Ron Taylor of Cumbria Designs sent me an email this afternoon with a suggestion to cure the frequency shift on transmit of my MFJ Cub, which occasionally causes the X-Lock board to lose lock.
He wrote: " A further thought from looking at the MFJ Cub diagram; the JFET buffer that is used to couple the VFO into the transmit mixer might be pulling the VFO slightly on Tx.
The reasoning is that the J310 has something like 5pF gate to source capacitance. The impedance presented by the NE612 will change when powered up on Tx changing the J310 source loading. The J310 is coupled to the VFO with a 4.7pF capacitor which. Any changes to the J310 loading will have some effect on the VFO frequency via C35 the 4.7pF capacitor. If it pulls the VFO by more than 40Hz, the X-Lock will come out of lock and attempt to lock to the new frequency.
The reactance of the C35pF is very low at the VFO frequency but because of the J310 input impedance it could be made even lower with little or no effect on Tx output power. By reducing C35 the loading effect on the VFO will be reduced and hence the frequency excursions will be reduced. If you can reduce the frequency excursion below 40Hz, the VFO will remain in lock.
If convenient, it might be worth reducing C35 to 2.2F or even lower to 1.5pF to see if this improves matters."
Unfortunately I don't have any such small value capacitors in my parts box, and the smallest value stocked by the local supplier Maplin is 2.7pF, so I can't try Ron's suggestion right now or even in the near future. C35 is an SMT component on a tightly packed board, so any replacement would have to be tiny. But when I can get hold of a suitable part I'll certainly give it a try.
But kudos to Ron for going beyond the normal definition of product support and providing a solution to a flaw in the MFJ Cub design that would enable me to get the full benefit of the X-Lock. On the basis of my experience with this kit I can recommend Cumbria Designs unreservedly.