Here now the reworked original cushcraft balun.

              

It was an awful view after opening the balun. I found not only the spider cocoons. Also ants must have nested

inside and they brought a lot of sand and mud into the balun. I really don't know if the balun had worked after

some days of clammy weather. Nearly half size of coil laid in that mud. Also the ferrite core stick only 3/4 inside

the coil. After carefully cleaning all it looks quite good now and I will use the balun for the ATB-34. 

 

I made a short balun test.

I used my MFJ-259 HF/VHF Antenna/SWR/RF Analyzer and a 50-Ohm dummy load. See pictures! I checked

the frequency response from 10 - 30 MHz.

 

1st, I checked the VSWR of the dummy load. By direct connection of the Analyzer to the 50-Ohm dummy I read

50-Ohm on the resistance meter and 1.1 on the SWR meter. No changes by tuning from 10 too 30 MHz occur.

 

2nd, I connected the balun between the Analyzer and the 50-Ohm dummy. Results are exactly the same like in

the 1st test.

3rd, I made exactly the same test with the HyGain BN 1000 Balun. Here I found that the SWR from 21 MHz up is 1.2!

And from 10 to 21 MHz the SWR is just a needle's breadth higher than the Cuschcraft balun.

Les than 10 MHz and up over 35 MHz the SWR is increasing and also the resistance changed below 50 Ohms.

 

Cushcraft balun:

7 turns trifilar winding, lenght 45mm, core length 62mm, core diameter 12mm, wire AWG 12 or 13, (1,8-2,05mm)

HyGain Balun:

9 turns trifilar winding, length 55mm, core lenght 70mm, core diameter 12mm, wire AWG 12 or 13 (1,8-2,05mm)

 

Shematic:

 

So the results I found on the Cushcraft balun are better than I expected, therefore I will use the balun like it is.

________________________________________________________________________________________

 

 

                                        Checking all traps before assembling the elements. 

 

The Cushcraft traps are simple mechanical construction (see pic. 6). A fibre tube carrying the coil and the element tubes.

The capacitor made from a piece of coax and rubber end caps used to seal the coax against dampness.

 

Test: How to do?

Simply because the traps are tuned circuits; I used my old transistor dip-meter to check the resonant frequency of the traps.

I got clear resonance dip's by using a link coupling. One turn around the trap coil and two turns on the dip-meter coil.

Reflector 15m trap                                                              only loose link coupling is needed                                 

  

Driven element 10m trap                                                         Driven element 15m trap                                                      

  

No differeces in measurement results found if the link coupling is direct on the coil or over the plastic slipcover.

 

Here are the found trap resonance frequencies:

reflector el. resonance   15m trap left 19,9 mc  (20,5 mc) reflector el. resonance  15m trap right 19,9 mc  (20,5 mc)
driven el. resonance      10m trap left 26,5 mc  (27,1 mc) driven el. resonance     10m trap right 26,5 mc  (27,1 mc)
driven el. resonance      15m trap left 19,5 mc  (20,1 mc) driven el. resonance     15m trap right 19,5 mc  (20,1 mc)
director el. resonance    10m trap left 27,0 mc  (27,6 mc) director el. resonance   10m trap right 27,0 mc  (27,6 mc)
director el. resonance    15m trap left 19,1 mc  (19,7 mc) director el. resonance   15m trap right 19,1 mc  (19,7 mc)
During measuring the dip frequencies  all is checked with  the a receiver:  
Scale-reading precision for the test: within 10 kc Repeat accuracy for the test within 50 kc

I suppose, that the exact 'only' trap (coil and capacitor only) resonance frequency is about 500 kc - 700 kc higher!

Like you see on the pictures, I've done the measuring on the complete assembled traps. Additional inductance and

capacity from the element tubes cause a lower resonant frequency. 

   

Last I checked the Q of the trap circuits by comparing the clear of the dip's on the dip-meter with a homemade

LC-circuit. Nearly the same dimensions of the coils are used, but an air-capacitor instead of the coax-capacitor. Also I 

used the same link coupling. Only very minor deviation found in the clearness and depth of resonance dip's on the meter.

I conclude from the fact that the Q of the traps is very good.

   

Sure these tests are very simple but sufficiently exactly for Ham Radio I suppose.

 

And now, make a beam antenna:

          All assembled now and stored in the back of the garden.

We will bring the antenna up if the the next suny day let's work outside.

 

 

... soon more

Zurück zur Hauptseite/go main page