Above — This blog post supports the video shown above
[1] SIGNAL GENERATOR
Steve AA7U & Everett N4CY, built gear -- plus a procedure to test Intermodulation Distortion (IMD) on a loop amplifier using a Siglent SDG2042X generator and SSA3021X spectrum analyzer. Click on this hyperlink to read about it. I'm a fan of Siglent test equipment.
My strategy employs a 50.0 MHz crystal oscillator-based signal generator plus a 50.1 MHz VFO as the second signal source. My VFO tunes from about 49.6 to 51.8 MHz via a front panel air variable capacitor.
My 50.1 MHz VFO
Above — VFO schematic. Although I had worked out the low-pass filter L and C values, I built this VFO without a schematic and perhaps would build it differently if I needed to make another. I might consider tuning the output of the differential amplifier buffer for more output power and less harmonic energy.
I thought mostly about temperature drift when making this -- I started with JFET amp as the oscillator and struggled to make it work. This would be wise since a JFET offers better temperature drift over a BJT and gives a cleaner output signal with lower phase noise. However, I only had 1 day for this entire project and got frustrated. I deployed a common base PNP BJT local oscillator (LO) that never fails for me.
Both the LO and its buffer get regulated, well filtered DC. The LO gets temperature compensation/separation from the 8.2 volt Zener diode-based voltage regulator by way of 2 R C low-pass filters. I applied several C0G caps to resonate the tank and ran 2 air variable trimmer capacitors -- 1 as the main board frequency trimmer, the other as the front panel tuning control.
The LO gets lightly coupled via 1 pF to a differential amplifier emitter fed 10 mA with a current source. Differential amps offer strong reverse isolation, plus a reduced 2nd harmonic if the BJT balance is OK. The BC546 pair offer reasonable balance right out of the bin (without matching) & the BC546C serves as my go-to differential amp BJT from DC to ~ 100 MHz. The 10 mA current source, plus the 21 mA current in the final feedback amp provide heat for my temperature compensation scheme.
Low-pass filters built using T30-10 toroids worked OK. This was a board cram -- so the inductors are not spaced apart as much as when more board space is available. The 22 gauge air inductor measured ~ 374 nH & seems well anchored to the main 1-sided board with J-B Weld epoxy, plus the grounded coil lead soldered to the main board. The main board = 1/16″ (1.60 mm) Half Ounce 500 Series Copper Clad Board from MG Chemicals.
Above — Copper board under test. To simulate the front panel capacitor, I've got the air variable front panel tuning cap in a small bracket that I got from a local Builders merchant. I have several for holding caps, jacks and potentiometers during test phase circuit development. Mine are all pre-drilled with the proper sized holes to fit pots jacks, or air-variable trimmer caps.
Above — Close up of the tank coil secured with a messy application of epoxy.
Above — Side view. The actual front panel capacitor leads were this long to allow slack to put on the herring tin cover. The Herring Tin lid added much difficulty with temperature compensation and construction tactics -- but I got it done! The idea of the herring tin cover came from this blog post
Above —View for the VFO showing the DC input port ( an RCA connector ) plus the SMA RF output port. 2 bolts hold the tin to the copper clad board.
Above — My 50.0 MHz xtal based oscillator next to the Herring made VFO. Ready for 2 tone testing. The front panel tuning capacitor is front left. The front panel bolt just fills in a hole I drilled by mistake.
If I want to drive a DUT such as a high IP3 amp -- or say a diode ring mixer ( I rarely use them anymore), I'll chain up 1 of 3 separate, sealed up wide band amplifiers that range from 12 dB to 26 dB gain (up to 150 MHz or so). I also have a plethora of low-pass and band-pass filters in sealed Hammond cases that go from 5 MHz to microwave if needed.
[2] 6 dB HYBRID COMBINER
Above — The VHF targeted hybrid combiner is also a return loss bridge and vice versa. No experimenter bench should likely be without a return loss bridge or 3. I built with standard 1/4 watt 1% metal film resistors and tried several different coils as the transformer. After many versions, I settled with 3 stacked BN61-2402 ferrites with 4 total turns of lightly twisted wire. I twisted the wires only enough so they would stay together during winding. Because of only 4 turns, I was able to use 28 gauge wire. I measured 43 dB port isolation at 50 MHz.
Above — The applied transformer.
Above — Boxed up combiner/return loss bridge with a Mini-Circuits Lab 50 Ω SMA resistive load attached.
Above — Another view of the hybrid coupler
Above — My favourite design project of 2014: a wide band return loss bridge with directivity >= 30 dB from 5 MHz to 1.5 GHz. You may read more about it in the old site pops.net archive: Topics 2012 - 2014 : Caitlyn 310 — UHF Beginnings : 3. Return Loss Bridge Experiments : Bridge #4