Dear OM Peter, G3LDO,

I also had bad results with VXO drift both in single X-tal  and in mixed 
settings. I found out that the old big type X-tals in the range between 5 and 9 
MHZ are most suitable for use in VXO's. These x-tals give a tuning range well 
over a 100kC's where the small type X-tals will not tune over a range of more 
than 30-40kC's The coil in series with te X-tal can play tricks as far as 
stability is concerned. This can be observed by putting your finger near the 
inductor. I found out that when the VXO  does not return rapidly to its 
original frequency i had constructed a VFO instead. I had better results when 
two or more inductors (range at least 1:10 in uH) are put in series, this also 
limits the influence of paracitic capacitance in the inductors. An other test i 
found out is to warm one of the x-tals with a finger and observe the drift back 
to the original frequency when 'cooling off'. When the original frequency is 
reached i am sure the X-tals have the main influence on the osc
illating frequency.

An other point a observed is the value to be used for the capactors dividing 
the voltalge in the collpits should be carefully designed. To mutch 
amplification makes the oscilator easy to start with a rock-solid frequency. To 
little amplification will not allow the circuit to oscilate at all. Sometimes a 
extra buffer/amplifier stage is needed to make sure the mixer works better but 
this does not influence stability.

Good luck on LF and best 73's

Mike, PA3BSH
Hoogeveen, The Netherlands
e-mail also to <[email protected]>