Pump up the Volume (in various projects)

[hi30u]

But an attenuator is not an essential capability of a power amplifier. The OP is confused, thinking the lack of an attenuator is an amplifier class thing.

I also get the feeling they believe that amplifier variable gain is achieved through adjusting the rail voltages (perhaps they are thinking of Class H etc.) or through changing the PWM behaviour of Class D.

You are right, it is not supposed to be there. Otherwise it would not be a monobloc but an integrated...
... which is exactly what I am trying to achieve. Or at least try to know if it is possible.

The first answer in this thread is very important I believe. If my $200 DAC goes wild, I will lose my $2,000 speakers and what's left of my ears!
So, unless I go for FDA (and sell my DAC), I will make the choice of using a safe way to control the volume.
So between the DAC and the monoblocs, I can use a DIY passive preamplifier or a cheap/good Topping pre-amp, one with relays and resistors.

Or, find a whatever class digital amp with variable gain in its output stage. Does this exist or is it impossible?
Digital amp because it is energy efficient and it has great review on ASR (in class D AFAIK).

 

[Killingbeans]

What about a modern 32 bit DAC?

It becomes a complete non-issue :

 

[hi30u]

It becomes a complete non-issue :

Thanks for finding and sharing this ESS document.

Here I copy and paste the ESS "Conclusion":

• Analog volume controls easily outperform digital, unless the digital control has access to the data path of the DAC (ie is internal to the DAC)
• Exquisitely well designed analog volume controls can still beat even the very best internal digital volume controls if they have a lower noise floor than the DAC itself
  • The -135dB of the ESS Sabre DAC would need an exceptionally low noise analog volume control to beat its internal digital one

Therefore, as many suggest on ASR, using the volume control feature of a recent DAC is indeed the best approach.
So, to answer the third case of my first post, for sure, when using a Class A/B dedicated power amp (or two A/B monoblocs), I should connect the DAC directly to the power amp.
I just wish my DAC had a dedicated volume knob or some sort of physical proof that the volume is set low before I turn it on.

Three questions remain.

One: is whether one can DIY a "Exquisitely well designed analog volume controls" passive attenuator with -130dB low noise?

Two: For a "digital" monobloc amplifier, such as Class D or another class, is it possible to change paradigm from controlling the volume "before" or "pre" (as in pre-amp) to controlling the volume inside the amplifier (by adjusting a constraint in the way the power is pulse-switched on and off)?

Three: For an FDA Amplifier (Power DAC) such as Sabaj A30A or SMSL VMV A2, I just found some info about how the volume is controlled.
They use a AX5689 Digital audio converter and amplifier controller

AX5689 - Axign

www.axign.nl

Question: does the AX5689 integrated circuit works using 32 bits in the same way as an ESS circuit in a DAC, therefore matching the -135dB of the ESS Sabre DAC?

 

[Killingbeans]

For an FDA Amplifier (Power DAC) such as Sabaj A30A or SMSL VMV A2, in the event that the volume is controlled before the Power Stage, does the FDA's integrated circuit works using 32 bits in the same way as an ESS circuit in a DAC?

Most likely. You'd have to ask Axign to be sure.

 

[MaxwellsEq]

Two: For a "digital" monobloc amplifier, such as Class D or another class, is it possible to change paradigm from controlling the volume "before" or "pre" (as in pre-amp) to controlling the volume inside the amplifier (by adjusting a constraint in the way the power is pulse-switched on and off

Theoretically you could vary an amplifier's gain by tweaking the feedback loop (classes A, A/B) or modifying the switching processes / frequency in a Class D amplifier, but both approaches would result in appalling degraded performance as well as very unpredictable stability and radio frequency behaviours. So nobody would do that in the baseband audio domain where you want a control to go from zero to full beans and where reliability is generally a requirement!

You might want to read up on class G and H amplifiers which are like class A an A/B amplifiers but with switched or tracking voltage rails. As power consumption increases, the amplifier can call on higher rail voltages to enable more output swing.

 

[Doodski]

Two: For a "digital" monobloc amplifier, such as Class D or another class, is it possible to change paradigm from controlling the volume "before" or "pre" (as in pre-amp) to controlling the volume inside the amplifier (by adjusting a constraint in the way the power is pulse-switched on and off)?

