On 11/10/2015 4:55 PM, John S wrote:
On 11/10/2015 11:40 AM, amdx wrote:
On 11/10/2015 9:30 AM, rickman wrote:
On 11/10/2015 9:12 AM, amdx wrote:
On 11/10/2015 1:41 AM, rickman wrote:
On 11/9/2015 12:08 PM, amdx wrote:


The silver is simply optimization. If you can make your coil
wire 5%
bigger, you should have already done that. Then if you want to
optimized
1 + 0.05, silver plate it.

Why can't you make the wire 5% bigger again? Where exactly is the
cost?

You can.
I don't think ultimate optimizing is for you.
Just make your coil with as much surface area as room will allow.
Then know it could have just a tiny bit less loss if you had the
silver.
That tiny bit of loss will not be noticed in use, except for that
little
nagging thought...

I once had a 6 or 7 turn loop made with 1/4" copper tubing*, Q was
about
800 at 1MHz. I could have made it with 1/2" tubing, probably would have
had higher Q.
I guess the limits are money and how you want to limit physical size
and
maximum inductance you can use. Did I miss any?

People seem to go nuts with ideas that you need to optimize every little
thing without any evidence to show the significance of the impact on
performance. Your example is perfect. Increasing the copper tube from
1/4 inch to even just 3/8 inch would more than make up for silver
plating and not really cost that much more. There are guys who talk
about using single piece, 3 inch copper tube bent into a loop to avoid
having solder joints when using straight pieces even though those solder
joints will be about the same resistance as a quarter inch of the tube
or a microscopic increase in the resistance. Then they conveniently
forget about the resistance of the clamp connection to the vacuum
variable capacitor swamping out the solder joint resistance even more.

It makes me want to scream, "Enough of the maddness"!


You want to see guys reaching for lowest losses, highest Q, follow a
crystal radio group.
Some there claim Q's of 2000, (I believe them) Using 2 and 3 parallel
pieces of 660/46 Litz wire. AMBCB.
The 2000Q thread; http://tinyurl.com/q2hc6ga

More info, or a way to get to page two and three;

"Meanwhile, you can do the following search on Google:
site:www.midnightscience.com/rapntap/ 2000Q
If the main blue link doesn't work, then go down to the green link one
line below, and the little drop down menu at the right, and select
"cached" which will show you Google's cached copy of the page."

I didn't know that Google trick!


Recently a ferrite material out of China has made inductors hitting
1300 Q's easy. AMBCB
http://theradioboard.com/rb/viewtopic.php?f=5&t=6259
Also rods of the same material.

Mikek

I'm curious about the 2000 Q restricting bandwidth. At 1MHz, the
bandwidth would be only 500Hz. Does that affect the quality of the
audio? That is, does it sound very bass-y?

Jeff explained the loading, but I'll add, the rest of the circuitry to
extract the audio signal and drive the headphones to create sound drops
the Q. It can be as bad as a 2000 ohm headphone or as good as a
transformer with very high input impedance, over 1.5 Megaohms according
to Ben Tongue.

http://www.bentongue.com/xtalset/5hpXform/5hpXform.html

All 29 of his crystal radio research papers are here.

http://www.bentongue.com/xtalset/xtalset.html

Tremendous resource for Crystal Radio devotees.
Mikek