Direct DSD256-raw signal input to GaN E-mode HEMT Class-D audio amplifier.

[Bunpei]

Professor Yamasaki of Waseda University and Bunpei tried applying not PWM but PDM (usual DSD) to an evaluation board provided by GaN Systems and preliminarily obtained a prospective results.

Please read this post if you are interested in DSD;

http://www.diyaudio.com/forums/class-d/323714-direct-pdm-input-gan-mode-hemt-class-amplifier.html

[jabbr]

On 6/10/2018 at 6:49 PM, Bunpei said:

Professor Yamasaki of Waseda University and Bunpei tried applying not PWM but PDM (usual DSD) to an evaluation board provided by GaN Systems and preliminarily obtained a prospective results.

Please read this post if you are interested in DSD;

http://www.diyaudio.com/forums/class-d/323714-direct-pdm-input-gan-mode-hemt-class-amplifier.html

 

To be clear this is not a Class D board, rather a driver "optimized for Class D" but rather direct digital. In Class D there is an analog to digital conversion step and analog feedback.

[Bunpei]

Hi, jabber!

 

I appreciated your comment!

 

This board is not indicated as “Class-D amplifier” by the manufacturer. They just say “Optimized for Class D Amplifiers”. In this context, you are right.

On the other hand, as for the definition of the “Class D Amplifer”, Wiki says “the amplifying devices (transistors, usually MOSFETs) operate as electronic switches, and not as linear gain devices as in other amplifiers.” I understand an ADC module is not essential. Moreover, I was surprised that there is a member who sticked to an analog input in this forum though it depend on his sense.

 

Professor Yamasaki and I are “full-digital” orienting audiophiles” in a sense.

[Maldur]

Hi, Bunbei.

"Bunbei speculates the experiment by Professor Yamasaki is the first achievement of class-D audio amplifier with direct PDM input."

 

If I remember correctly, in Japan some entusiast tries that concept in about 2011, with DSD64 input signal. Volume regulation was done via high quality regulated power supply, where increase or decrease of power voltages changes the volume level. THD level was in 0,1% class in whole audio band. Not the cleanest performance, but good enough for first try. 
This is very interesting idea, I must say. I have Tascam DA-3K too, and one more Amanero Combo384 for testing purposes, but sadly, I have not much spare time for that hobby. 

 

All the best,
Maldur

[Miska]

Sounds similar to what Estelon LYNX speakers are doing... But Estelon has their completely own amplifier.

[Bunpei]

On 6/13/2018 at 7:24 AM, Maldur said:

If I remember correctly, in Japan some entusiast tries that concept in about 2011, with DSD64 input signal. Volume regulation was done via high quality regulated power supply, where increase or decrease of power voltages changes the volume level.

 

Yes, your memory is correct!

I asked Professor Yamasaki about that. He said, “applying GaN E-mode HEMT as a high-power switching device for 1 bit delta-sigma modulated audio signal this time might be the first trial”.  He added, “Realization of such a simple & compact and energy-efficient class D audio amplifier is very valuable”.

[Bunpei]

On 6/13/2018 at 5:51 PM, Miska said:

Sounds similar to what Estelon LYNX speakers are doing... But Estelon has their completely own amplifier.

Thank you very much for the information.

They say just “FET”. I am eager to know whether their device is of Si based or GaN based.

[Maldur]

On 6/13/2018 at 11:51 AM, Miska said:

Sounds similar to what Estelon LYNX speakers are doing... But Estelon has their completely own amplifier.

I feel myself somewhat "responsible" for that, because I promoted DSD and that DSD-amplifier idea in one other forum to Estelo'ns co-worker, sadly that person no longer works for Alfred anymore, I don't know why.