[i]"I'm still thinking about what to do with the wire to the aerial. I had been considering using shielded cable with the shield earthed. Obviously this increases the capacitance of the antenna but at least this capacitance should be fixed and I think it should still be small in comparison to the capacitor in the tuned circuit. Time to break out the calculators, perhaps" - Kitwolf[/i]
Not a good idea to use screened cable for antennas! - I went down this route briefly, some years back.. Even using RF Co-ax with low C is a bad idea (IMHO).
Before the EQ inductor, the wire is not as sensitive to capacitance change - but even for this, co-ax causes problems.. After the EQ inductor, even 0.1pF is extremely significant... Worse, the capacitance of the co-ax cable is quite severely affected by temperature.. Nothing dignificant for normal use of the cable, but HIGHLY significant if it forms part of the antenna circuit which is 'looking' for tiny capacitance changes.
If you use thin wire (I tend to use single strand 35 SWG enammelled copper wire or sometimes thinner) you should not have any problems.. Ideally, route the wire on plastic stand-off pillars of some sort.. I often have grounded aluminium screening on my boxes, and fixing the position of the wire makes problems less likely.
The EPE design has some imperfections.. It is based on the S/C design which did not have an equalizing coil.. Adding an EQ coil was the right move, but (alas) all that was done was to simply add this coil to the existing circuit..
If one analyses the Moog EW design closely, one finds that the circuit and component values are carefully selected so that the oscillator is correctly loaded by the Equalizing circuit .. If you carefully analyse the EPE design, one can see that no care has been taken at all, and the loading of the "tacked on" EQ is mismatched (and not easily corrected).. Matters are made much worse by the appalling choice of EQ inductor.
The EPE oscillator can 'latch up' close to resonance, and setting the optimum tuning is a nightmare. Some units are ok, but a 'bad one' could drive you crazy .. "bad" is, I think, down to component tollerances - particularly the EQ inductor.
Some good replacement inductors can be found at Digi-Key.. I use the 2.5mH versions, usually 3 in series (4 if you have a thin / short antenna) but you can reduce cost by using 5mH or 10mH inductors.. You usually need something between 5mH and 10mH.
I have found that these Inductors (http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=M10164-ND) are well suited to antenna Equalization, having a reasonably low capacitance (particularly if one uses several in series to get the required inductance).
I think, perhaps, that I was not clear in my last..
One wants to run the (reletively) INSENSITIVE wire to the equalizing coil, and have this coil as close as possible to the antenna (ideally, no connecting 'wire' from this coil to the antenna - on my 16 H1 units at HO-2010, I had the EQ inductors bolted directly to the bottom of the antennas.. Internal wiring was a shambles, with loudspeaker and power / ground wires flapping about close to the circuit board.. But not having any highly sensitive antenna wires anywhere allowed the units to operate without problem).. There is no 'insensitve' antenna for the volume circuit, so I advise positioning the board close to the volume antenna, so that you can keep this wire short.
[i]"small in comparison to the capacitor in the tuned circuit"[/i]
There are 2 seperate "tuned circuit"'s .. You are looking at the "tank" capacitance (560pF I think) - This, with the "tank" inductance (all part of T1) form the oscillator.. And yes, a few pF change here is not too significant.. (a change of 1pF will change the frequency by about 400Hz - and one can compensate for this by adjusting T2) But most of the pitch change is not the result of these components
