Here is some additional information on the RF current meter I made.
This was prompted by GM3SEK's comment about core loss. His comments
were correct, so that prompted me to go back and see why my measurements
indicated such high losses when using the split ferrites intended for
RFI suppression. Plus I hate having measurements that produce
unexplained results. I traced the error back to a cantenna load I was
using for some of the tests (when using power). It had some impedance
bumps, particularly bad around 14 MHz. I had other loads I was using
which were good, but the cantenna is the one that caused the problem.
Repeating some of my previous measurements I found that some of these
cores were acceptable. The 1x1x1 split core in #43 material was pretty
good but the 1/2 inch hole was really too small. The large core I
originally wanted to use with a 3/4 inch hole started to have problems
around 25 MHz.
So my original conclusion about lossy ferrites was not correct. Some of
them may be acceptable. Before building this it would be best to take
the ferrite you wish to use and check it. The best way to do this is to
use an antenna analyzer with the terminated transformer snapped on the
center conductor of a 50 ohm line, terminated in 50 ohm load, and check
the SWR versus frequency. The open loop where the transformer is
attached needs to be short so that the impedance bump due to the open
wire is not too big.
>Ian White GM3SEK wrote:
>>It is important to terminate the secondary winding with a low-value
>>resistor. If you do, then even an RFI suppression bead will act as a
>>transformer with a very low insertion impedance in the main line. For
>>example, if a 10-turn secondary is terminated in 47 ohms, the insertion
>>impedance in the main line is theoretically 0.47 ohms; and in practice
>>it is only a few ohms.
>>If you leave the secondary with a high-impedance load, *then* you will
>>see the full insertion impedance of the ferrite bead. (A simple
>>impedance meter like the MFJ-259B will show this effect quite well.)
>Thanks for the reply Ian. Yes, this makes sense. I guess I wasn't
>thinking too clearly when building this, however I did test some of the
>large ferrite split beads to see what they did to the circuit being
>measured. I used a 20 turn transformer terminated in 100 ohms, placed
>on a 50 ohm line which was terminated in 50 ohms, and then measured the
>change in SWR on the line, using about 10 watts. I also checked it with
>an antenna analyzer. I found all of them produced an unacceptably high
>SWR. Maybe the reason is something other than core loss. A very thin
>toroid that contains only a small amount of material worked much better.
>TowerTalk mailing list
TowerTalk mailing list