Bypass caps to Vcc?
 

Hi gang,
Looking over an old AM/FM radio kit I built some years ago, I notice
that instead of bypass caps going to ground (negative side of 9V), they
all tie into the Vcc line, which has a single bypass cap to ground with
an associated decoupling resistor. Any thoughts on this?

Bob WB0POQ

The whole purpose of bypass caps is to return the RF currents to where they
started from. Since it's typical to connect the negative supply to the
chassis and use that as a 'common' point for V+ bypassing, many people think
that bypasses should be connected to chassis 'ground'. In point of fact,
you're really better off bypassing the V+ line (or V-, as the case might
be) directly to the V- line (or V+, as the case might be). This is not a
straightforward matter, and the choice of where's the best place to connect
the bypass is often a complex issue involving the impedance of the return
current path, as well as what other currents from lower level stages might
be sharing that same current path. It's always good to prevent output stage
currents from sharing the same return path as input stage currents. As was
popularized in the movie Ghostbusters, it's not wise to "cross the streams".

Joe
W3JDR


"Bob Liesenfeld" wrote in message
...
Hi gang,
Looking over an old AM/FM radio kit I built some years ago, I notice
that instead of bypass caps going to ground (negative side of 9V), they
all tie into the Vcc line, which has a single bypass cap to ground with
an associated decoupling resistor. Any thoughts on this?

Bob WB0POQ

On Wed, 30 Jun 2004 00:35:41 GMT, "W3JDR" wrote:

The whole purpose of bypass caps is to return the RF currents to where they
started from. Since it's typical to connect the negative supply to the
chassis and use that as a 'common' point for V+ bypassing, many people think
that bypasses should be connected to chassis 'ground'. In point of fact,
you're really better off bypassing the V+ line (or V-, as the case might
be) directly to the V- line (or V+, as the case might be). This is not a
straightforward matter, and the choice of where's the best place to connect
the bypass is often a complex issue involving the impedance of the return
current path, as well as what other currents from lower level stages might
be sharing that same current path. It's always good to prevent output stage
currents from sharing the same return path as input stage currents. As was
popularized in the movie Ghostbusters, it's not wise to "cross the streams".

You're no fan of the 'star ground' then?
--

"What is now proved was once only imagin'd." - William Blake, 1793.

For audio, maybe; for RF, no.

It's often not feasible at RF, especially at VHF & UHF. The lead lengths
required to take all bypasses to a single point can introduce significant
impedance which prevents the bypass from doing its job. In general, it's a
good idea to keep 'circulating' RF currents local to the stage where they
are generated so they don't co-mingle with other stages' currents.

Joe
W3JDR

"Paul Burridge" wrote in message
...
On Wed, 30 Jun 2004 00:35:41 GMT, "W3JDR" wrote:

The whole purpose of bypass caps is to return the RF currents to where
they
started from. Since it's typical to connect the negative supply to the
chassis and use that as a 'common' point for V+ bypassing, many people
think
that bypasses should be connected to chassis 'ground'. In point of fact,
you're really better off bypassing the V+ line (or V-, as the case might
be) directly to the V- line (or V+, as the case might be). This is not a
straightforward matter, and the choice of where's the best place to
connect
the bypass is often a complex issue involving the impedance of the return
current path, as well as what other currents from lower level stages
might
be sharing that same current path. It's always good to prevent output
stage
currents from sharing the same return path as input stage currents. As
was
popularized in the movie Ghostbusters, it's not wise to "cross the
streams".

You're no fan of the 'star ground' then?
--

"What is now proved was once only imagin'd." - William Blake, 1793.

Good summary, Joe. Return the current to the proper place is THE correct
answer. Ghost Busters analogy is interesting...

....and don't forget circuit board layout issues. With a one (or two) sided
board it may be expediant to use one or the other.

The IF (with a high gain) is usually the most susceptible to oscillation.

I would expect that the AM/FM is one of the "less" critical applications,
acknowledging, however, that it is easy to get into trouble anywhere.
--
Steve N, K,9;d, c. i My email has no u's.


"W3JDR" wrote in message
...
The whole purpose of bypass caps is to return the RF currents to where
they
started from. Since it's typical to connect the negative supply to the
chassis and use that as a 'common' point for V+ bypassing, many people
think
that bypasses should be connected to chassis 'ground'. In point of fact,
you're really better off bypassing the V+ line (or V-, as the case might
be) directly to the V- line (or V+, as the case might be). This is not a
straightforward matter, and the choice of where's the best place to
connect
the bypass is often a complex issue involving the impedance of the return
current path, as well as what other currents from lower level stages might
be sharing that same current path. It's always good to prevent output
stage
currents from sharing the same return path as input stage currents. As was
popularized in the movie Ghostbusters, it's not wise to "cross the
streams".

Joe
W3JDR


"Bob Liesenfeld" wrote in message
...
Hi gang,
Looking over an old AM/FM radio kit I built some years ago, I notice
that instead of bypass caps going to ground (negative side of 9V), they
all tie into the Vcc line, which has a single bypass cap to ground with
an associated decoupling resistor. Any thoughts on this?

Bob WB0POQ