I'm not sure why you'd build any other type. If you want to colour the sound coming out of your amplifier use a pre amplifier with tone controls. Don't build something with known shortcomings and then hope that it will produce sound to your liking.
So, basically: "I want a superior amp, but it MUST use this inferior technology". 🤷♂️
There. I said it.
But you did say "all thoughts would be most welcome"...
What if the sound of 'sweetness' is not simply due to a single commonly published 'figure of merit'? What if its a combination of multiple factors, some of which are not commonly measured?
I personally feel that S/N is underrated. That was why I took the "Is Your System Silent?" survey.
Ed
Ed
I took the test, however, the third sample sounds more distorted than the first two, I could not listen after that.
I took this test as well.
I can detect 1% distortion, however I can tolerate upto 7% distortion, which is a plus since I can listen to music until better equipment comes along.
In any case speakers add distortion and my mini-amp is less than 1% at less than 2W. 2 Watts.
There is another test you can try:
My results:
"You reached an audibility threshold of -18 dB. This value is near the mean value of the audibility threshold of all listeners participated in the test."
I took this test as well.
I can detect 1% distortion, however I can tolerate upto 7% distortion, which is a plus since I can listen to music until better equipment comes along.
In any case speakers add distortion and my mini-amp is less than 1% at less than 2W. 2 Watts.
There is another test you can try:
My results:
"You reached an audibility threshold of -18 dB. This value is near the mean value of the audibility threshold of all listeners participated in the test."
THD measurements would be so much more informative if some of the other variables were recognised as important and were controlled, instead of being left as random.
One in particular I have in mind: low output impedance. It typically goes down as the amplifier topology is tuned for lower THD. Given that 99.9x% of all speakers are duplex voltage-generating microphones (for simplicity, I'm just going to focus on what nearly everyone uses, not electrostatics or piezoelectric types for now), I think it's important to recognise the potential for unwanted interactions if the damping factor is too high.
E.g.:
1)the amplifier outputs an impulse.
2)it's converted to mechanical motion with a little bit of distortion.
3)a transverse wave travels across the cone like the proverbial ripple on a pond.
4) the wave hits the 1st major point of reflection, which is the speaker structure (surround and/or spider) and goes back to the voice coil.
5) The reflected wave applies force to the VC, but the damping factor is tuned for bass circa 50 Hz, not tiny midrange ripples around 2-5 kHz. The amplifier detects tiny voltage "errors" on the line and its feedback system automatically adjusts the current to "fix the voltage".
5i) Notice how the amplifier's negative feedback is no longer sticking to its day-job of keeping the amplifier linear, but it's now moonlighting as a short-circuit for the spurious voltages generated by the speaker.
5ii) Notice, after just one cycle of reflection, the signal has already been distorted twice. Electrical to mechanical, and then back to electrical again. This the basic recipe for foul-sounding cone 'break-up', which is often written off as a flaw inherent to the speaker on its own, without looking into the type of amplifier or whether not there are any additional components like resistors or inductors in between.
--
Looking at the systems people build, try using a 100+W class-B or AB amplifier with ultra-low THD together with high sensitivity pro woofers, and active-only crossover. It's going to be hard work to make it come together. Yet a "high distortion" amplifier will most probably sound a lot better, but it has nothing to do with the "sweetness" of the 2nd harmonic or anything like that, just a big improvement in the other paramters like the rate of energy decay when an impulse is applied.
One in particular I have in mind: low output impedance. It typically goes down as the amplifier topology is tuned for lower THD. Given that 99.9x% of all speakers are duplex voltage-generating microphones (for simplicity, I'm just going to focus on what nearly everyone uses, not electrostatics or piezoelectric types for now), I think it's important to recognise the potential for unwanted interactions if the damping factor is too high.
E.g.:
1)the amplifier outputs an impulse.
2)it's converted to mechanical motion with a little bit of distortion.
3)a transverse wave travels across the cone like the proverbial ripple on a pond.
4) the wave hits the 1st major point of reflection, which is the speaker structure (surround and/or spider) and goes back to the voice coil.
5) The reflected wave applies force to the VC, but the damping factor is tuned for bass circa 50 Hz, not tiny midrange ripples around 2-5 kHz. The amplifier detects tiny voltage "errors" on the line and its feedback system automatically adjusts the current to "fix the voltage".
5i) Notice how the amplifier's negative feedback is no longer sticking to its day-job of keeping the amplifier linear, but it's now moonlighting as a short-circuit for the spurious voltages generated by the speaker.
