Experiments with LC circuits  part 4

<< Go to part 3    Go to part 5 >> 
Back to the index

From tuning capacitor C1b I now removed the insulation rings which keeps the stator on it's place.
I found that the used nylon screws gave some dielectric losses which reduces circuit Q.
Now I use instead a polyethene block which supports the stator from the underside.
In the following drawings this is shown., this version of tuning capacitor I call: C1c.

Tuning capacitor C1b 

1= frame of tuning capacitor
2= plastic insulation ring
3= nylon screw
4= polyethene insulation block
5= copper connection bar for stator plates
6= stator plates

Tuning capacitor C1c

1= frame of tuning capacitor 
2= copper connection bar for stator plates
3= metal screw
4= polyethene block
5= polyethene bottom plate
6= nylon screw
7= stator plates

The plastic insulation blocks of tuning capacitor C2 are replaced by polyethene blocks (which are the green blocks on the next picture).

Left: C1c  (495 pF).

Right: C2a (390 pF).

After these changes, I measured the Q's again, they were increased.
The percentages of increase are also mentioned, compared to measurement 21 and 22.

Measurement 
number

LC combination 

F min 
kHz

F max 
kHz

Q
600 kHz 

Q
900 kHz 

Q
1200 kHz 

Q
1500 kHz 

23

L9 C1c 

542

2349

1072
(+10.7%)

947
(+7.3%)

815
(+19.3%)

658
(+12.8%)

24

L9 C2a

606

2400

1143
(+24.9%)
(606 kHz) 

1000
(+32.2%)

833
(+46.6%)

632
(+55.3%)


Now I have drilled a lot of 2 mm holes in the polyethene insulation blocks of tuning capacitor C2a.
Hoping that removing insulation material will reduce dielectric losses even more.
This version of tuning capacitor is called: C2b.

Measurement 
number

LC combination 

F min 
kHz

F max 
kHz

Q
600 kHz 

Q
900 kHz 

Q
1200 kHz 

Q
1500 kHz 

25

L9 C2b 

600

2382

1052
(-7.9%)

1000
(0.0%)

875
(+5.0%)

714
(+12.9%)

The percentages mentioned are compared to measurement 24.
At 600 kHz there is a decrease in Q, how that comes, I don't know.
Maybe there was a measuring error in measurement 24, normally the accuracy of the measured values are I think between +5% and -5%, but maybe this time the error is higher.

Or maybe the Q is really reduced at 600 kHz, this is no problem, at low frequencies a Q above 1000 is more than enough.


"Tot lering en vermaak"  That's an old Dutch expression which means "for learning and fun".
And that's why I made the following two measurements with lower Q coils, often used in crystal receivers.

Coil L10
13 meters litzwire 40x0.07mm on a former made of a CD.
The shiny aluminium layer is fully removed from the CD.

Coil L11
70 turns on a toilet paper roll.
Wire: massive 0.8mm enamelled copper wire.
Wire length: about 10.5 meters.
The windings are not laying nice side by side, because it was used (old) wire with lots of bendings.

The used tuning capacitor is C2b.

Measurement 
number

LC combination 

F min 
kHz

F max 
kHz

Q
600 kHz 

Q
900 kHz 

Q
1200 kHz 

Q
1500 kHz 

26

L10 C2b

530

*

358 371 346 304

27

L11 C2b

761

3270

 *

211

234

241

* = not measured.
The measured values are not bad at all, but keep in mind that a very good tuning capacitor is used.
So, a good comparison with measurement 1 to 16 is not possible, also because the test method used now is much better.

<< Go to part 3   Go to part 5 >>
Back to the index