RF transformer bandwidth measurement is an easy task using a vector network analyzer. I was curious about the bandwidth that
could be achieved with the simple trifilar winding home made transformer used here:
LVDS to single ended output transformer for the SI570.
In order to test that transformer into a 50 ohms measurement system, due to the 50:200 ohms transformation ratio, two identical transformers should be built and connected "back to back". That way, S11 and S21 parameters could be measured. True S21 value would be half the measured value.
But today I was a bit lazy and built a 1:1 transformer, just to test.
I had in my junk box a BN-43-302 binocular core from Micrometals. Having in mind that the transformer should have enough reactance to work down to 3.5 MHz, I choosed the lowest number of turns that generated a reactance of at least 4 times the system impedance. I needed the required number of turns to have a reactance of at least 200 ohms. We have an useful WEB toroid calculator at http://toroids.info/BN-43-302.php
The core has an AL=1.28 uH per square turn. Three turns on the core has an inductance of 3^2*1.280 = 11.52 uH
At 3.5 MHz, the reactance is X = 2*pi*3.5e6*11.52e-6 = 253.34 ohms.
I connected the transformer to my network analyzer through two SMA female connectors and made a frequency sweep from 1 to 432 MHz:
Now a detailed picture at low frequencies, sweeping from 100 kHz to 3 MHz:
Measured -3dB bandwith from 267 kHz to 295 MHz. There is a resonance at 96 MHz. SWR is low at the full 1.8 MHz-50 MHz range, with losses about 0.32-0.40 dB.
Not too bad, isn't it?Copyright (c) 2013 Ramiro Aceves . Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license is included in the section entitled "GNU Free Documentation License".