The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: firstname.lastname@example.org (David)
Subject: Re: Choosing the right transformer
Date: 20 Jan 2003 18:44:54 -0800
References: <3E2AEA6B.1A9BBDD1@webaccess.net> <3E2AF34D.4100ABAF@webaccess.net>
NNTP-Posting-Date: 21 Jan 2003 02:44:54 GMT
Nice explanation. I see what you mean but don't understand how it
could possibly work.
Let's say the dsp is working at 64 times oversampling (2.8Mhz). There
is then one output bit every 0.35us that goes through the H-bridge. If
it is a transition between a 0 to a 1, the bridge has to switch to say
+20V. That 20V goes through the analog filter and then to the speaker.
How come this signal could be digitized and analysed by the dsp fast
enough to adjust its internal filter in time for the next bit? It
looks to me the delay is too long, maybe I'm wrong.
I also wonder why some people desperately wants to avoid feedback in
audio systems. As said earlier in this thread by Winfield Hill :
> But the alternate choice to use output feedback with an unregulated
> supply can spoil the high-quality class-D output that's otherwise
> possible with no feedback.
What does he mean? Everyone seems to say feedback is the only way to
Thank you very much for your time.
Chuck Simmons wrote in message news:<3E2AF34D.4100ABAF@webaccess.net>...
> Greg wrote:
> > Here's the block diagram of the system
> > CD player -->digital out-->(DSP or FPGA)-->one-bit
> > output-->Hbridge--->Filter--->Speaker
> One bit output is a bit misleading because it is PWM output. Let's put
> the feedback in.
> CD player -->digital out-->DSP-->one-bit-output-->(1)
> | A/D |
> In the above, the DSP compares the data word from the CD with the
> digital word from the A/D converter. If they are different, the pulse
> width is corrected accordingly.
> This is expensive for audio because of the high resolution of the A/D
> but it can be done. In servo systems having very similar block diagrams,
> closing the loop on class D amplifiers is always done when quality
> considerations do not forbid class D (most engineers will not consider
> using class D amplifiers in the same system with low level signal
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup