The Xtal Set Society
The Xtal Set Society
profile | register | search | main site

email us
Forums | Crystal Radio Think Tank ** Advanced Forum | Double slider question. Post Reply Send Topic To a Friend
Author Topic
Bob WeaverPosted - 27 February 2013 23:40  Show Profile  Email Poster  Edit Message
Yes, I read the posts. They seem to indicate that shorting turns either makes things worse, or else no difference, but no details are given about the types of coils that were used for the experiments. In any event, no one mentions shorted turns making things better, which is pretty much what I was saying.

I've done experiments with ferrite toroids, and as mentioned above the Q plummets when just a single turn is shorted. I believe that the significant factor is how well coupled the shorted turns are to the rest of the coil. And that would be my line of investigation if I were to do any further experiments.

gzimmerPosted - 28 February 2013 0:18  Show Profile  Email Poster  Edit Message

We discussed this at length a while back.

As you say, a shorted turn on a toroid is disastrous because of the very tight coupling.

On the other hand, shorted turns are sometimes used for tuning high power BC transmitters (where efficiency is extremely important). This suggests that shorted turns are not inherently lossy with air-cored coils.

As for unused turns, I'm guessing that if they are lossy it's for the same reason that taps are lossy.
Any unwanted material at the high impedance end of a coil adds capacitance. If the surrounding dielectric is not perfect then that added capacitance will be lossy.

....... Zim

Edited by - gzimmer on 2/28/2013 1:30:24 AM

Garry NicholsPosted - 28 February 2013 6:39  Show Profile  Email Poster  Edit Message
When John said, "Open end turns on a hi Z radio develop substantial voltages, acting like a Tesla coil.", I take that to mean an increase in dielectric and environmental losses due to an increase in Z that is not intentional from the non-contributing section of the coil.

It is the elevated Z, no matter what the operating voltage that would be the problem. The Tesla coil example was used to indicate transformer effect elevating the Z.

He is not talking about the crystal set bursting into flame, as a transmitter might!

Did I get that right John?


Electronic TechnicianPosted - 1 March 2013 22:16  Show Profile  Email Poster  Edit Message
"Supposing there are *actual* (measureable) losses involved with "dead-end" turns, if a double slider were adjusted, such that there are always ~equal numbers of "dead-end" turns at both ends of the coil (for the required inductance), would that eliminate the problem? (ie. "phasing" and all that)."
If you must change the inductance, the best way to do this is probably the contra-winding since the lower inductance setting actually increases Q rather than decreasing it as would be the case for shorted or open windings. Other ways to do this would be using vintage ferrite as a slug tuner to change inductance or use a variometer to accomplish inductance change. Perhaps in an ideal world, you might have separate sets for different portions of the BCB each with its own optimal coil. In any case, I would suggest that using a slider for main signal tuning is much poorer than using a tuning capacitor if that is the aim for your question.
It can be difficult to decide if your coil Q has suffered in practice. If the loaded Q of the set is fairly low compared to the unloaded Q of the tank, then modest unloaded Q changes can be hard to detect. Loss is related to Q unloaded to Q loaded ratio. Loss is high when these two are near equal and loss would be fairly low if loaded Q is an order of magnitude lower than unloaded Q. That is one reason we like 660/46 litz.

_J_Posted - 4 March 2013 12:55  Show Profile  Email Poster  Edit Message
Tuning with variable inductance resists efforts to tune sharp. But for simple low performance thrills, it is unbeatable and for that worrying about dead end turns is not very important. Tuning with variable C has to be well architected to be beneficial, it can be worse.

John Davidson

golfguruPosted - 4 March 2013 16:30  Show Profile  Email Poster  Edit Message  
>>>>>>> Tuning with variable inductance resists efforts to tune sharp.

More info please.


_J_Posted - 5 March 2013 11:21  Show Profile  Email Poster  Edit Message
The paraphernalia to make the inductance variable almost always hurts Q (like taps, slider, end turns, variometer losses) more than that for variable capacitance.

But simple antennas are often short and can't be tuned with variable capacitance alone. Capacitors don't usually have enough range to tune the BC band if we simply add a fixed inductor across the capacitor. So capacitive tuning gets complicated.

John Davidson

golfguruPosted - 5 March 2013 16:17  Show Profile  Email Poster  Edit Message  
Wondering about ferrite core variable inductors. They should be able to tune the full MW band against a reasonable value of antenna C (150pF plus).

Q (which will be low anyway) should improve marginally as inductance is reduced (ie. Hot end of coil can "increasingly" become an "air-coil" at higher freqs.

Not much paraphernalia required there. Unloaded Q of a ferrite coil can be higher than a magnet-wire air-coil as well.


Edited by - golfguru on 3/5/2013 4:55:16 PM

RichardPosted - 5 March 2013 18:30  Show Profile  Email Poster  Edit Message  
"Wondering about ferrite core variable inductors. They should be able to tune the full MW band against a reasonable value of antenna C (150pF plus)"

Yes, ferrite rod core variable inductors can tune the entire MW band when coupled with a suitable fixed capacitance. Using close wound litz wire, spaced on low loss forms with a smaller diameter ferrite rod core, I've achieved very respectable Q's in the several hundreds.

