• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Looking beyond the single headline SINAD value...

Ufff SINAD at 15 KHz? SINAD is THD+N in Db scale. This doesn't apply. Second harmonic of 15 KHz is 30 KHz. So it's desirable that the noise stays low but THD is not relevant.

What the graphs show is the THD + N ratio expressed in dB (aka SINAD) when driven at 15kHz or 10kHz or 5kHz:
1000025990.png

When driven with a test tone at 5kHz the THD + N ratio (dB) would be only -70dB in the above. (Compared with the expected performance of -80dB, based on the 1kHz curve).

If the intermodulation test was done with tones at 4KHz and 5kHz (which could easily occur in music) then you would have IMD sidebands at 3kHz an 6kHz (well inside the audible range). My conjecture is that the IMD (as signified by the sidebands) if at sufficient levels could cause the mushiness of sound that people often comment on.
 
Last edited:
However, as you can see from the IMD graphs above there are a lot of components in the human hearing range, and these are comparable in magnitude to the ultrasonic components that you refer to.

The magnitude is also 90dB or more below the reference signals. Hardly anything worth worrying about.
 
Interesting new video about THD..
  • Distortion profile much more important than %THD value
  • H2,H4 and H3,H5 can be benign
  • odd order >=H7 can be objectionable
  • some people may prefer more H2,H4
  • some people may prefer more H3,H5
  • NFB trigger higher gain requirement, resulting in problems
...then Darko took the opportunity to take a swipe at SINAD (29:40) :)
...then took a swipe at Topping DACs (33:30) :)
(Joker ... looks like he is ignorant that the SINAD chart do show the profile ...)

 
Last edited:
Interesting new video about THD..
  • Distortion profile much more important than %THD value
  • H2,H4 and H3,H5 can be benign
  • odd order >=H7 can be objectionable
  • some people may prefer more H2,H4
  • some people may prefer more H3,H5
  • NFB trigger higher gain requirement, resulting in problems
...then Darko took the opportunity to take a swipe at SINAD (29:40) :)
...then took a swipe at Topping DACs (33:30) :)
(Joker ... looks like he is ignorant that the SINAD chart do show the profile ...)

I caught this. He started by describing the harmonics which make different musical instruments sound different when playing the same fundamental as "harmonic distortion". At least his guest (from 6 moons I think) put him right on that!

But the guest then went on to "explain" why a tube amp with 10% THD might sound better than a solid state amp with THD of a few hundredths of a percent - it's all about high orders of distortion, particularly above 5th, apparently. He then went on to wax lyrical about the ability to "tune" speakers by adding warmth (with even order distortion) or sharpness (with odd order distortion) from a tube amp.

Of course, in theory, different types and mixes of harmonic distortion are going to sound different, but two massive issues were left unanswered: firstly, no discussion on the orders of magnitude greater THD from a tube amp, however benign (or looked at the other way, even if all of the say 0.01% THD of a modern amp is of the "nasty" type, how audible is that likely to be?). And secondly, no mention of the actual frequencies which would be involved in 9th, 11th, 13th order harmonic distortion.

Then Darko said the current fashion for SINAD measurements is even worse than the competition between manufacturers in the '70s and '80s for good measurements, because this time it's not driven by manufacturers but by idiots on the internet, and finished off a discussion which had been all about amplifiers by transferring the same theory to DACs to tell his audience how it explained why he didn't like Topping Dacs much, at least when compared to much more expensive ones.

It sounded to me like a massive apologia for expensive, badly-measuring tube amps on the part of the guest, and from Darko, a masterclass in passive aggression aimed at sites like this one.

It also seemed instructive that the entire segment (maybe the whole video, I stopped watching after the THD part) had the two talking heads in little boxes, with the vast majority of the screen taken up with what seemed to be adverts for expensive looking tube amps.
 
My 2 cents: Pick any well designed amp with...low enough residual noise to make tweeter hiss a non-issue,
How low is low enough? I saw AVRs with SINAD 98 dB, even a 107 dB horn should have hiss like 10 dB, inaudible. OK that's not technically correct, I am massively conflating some things, but looking at numbers it seems hiss should never be a problem. And yet it IS, for instance at this thread
https://www.diyaudio.com/community/...-amp-for-high-sensitivity-horn-drivers.409494
I don't get the discrepancy.

I'm also puzzled by the SINAD data themselves-my distinct recollection is the Denon AVR-X3600H and some immediate successors at like 98 dB. "Better than CD!" again not really but kind of. Now I see their AVRs like 88 dB in the latest charts-?!? What the hell happened? Is @amirm measuring differently now and I missed it?
 
Was the hissing heard at 1m? Or with ear right up at the driver (1-2cm)?

System SINAD of 90dB should not hiss at 1m.

I did a quick check with online spl distance calculator. (omni calc)

-90dBSPL (1m) ==> -50dBSPL (1cm)

-50dBSPL could be audible with ear right up against driver (1cm)
 
Last edited:
Was the hissing heard at 1m? Or with ear right up at the driver (1-2cm)?

System SINAD of 90dB should not hiss at 1m.

I did a quick check with online spl distance calculator. (omni calc)

-90dBSPL (1m) ==> -50dBSPL (1cm)

-50dBSPL could be audible with ear right up against driver (1cm)

90 dB SINAD at what level? I find calculating system residual noise in uV at the speaker terminals much more useful in evaluating hiss as it is independent of level. Of course if you know speaker sensitivity you can calculate dBSPL hiss from residual noise.

How low is low enough? I saw AVRs with SINAD 98 dB, even a 107 dB horn should have hiss like 10 dB, inaudible. OK that's not technically correct, I am massively conflating some things, but looking at numbers it seems hiss should never be a problem. And yet it IS, for instance at this thread
https://www.diyaudio.com/community/...-amp-for-high-sensitivity-horn-drivers.409494
I don't get the discrepancy.

I'm also puzzled by the SINAD data themselves-my distinct recollection is the Denon AVR-X3600H and some immediate successors at like 98 dB. "Better than CD!" again not really but kind of. Now I see their AVRs like 88 dB in the latest charts-?!? What the hell happened? Is @amirm measuring differently now and I missed it?

If you follow that thread to completion you will see where I show the OP how to calculate residual noise, see here -> https://www.audiosciencereview.com/...s/minidsp-flex-htx.49512/page-17#post-1906197. Again, residual noise is much more valuable in assessing hiss.

Michael
 
Back
Top Bottom