Ralph Karsten's Theory on Voltage and Power Paradigms in Layman's Terms

[hopkins]

Yes- from my perspective, the engineering is better than the marketing. Can you change the volume? If yes, then amplification is occurring.

The volume is changed through a combination of power supply variation and some bit-shifting. I don’t know if this means that there is “amplification”. It could be understood as “attenuation” as well.

From my perspective, there is no amplification because the output signal is not produced using an amplified input (digital) signal.

You can read over their blog post - my summary is probably too simplified…

It all sounds fairly simple, once you spend a little time to understand it, but in practice they have gone through several iterations in the past few years, and are still refining some aspects.

 

[Atmasphere]

The volume is changed through a combination of power supply regulation and some bit-shifting. I don’t know if this means that there is “amplification”. It could be understood as “attenuation” as well.

The means is unimportant. What is important is you can change the volume; therefore it does amplify. But certainly not in the conventional sense.

 

[hopkins]

The means is unimportant. What is important is you can change the volume; therefore it does amplify. But certainly not in the conventional sense.

Read over their blog, when you have some time, and I think you will agree that the term “amplification” is really not a good way of characterizing things.

Anyway, it’s a fun and interesting product to use, if you can work with the restrictions I mentioned.

 

[Atmasphere]

Read over their blog, when you have some time, and I think you will agree that the term “amplification” is really not a good way of characterizing things.

Anyway, it’s a fun and interesting product to use, if you can work with the restrictions I mentioned.

I read the blog at the link you posted last week.

 

[Holmz]

….
It amplifies for sure, since taking a small signal and making it bigger is amplification. Put another way, you can't drive a speaker directly with a digital signal, and if I'm not mistaken its possible to change the volume. The system might not be done the way amplification is done in a conventional sense but it is amplification nevertheless.

Ralph it was their material that said it was not amplifying.
So it would be like comparing an RCA to and XLR and one has more signal voltage.

And changing the voltage of the powersupply results in the amplification.

I am not a technician, but this is my understanding:

The digital signal itself is not amplified.

It is being bumped up and converted to some higher voltage level.

The digital signal is split into parallel signals corresponding each to a given bit of the digital signal (to minimize the impact of bit errors, the most significant bit is actually split into multiple bits - hence the term “fractal” DAC they coined).

Maybe? They start off with some number of bits and have 18 bits of output.
So it starts off with 24 bits, then goes to 18… something is lost, albeit at a low level.

Each bit”s signal controls a gate allowing the power supply to go through (or not) a specific resistor value.

The resulting current that has passed through each gate and resistor is summed and directly powers the speaker drivers.

All the gates are opened/closed using a clock whose frequency is determined based on the incoming digital signal. The clock signal embedded in the Toslink signal is not used/reclocked. The clock signal is completely independant from the Toslink signal.

It is all pretty ingenious!

^Maybe^. It seems like low level signal would have a very high output impedance, and higher level signals a lower level output impedance?

 

[hopkins]

It is being bumped up and converted to some higher voltage level.

No, the incoming digital signal is not bumped up and converted to a higher voltage level. In fact, the incoming digital signal is not in the signal path to the speaker…

Changes in the power supply do not result in “amplification” because there is no amplification

You can talk about the sound level being higher, but the term “amplification” is misleading and ill-suited in this case.

 

[RCanelas]

Well, having a device that outputs a higher power signal than the input and trying to call it a 'not amplifier' is the misleading part, IMO. It is doublespeak, and it serves no purpose other than to turn off people that see through it.

Regardless of how it is done (think of it as a black box), if you get 1 at the input and 10 at the output, this black box is an amplifier of some type. It is just a class definition. You can then open the box, and further classify its type.

Virtually all designs are just a power supply output modulated by a signal in some form or another. The case with the ECdesigns device is another one of those forms. Some keep the original signal in the path, some do not. Some are differential, others use PWM, some use carrier signals, others output transformers, some a single output device, other dozens. It really doesn't matter, they are all amplifiers if 1 gets turned into something more than 1.

 

[hopkins]

Well, having a device that outputs a higher power signal than the input and trying to call it a 'not amplifier' is the misleading part, IMO. It is doublespeak, and it serves no purpose other than to turn off people that see through it.

Regardless of how it is done (think of it as a black box), if you get 1 at the input and 10 at the output, this black box is an amplifier of some type. It is just a class definition. You can then open the box, and further classify its type.

Virtually all designs are just a power supply output modulated by a signal in some form or another. The case with the ECdesigns device is another one of those forms. Some keep the original signal in the path, some do not. Some are differential, others use PWM, some use carrier signals, others output transformers, some a single output device, other dozens. It really doesn't matter, they are all amplifiers if 1 gets turned into something more than 1.

