Hi Q Litz coil for MW BCB?
JasH
As a member of a club devoted to high quality RF powered Crystal Radio
Sets, I am intrested in low loss induction coils.
I noticed that Winfield Hill and others had several threads about the
subject in Fall 1997, but some of the material seems to be lost in the
stacks due to crashes.
I seem to get the idea that the best coil was about 400/44 litz for a
coil to go up to 1MH, wound on a 3-4" low loss form, 1 layer solenoid,
spaced 1 wire diameter. Another member has seen references that say
600/44 is better. Please voice your opinions on litz or other coils.
Mulitiple coils to cover MW are OK.
And please recommend methods for measuring high Q coils.
Please feel free to visit us at this link and see what we get up to.
Next annual DX contest is in April.
http://clubs.yahoo.com/clubs/thecrystalsetradioclub
Thank you,
Steve Holden aka Variocoupler
AMDX
a 240 uh coil. I'll see if I can find the companies I got info from after
work. Got to go now!
--
Mike Knowlton
Panama City Fl.
AMDX
http://www.wiretron.com/litz.html
www.neewcweb.com/litz/litz.htm
russell shaw
is nothing, so don't bother.
The best crystal set i made used thick automotive wire wound into a 'bundle'
coil 1m in diameter (10-20 turns), connected across a dual-gang 415pF metal
tuning cap., then the usual OA91 diode, and crystal ear-piece (can't remember
if shunt resistor was needed or not).
This set tuned 5 stations *sharply*, with good high audio, and without an
external antenna or earth! Whats more, you could 'wear' the coil as a portable
set!
--
*******************************************
* Russell Shaw, B.Eng, M.Eng(Research) *
* Electronics Consultant *
* email: rus...@webaxs.net *
* Australia *
*******************************************
K7ITM
>I am intrested in low loss induction coils.
What sort of Q are you hoping to get? I doubt that you'll see a significant
difference between the two wires you mention, but I could be wrong. Have you
visited any web sites looking for info about Litz wire? There are a couple
I've found, at least one of which has some applications info, though not very
specific to MW coils.
Where I worked about 15 years ago, we made a 1MHz RF tank circuit with 1 inch
diameter forms and managed a Q close to 300, as I recall. The guy who did it
spent a lot of time tweaking and fine-tuning things. It exceeds what I'd
expect to be able to do from my rules of thumb about RF coil Q.
600/44: that sounds like about 16.3AWG equivalent. That's pretty big wire.
Sounds like you need about 120 turns at 4" long and 4" diameter for 1mH. Guess
that's not too unreasonable. At that frequency, I'd expect to get higher Q
with a physically larger coil. The limit is radiation effects, but you should
be quite a ways away from that limit. Below that limit, I'd expect Q to scale
nearly linearly with coil diameter.
Cheers,
Tom
kize...@ced.utah.edu
broadcast band radio for kids to build. I have done considerable
experimentation with loopstick antennas, and I can tell you litz wire
does make a difference, a huge difference.
I tried in vain for days to duplicate a litz wire loopstick with magnet
wire. I used the same core for all tests. The original coil was
purchased from Electronic Goldmine. I used the same core for all my
tests. All coils were measured on an HP 4275A Multi-Frequency LCR Meter
at frequencies between 400 KHz and 2MHz, and then evaluated in the radio
circuit.
Even though the stock coil was wound with very fine (and fragile) litz
wire I could not duplicate its performance with various sizes of magnet
wire. Getting the inductance value was easy. Getting an equivelant Q was
not, but Q is only part of the problem. Coils wound with single strand
exhibited much higher inter-winding capacitance. As a result coils wound
with the magnet wire would not tune as wide a bandwidth, and would not
tune the upper end of the broadcast band at all.
I am now looking around for some surplus litz wire in order to finish my
coil design. Even litz wire with only a few strands will make a big
difference in coil performance.
Hope this helps.
