How Class D Converts Audio to Square Wave part 2 with EPC GaN FETs

[@jeanious2009]

Thanks for sharing, showing and explaining how class D works.

[@KissAnalog]

Thanks for your great feedback!

[@mosfet500]

Thanks Eddie, I've been playing around with these class D's now and they are very impressive amps. Thanks for putting me on to them.

[@KissAnalog]

You bet! Thanks for the feedback! More toys;)

[@abusiveben]

Auto liked and adding to queue. I just noticed this is part 2 lol. I'm going to go watch pt 1 first

[@KissAnalog]

Awesome - thanks!

[@versace885]

Hi Eddie, thank you for the great tutoring an interesting demonstration. Happy Fourth of July.

[@KissAnalog]

versace885 Thank you! I appreciate you.

[@jeanious2009]

Thanks for sharing and explaining.

[@KissAnalog]

Thank you!

[@rrowan327]

Cool stuff, Love you're O-scope.. Good luck with the testing. Hoping ALL is well

[@KissAnalog]

Thanks - I appreciate you! Yes, all is well and I have a new amp to test - can't wait;)

[@ampdaptive9988]

Awesome. Can you do video testing this PowerIce amplifiers? Thanks extremely informative. Keep it commin

[@KissAnalog]

Thank you! I'll check into them. Thanks for the tip!

[@andyj2106]

Really interesting demonstration and explanation. How interesting would it be to see the full path of the input signal and its transformation - from input to load? I don't know how it might change but it might make a good final demo. We only get news filtered through whatever the media organisations decide to say, but it looks pretty rough over there in the States. Hope you and yours are all ok - stay safe.

[@KissAnalog]

Andrew Johnson Thanks Andrew - that does sound like a great idea. I started that in the beginning - but it was so long - I think over an hour. I worry about my long videos... Thank you - we are good. I don’t know too many people personally that have been tested positive but a couple went thru flu symptoms but most only lost the senses of taste and smell. Great to hear from you!

[@nurahmedomar]

Great video. I learned a lot.

[@KissAnalog]

Glad it was helpful! Thank you for the feedback;)

[@11orchids]

Another informative video.. Thank you. Happy 4th of July weekend.

[@KissAnalog]

Thanks I appreciate you!

[@wbetet6088]

In a previous video, you mentioned that the amp switches at 400 khz. Is there some optimization function to determine switching frequency? For example, I know audio usually goes up to about 44 khz, so looks like an approximate 10 to 1 ratio. Does higher frequency increase noise and efficiency? Is there an inverse relationship between noise, and efficiency?

[@KissAnalog]

You are asking a very good question and is really the difference in a class D amp. But I think the 44 kHz is a sampling rate which provides almost 2.5 samples per cycle at the max audio frequency of 20 kHz. They do this to meet Nyquest theory and also to be sure all the samples don’t end up at the zero volt line. The switching frequency is generally chosen to be as fast as practical for the components and controller. This drives the inductors and capacitors smaller. There is a trade off for efficiency vs size. The noise left over is hopefully at a very high frequency - far outside the hearing range. Thanks for the great question!

[@omsingharjit]

13:02 but i heard that in class D it's Triangular wave which modulate the pwm signal ?

[@thomassorensen7907]

Hi Very nice video(s). Just a question for my understanding. I was thinking how this circuit start up. From your explanation it seems that the PWM is generated from the feedback that is summed with the audio. But how does the PWM start, when there is no squarewave at the feedback? Is there a squarewave generator not shown? Hope my question make sense. Best regards Thomas

[@KissAnalog]

I'll have too show this in a better video just on this aspect as it is so important. Great question - thank you! I'll do this very soon Thomas!

[@gkdresden]

This HF triangle LF shifting method with zero detection is different from the method of comparizon between the LF and the HF triagle signal. Are there specific advantages / disadvantages with both methods of PWM generation?

[@KissAnalog]

Thanks for asking. I'll have to do a new video soon and talk about the design.

[@waynegram8907]

What do you mean by using the oscilloscope to use "cursor gating"? I'm not sure what the "gating" is doing to the measurement or what the benefits of using Gating, can you explain more about it?

[@BjornV78]

I have got a trick question (as usual) :-) imagine you have a wall with 2 audio sockets, an input and an output (they are labeled), is there a way to tell, by using some signal (music, sine wave , square wave etc...) on the input, how to determine which amplifier type (A, B, A/B, G&H or D) is behind the wall ? I know with a voltage source, a SMPS will have more ripple then a linear powersupply, and if there is almost no ripple, it must be a battery or capacitor. Is such a conclusion possible with a audio source (i know it's also a voltage source).

[@KissAnalog]

Bjorn V That is a great question! First, I think we could tell if it was a switcher like Class D. They seem to have low distortion in the Audio frequency range, but we could most likely see artifacts in the high frequencies. So, an FFT would probably be the best way to look at it. Class B would be obvious with the crossover distortion. The type of harmonics could give some topologies away, but some designers like Nelson Pass know how to Taylor the harmonic content. Thanks for the trick question - did I pass? ;)

[@BjornV78]

@@KissAnalog , you passed with flying colors ;-)

[@VandalIO]

There’s a typo in the thumbnail… it says sign to square …. Instead of sine to square

[@KissAnalog]

Thanks so much! I'll have to see if I can find that and fix it;)

[@brockallentaylor]

Referring to everything as "This Guy" or "That Guy" and "He and him" rather that what it actually is, really adds to the confusion. Thanks for the effort, but this is painful to watch.

[@KissAnalog]

Thank you for that feedback! It’s funny, I was listening to myself during edit and thought the same thing :-) I really appreciate the feedback! I really appreciate the feedback!