EMRFD Message Archive 14608

Message Date From Subject
14608 2018-04-07 15:53:20 itsbeen29years Butler VXO from EMRFD for Intermod Testing of RX


 I am building the Oscillator from EMRFD, page 7.19 Fig. 7.32.  I have reviewed the corrections from the W7ZOI EMRFD Errata site.

 Designing for 7 MHz, I plan to adjust by using the formula Fold/Fnew and multiply (i.e. by 2) all the required LC components.  This works for the Pi filter, fine enough.

My question: 

(1) how do I modify the Transformer that is in the Oscillator Collector circuit?  I see that for 14 MHz, the inductance is actually giving a 44 Ohm impedance; would I simply calculate the primary inductance of the transformer, to try and obtain the same impedance, but for 7 MHz? 

(2) I haven't found much info on Butler oscillator design,  and how its LC values are set, and cannot find a good site on the web to refer to.  Can I use the same 220 pF and 440 pF capacitors?  Butler VXOs seem to be often used for overtones of the crystal.   

(3) why are there 2 x 0.1 microFarad bypass capacitors on the Base of the Final transistor?   


14609 2018-04-07 16:57:59 w7zoi Re: Butler VXO from EMRFD for Intermod Testing of RX
Hi Nigel,

That oscillator has been a real work horse for me.  I first built it (actually, the matched pair)  in the late 1970s and have used them for IMD measurements ever since.   They still work! Many of the IMD results quoted in EMRFD were done with those circuits.

The transformer in the oscillator collector is just part of a tuned circuit.   The exact value is not critical.   The 13 turns has an inductance of 0.5 uH, which resonates at 14 MHz with 255 pF.     The differential amplifier is a means to extract a signal with minimal loading.  The load on the tuned circuit is provided by the 100 Ohm R between bases of the diff amp.   If the coupling were perfect, the 13 to 4  turn winding would reflect a load resistance to the collector of 100(13/4)^2, or just over 1K.   This would load the transformer to a Q of 24.   That's a bit higher than I have used in VHF designs.   A more typical value might be 10 or 15.  You want enough Q here to be sure to select the right overtone, but nothing more is needed.  The real oscillator Q is provided by the crystal.    So yes, you can probably add turns to the transformer to get the L needed for 7 MHz and then just stick with the 220 and 470 pF caps.

This is not a VXO.   That is, there is no means to tune the crystal frequency, even a little.  You could do that with a trimmer in series with the crystal.   The 90-400 pF trimmer is tuned to resonate the tank at 14 MHz. 

A good way to adjust this circuit would be to add turns to the transformer while using about the same capacitors.    A 90 to 400 pF trimmer might be hard to find, so use a common one with lower C.   One of the Sprague 65 pF jobs would be fine.   Put a resistor in the circuit in place of the crystal.   Then trim the coil to get it to 7 MHz.    This is described in the oscillator chapter.  

The output amplifier is a common base stage, used because it has good reverse isolation.  There is just one base bypass, a .01 uF.   There is another .01 in series with 47 that comes from the 6 dB pad.    There are two 0.1 uF in parallel to form the bypass for the output transformer for this amplifier.   The use of two in parallel is a way to enhance the wide band performance.   See the discussion beginning on emrfd page 2.28.

There is a 3 pole low pass filter in the output of the diff amp.  That circuit will have to be changed to 7 MHz.   

Good luck with your experiments.

73, Wes