Bruce Kizerian
Centerville, Utah
Sent via Deja.com http://www.deja.com/
Before you buy.
dang...@earthlink.net
>Here's a couple of sites I found with litz info.
>
>http://www.wiretron.com/litz.html
>
>www.neewcweb.com/litz/litz.htm
Skin effect at 1kHz?
Hmm...
Regards,
Dangerdave
dang...@earthlink.net
>I have seen figures of like 5% Q improvement from litz wire. In dB, this
>is nothing, so don't bother.
I've looked at some Litz wire loss charts as well. In a narrow
frequency window, around IF, the Litz construction offers a small loss
advantage. At higher or lower frequencies, no advantage.
Regards,
Dangerdave
Mike Monett
Yup. Anything above DC. Isn't that why power lines use four separate
conductors instead of one big one?
Best Regards,
Michael R. Monett
mailto:mkmo...@hotmail.com
James Meyer
>
>Skin effect at 1kHz?
>
>Hmm...
>
>Regards,
>Dangerdave
It's a real physical phenomenon. The frequency only determines the
depth at which the current density falls off.
The skin effect depth is about a half an inch at 60 Hz power line
frequencies. That's why you don't see too many long distance power distribution
lines with a diameter of much over one inch. The copper inside a really big
cable would be wasted.
Jim
dang...@earthlink.net
>Hope this helps.
>
>Bruce Kizerian
>Centerville, Utah
Interstrand capacitance makes more sense than skin effect arguments,
using Terman's method to calculate skin effect vs frequency vs wire
diameter.
The difference in insulation thickness, Litz vs magnet wire, would
support parasitic capacitive arguments (Litz insulation can be
comparatively thick). Wire-wrap wire has comparatively thick
insulation too. Have you tried that? Ditto for the cheap magnet wire
that uses thick plastic insulation. Thinly varnished or enamel coated
magnet wire would seem to be the worst for parasitic capacitance,
based on C=kA/d. I believe the principle with that type wire is
high packing density, not low interstrand capacitance.
Regards,
Dangerdave
dang...@earthlink.net
>dang...@earthlink.net wrote:
>>
>> >On Wed, 19 Jan 2000 20:49:19 -0600, "AMDX" <AM...@worldnet.att.net> wrote:
>>
>> >Here's a couple of sites I found with litz info.
>> >
>> >http://www.wiretron.com/litz.html
>> >
>> >www.neewcweb.com/litz/litz.htm
>>
>> Skin effect at 1kHz?
>>
>> Hmm...
>>
>> Regards,
>> Dangerdave
>
>Yup. Anything above DC. Isn't that why power lines use four separate
>conductors instead of one big one?
>
>Best Regards,
>
>Michael R. Monett
>mailto:mkmo...@hotmail.com
Right. Steel core, multistrand, because there is only skin conduction
at 50/60 Hz <g>.
Now, for some swamp land in Florida....
Regards,
Dangerdave
bkiz...@my-deja.com
dang...@earthlink.net wrote:
> >On Thu, 20 Jan 2000 14:29:00 GMT, kize...@ced.utah.edu wrote:
>
> >Hope this helps.
> >
> >Bruce Kizerian
> >Centerville, Utah
>
> Interstrand capacitance makes more sense than skin effect arguments,
> using Terman's method to calculate skin effect vs frequency vs wire
> diameter.
>
> The difference in insulation thickness, Litz vs magnet wire, would
> support parasitic capacitive arguments (Litz insulation can be
> comparatively thick). Wire-wrap wire has comparatively thick
> insulation too. Have you tried that?
I did, in fact, try winding a coil with wire wrap wire for that reason.
It did not do very well, making me wonder if there is not more than one
cause for the poor performance. However, since my memory is sometimes
not great, I'll bring the coils in to work tomorrow and test some of
them again. I'll post a message after that. One of the problems with
this is that the wire insulation has a higher dielectric constant than
air, so inter-winding capacitance might not go down that much. It was my
intention to wind a coil with the magnet wire turns spaced apart (with
an air gap in between), but my ferrite rods were not long enough to get
an inductance comparable to the other coils.
Bruce
Mike Monett
> Right. Steel core, multistrand, because there is only skin conduction
> at 50/60 Hz <g>.
>
> Now, for some swamp land in Florida....
>
> Regards,
> Dangerdave
Of course, it's stranded with a steel core. But why go to all the trouble
to make four of them instead of one big one?
A simple calculation for skin depth at 60 hz:
; Terman p. 21 Skin Depth where I(s) = 38.6% of I(surf)
; Resistance assumes uniform current distribution
; See Also ITT p. 6-9
F = 60 ; frequency in Hz
s = 6.62 / sqrt(F) ; skin depth in cm
n = s / 2.54 ; skin depth in inches
The solution is
F = +60.000000000000000
s = +0.854638325063103
n = +0.336471781520907
or about one/third of an inch.
Alan Douglas
>Isn't that why power lines use four separate
>conductors instead of one big one?
No, I believe that is to give a larger effective diameter, so the
voltage can be raised without causing corona.
73, Alan
JasH
Thank you for commenting.
One of the members (Skywave_99) at The Crystal Set Radio Club found some
rolls of 3/7/44 Litz as surplus. I see he is selling whats left (after
we got ours) on Ebay. You could e-mail him from the Club links page and
ask if any is left. In the back posts are 2 or 3 links to surplus
dealers who have some for small orders.
Steve Holden
Mike Monett
> Right. Steel core, multistrand, because there is only skin conduction
> at 50/60 Hz <g>.
>
> Now, for some swamp land in Florida....
>
> Regards,
> Dangerdave
Don't feel bad. I thought the same you did until one day I was
talking to an old clerk in a hardware store. I forget how he knew,
but I checked when I got home, and sure enough, he was right.
Now, how would a clerk in a hardware store selling plumbing know
about skin effect at 60 Hz?
The next time he saw me, I was in the parking lot looking under my
car for a coolant leak.
He asked what the problem was, and I said I had an intermittent loss
of coolant.
"Loose hose clamp?", he said.
Nah, I replied. I had just changed all the hoses and installed
brand-new clamps. They were tight. It had to be something else.
He just smiled.
Well, darn if there wasn't a clamp I had missed. The rad outlet hose
was in a difficult spot to reach, and pointed down so you couldn't
notice any leak.
Smart clerk!
dang...@earthlink.net
> Don't feel bad.
I appreciate your concern.
> Now, how would a clerk in a hardware store selling plumbing know
> about skin effect at 60 Hz?
I don't have the slightest idea. I've never asked hardware clerks for
EE design theory.
Regards,
Dangerdave
Roy McCammon
>
> For the past year or so I have been developing a simple regenerative AM
> broadcast band radio for kids to build.
>
> wire I could not duplicate its performance with various sizes of magnet
> wire. Getting the inductance value was easy. Getting an equivelant Q was
> not, but Q is only part of the problem.
At AM radio frequencies, losses in the dielectric (insulation) can be
greater than losses due to skin or proximity effect.
Opinions expressed herein are my own and may not represent those of my employer.
kize...@ced.utah.edu
> experimentation with loopstick antennas, and I can tell you litz wire
> does make a difference, a huge difference.
>
> I tried in vain for days to duplicate a litz wire loopstick with
magnet
> wire. I used the same core for all tests. The original coil was
> purchased from Electronic Goldmine. I used the same core for all my
> tests. All coils were measured on an HP 4275A Multi-Frequency LCR
Meter
> at frequencies between 400 KHz and 2MHz, and then evaluated in the
radio
> circuit.
>
> wire I could not duplicate its performance with various sizes of
magnet
> wire. Getting the inductance value was easy. Getting an equivelant Q
was
> exhibited much higher inter-winding capacitance. As a result coils
wound
> with the magnet wire would not tune as wide a bandwidth, and would not
> tune the upper end of the broadcast band at all.
>
> I am now looking around for some surplus litz wire in order to finish
my
> coil design. Even litz wire with only a few strands will make a big
> difference in coil performance.
>
>
> Bruce Kizerian
> Centerville, Utah
>
As a follow-up to the above messages I retested the litz wire loopstick
and the wire wrap wound loopstick.
Electronic Goldmine litz wound loopstick:
400KHz: L=222microH Q=170
1000KHz: L=223microH Q=310
2000KHz: L=227microH Q=400
Coil wound on same ferrite rod with #30 kynar insulated wire wrap wire:
400KHz: L=210microH Q=160
1000KHz: L=210microH Q=122
2000KHz: L=210microH Q=140
The wire wrap coil could not be made to work well in the regenerative
radio circuit. Tuning range was greatly restricted and sensitivity was
significantly reduced.
As I said before the litz wire used in the Electronic Goldmine coil is
very fine (and fragile!), much finer, in fact, than the litz wire used
in the half dozen or so old AM transistor radios I have dismantled.
Using the formula provided by Mike Monett for skin depth:
F=1,000,000 Hz
s(in cm)=6.62/sqrt(F)
s (inches)=s(in cm)/2.54
the solution is:
F=1,000,000 Hz
s(in cm)=6.62/1000=.00662cm
s(in inches)=.00662/2.54=.0026 in.
#30 wire wrap wire is Approximately .010 inch in dia., so skin effect
may be significant when using single strand wire larger than .005 inches
in diameter, and litz wire should have an advantage.
Your comments and help will be appreciated.
kize...@ced.utah.edu
> At AM radio frequencies, losses in the dielectric (insulation) can be
> greater than losses due to skin or proximity effect.
>
Thanks for this information. Litz wire used in AM broadcast band
loopsticks probably has a lower dielectric loss than wire wrap wire.
dang...@earthlink.net
was to compare ac wire resistance to dc resistance, which he termed
the ac/dc wire resistance ratio. The ratio is calculable at different
wire diameters and frequencies, for different materials.
As the wire diameter becomes larger, the circumference becomes
greater, and so the amount of conducting material. Works for any
frequency [1]. What Terman's resistance ratio indicates is the
amount (if any) of unutilized conductive material.
At 60 Hz, using solid cylindrical pure Copper conductor:
For 1% non-conducting cross-section, max wire diameter = 0.571 inches
For 10% non-conducting cross-section, max wire diameter = 1.00 inches
For 0.57" diameter, current capacity = 457A
For 1.00" diameter, current capacity = 1157A
For 2.00" diameter, current capacity = 1862A
and so on...
(700CM / A, conductive cross-section, 60Hz)
Larger cable ==> greater current capacity. Would you agree?
Strangely enough, I believe there may be more to power distribution
engineering than skin effect. Weight, cost, tensile strength,
wind/ice/tree limb load capacity, wire availability vs diameter, line
weight handling considerations, local power demand, voltage vs current
transmission mode?
Regards,
Dangerdave
[1] at RF, the Terman model is not good enough for consumer work, but
may be adequate for government work.
dang...@earthlink.net
>Electronic Goldmine litz wound loopstick:
>
>400KHz: L=222microH Q=170
>1000KHz: L=223microH Q=310
>2000KHz: L=227microH Q=400
>
>Coil wound on same ferrite rod with #30 kynar insulated wire wrap wire:
>
>400KHz: L=210microH Q=160
>1000KHz: L=210microH Q=122
>2000KHz: L=210microH Q=140
>Your comments and help will be appreciated.
>
>Bruce
The problem might benefit from calculating resistance loss effects.
For 30AWG, 400kHz, Ratio AC/DC wire resistance = 1.04
For 30AWG, 1MHz, Ratio AC/DC wire resistance = 1.22
and Q = 2pi * f * L / R
where; R = wire resistance at frequency
Skin effect causes R to increase with frequency, so Q decreases with
frequency if skin effect is the controlling loss factor.
Since Q in both your data sets either 1) increases with frequency, or
2) both increases and decreases with frequency, it is difficult to see
how skin effect could be dominating Q.
Anyone got a better suggestion?
Regards,
Dangerdave
Mike Monett
[...]
> Larger cable ==> greater current capacity. Would you agree?
Of course, but it also means more inactive material and increased weight.
It costs money for stronger towers to support it.
> Strangely enough, I believe there may be more to power distribution
> engineering than skin effect. Weight, cost, tensile strength,
> wind/ice/tree limb load capacity, wire availability vs diameter, line
> weight handling considerations, local power demand, voltage vs current
> transmission mode?
I'm sure power companies have these tradeoffs figured to the penny,
and can dictate to the cable manufacturers exactly what size they want,
how many strands, and when they want it.
> Regards,
> Dangerdave
>
> [1] at RF, the Terman model is not good enough for consumer work, but
> may be adequate for government work.
Terman assumes uniform current distribution, which disqualifies it for
stranded wire or the displacement effect of parallel wires close
together.
There are some complaints that Terman's data is pretty old and becoming
less useful for modern analysis.
I just use it for quick ballpark info - for example, the skin depth is
0.001 inch at about 7 MHz, which means you have to start thinking about
skin effect at a surprisingly low frequency for pcb traces on one oz
copper. This means short leads, which we need anyway to minimize prop
delay in fast logic.
Dr. Howard Johnson has a nice analysis of skin effect on his web site,
and there is a wealth of information dealing with skin effect, from foil
power transformers on up through light. Everyone seems to come up with a
slightly different formula.
But nobody seems to have found a clever way to eliminate it.
russell shaw
Litz wire is lots of *insulated* fine strands that have an effective skin
surface area greater than that of the single strand (for same overall diameter),
so total resistance is lower.
Large power lines use a steel multi-strand core for tensile strength, and a multi-strand
over-wrapped aluminium sheath for high conductivity. Due to skin effect, there
is no need for the steel core to be highly conductive.
dang...@earthlink.net wrote:
>
> Oddly, I get different results using Terman. Terman's basic approach
> was to compare ac wire resistance to dc resistance, which he termed
> the ac/dc wire resistance ratio. The ratio is calculable at different
> wire diameters and frequencies, for different materials.
>
> As the wire diameter becomes larger, the circumference becomes
> greater, and so the amount of conducting material. Works for any
> frequency [1]. What Terman's resistance ratio indicates is the
> amount (if any) of unutilized conductive material.
>
> At 60 Hz, using solid cylindrical pure Copper conductor:
>
> For 1% non-conducting cross-section, max wire diameter = 0.571 inches
> For 10% non-conducting cross-section, max wire diameter = 1.00 inches
>
> For 0.57" diameter, current capacity = 457A
> For 1.00" diameter, current capacity = 1157A
> For 2.00" diameter, current capacity = 1862A
>
> and so on...
>
> (700CM / A, conductive cross-section, 60Hz)
>
> Larger cable ==> greater current capacity. Would you agree?
>
> Strangely enough, I believe there may be more to power distribution
> engineering than skin effect. Weight, cost, tensile strength,
> wind/ice/tree limb load capacity, wire availability vs diameter, line
> weight handling considerations, local power demand, voltage vs current
> transmission mode?
>
> Regards,
> Dangerdave
>
> [1] at RF, the Terman model is not good enough for consumer work, but
> may be adequate for government work.
--
James Meyer
>Dr. Howard Johnson has a nice analysis of skin effect on his web site,
>and there is a wealth of information dealing with skin effect, from foil
>power transformers on up through light. Everyone seems to come up with a
>slightly different formula.
>
>But nobody seems to have found a clever way to eliminate it.
I thought Litz wire was a pretty clever way to eliminate it. I haven't
seen Litz PC board traces yet. Perhaps you can't make them on a 3-D board.
Anybody got a free layout program that does hyperspace layers?
Speaking of Litz wire... I've got a MicroSim v6.2 PSpice simulation
file that demonstrates the skin effect and the way that Litz wire overcomes the
effect. I could post it to my web page or e-mail it to people if there's any
interest. The file uses the standard library built in to the free evaluation
software. No "trick" components.
Jim
dang...@earthlink.net
>A single strand wire has wasted copper in the centre due to skin effect.
Could you be less specific?
Regards,
Dangerdave
Mike Monett
[...]
> Speaking of Litz wire... I've got a MicroSim v6.2 PSpice simulation
> file that demonstrates the skin effect and the way that Litz wire overcomes the
> effect. I could post it to my web page or e-mail it to people if there's any
> interest. The file uses the standard library built in to the free evaluation
> software. No "trick" components.
>
> Jim
Why don't you post it, along with some GIF's for those who don't run
PSpice?
The subject appears in s.e.d from time to time. It would be nice to have
a complete library of the excellent responses all in one place.
Are you busy this weekend?
bkiz...@my-deja.com
>
> The problem might benefit from calculating resistance loss effects.
>
> For 30AWG, 400kHz, Ratio AC/DC wire resistance = 1.04
> For 30AWG, 1MHz, Ratio AC/DC wire resistance = 1.22
>
> and Q = 2pi * f * L / R
>
> where; R = wire resistance at frequency
>
> Skin effect causes R to increase with frequency, so Q decreases with
> frequency if skin effect is the controlling loss factor.
>
> Since Q in both your data sets either 1) increases with frequency, or
> 2) both increases and decreases with frequency, it is difficult to see
> how skin effect could be dominating Q.
>
> Anyone got a better suggestion?
>
> Regards,
> Dangerdave
>
So, let's try and go a bit further with this. Since I am the "hands-on"
type with a very short attention span, I do not grasp things until I
have spent some time on the workbench.
The HP LCR Meter I am using will also measure the following parameters:
|Z|, R, ESR, G, X, and B.
Some of these parameters I understand, but others are still a bit of a
mystery. Would measurement of any of these give more information that
would help us understand what is happening? My goal is to design a
better coil for the kids AM Broadcast radio kit I have been designing.
Any help with this problem would be appreciated.
Bruce Kizerian
Mike Monett
Mike Monett wrote:
[...]
> There are some complaints that Terman's data is pretty old and becoming
> less useful for modern analysis.
But Terman is still the standard.
Dr. Johnson addresses flat rectangular conductors (microstrip) at
http://www.sigcon.com/news/skineffect.htm
He gets the same result for skin depth at 1 GHz (8.2E-05 inches) as
Terman.
Kevin Schmidt, W9CF, derives the equations for rectangular cross
section conductors at
http://fermi.la.asu.edu/w9cf/articles/equiv/equiv.html
His results agree with Terman to within 10%
Kevin Schmidt states
"So mostly my code agrees with Terman within 5 to 10 percent. The
Terman data are old and not readily available."
dang...@earthlink.net
>> less useful for modern analysis.
Has electromagnetic physics changed recently?
For simple effects, like wire resistance vs. frequency, the basic
Terman models should be timeless. Terman's wire resistance model was
derived from simple first principles. It doesn't depend on measured
data, excepting fundamental material properties like permeability. If
you need more accuracy, it comes in the usual way - a more complex
model, including other effects specific to the application. For
example, an important wire resistance effect, not included in the
Terman EM model, would be conductivity change vs. temperature.
Regards,
Dangerdave
dang...@earthlink.net
>
>|Z|, R, ESR, G, X, and B.
>
>Some of these parameters I understand, but others are still a bit of a
>mystery. Would measurement of any of these give more information that
>would help us understand what is happening?
