• 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!

God of SINAD vs. reality we get from most available music files

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
It is interesting that it was Nakamichi - as the Stasis amps were a development and derivation of Quads current dumping....

So they would both have been touting different takes on a very similar underlying principle!

Not based on current dumping at all (I checked the schematics of several Statis amps) It was more like class-G type designs (splitted rail)
The only brand that did something similar-ish was Technics (for a few years and then abandonned that design) and they too used modulated rails driving a smaller amplitude class-A amp.
 

dlaloum

Major Contributor
Joined
Oct 4, 2021
Messages
3,092
Likes
2,353
Not based on current dumping at all (I checked the schematics of several Statis amps) It was more like class-G type designs (splitted rail)
The only brand that did something similar-ish was Technics (for a few years and then abandonned that design) and they too used modulated rails driving a smaller amplitude class-A amp.
I believe the underlying principle of the Stasis amps, same as for the Quad current dumpers was a high quality Class A voltage driver, supported by a Current driving circuit which could be Class B or Class G or Class D for that matter... the principle was called current dumping by Peter Walker - but "Feed Forward" on some of the later derivatives.... Quad used a class B circuit for the heavy lifting - Devialet uses a Class D I believe so the heavy lifting circuit has varied over the years but the principle of a small high quality Class A circuit with heavy lifter supporting it has become common.

The AHB2 is another example of the genre...
 

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
15,891
Likes
35,912
Location
The Neitherlands
Yes, these are not current dumping though. But rather class-G (variants).
The Quad works via a wheatstone bridge where the OP drivers are driven by a class-A driver stage which also acts as the small part where it provides class A (and the class-B still does nothing)

The other designs are fundamentally different and have low voltage (but high idle current) class-A output stages with the power supply of that amp being modulated by a class AB amp.
The weird part is that the current always comes from the class-AB output stage (as it basically is in series with the class-A) so solves nothing except it makes for a nice story and one can say there is a class-A design (high idle current) in there.

The Quad is very different though. Only the extra current is provided by the output stage and it thus is only class-A near the 0 crossing. The idle current for the class-A part (basically the output driver stage) can thus be very low.
All the other designs do not do it this way and are fundamentally different and thus not based on the current dumping principle.
Only Technics used a (different config) wheatstone bridge and took feedback from another point.

All of this in an attempt to lower crossover distortion (which is not an issue these days) and to have high output power and low heat generation or to use the magic 'class-A' label. That seems to attract class-A believers.
 
Last edited:
OP
pma

pma

Major Contributor
Joined
Feb 23, 2019
Messages
4,591
Likes
10,727
Location
Prague
I prefer output stage error correction circuits fighting with crossover distortion (see Cordell, Hawksford, Bruce Candy) to various attempts with pseudo-class A. Below please see a result with a simple class AB amplifier with MOSFET output stage with moderately low idle current, with error correction circuit. This results in fast decay of high order harmonics. Without the error correction circuit, we have the well known "comb-like" spectrum.

PM-AB2_5W_4ohm_1kHz.png


PM-AB2_25W_4ohm_1kHz.png


Another example - below see the Halcro patented circuit

1660638899817.png
 
Last edited:

SoundGuy

Active Member
Joined
Aug 13, 2022
Messages
115
Likes
120
Here is the F208 I referred to:

View attachment 224271


And Amir, are you trying to make my point for me? the charts you posted of the 328Be show THD peaking at around 0.6% above 50Hz...

Your own comment on one graph is how well it is performing at only 3% THD @ 30Hz... which unless I am much mistaken, calculates to a SINAD of around 30?... and yes if we ignore the bass the rest of the frequency range manages a very decent 0.2% THD.... (SINAD 53...)
I don’t think harmonic distortion can be conflated with noise. It is a different animal. The Revel results are excellent. A big thing about speakers is actually how the transducers decay. Intrinsically damped drivers and boxes that don’t resonate are very important. The room too (as Sabine discovered). Phase is also very important in speakers. Ideally measurements should focus on factors that affect how we hear. For me, a speaker waterfall plot is very important. Flat frequency response and even dispersion are important to portraying a natural sound but if that was all that mattered then many speakers would sound identical and they don’t. I don’t like metal or rigid drivers. I hear problems with nearly all of them. I believe it is the added ringing which I object to - I much prefer damped cones that ‘technically‘ aren’t as good because I can hear more details when a cone decays quickly and without a characteristic ringing associated with the cone itself. Technically some cones are so light and hard that they ring outside the audible range - but I can still hear issues with them that nobody has measured.

All to say, measurements are extremely important - especially for Design and QC. However, we don’t know all there is to know about human hearing to be able to define specific rules about what the collective importance and balance of measurements should be. A lot of things we don’t even measure….all we can say is that better SINAD is definitely desirable as long as it doesn’t result in other audible compromises.
 

krabapple

Major Contributor
Forum Donor
Joined
Apr 15, 2016
Messages
3,168
Likes
3,713
No, I covered two scenarios. One where the product is expensive, and one where it is not. Think of me as the food inspector. I don't care if you are not going to get sick if there is a violation of food safety. My job is to make sure that doesn't happen. You as a consumer can decide if you still want to eat at a restaurant with such violations. Don't tell me to not do my job.

I'm not telling you to not do your job. I'm 'telling' those who are reflexively dismissive when they see a meh SINAD number, to reconsider.
 

Palladium

Addicted to Fun and Learning
Joined
Aug 4, 2017
Messages
630
Likes
769
My fault for over-generalising,

When I think back on classic gear I have owned... the Quad gear is particularly "repairable/maintainable"...

