I decided to get myself active on 60m. I replaced an exiting 40m Ground Plane with a dual band 40m / 60m element. My tree support was not high enough to support the full 60m element, so I bent the top 1.2m back towards the 40m element. The elements are spaced with the excellent DXwire SP-200 ladder line spacer. Two elevated radials were added for each band.
GM0GAV ADSL Filter
For several years I have used the popular OZ7C ADSL filter to prevent dropping my broadband when transmitting on 160m. It worked well, however reduced my download speed by 1 Meg indicating considerable loss in the passband below 1.1Mhz. Steve G3VMW improved the design by getting the roll off on the correct frequency, the downside was it needed custom inductors on iron powder cores.
I used the excellent Elsie and LT-Spice to come up with a design which is low loss and easy to build with off the shelf inductors. It’s an m-derived low pass filter, designed to match 100 ohm lines. Construction is very straightforward as all inductors are 10uH, so you only need 3 different components. The loss is typically 0.2dB across the ADSL band up to 1.1Mhz.
LT-Spice plot of performance using real inductors. Loss > -50dB on 160m and -38db on 80m.
Blue = Insertion Loss Green = 100 Ohm Return Loss
I have successfully built filters using Murata 2200R series & Epcos LBC series inductors, both have reasonable Q at 1Mhz. Capacitors should be 100v working voltage or greater, I used NOS / surplus Silvered Mica which are readily available on Ebay. If you have access to an L/C meter it’s worthwhile picking component combinations which resonates the input / output series tuned circuit around 1850 Khz.
My Buffalo ADSL router appears to measure line loss accurately, adding the filter increases the line attenuation by 0.2dB which agrees with LT-Spice. I added and removed different filters several times, the result is always 0.2dB loss with no effect on sync speed.
Common mode choke
When building a filter you should add a common mode filter, easier than wrapping the cable through ferrite at a later stage. I added a choke using 10 turns of thin twisted pair on a FT50-75 core. This results in a a choke > 1000 ohms between 0.5 and 14 Mhz. The twisted pair can be taken from a scrap CAT5 cable or formed from thin hookup wire. Two of these chokes could be cascaded or a second choke with 10 turns on FT50-43 added to improve HF performance.
Good common mode choking is important if you have a long overhead line to your property, it may improve ADSL sync speed if your ADSL router has poor common mode rejection.
I needed to build some small baluns for vertical dipoles and arrays, to be used on the T2GM dxpedition. We required the following:
- Handle 500w
- good choking impedance 14 to 30Mhz
- Lightweight and waterproof
First I tried a W1JR type on a FT140-43 core. This was a nice balun, but was not easy to form even with thin teflon coax and the performance dropped off on 10m. After reading GM3SEK’s article I headed to the junkbox to see what other ferrite I had to hand. I have a good number of Ferroxcube CST19/10/29-4S2 cores which I picked up on ebay for a song. 4S2 material is very similar 43 material.
Two cores were glued together to form a binocular, 3 turns of RG316 teflon coax fits through neatly. RG316 is only 3mm diameter but can handle 1.1Kw on 10m! The choking impedance is > 2000 ohms 14 to 30Mhz, reaches 2550 ohms on 10m. This was fitted into a small 60 x 65mm polycarbonate waterproof box with a 6m flying lead of Aircell 5 coax.
Two identical baluns were connected back to back and tested for extended periods 500w key down. Both the cores and the RG316 coax became warm as you would expect, but nothing to get worried about. I am confident they will survive anything when used at this power level.
I have completed the 6X2 7/16DIN switch and works as expected. Mike has done a great job with the design and the PCBs are excellent quality. The case is made by Rose Bopla p/n 012328118, available from Digikey. Rather expensive but perfect for this project. Measurements made with my N2PK VNA :
Insertion Loss Radio 2 <=> Port 6 0.066dB @ 28.477 Mhz
Return Loss Radio 2 <=> Port 6 24.5dB @ 28.477 Mhz