[Amps] bifilar wound toroid core frequency response

[Dan Sawyer]

Hello, this is a question relating to a bifilar winding around various 
torroid cores. The winding was made from two pieces of #26 enamel wire 
about 6 inches long and wound at 3 to 4 turns per inch. The toroids were 
either size 37 or 50 and made from #43, #61, and #67 material. The wire 
winding was then wound around the toroids.

The winding cores were then tested for frequency response. The #43 core 
had the lowest response and the least loss, the #61 was second, and the 
#67 poorest. The #43 had a loss of .4 to 1 db between 160 and 6 meters, 
the #67 had a loss of 1.5 to 2.4 db.

This all seems reasonable. There is a small response glitch in response 
at about 25 MHz, for about a MHz. This is common on all toroids. The 
question is: What is the cause of this glitch?

The measurement system is an HP 8505a network analyser. The power is set 
to -20db, the network is RF out to an mini-circuits HF splitter, one leg 
to the R input and the second to channel A. The measurement system does 
not have the glitch measuring other configurations.

Thanks in advance - Dan kb0qil