance that low.

The 'almost part' is the THD of the Lamm, which is unusual. But without knowing the distortion signature you can't just say that 'lower THD is better'. Lower THD and a proper distortion signature such that higher orders are masked is what will result in an amp that is musical and not bright. Low THD on its own can easily result in a bright sounding amp. You can try to make the argument that such is subjective, but its also well-known that humans find the higher orders unpleasant and that understanding was published way back in the 1930s. If you're going to subject an amplifier (including the Lamm in this case) to a situation where you know its going to make those higher orders you can count on the result being less musical and you don't have to hear it to know that. All you have to know is the science of how the amps work and how that interacts with human hearing perceptual rules.

If you design an amp without paying attention to the aspects of human hearing you stand a good chance of making something that will sound bad. Its not subjective so much as its engineering applied to physiology.

BTW , your suggestion of simple change of series / parallel to increase impedance load is nothing more than band aid for what is really an amp mis-match situation , you should investigate how series coils affects the sound vs parallel circuits, in regards to loudspeakers ..

I have investigated that and it works fine as long as all the drivers are the same kind and interacting in the same way in their enclosure.

 

[a.wayne]

Well i guess everyone hears the same and like the same sound, it’s settled then, err , if not for the unsettling Irony ...




Regards

 

[atmasphere]

Well i guess everyone hears the same and like the same sound, it’s settled then, err , if not for the unsettling Irony ...

Regards

That's not what I'm saying. I am saying that the rules that govern how we perceive sound are the same rules- this is how MP3s are possible. They rely on the masking principle to compress the file size. And its also true that humans don't like higher ordered distortion components - that's why most guitar amps are tube amps, although to be clear in the case of the OP I'm not talking about an overdriven amp.

 

[a.wayne]

You’re obfuscating ,

It’s apples and oranges , the discussion has nothing to do with distortion harmonic (complete different discussion) structure its all academic really, these are all known knowns , yet you keep wanting to keep this circular Strawman argument, waffling between objective and the subjective ..

Whats your thoughts on Specialized vs Trek ..?


Regards

 

[a.wayne]

That's not what I'm saying. I am saying that the rules that govern how we perceive sound are the same rules- this is how MP3s are possible. They rely on the masking principle to compress the file size. And its also true that humans don't like higher ordered distortion components - that's why most guitar amps are tube amps, although to be clear in the case of the OP I'm not talking about an overdriven amp.

Your good entertainment Ralph , Whats your thoughts on why most amplifiers sold are class D....

 

[a.wayne]

The Anthem P2 does not get hot beyond ambient air temperature, I have used a P2 amp for over 8 years and have never had it get hot or shut down and I have driven speakers down around 1-2 ohms for 10-12 hours at a time at 3/4+ power and this amp will just keep going. If you want look up the reviews for this amp they have been made for at least 10 years.
This amp is basically 2 mono amps in one chassis, no fuses, and I have never had the reset buttons pop to reset this amp, they run around 5k new but used can be had from 2-3k

You have 1 ohm speakers .? Or does your speakers have a 1-2 ohm dip at a particular Frequency ..?



Regards

 

[atmasphere]

You’re obfuscating ,

It’s apples and oranges , the discussion has nothing to do with distortion harmonic (complete different discussion) structure its all academic really, these are all known knowns , yet you keep wanting to keep this circular Strawman argument, waffling between objective and the subjective ..

Whats your thoughts on Specialized vs Trek ..?


Regards

If you think I'm obfuscating or waffling, then you've missed my point.

Class D offers a way (through self-oscillating design) to allow for a large amount of feedback (+35dB) that is otherwise difficult to design into traditionally-designed solid state or tube amplifiers. This is done by intentionally exceeding the amp's phase margin, so it oscillates. The oscillation is then used as its switching frequency. This allows for an amp that lacks the typical brighter/harsher presentation than the prior art (otherwise to avoid this you have to avoid feedback altogether). A very nice benefit is they are also cooler running, more efficient and lower cost. Class D amps have a different artifact that is actually similar in some ways to tube amps. Dead time causes distortion; but it tends to be of the lower harmonic variety, as do non-linearities in the encoding scheme. This allows for an amp that not only has the right distortion signature, but also low THD, the best of both worlds. Of course if you overdrive it, it will clip like any other solid state amp. So don't clip it.

 

[Dizzie]

Right now we're working on a class D amp of our own design.

Could you hurry that up please? :thumbsup: I would love a cool-running, high-power, great-sounding amplifier that I can lift.

All too frequently it is over 100 degrees here so no Class-A. Tubes are relegated to winter use but swapping amps is getting harder every year. Every Class-D amp I have tried did everything right--except make music. :dry: At least to my ears.