On Mon, 28 Sep 2009 19:33:12 -0400, Bill M
wrote:

Peter Dettmann wrote:
On Sun, 27 Sep 2009 22:28:33 -0700 (PDT), K7ITM wrote:

On Sep 26, 8:25 pm, brian whatcott wrote:
rtfm wrote:
On 2009-09-24, Tomylavitesse wrote:
HI,
OK?

snip

As well, it would be interesting to see the effect of stray
capacitance, as the reason for shorting unused turns is to reduce the
problem of stray capacitance and the inductance having unwanted
parasitic resonances in the system, That is why shorting unused
sections is normally advocated.

Peter Dettmann


Well, its the better of the options but the Q of the 'desired' coil
still takes a hit because of that nearby lump of copper.

It works out ok in practice, though.

-Bill

Exactly Bill, and the whole thing really is a matter of just what is
the needed, or acceptable Q. If the Q is satisfactory with the
shorted turns, then that is the way to make your design (keeping in
mind the disadvantages of open circuit unused turns).

However should you need the highest obtainable Q, then a single
isolated coil is called for which has the needed turns.

Peter

On Sep 30, 3:46*am, Peter Dettmann wrote:
On Mon, 28 Sep 2009 19:33:12 -0400, Bill M
wrote:





Peter Dettmann wrote:
On Sun, 27 Sep 2009 22:28:33 -0700 (PDT), K7ITM wrote:

On Sep 26, 8:25 pm, brian whatcott wrote:
rtfm wrote:
On 2009-09-24, Tomylavitesse wrote:
HI,
OK?

snip

As well, it would be interesting to see the effect of stray
capacitance, as the reason for shorting unused turns is to reduce the
problem of stray capacitance and the inductance having unwanted
parasitic resonances in the system, * *That is why shorting unused
sections is normally advocated.

Peter Dettmann

Well, its the better of the options but the Q of the 'desired' coil
still takes a hit because of that nearby lump of copper.

It works out ok in practice, though.

-Bill

Exactly Bill, and the whole thing really is a matter of just what is
the needed, or acceptable Q. * If the Q is satisfactory with the
shorted turns, then that is the way to make your design (keeping in
mind the disadvantages of open circuit unused turns).

However should you need the highest obtainable Q, then a single
isolated coil is called for which has the needed turns.

Peter- Hide quoted text -

- Show quoted text -

All this very interesting; because many, many, (like 50+) years ago I
built a signal generator, that didn't work! Nice job in an ex-
Admiralty copper lined wooden box!

Came across a complete set of coils with a multi position switch which
covered a wide range of frequencies. Tested it out breadboard fashion
and it worked fine. Then built it into a double layer chassis
constructed from two thick and therefore potentially rigid and
frequency stable pieces of Aluminum!

But what I had done was dismount a set of coils from a plastic frame
and mounted them in holes drilled into the two opposite panels of
aluminum. Nice job mechanically but what must have hap pend is the Al.
panels provided shorted turns in the same plane as the winding of the
coils and reduced their Q to where the (tube, inductive feedback)
circuit wouldn't work.

Never got time to work on it because I then emigrated to North America
and the unfinished item is long gone. Had a nice Muirhead slow motion
drive on it too!

Live and learn!

On Oct 8, 8:17*am, stan wrote:
On Sep 30, 3:46*am, Peter Dettmann wrote:



On Mon, 28 Sep 2009 19:33:12 -0400, Bill M
wrote:

Peter Dettmann wrote:
On Sun, 27 Sep 2009 22:28:33 -0700 (PDT), K7ITM wrote:

On Sep 26, 8:25 pm, brian whatcott wrote:
rtfm wrote:
On 2009-09-24, Tomylavitesse wrote:
HI,
OK?

snip

As well, it would be interesting to see the effect of stray
capacitance, as the reason for shorting unused turns is to reduce the
problem of stray capacitance and the inductance having unwanted
parasitic resonances in the system, * *That is why shorting unused
sections is normally advocated.

