Mar 26, 2009

Receiving splitter II

Another one?
Yes; in my previous blog of January I mentioned needing another splitter.
It would need a better spec at 28Mhz. The IF output frequency of my 144Mhz & 50Mhz transverters.
I have plans for:
- analysing signals during reception
- comparing different IF rigs at readability
- combining different IF rigs to create sort-of-diversity reception

And not for?
This blog is not for writing a paper on already existing applications, home built equipment and such. There are lots of other, better web resources for that, like: W8JI, VK1OD, hard-core-dx to name a few.
I just want to record my findings and describe the equipment used while in the process. It will serve me when picking up a project again in the future.

What do we have so far?
A quick & dirty W8JI style Magic-T combiner, using #73 ferrite binocular.
The frequency sweep shows the port to port isolation from 1 to 30Mhz.

W8JI version Magic-T splitter_MKII
A more serious attempt this time, using:
- alum die cast box
- #43 ferrite binoculars
- 24 AWG red enamelled wire
- BNC chassis parts, pop riveted to the box
- 1W 100 Ohm metal film resistor

I have ordered carbon composite resistors. If that offers a significant better specification at 28MHz I will do some more testing and create additional versions of the splitter. When time allows.

How do they compare?

attenuation port to port isolation
1.8Mhz 28Mhz 1.8Mhz 28Mhz

MKI -3.52 -4.33 28.0 15.5
MKII -3.05 -3.30 38.0 19.2

Remarkable that using #43 ferrite brings better specs at 1.8Mhz, right?
There sure are other possible reasons for that: using a shielded box and different wire.

Other remarks:
Port 1 to Port 2 isolation and vice versa differ 1 to 1.5 dB.
The measuring coax cables have been calibrated in the test set-up by the HP network analyser.
The cables (RG58) might have a small negative influence on the isolation measurements due to the limitation of their shielding.

Yes; it has low attenuation, sufficient port-to-port isolation and it is easy to reproduce.
Let's get using it and do some of the tests I've on my mind.
I'll be back.....

No comments: