My thoughts on the Vibelink Amp

For me it's 48V.
Ah, so you have measured that in the vibelink ? Thruth is you do loose some power when using PFFB.
Lowest distortion without PFFB with a single tpa3255 is at 34V according to some measurements done at diyforum.
 
Last edited:
No extensive tests this time, just some thoughts and findings instead. Incrementally, of course.

First case, how to make sure that the analog inputs are not digitized? Any ideas?


This time there is almost no delay. And no sign of the conversion affecting the wave, except the overshoot of the amp.
lol...

first stage of an class d amp: digitize signal....
 
lol...

first stage of an class d amp: digitize signal....
Texas Instruments who produce the TPA3255 that the Vibelink uses describe it as a “315-W stereo, 600-W mono, 18- to 53.5-V supply, analog input Class-D audio amplifier”. Do you want to tell them they’re wrong? ;)



 
Texas Instruments who produce the TPA3255 that the Vibelink uses describe it as a “315-W stereo, 600-W mono, 18- to 53.5-V supply, analog input Class-D audio amplifier”. Do you want to tell them they’re wrong? ;)



The link in post #103 also shows an analog input 🙂
 
Texas Instruments who produce the TPA3255 that the Vibelink uses describe it as a “315-W stereo, 600-W mono, 18- to 53.5-V supply, analog input Class-D audio amplifier”. Do you want to tell them they’re wrong? ;)



First step in the class D amplifier is a PWD encoding of the analog input. Not really a digitized (binary) signal but a kind of. It's this pulse signal that drives output stage.
 
i just don't get you first use-case for testing; "how to make sure that the analog inputs are not digitized?"
why bother really?
- the ESS is a D/A converter, not an A/D converter. WIIM does not mention any A/D converter being implemented (besides the class-D-amp itself)
- WIIM would be silly to include more components the necessary (as it increases cost)
- as I stated; de D-class amplifier is a form of digitized-signal-amplifier-by-design. so the answer to your question is; i can assure that they are... otherwise the amp would not work

you might question; do we hear induced noise by the D/A pre-stage when using the analog inputs of the amp?
that would be a probable 'no'; with an S/N of the chip of about 130DB you wont hear it. (welll, actuallyonly when WIIM messes up the supporting components or bad design (grounding and such) )

if we want cutting edge questions....why does WIIM convert the digital signal back tot analog using the ESS at all?
why not convert it to the needed PWD stream for amplification directly? ;)
 
Last edited:
i just don't get you first use-case for testing; "how to make sure that the analog inputs are not digitized?"
why bother really?
- the ESS is a D/A converter, not an A/D converter. WIIM does not mention any A/D converter being implemented
- WIIM would be silly to include more components the necessary (as it increases cost)
- as I stated; de D-class amplifier is a form of digitized-signal-amplifier-by-design. so the answer to your question is; i can assure that they are... otherwise the amp would not work

you might question; do we hear induced noise by the D/A pre-stage when using the analog inputs of the amp?
that would be a probable 'no'; with an S/N of the chip of about 130DB you wont hear it. (welll, actuallyonly when WIIM messes up the supporting components or bad design (grounding and such) )

if we want cutting edge questions....why does WIIM convert the digital signal back tot analog using the ESS at all?
why not convert it to the needed PWD stream for amplification directly? ;)
I think you are getting it too serious. It was just a funny test to verify if an ADC was used or not. BTW, WiiM didn't mention that 2 DACs were used in the WiiM Amp.
If I remember correctly 3255 does not accept the digital signal directly, so some sort of the external conversion is needed anyway. If the "true digital" amp is planned, infineon chip is rather used. Has anyone implemented a solution with 3255 that bypasses the conversion to analog?

Edit: my mistake, AX5689 is not from Infineon.
 
Last edited:
why not convert it to the needed PWD stream for amplification directly? ;)
Probably because direct conversion of a 24-bit PCM sample to PWM would require a PWM clock frequency many times higher than the audio sampling rate (at 192kHz that could be 50MHz) and there may then be problems with noise shaping. Using the well known (and cheap) components for D/A conversion and Class D amplifier instead, makes the design robust.
 
Last edited:
I think you are getting it too serious. It was just a funny test to verify if an ADC was used or not. BTW, WiiM didn't mention that 2 DACs were used in the WiiM Amp.
If I remember correctly 3255 does not accept the digital signal directly, so some sort of the external conversion is needed anyway. If the "true digital" amp is planned, infineon chip is rather used. Has anyone implemented a solution with 3255 that bypasses the conversion to analog?

Edit: my mistake, AX5689 is not from Infineon.
ok maybe I took that too serious indeed :)

nah...would need a different design indeed, way to costly. can't be done at this price point (yet)
at current price, design seems solid
 
ok maybe I took that too serious indeed :)

nah...would need a different design indeed, way to costly. can't be done at this price point (yet)
at current price, design seems solid
You will also have the analog feedback from the amplifier output that needs to be compared with an analog input (HHS Loop). So by design the DAC is needed.
 
You will also have the analog feedback from the amplifier output that needs to be compared with an analog input (HHS Loop). So by design the DAC is needed.
You mean an ADC in this case?
 
No. The analog output of the amplifier is compared to its analog input, that is the output of the DAC.
What type of amplifier are we talking here? I thought it was about digitally controlled Class D amps. :unsure:

If it is the classical analogue input Class D amp, then output and input signal are both analogue already. What would we need a DAC for? We just feed back a (filtered).portion of the output signal to the input comparator.
 
What type of amplifier are we talking here? I thought it was about digitally controlled Class D amps. :unsure:

If it is the classical analogue input Class D amp, then output and input signal are both analogue already. What would we need a DAC for? We just feed back a (filtered).portion of the output signal to the input comparator.
The DAC is for the analog input to the amp. The initial question was why it was not implemented to go directly from PCM data to PWM, without the use of a DAC.
 
The DAC is for the analog input to the amp. The initial question was why it was not implemented to go directly from PCM data to PWM, without the use of a DAC.
Well, yes. But the need for a DAC is not directly related to feedback. It stems from the fact that the amplifier module in use here doesn't have a digital input (which is not a bad thing).

Feedback really is an issue of concern with digitally controlled Class D amps. Direct PCM to PWM conversion is mostly suitable to low performance, high efficiency applications. The famous TacT Millenium was able to reach good (but not exceptional) technical characteristics with no feedback at all. Other, newer designs are designed with analogue to digital feedback implementations, which is pretty hard to do.
 
The DAC is for the analog input to the amp. The initial question was why it was not implemented to go directly from PCM data to PWM, without the use of a DAC.
From what I have seen, "true digital," power amps (frequently called, "power DACs") cannot be designed with feedback, so that rules out PFFB and also rules out load-independent-frequency-response.

-Ed
 
From what I have seen, "true digital," power amps (frequently called, "power DACs") cannot be designed with feedback, so that rules out PFFB and also rules out load-independent-frequency-response.

-Ed
Abd as the WiiM AMP Ultra has PFFB it must have a DAC.

So thats the answer to the question.
 
Abd as the WiiM AMP Ultra has PFFB it must have a DAC.

So thats the answer to the question.
Well it's also known that WiiM Amp, Amp Pro, and Amp Ultra all utilize Texas Instruments' TPA3225 amplification chip, which is designed specifically with an analog input circuit, so those specifically need a DAC if the source does not have analog outputs.

-Ed
 
Back
Top