Peter Dettmann

Well, its the better of the options but the Q of the 'desired' coil
still takes a hit because of that nearby lump of copper.

It works out ok in practice, though.

-Bill

Exactly Bill, and the whole thing really is a matter of just what is
the needed, or acceptable Q. * If the Q is satisfactory with the
shorted turns, then that is the way to make your design (keeping in
mind the disadvantages of open circuit unused turns).

However should you need the highest obtainable Q, then a single
isolated coil is called for which has the needed turns.

Peter- Hide quoted text -

- Show quoted text -

All this very interesting; because many, many, (like 50+) years ago I
built a signal generator, that didn't work! Nice job in an ex-
Admiralty copper lined wooden box!

Came across a complete set of coils with a multi position switch which
covered a wide range of frequencies. Tested it out breadboard fashion
and it worked fine. Then built it into a double layer chassis
constructed from two thick and therefore potentially rigid and
frequency stable pieces of Aluminum!

But what I had done was dismount a set of coils from a plastic frame
and mounted them in holes drilled into the two opposite panels of
aluminum. Nice job mechanically but what must have hap pend is the Al.
panels provided shorted turns in the same plane as the winding of the
coils and reduced their Q to where the (tube, inductive feedback)
circuit wouldn't work.

Never got time to work on it because I then emigrated to North America
and the unfinished item is long gone. Had a nice Muirhead slow motion
drive on it too!

Live and learn!

In at least some editions of "Reference Data for Radio Engineers,"
there's a graph of how much the inductance of a solenoid coil is
lowered by being placed inside a shield. If the shield isn't too
close to the coil, the percentage decrease in inductance is rather low
(the coil isn't affected very much). The graph doesn't tell you how
much the Q is lowered, but I did some comparisons of Q estimates from
a couple different solenoid coil programs I have and Q estimates for
helical resonators, and came to the conclusion that you can account
for most of the lowering of Q by assuming the RF resistance remains
constant and the inductance is lowered, per the R.D.R.E. graph
(attributed to RCA). That assumes a highly-conducting shield, I'm
sure. The observation helped dispel the "magic" aura of helical
resonators: their Q is actually lower than the Qu of the coil they
contain, if that coil is in free air. (A big advantage, of course, is
that they are fully shielded.)

Cheers,
Tom

K7ITM wrote:

In at least some editions of "Reference Data for Radio Engineers,"
there's a graph of how much the inductance of a solenoid coil is
lowered by being placed inside a shield. If the shield isn't too
close to the coil, the percentage decrease in inductance is rather low
(the coil isn't affected very much). The graph doesn't tell you how
much the Q is lowered, but I did some comparisons of Q estimates from
a couple different solenoid coil programs I have and Q estimates for
helical resonators, and came to the conclusion that you can account
for most of the lowering of Q by assuming the RF resistance remains
constant and the inductance is lowered, per the R.D.R.E. graph
(attributed to RCA). That assumes a highly-conducting shield, I'm
sure. The observation helped dispel the "magic" aura of helical
resonators: their Q is actually lower than the Qu of the coil they
contain, if that coil is in free air. (A big advantage, of course, is
that they are fully shielded.)

This gets to be a jello-y area if you follow the clues far enough. The
coil's Q, or better said, the circuit's Q has a lot to do with how much
spacing is required. Old timey Rule Of Thumb suggests one diameter
spacing away from enclosures but thats typically relative to lower Q
coils which are common rote. I make that comparison against high-Q
coils in dx crystal sets where 12-18 inch separation isn't uncommon.

The textbook Q formula doesn't address this...nor is it a necessarily a
biggie because most of the coils that we encounter in the situation
presented in the original post aren't of high enough Q so as to present
any problem.

But, it needs to be reminded that coil Q is definitely subject to its
surroundings. Re-tweaking the inductance value is easy enough but you
cannot recoup the loss of Q. It always depends on the specific application.

-Bill