However, as the ferrite is withdrawn coil Q's always decline, in step with decreasing inductance. This applies to air core coils as well.

One of the difficulties of using sliding core tuning is that small linear shifts in core position result in relatively large inductance changes, which aren't linear. Frequency calibration is no easy task using this method.

I've considered using a 1/1000 inch dial indicator or micrometer screw for tuning repeatability but consider that approach far too complicated and cumbersome to be practical. Sliding core tuning works great in a simple 'Rocket Radio' type set that's tuned by ear but that's about the limit as far as I can see. Vernier dials and SLF capacitors are far more practical for precision tuning.

Edited by - Richard on 3/5/2013 6:32:27 PM

golfguruPosted - 5 March 2013 20:28  Show Profile  Email Poster  Edit Message  
>>> small linear shifts in core position result in relatively large inductance changes, which aren't linear. Frequency calibration is no easy task using this method.

Ferrite rod inductance (507uH @ 500kHz) varied against fixed 200pF.

500-800 = 18mm (rod withdrawal measurement)
800-1100 = 9mm
1100-1400 = 6mm
1400-1700 = 4.5mm

Equivalent capacitance values (675pF @ 500kHz) versus percentage rotation for tuning a 150uH coil. (1000pF, calibrated, SL capacitor used.)

kHz = relative rotational difference
500-800 = 33%
800-1100 = 10.5%
1100-1400 = 4.5%
1400-1700 = 2.7%

These figures seem to indicate that, in this particular case, the ferrite tuning is 2-3 times friendlier than straight line capacitive tuning.


Edited by - golfguru on 3/5/2013 9:59:18 PM

_J_Posted - 5 March 2013 22:16  Show Profile  Email Poster  Edit Message
Agree, GG, ferrite variable inductors are hard to beat when you don't want to use a variable capacitor, or any capacitor. Many will cover the BCB.

John Davidson

Bob WeaverPosted - 5 March 2013 22:18  Show Profile  Email Poster  Edit Message
Automobile radios used reluctance tuning (sliding ferrite slugs) for decades, and were sufficiently repeatable that mechanical pushbutton tuning was not only practical, but worked very well.
RichardPosted - 5 March 2013 22:33  Show Profile  Email Poster  Edit Message  
Yep, old tube type car radios worked very well using variable inductors. However, the mechanics and radio dials were more suited to those times. Dial calibration was rather course at best.

Tune around until a station was found and set the mechanism for repeatable tuning. That was it. I still have one of those mechanisms with ferrite cores in my collection. It's a marvel of mechanical engineering.

BTW, the same type of push button mechanism was also used with variable capacitors in some home sets.

Edited by - Richard on 3/5/2013 10:35:44 PM

RichardPosted - 5 March 2013 22:37  Show Profile  Email Poster  Edit Message  

rod withdrawal measurement

0500-0800khz = 18mm
1400-1700khz = 4.5mm

Interesting data. Let's see, we have a range of 18mm divided by 30 channels at the low end of the MW band and 4.5mm divided by 30 channels at the high end.

On a linear scale (which it isn't) that yields an average of 0.6mm (0.024") / 10khz channel at the low end and 0.15mm (0.006") / 10khz channel at the high end.

That sounds about right to me. As I said, it's rather difficult to determine what channel one is on using that method to produce a variable inductance without resorting to a micrometer type readout. Doable I suppose but not easy.

OTOH, I've encountered no difficulty calibrating and separating individual channels with my BC-221 crystal set. The detector circuit uses a 50 turn dial, each turn with 100 divisions for a total count of 5000.

At the top end of the band the vernier dial indicates 30 divisions/ channel, that's 36 degrees of a full dial rotation with 100 divisions/ channel at the low end.

Equipped with high Q ferrite core litz coils and a two cap Tuggle antenna tuner the set achieves excellent selectivity. I find this approach imminently practical and quite elegant. It's all I could ask for in a cased but portable crystal set.

golfguruPosted - 5 March 2013 22:52  Show Profile  Email Poster  Edit Message  
I guess I am suggesting that adjustment of this small ferrite rod is no more fiddly than an SLC capacitor at "shaft diameter" level.

The knob (larger the better) gives the capacitor the ability to be tuned more easily.

A similar sized knob attached to a ferrite rod (to convert linear to rotary, somehow) should yield no worse adjustment ability.

>>>> 0.15mm (0.006") / 10khz channel at the high end.

Using 0.6mm threaded rod would give an average of 4 chans per 360 degree turn
in the "high quartile". Obviously way less at the very top end but still achievable? (a few degrees per channel?).

A vernier dial attached to the "nut" should definitely fix any problems.


Edited by - golfguru on 3/6/2013 4:08:51 AM

Topic is 4 Pages Long:    1 2 3 4

Click Here To Close Thread, Administrators & Moderators Only.

Show All Forums | Post Reply