I understand that point of view, and that may be fine for people who do see hi-fi equipment as "black boxes", and don't care to know how things operate, but this thread is about the technical aspects of amplification, and in that context, lumping everything together in that way is of little interest. Personally, I do see a fundamental difference with other amplifier types. That being said, since there is no other term available in the audiophile lexicon, "amplifier" will have to do! I guess we need a new "class" type.

 

[Holmz]

No, the incoming digital signal is not bumped up and converted to a higher voltage level. In fact, the incoming digital signal is not in the signal path to the speaker…

If we unplug the cable, then the magic likely stops.
It is basically a switching power supply or a digital amp.

Changes in the power supply do not result in “amplification” because there is no amplification

There is no way to “alternate truth” the thing as being something other than a device that sits between a source and a speaker.

You can talk about the sound level being higher, but the term “amplification” is misleading and ill-suited in this case.

OK… whatever you say.
It is often said that with amplifiers one is listening to the powersupply.
This example sort of takes that saying to the extreme.


It may be possible to have the impedance be at 2ohms for all signal levels, if the resistors are in pairs summing/adding to 2ohms… not overly efficient, but it seems possible.

 

[hopkins]

It is often said that with amplifiers one is listening to the powersupply. This example sort of takes that saying to the extreme.

Yes, you are correct, and that sounds like something anyone would want to avoid given the extremes that audiophiles are willing to go to in trying to get “clean” power supply in their systems.

With this system, however, you completely avoid ground issues. There is no incoming analog signal, interconnects, etc… You cannot compare it to traditional audio equipment.

 

[hopkins]

It may be possible to have the impedance be at 2ohms for all signal levels, if the resistors are in pairs summing/adding to 2ohms… not overly efficient, but it seems possible

Yes it is possible.

 

[Atmasphere]

You cannot compare it to traditional audio equipment.

You certainly can I'm sure that others will do compare it to PP and SE amplifiers.

Ralph it was their material that said it was not amplifying.
So it would be like comparing an RCA to and XLR and one has more signal voltage.

And changing the voltage of the powersupply results in the amplification.

I currently regard the ad material (blog) as marketing.

FWIW, a balanced output, if done along the lines of AES48, the balanced standard, will have the same Voltage as the RCA connection. Its only when two SE outputs are used, one out of phase with the other (a technique that is not AES48 compliant) that you get that 'doubled Voltage' thing. So maybe not the best analogy.

Since the volume can be changed, amplification is occurring. If the volume cannot be changed I might not say that although its still a bit of a stretch to take a signal at TOSLINK level and do something such that it can drive a speaker. Any engineer would look at that and say that amplification has occurred

You can talk about the sound level being higher, but the term “amplification” is misleading and ill-suited in this case.

In my opinion saying that amplification doesn't happen is the misleading bit. The problem remains that the volume can be changed. We don't really care how, just that it does since that satisfies the definition. In digital audio, bits arranged in digital words represent an audio Voltage. The bits can be in two states, 1 and 0, where 0 is typically zero Volts. The 1 might be 3-5 Volts in most digital circuits. In theory it can be bigger, for example the 1 might be 30 Volts. This is amplification when compared to 3V. When the 30Volt bits, assembled in words, are applied to a purpose-built DAC for this application, the result is an output Voltage, in this case one that can drive a loudspeaker.


.

 

[hopkins]

You certainly can I'm sure that others will do compare it to PP and SE amplifiers.

Obviously, it was a comment on the technology.

I currently regard the ad material (blog) as marketing.

Why? It simply describes how it works. There is nothing in there that is not factual. I suggest you read it again.

Do I brush off any of your posts on this forum as marketing? You are a manufacturer and have no issue providing your opinion on a number of topics. Do all audio engineers agree with your opinions?

Since the volume can be changed, amplification is occurring. If the volume cannot be changed I might not say that although it’s still a bit of a stretch to take a signal at TOSLINK level and do something such that it can drive a speaker. Any engineer would look at that and say that amplification has occurred

Sorry to say, but you clearly have made no effort to understand how this works.

In my opinion saying that amplification doesn't happen is the misleading bit. The problem remains that the volume can be changed. We don't really care how, just that it does since that satisfies the definition. In digital audio, bits arranged in digital words represent an audio Voltage. The bits can be in two states, 1 and 0, where 0 is typically zero Volts. The 1 might be 3-5 Volts in most digital circuits. In theory it can be bigger, for example the 1 might be 30 Volts. This is amplification when compared to 3V. When the 30Volt bits, assembled in words, are applied to a purpose-built DAC for this application, the result is an output Voltage, in this case one that can drive a loudspeaker.

So how a certain volume change is achieved is completely irrelevant? If your only comment about all this is to obsessively deny the fact that there is no conventional amplification, then I am a little disappointed… let’s move on.

 

[Atmasphere]

Obviously, it was a comment on the technology.

