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: Chuck Simmons
Organization: You jest.
X-Mailer: Mozilla 4.61 [en] (X11; U; Linux 2.0.33 i586)
Subject: Re: Choosing the right transformer
References: <3E2A5B59.192EAE31@webaccess.net> <3E2AF5C9.FE4B9A0C@webaccess.net>
Date: Sun, 19 Jan 2003 21:16:54 GMT
NNTP-Posting-Date: Sun, 19 Jan 2003 13:16:54 PST
Winfield Hill wrote:
> Chuck wrote...
> > The H-bridge is like a logic gate in this case. When the supply varies,
> > what is going into it is already wrong so it must be corrected. That can
> > only be done in the modulator. In other words, the feedback must be to
> > the last element which contains a linear representation of the signal.
> Surely such feedback comes at a price? Besides the introduction of
> artifacts into an otherwise conceivably-perfect system, there's a
> possible limitation of bandwidth. Some high-performance modulator
> schemes have a considerable time delay, which would create trouble
> inside a closed loop at the upper end of the audio band. Or can
> the output feedback somehow be limited to low frequencies, where
> the unregulated power rails are vulnerable to ac-line ripple, etc.?
The switching systems are not so perfect. Rise time is non-zero and turn
on and turn off delays tend to be different. Thus I don't believe that
class D open loop is essentially perfect. OTOH, designing a class D
amplifier that meets certain quality specifications can be done and may
be open loop given tight power and fast parts.
In the simple case where the average output of the PWM is to match the
input, the delay is an implementation issue but it is nearly always
small enough to close a loop. If there is a FIR in the block diagram,
there is trouble. The input of the filter must ripple through to the
output and the delay may make loop closure impossible. In such a case,
an IIR with response close to the desired FIR might be used to fix the
Another approach is to measure the supply and use a correction table.
This is a kind of feed forward or anticipation. Anticipation can work
well in some systems. Basically the dynamics have to be well known. In
this case, the dynamics can be calculated because they reside in
software. Such an anticipation approach might be cheaper than power
I have used anticipation with success and IBM used anticipation in some
of their large disk drives in the late '70s and '80s. The context was a
bit different but the idea was the same. To remove the effect of
dynamics not included in the feedback loop.
... The times have been,
That, when the brains were out,
the man would die. ... Macbeth
Chuck Simmons email@example.com
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup