Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Hi resolution audio circuit

Status
Not open for further replies.

deepsetan

Advanced Member level 4
Joined
May 8, 2013
Messages
119
Helped
6
Reputation
12
Reaction score
5
Trophy points
1,298
Location
Malaysia
Activity points
2,137
Hi,

How do we implement hi res audio function in our circuit? Does special ICs needed to do that?
 

Hi,

I'd say "analog" is high resolution...

Klaus
 

What is the purpose of this audio circuit; voltage amplification or driving a speaker (or other)?
 

People usually call them H-Fi (whatever that may mean) and I guess you mean the same.

I usually understand by Hi-Fi as good freq range (response), good distortion (THD) and good noise. We often use audio amplifiers in many scientific applications.
 
Your "hi res" circuit needs audio frequency response, low distortion and low noise. Audio opamps are made with those spec's. But you forgot to tell us what is the function of the circuit. A preamp to feed a power amp? A power amp to drive a speaker? Then the power amp also must have those spec's.
 
Actually I'm trying to combine both preamp and power amp together to drive my 1 unit speaker~ 4.0 ohm. This is just rough sketches for my block diagram. I knew that the range of human ear is 20Hz to 20KHz and I'm trying to amplify the signal at higher frequency so that I can hear the signal at this frequency. Please add anything that can improve this diagram. How do we analyze our frequency response is good or not? I mean at which level I should measure it? The speaker level itself? Capture.PNG

- - - Updated - - -

Your "hi res" circuit needs audio frequency response, low distortion and low noise. Audio opamps are made with those spec's. But you forgot to tell us what is the function of the circuit. A preamp to feed a power amp? A power amp to drive a speaker? Then the power amp also must have those spec's.

Actually I want to combine both, preamp to power amp, and power amp to a speaker :)
 

Hi,

I don´t see a preamp, nor a power amp.

What exactely do you refer to?
You ask about high resolution. Your didn´t specify "high resolution" and you don´t answer our other questions...

***
The sketch is almost digital... and at first sight i´d call it an overkill to use 3 processors. Why?

Klaus
 
Hi,

I don´t see a preamp, nor a power amp.

What exactely do you refer to?
You ask about high resolution. Your didn´t specify "high resolution" and you don´t answer our other questions...

***
The sketch is almost digital... and at first sight i´d call it an overkill to use 3 processors. Why?

Klaus

Hi KlausST,

Actually my real block diagram is more complex than this. In this thread, I try to simplify it and focus more to audio portion. I'm thinking to use ICs that support wireless and can improve quality for audio system. Yes, for sure I need to use some of digital because that is the easiest way for me to control and increase the sampling frequency. I don't know the real definition of Hi res but for me, I want my device be able to play a 96KHz or above at 24-bit. Before this I made a audio circuit that using 24-bit ADC/DAC (about 90dB dynamic range) but when it plays a hi res music files, the output becomes distorted. That's why I'm asking which level should I measure to improve it?
 

Hi,

and can improve quality for audio system
This is meaningless as long as you don´t tell WHAT exactely you want to improve.
****

Yes, for sure I need to use some of digital because that is the easiest way for me to control and increase the sampling frequency.
I don´t understand.
If you have analog signals: they have no sample frequency....
If you have digital signals: then they already are converted with a dedicated sample frequency.
In both cases you can´t increase the sample frequency.

In case you mean "sample frequency conversion", then I have to disappoint you: It won´t omprove quality.
****

I want my device be able to play a 96KHz or above at 24-bit.
Again I have to disappoint you. I don´t know any device that may convert digital audio signals with 96kHz to analog with 24bit resolution on the analog side. In my eyes it´s more a marketing number.
BUT it makes sense to process digital audio data inside a digital system with 24 bit resolution to keep errors low.

****
but when it plays a hi res music files, the output becomes distorted.
My opinion: Enigneering means to find out WHY it is distorted rather than build a new device. .. without specifying your expected system performance before.
****

That's why I'm asking which level should I measure to improve it?
Specify and measure your device from input to output.
If your output is electrical then measure the electrical output. If the output is a speaker, then specify/measure the sound quality including speaker.


Klaus
 

A 24 bit of dynamic range equals 144 dB of dynamic range.

Dynamic range is defined from the noise floor to the distortion ceiling. To improve the range one must either reduce the floor or increase the ceiling.

Someone else may correct me, but I have never seen any power audio amps with more than 100 dB of dynamic range, across the full audio bandwidth.
Heck, most IC audio amps are only between 70 to 80 dB. You may require to build an all-discrete audio amp.

You may want to ask in the diyaudio forum, they have some esoteric circuits there.

- - - Updated - - -

Also...the 96 Khz is only the sampling rate. It is oversampled to reduce the requirements of the anti-alias filter and the post filter. You don't need to reproduce that high frequency. 20 Khz is the accepted top frequency limit.
 

, I want my device be able to play a 96KHz or above at 24-bit. Before this I made a audio circuit that using 24-bit ADC/DAC (about 90dB dynamic range) but when it plays a hi res music files, the output becomes distorted.

Did you investigate the reasons, in particular for the distortion?

In general, we mean equal response to be equal energy per cycle but with higher frequencies that leads to very high responses (power output) for high frequencies. So we reduce the amplitude for high frequencies so that the power output at higher frequencies will actually get reduced resulting in the reduction of amplitude at high frequencies. Similar considerations apply for low frequencies.
 

12 Bit resolution is suffisant for undistorted clear sound quality, a bit more for a high dynamic range. If you hear distortion, there's something terribly wrong with your circuit.
 

I think our brain cancels distortion and frequency response humps that we hear frequently but is sensitive to those things that are new.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top