5ii) Notice, after just one cycle of reflection, the signal has already been distorted twice. Electrical to mechanical, and then back to electrical again. This the basic recipe for foul-sounding cone 'break-up', which is often written off as a flaw inherent to the speaker on its own, without looking into the type of amplifier or whether not there are any additional components like resistors or inductors in between.
--
Looking at the systems people build, try using a 100+W class-B or AB amplifier with ultra-low THD together with high sensitivity pro woofers, and active-only crossover. It's going to be hard work to make it come together. Yet a "high distortion" amplifier will most probably sound a lot better, but it has nothing to do with the "sweetness" of the 2nd harmonic or anything like that, just a big improvement in the other paramters like the rate of energy decay when an impulse is applied.
Last edited:
THD quoted by manufactures is measured under controlled conditions to achieve the best results. Connect the same amp to a live rig and it is going to behave very differently, especially at either end of the spectrum. Personally I have never been impressed with Lab Gruppen either for sound quality or reliability. Switched to ASC amps for the big installs several years back, much cleaner sound and no failures.. A major problem with modern PA systems is they are almost always played too loud. As human hearing starts to shut down it restricts both frequency and dynamic range. Knock the level back a few dBs and you will notice improved clarity and a much happier audience.
Not really topology war.
More specific to the design and what comes with it.
Having used D class, for live Bass Guitar
Majority have filtering, limiters, weak power supplys
overheat with distortion effects.
Advanced safety features, AKA shut down in middle
of set.
Might be plenty of good examples.
Ive shut down every backlined turd.
No snap to note attack.
Freakin hate the things
Nope
More specific to the design and what comes with it.
Having used D class, for live Bass Guitar
Majority have filtering, limiters, weak power supplys
overheat with distortion effects.
Advanced safety features, AKA shut down in middle
of set.
Might be plenty of good examples.
Ive shut down every backlined turd.
No snap to note attack.
Freakin hate the things
Nope
Even as I design PPM amps , I don't feel that is what makes them sound good. For a full range system - the IMD distortion , slew rate , and
damping factor are far more important. Being fast enough to recreate a complex waveform with multiple frequencies and negotiate the "jungle" of
impedances that the speakers present is the top consideration.
Layout that does not create a bad environment for audio is also important. Some SMPS's do this , but can be isolated by adhering to the "rules" of
keeping your analog and digital separate.
If you ignore all the OEM garbage coming out of Asia these days , we are lucky. SMD , better components , and lots of design information make designing
utter perfection (Wolverine) possible even for a pitchfork villager like me..... he he.
Only in 2020+ can we put 3ppm/20Khz-300V/us ... at 100V p-p on a credit card sized gold plated FR-4 for <10$ (below).
This is my next design to give to the community. (below 1) 60W/120W EF3 to go with it.... or lateral FET ??
And what a good bass/sub amp ... overdrive it and get very little content after 1k ! my cascode acts just like a valve stage... (below)
"hellraiser mini amp" will be the thread heading.
OS
damping factor are far more important. Being fast enough to recreate a complex waveform with multiple frequencies and negotiate the "jungle" of
impedances that the speakers present is the top consideration.
Layout that does not create a bad environment for audio is also important. Some SMPS's do this , but can be isolated by adhering to the "rules" of
keeping your analog and digital separate.
If you ignore all the OEM garbage coming out of Asia these days , we are lucky. SMD , better components , and lots of design information make designing
utter perfection (Wolverine) possible even for a pitchfork villager like me..... he he.
Only in 2020+ can we put 3ppm/20Khz-300V/us ... at 100V p-p on a credit card sized gold plated FR-4 for <10$ (below).
This is my next design to give to the community. (below 1) 60W/120W EF3 to go with it.... or lateral FET ??
And what a good bass/sub amp ... overdrive it and get very little content after 1k ! my cascode acts just like a valve stage... (below)
"hellraiser mini amp" will be the thread heading.
OS
Attachments
(Many pages have passed and i have not read any of it).
No, THD is not very useful. It is the scalar produced by collapsing the distortion spectrun and the instance it is collapsed the important information is lost and the resulting number becomes largely meanngless.
A good designer can use it during development, but as an end user i pay little attention.
dave
There is indeed precisely such a thing. Harmonic profile is more important than a single THD number.
How much THD/noise does a potentiometer add?
I was shocked to find out.
I was shocked to find out.
