TX Experiments

[Silver Dollar]

Disregard that data above. I had 2 coils on my bench and used the wrong leads
after changing over to the pnp's. Now putting the 440 coil on it gives dismal results.
The 12.6 mh has too much resistance to get any power going without higher voltages.
Maybe make a doubler?

[Davor]

What about that 470 ohm in between the bases. It reduces circuit gain severely, and perhaps if tweaked correctly you'd get enough gain to supply coils with poor Q as well.

[Silver Dollar]

I tried changing that in the sim and it didn't seem to make much difference.
For some reason the sim doesn't give the same results as I'm seeing. With
the Deep Search coil (a Tesoro coil with CT RX ((using the RX as TX))) it
agreed perfectly to the sim with the other coils specs.

Deep Search is 6.1mh ct
440 Coil is 12.6mh ct

Not sure what is going on....

[Davor]

No idea. Did you measure the coil resistance and corrected it in sim?

[Silver Dollar]

Yes I'm using the correct resistance.
I remeasured everything today.

440 Coil
12.6 mh total
3.2mh 14.3 ohms per side
W 6.8n cap oscillates at 14.9khz

@7.35V in
3.1ma in
25.7v P/P out
16ma per side

@5.45
2.3ma in
18v P/P out
11ma per side

DeepSearch Coil
6.1mh total
1.2mh 8.5 ohms per side
with 18n cap oscillates at 14.5khz

@7.35v in
4.0ma in
25v P/P out
40 ma per side

@5.45v in
2.9ma in
17v P/P out
29ma per side out

I retested the Aziz oscillator which I had rejected due to excessive current draw
and found I was powering it wrong. It was drawing 5ma from the positive and negative
supplies but I think it was meant to float between the power rails and not run from +-V.
So I applied +9v and ground to the power pins letting the gnd find itself and Voila!

440 coil 9V in
1.5ma in
30v P/P
35ma per side in the coil!

Though there was a glitch in the current waveform not sure if it will be a problem.
Here's a picture from my current probe. It's 10ma per mv...


Also not sure about ground referencing on this design.

Here's the new drawing;

[Davor]

If the frequency gets severely dragged by bias or supply changes, yes it may be a problem. With a better Q you'd most probably miss it.