I also had/have Revox gear and various Japanese components - and in terms of repairability / complexity, I think they were probably much of a muchness

But current trends towards very small, surface mounted components, and a very high density of components, does not bode well for repairability down the line... And the trend towards replacement of entire circuit boards (or entire components) rather than doing repairs, is very very clear.

Started by Motorola in the early 90's after they worked out that the cost, to them, of a replacement mobile phone, was less than the cost of receiving and shipping, in addition to the logistical labour, and the physical repair time combined... in fact it cost them several times the value of a new phone to repair an old one...

Onkyo never even tried to do any component replacement on my failed 2008 HDMI board (which according to many DIY postings, could be fixed by replacing the caps) - they provided me a replacement board.... and some years later when that failed - offered me another replacement board for $900.... (!!) - I feel that this is now the norm.
Other than the occasional greybeards out there.... ( a few of whom are on here) - how many current AV techs/engineers even bother with component level repairs? :(

Yeah times have changed; even manually replacing BGA chips is a crapshoot under the best of circumstances, and who is really gonna do that on let's say a $60 PC motherboard?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,384
Location
Seattle Area
I'm not telling you to not do your job. I'm 'telling' those who are reflexively dismissive when they see a meh SINAD number, to reconsider.
Looking the other way on SINAD is what has gotten us all these DACs in red and orange:

index.php


And wrong ideas about audio design such as discrete, low or no feedback, tubes, etc.

Remember, as I have said repeatedly, you will hear the quieter noise floor of higher SINAD DACs. They will be audible with sensitive IEMs and speakers. Or at elevated playback level.

Nothing good comes out of ignoring lackluster SINAD. It encourages worse designs, rewards companies building the same and no benefit whatsoever for consumers.
 

Lambda

Major Contributor
Joined
Mar 22, 2020
Messages
1,785
Likes
1,519
But current trends towards very small, surface mounted components, and a very high density of components, does not bode well for repairability down the line... And the trend towards replacement of entire circuit boards (or entire components) rather than doing repairs, is very very clear.
This is not the problem not that things get smaller.
People adapt to this very fast.

The problem is intentionally obfuscating the design. removing marking, poting parts in black epoxy, and generally not providing documentation or parts.


I have 20-30 year old gear that comes with schematics and has labeled test points. it's made for maintenance.
wonder how mush of today's tech will be usable in 30-40 years.

Don't think making things more repairable would cost performance. but it would maybe add cost.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
This is not the problem not that things get smaller.
People adapt to this very fast.

The problem is intentionally obfuscating the design. removing marking, poting parts in black epoxy, and generally not providing documentation or parts.


I have 20-30 year old gear that comes with schematics and has labeled test points. it's made for maintenance.
wonder how mush of today's tech will be usable in 30-40 years.

Don't think making things more repairable would cost performance. but it would maybe add cost.
The first Audio Research solid state preamps had potted modules and obscured parts labeling. So the company known for vacuum tubes was using Op-amps. They didn't want that widely known.
 

theREALdotnet

Major Contributor
Joined
Mar 11, 2022
Messages
1,181
Likes
2,036
The first Audio Research solid state preamps had potted modules and obscured parts labeling. So the company known for vacuum tubes was using Op-amps. They didn't want that widely known.

It wouldn’t be hard to hide an opamp between the plates and package it all inside a glass tube, if that’s what people want to buy… :)
 

theREALdotnet

Major Contributor
Joined
Mar 11, 2022
Messages
1,181
Likes
2,036

GXAlan

Major Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
3,866
Likes
5,953
PS Has anyone seen a Topping service manual?

@JohnYang1997 ? Can you share a service manual for an out of production Topping?

I have 20-30 year old gear that comes with schematics and has labeled test points. it's made for maintenance.

Amir currently has in his posession my JBL SA600 which was an estate sale find that came with an original, non photocopied instruction manual from 1965. Not only was the theory of operation published for others to review, the service manual is extremely detailed includes internal RCA jacks for testing, etc. The specific transistors use JBL specific numbering it is also straight forward. So there is something that is almost 60 years old that is repairable!

Denon and Marantz are also extremely good at documentation whether you are looking at the Reference series almost 20 years ago or contemporary AVRs like the X8500H.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
It wouldn’t be hard to hide an opamp between the plates and package it all inside a glass tube, if that’s what people want to buy… :)
Oh they admitted it was solid state, but wanted everyone to assume it was some great discrete circuit. It in fact was just opamps. Which they called Analog Modules. They put them inside this big plastic cover and filled it with epoxy. This was late 1970s to early 1980s. Like this.
1660707910411.jpeg
 

theREALdotnet

Major Contributor
Joined
Mar 11, 2022
Messages
1,181
Likes
2,036
They put them inside this big plastic cover and filled it with epoxy.

Does anyone know what’s inside PS Audio’s ”gain cell” plastic modules? I’m guessing a Jones/Gilbert mixer of sorts, for voltage controlled amplification?
 
D

Deleted member 50971

Guest
So if I’m getting the gist of this, is this a fair statement;

The best our gear can do is passthrough what it’s given, IE: the music.

Would that be achievable by gear with the same SINAD as the music, or would it take transparent gear for this to happen?
 
D

Deleted member 50971

Guest
the signal. Versus adding noise and distortion to it
Thanks. So a SINAD (the sum of our gear) equivalent to the music or greater will accomplish this?

What I’m getting at is we can have this great measuring gear, but if the musics SINAD is 70 it appears that it’s wasted. Which I believe is @pma point.
 
  • Like
Reactions: pma
Top Bottom