One habit I have is to glance at the THD measurement, and use that to gauge how much I trust the Wattage promised. If the THD is really high, then my judgement is that I should treat the amplifier as if it's usable Wattage rating is significantly lower than advertised. Conversely, if the THD is really low, then I can probably safely use the full Wattage promised. e.g. A 60W amp with 0.01% THD is probably going to sound better than a 1000W amp with 1% THD, even with the same speaker. In general, I believe that THD goes up with amplitude, so you can enjoy lower THD on most amps by simply turning them down. That may not be true for all amplifier circuit topologies, though.
But to answer the original question, THD is definitely not the only specification that you need to be concerned about. THD is not even measured and reported consistently, so even if it were the magical one-spec-to-rule-them-all (which it isn't) you couldn't compare different amplifiers based solely on THD because it's not even apples to apples.
That said, I believe THD is "useful" albeit only mildly useful, and sometimes it's even absent from the specs that you can find (for vintage amps).
But to answer the original question, THD is definitely not the only specification that you need to be concerned about. THD is not even measured and reported consistently, so even if it were the magical one-spec-to-rule-them-all (which it isn't) you couldn't compare different amplifiers based solely on THD because it's not even apples to apples.
That said, I believe THD is "useful" albeit only mildly useful, and sometimes it's even absent from the specs that you can find (for vintage amps).
@rsdio "But to answer the original question, THD is definitely not the only specification that you need to be concerned about. THD is not even measured and reported consistently, so even if it were the magical one-spec-to-rule-them-all (which it isn't) you couldn't compare different amplifiers based solely on THD because it's not even apples to apples.
That said, I believe THD is "useful" albeit only mildly useful, and sometimes it's even absent from the specs that you can find (for vintage amps)."
THD is just another overly hyped area indeed, important to people insanely worried about trivialites, aka the nitpickers who insist on Holy Grail perfection.
Which, thankfully, are in the minority.
That said, I believe THD is "useful" albeit only mildly useful, and sometimes it's even absent from the specs that you can find (for vintage amps)."
THD is just another overly hyped area indeed, important to people insanely worried about trivialites, aka the nitpickers who insist on Holy Grail perfection.
Which, thankfully, are in the minority.
So what is to be done? Choose by listening and matching your speakers to you amplifier, source material and listening levels?
One of my biggest nightmares (imagined) is purchasing an amplifier and speakers and having it sound like garbage.
One of my biggest nightmares (imagined) is purchasing an amplifier and speakers and having it sound like garbage.
Don't forget the dac, if you are using one. In the consumer market measurements and reviews in some mix may be all you can do if there is nowhere you can go to listen to what you are considering. Buying from someplace like Amazon may give you more practical return options as versus something like Aliexpress.
If in the high end market, then carefully considered reviews may be more telling than measurements, in balance of course. Both resources can be useful to point out red flags, its just that one may chose to give reviews more weight than if in the consumer market. Still probably wise to have some return option too.
I don't think it's fair to dismiss THD completely, just because the minority give it too much weight. In other words, to avoid giving THD too much weight, we should not overreact and give it absolutely no weight at all.
High distortion amplifiers sound bad. In my experience, it's possible to find bad amplifiers if you spend too little on the technology. At other other end, though, once THD gets into the range where it's "low enough," you might find an amplifier with "more" distortion sounds better than one with "less" distortion. This is complicated because you might find that you can turn down the amp with high distortion because it has higher gain, internally, and then you end up experiencing less distortion than the exact THD figure quoted on the spec sheet. As others have pointed out, there are many kinds of distortion and the harmonics can appear in different frequency ranges.
I don't like the advice to "just use your ears" because I don't think anybody has time to listen to every amplifier on the market. I look at specs and cross out the models that don't meet the thresholds that have worked for me in the past. Then I'm left with the amps in my price range that meet my very general requirements. Of course I listen, but I spend far more time listening to different speakers than I dedicate time to listening to a large number of different amplifiers.
Remember, although we might want a magic recipe that's incredibly simple, there isn't a short sound bite that will tell you everything you need to know about choosing audiophile gear.
When reading reviews, if there is a red flag it is usually in the last or the second to the last paragraph. The text may suggest that you might want to compare the reviewed device to another before deciding. Or it may describe where the reviewed device is a tradeoff in some way, some good, some not as good, relative to some other device. If you read enough reviews with that sort of thing in mind, you will learn to recognize when you are being given a mildly framed warning. Take it seriously.
Regarding amplifier distortion, IMHO some very perceptive commentary at:
https://www.diyaudio.com/community/threads/euphonic-mechanisms-in-amplifiers.397918/post-7336457
Regarding amplifier distortion, IMHO some very perceptive commentary at:
https://www.diyaudio.com/community/threads/euphonic-mechanisms-in-amplifiers.397918/post-7336457