Yes. The wire length is also useful to know.
>My goal is to design a
>better coil for the kids AM Broadcast radio kit I have been designing.
>Any help with this problem would be appreciated.
>
>Bruce Kizerian
I'm probably not the best person to advise you. Many factors
influence the Q of an iron-core inductor. Just from design theory,
they would include core permeability vs frequency, core eddy current
loss, wire DA and DF losses, wire R loss, parasitics and stray
couplings.
Someone who winds IF/RF coils for a living, or some of the dedicated
amateurs in the radio newsgroups might be able to narrow it down more
from experience.
Regards,
Dangerdave
Mike Monett
I said nothing about Terman's models. His data is old.
dang...@earthlink.net
>I said nothing about Terman's models. His data is old.
>
>Best Regards,
>
>Michael R. Monett
>mailto:mkmo...@hotmail.com
I believe another poster introduced Terman's dusty data, not you.
Sorry if you took the reply to heart.
Terman's derivation/data are public domain. But I can understand how
discussions would invoke personal associations. We all have our
favorite methods and authors :-)
Regards,
Dangerdave
Mike Monett
>
> >On Mon, 24 Jan 2000 02:54:09 GMT, Mike Monett <mkmo...@hotmail.com> wrote:
>
> >I said nothing about Terman's models. His data is old.
> >
> >Best Regards,
> >
> >Michael R. Monett
> >mailto:mkmo...@hotmail.com
>
> I believe another poster introduced Terman's dusty data, not you.
T'was I.
> Sorry if you took the reply to heart.
Nope. Just keeping it short to minimize the chance for misunderstanding.
[...]
> Regards,
> Dangerdave
Anyway, glad you now consider skin effect real below 1 khz. Did we drain
the swamp?
Martin Pickering {UK}
dang...@earthlink.net wrote:
>>My goal is to design a
>>better coil for the kids AM Broadcast radio kit I have been designing.
>>Any help with this problem would be appreciated.
I investigated this at university for a 30MHz coil back in 1972.
If "better" means "high 'Q'" then an air core is best at that frequency.
In addition, windings should be made with a gap the width of the wire
diameter between each pair of turns. Wire should be as thick as possible.
Those factors were the most important. Any introduction of a former or iron
core was found to reduce the 'Q'.
However, for AM broadcast, the frequency is lower. You might try using a
plastic film cassette container as the former. This is about 30mm in
diameter and about 50mm in length. Its smooth surface allows the wire to be
wound easily by hand and it is mostly air dielectric inside. You can get
these empty plastic containers by the bucketful at any camera shop that
takes in films for processing.
This factor is useful because you can get them free - which reduces the
cost of your kit. I already use them for my own crystal set kit.
http://www.satcure.co.uk/crystal.htm
Martin Pickering
http://www.satcure.co.uk
dang...@earthlink.net
>In article <LcmKOPAQIeGnWH...@4ax.com>,
>dang...@earthlink.net wrote:
>
>>>My goal is to design a
>>>better coil for the kids AM Broadcast radio kit I have been designing.
>>>Any help with this problem would be appreciated.
>Martin Pickering
>http://www.satcure.co.uk
>
I'm not the poster requesting help, Martin. But IIRC, the coil was
being optimized for Q over the frequency range 400kHz to 2MHz.
Regards,
Dangerdave
Robert Strand
to get side-band attenuation of the information signal. The use of rods
appears to be optimum for achieving a good Q with the bonus of more voltage
output. The only simple way to increase selectivity without undesirable
attenuation of the signal band is to use a higher order filter - ie two stages
(or high order IF).
Regards
Rob
Martin Pickering {UK} wrote:
> In article <LcmKOPAQIeGnWH...@4ax.com>,
> dang...@earthlink.net wrote:
>
> >>My goal is to design a
> >>better coil for the kids AM Broadcast radio kit I have been designing.
> >>Any help with this problem would be appreciated.
>