Hello @hi30u. Class D amplifiers are not digital they are an analogue operation.

 

[hi30u]

Hello @hi30u. Class D amplifiers are not digital they are an analogue operation.

True.
But they switch (0 and 1) some kind of pulse (samples) at a frequency that complies with the Nyquist–Shannon sampling theorem .
I am not too sure that that theorem is relevant here...

So they behave "digital", correct?
Do their success in DIY kits and good reviews by Amirm come from that BTW?

 

[Doodski]

True.
But they switch (0 and 1) some kind of pulse (samples) at a frequency that complies with the Nyquist–Shannon sampling theorem .
I am not too sure that that theorem is relevant here...

So they behave "digital", correct?
Do their success in DIY kits and good reviews by Amirm come from that BTW?

No, they are analogue in all aspects. Forget all about thinking because there are square waves that are then filtered down to the audio output. It is analogue.

 

[Doodski]

Do their success in DIY kits and good reviews by Amirm come from that BTW?

I don't understand the question.

 

[JayGilb]

True.
But they switch (0 and 1) some kind of pulse (samples) at a frequency that complies with the Nyquist–Shannon sampling theorem .
I am not too sure that that theorem is relevant here...

So they behave "digital", correct?
Do their success in DIY kits and good reviews by Amirm come from that BTW?

No, class D amplifiers use a PWM (pulse width modulation) scheme in which the internal signal is either on or off and just the width of the signal pulse changes.
A final low pass filter converts it back into the amplified analog output signal.

 

[hi30u]

No, class D amplifiers use a PWM (pulse width modulation) scheme in which the internal signal is either on or off and just the width of the signal pulse changes.
A final low pass filter converts it back into the amplified analog output signal.

So class D are not bit perfect, correct?
They could create distortion by a wrong (none accurate) width, or some kind of jitter, correct?
Then, FDA could too.
OK, never mind, out of topic!

Theoretically you could vary an amplifier's gain by tweaking the feedback loop (classes A, A/B) or modifying the switching processes / frequency in a Class D amplifier, but both approaches would result in appalling degraded performance as well as very unpredictable stability and radio frequency behaviours. So nobody would do that in the baseband audio domain where you want a control to go from zero to full beans and where reliability is generally a requirement!

You might want to read up on class G and H amplifiers which are like class A an A/B amplifiers but with switched or tracking voltage rails. As power consumption increases, the amplifier can call on higher rail voltages to enable more output swing.

This is the final answer to my question #2. What ever class, A/B or D, either use the DAC preamp function or build a "Exquisitely well designed analog volume controls". If ever possible!

 

[hi30u]

Do their success in DIY kits and good reviews by Amirm come from that BTW?

I don't understand the question.

I started the conversation in the DIY thread for a reason, which is to eliminate the weakest link (after the room and the speakers).
To check if I need to (or can) build a passive preamp.
My question is also:
If they are not digital or perfect by design, why are the class D kits so popular in DIY projects ?
Why class D amps have so good scientific results when Amir tests them?

 

[Doodski]

o check if I need to (or can) build a passive preamp.

Tell me about your abilities and I'll see if I have suggestions.

If they are not digital or perfect by design, why are the class D kits so popular in DIY projects ?

They run cooler on average, they cost less for high wattage output, they are more compact, they are more efficient with main power consumption for starters.

hy class D amps have so good scientific results when Amir tests them?

Because it is the nature of the design. The PWM power supply design makes for low distortion induced onto the audio output waveform. Yes, the switch mode power supplies used with class D audio amps both use PWM operation.

 

[hi30u]

Tell me about your abilities and I'll see if I have suggestions.

I will make mistakes if I need to solder 100 metal film resistors to an empty attenuator.
Otherwise, I can be meticulous.
About stepped attenuators with SMT resistors, do you recommend such technology?
SMT can be 0.5% which is better than DALE.
I plan to use two attenuators, one per channel.
Short length cables, etc...

 

[Doodski]

I will make mistakes if I need to solder 100 metal film resistors to an empty attenuator.
Otherwise, I can be meticulous.
About stepped attenuators with SMT resistors, do you recommend such technology?
SMT can be 0.5% which is better than DALE.
I plan to use two attenuators, one per channel.
Short length cables, etc...