And my comment (hence the smiley face) was a joke.

Why? It simply describes how it works. There is nothing in there that is not factual. I suggest you read it again.

I did. Several times in fact.

Do I brush off any of your posts on this forum as marketing? You are a manufacturer and have no issue providing your opinion on a number of topics. Do all audio engineers agree with your opinions?

I don't think you do. Nor am I brushing off your posts. I think though, if it can be shown that the circuit in question has variable volume capability, that any engineer would agree with my 'opinion' that there is amplification.

Sorry to say, but you clearly have made no effort to understand how this works.



So how a certain volume change is achieved is completely irrelevant? If your only comment about all this is to obsessively deny the fact that there is no conventional amplification, then I am a little disappointed… let’s move on.

Your first comment here is false.
FWIW I did not at any time deny there was no conventional amplification.
Amplification is clearly happening is all I'm pointing out. To drive a loudspeaker you need to swing some Voltage and have the current available to support that Voltage. If you connected the TOSLINK to a conventional DAC and then to a loudspeaker you'd see that it wouldn't play very loud at all. IOW how the amplification is achieved is quite different from 'conventional' amplifiers, but its amplification nevertheless.

This is directly from that blog:

The voltage can be set SLOWLY in order to change the output VOLUME of the PowerDAC-SX.

HIGHER supply voltage = MORE output power, LOWER supply voltage is LESS output power.

Put simply, you change the gain (volume) by changing the power supply Voltage as @Holmz mentioned earlier. So if you wanted to increase the gain, you'd increase the power supply Voltage. I'm sure its carefully limited to a maximum value to insure reliability. But if you went a bit higher without damaging it, the gain would be even higher.


am·pli·fi·ca·tion
/ˌampləfəˈkāSHən/


noun
noun: amplification; plural noun: amplifications

1. 1.
   the process of increasing the volume of sound
   
   
   

 

[hopkins]

And my comment (hence the smiley face) was a joke.

I did. Several times in fact.

I don't think you do. Nor am I brushing off your posts. I think though, if it can be shown that the circuit in question has variable volume capability, that any engineer would agree with my 'opinion' that there is amplification.


Your first comment here is false.
FWIW I did not at any time deny there was no conventional amplification.
Amplification is clearly happening is all I'm pointing out. To drive a loudspeaker you need to swing some Voltage and have the current available to support that Voltage. If you connected the TOSLINK to a conventional DAC and then to a loudspeaker you'd see that it wouldn't play very loud at all. IOW how the amplification is achieved is quite different from 'conventional' amplifiers, but its amplification nevertheless.

I think of it as an “attenuator” not an “amplifier”, but let’s agree to disagree on the terminology.

Are there benefits to this design? There are certainly restrictions which I made sure to point out. I have my own opinion on the sound quality, which I don’t need to share here. The overall principle is one thing, the current implementation is another - it may be improved upon in the future. I don’t know whether other manufacturers will be curious to develop something based on the principles outlined by ECDesigns. I certainly find them interesting.

Put simply, you change the gain (volume) by changing the power supply Voltage as @Holmz mentioned earlier. So if you wanted to increase the gain, you'd increase the power supply Voltage. I'm sure its carefully limited to a maximum value to insure reliability. But if you went a bit higher without damaging it, the gain would be even higher.

Yes, and the maximum output is limited by the resistance.specifications.

 

[Atmasphere]

I think of it as an “attenuator” not an “amplifier”, but let’s agree to disagree on the terminology.

That is why I mentioned the idea of increasing the power supply Voltage to increase gain at the end of my last post. But- OK.

Are there benefits to this design? There are certainly restrictions which I made sure to point out. I have my own opinion on the sound quality, which I don’t need to share here. The overall principle is one thing, the current implementation is another - it may be improved upon in the future. I don’t know whether other manufacturers will be curious to develop something based on the principles outlined by ECDesigns. I certainly find them interesting.

The problem its up against is marketplace share much as we've faced over the last 49 years, same as anyone making SETs. Since the output impedance is high with no feedback the speaker choice is important since it has to be one of the kind that is meant for that kind of 'amp'. IOW its power paradigm as its output is likely to be constant power with respect to load impedance rather than constant Voltage.

That's a limited market although in high end audio there are plenty of speakers that will work fine with that (its why we're still around after nearly 50 years).

 

[hopkins]

It is definitely a niche product!

 

[Holmz]

In other words you make zero effort to understand what is explained but feel the need to make a public comment that it is “BS”.

I think we have gone full circle here.
Three of us cannot even form an agreement on whether the thing amplifies or not.

I read the blurb a few times, and while being a somewhat interesting design, it is pretty much passive parts and maybe a better comparison to passive preamp… with the power supply being the main contrôlér of the output.

It’s sort of similar with a handful of resistors.