Get yourself solder rosin flux for a perfect solder job. You can get a pen applicator of no clean flux but after you will want to clean it all off with isopropyl alcohol and maybe if perfect rinse it with methyl hydrate for a perfect finish. Be careful because the liquids are flammable.

https://www.digikey.com/en/maker/blogs/2023/what-is-solder-flux-and-why-you-should-use-it

 

[MaxwellsEq]

So class D are not bit perfect, correct?
They could create distortion by a wrong (none accurate) width, or some kind of jitter, correct?
Then, FDA could too.
OK, never mind, out of topic!



This is the final answer to my question #2. What ever class, A/B or D, either use the DAC preamp function or build a "Exquisitely well designed analog volume controls". If ever possible!

Correct. But the "exquisite" adjective is unnecessary. Classic "preamplifiers" are rarer than they were, but there are some well designed devices. If you don't have long cables and your impedances are sufficient, a "passive attenuator" can be sufficiently good.

 

[MaxwellsEq]

So class D are not bit perfect, correct?

Class D amplifiers are not bit perfect, neither are rabbits, boulders, false teeth, trouser presses.

The only things that can be considered bit perfect are devices which read stored bit patterns and devices which convert bit patterns. A CD player can be bit perfect, or not. A streamer can be as well, or not. As can a DAC, or not.

A Class D amplifier IS NOT DIGITAL so has no concept of bit perfect, rather like cheese, trainers, windmills and scuba masks.

 

[ppataki]

No, class D amplifiers use a PWM (pulse width modulation) scheme in which the internal signal is either on or off and just the width of the signal pulse changes.

I think this picture here helps with the understanding:

PWM is on the right-hand side (analog data)

From: https://en.wikipedia.org/wiki/Pulse-width_modulation

I would not like to digress but in the same Wikipedia article they mention:

In more recent times, the Direct Stream Digital sound encoding method was introduced, which uses a generalized form of pulse-width modulation called pulse-density modulation, at a high enough sampling rate (typically in the order of MHz) to cover the whole acoustic frequencies range with sufficient fidelity. This method is used in the SACD format, and reproduction of the encoded audio signal is essentially similar to the method used in class-D amplifiers.

Does this mean that if DSD = PDM which is a generalized form of PWM (which is analog data) then DSD can also be considered as an analog signal at the end of the day?
(really don't want to start a storm here but could not resist posting this here when I was reading the article about PWM)

 

[hi30u]

Does this mean that if DSD = PDM which is a generalized form of PWM (which is analog data) then DSD can also be considered as an analog signal at the end of the day?

I think so. I know nothing, as many noticed here, but if I remember correctly, it is teated as such in a Full Digital Amplifier.
If the source is DSD, it is converted, it goes on a different path than would a 16bit/44.1 kHz file, for example.
BTW, the output stage of an FDA amplifier is similar to class D, analog, correct?
Not bit perfect either?
Power DAC is a better name then?

 

[hi30u]

Does this mean that if DSD = PDM which is a generalized form of PWM (which is analog data) then DSD can also be considered as an analog signal at the end of the day?
(really don't want to start a storm here but could not resist posting this here when I was reading the article about PWM)

I found the source and it says:
Why must DSD be converted to PCM in an FDA?
Because DSD is not a digital signal, it cannot be processed directly by the DSP (Digital Signal Processor of the FDA amplifier), especially for volume control.
DSD is a PDM signal derived from delta-sigma modulation.
PDM is a series of uncoded pulses like PWM, the difference being that the information is contained in the number of pulses per cycle and not in the pulse width. It is therefore a pseudo-analog binary signal that must be "decimated" to convert it into "numbers" for processing by a processor.

Another consequence is explained here:

Sabaj A30a announced

Maybe with your newly sent A30a there is also a problem with the copper not being attached 100%. Maybe this is why it is getting so hot? No there is no such thing. The housing gets hot, but the fan of the previous A30a started spinning after 5 minutes, but the fan of this A30a does not rotate...

audiosciencereview.com

Which brings me to ask another question to check if I understood correctly what that consequence implies for a DAC:
With the following: DSD file on Raspberry Pi --> USB --> DAC --> Class D monoblocs --> speakers
Then, I cannot use the volume control function of the DAC, correct ?