Tuesday, February 12, 2013

Speaky VLF/LF band module

Besides tying some loose ends on the Speaky I'm also having fun building/testing  a VLF/LF band module for this radio.

Speaky only has 80 to 10m (excluding 12m) band modules offered in kit form, so why not improving things a little bit.
All band modules share the same design, only changing tuned parts and oscillator values.


Band module basic schematic:
The PLL part diagram:




...and schematic:


Now the fun part....

If you put an 4Mhz crystal in the VXO (T1) you will have a nice tone at the IF frequency (8Mhz), that is the oscillator harmonic. ... so I placed a 4.194Mhz one but that way I would only cover 194 to 694Khz (500Khz band span) reception, VFO would cover from 8.194 to 8.694Mhz... no problem for now, it would serve...

...even more fun...:
With the 4.194Mhz crystal (or any other for that matter in the 4-4.5 range) due to some leakage on IC1 (a not perfect mixer)  the PLL loop comparator would get the 4.194 from the VXO, the mix of 4.194 with the LO oscillator and the DDS generated one, that way he would compare tree frequencies all in the 4-4.5Mhz pass band instead of the normal 2 needed for the PLL to work correctly.
That gave me some interesting results like sudden frequency jumps until I realized what the problem was, I know I should have foreseen this in first place but you know how it's like when you start soldering first and thinking last...

As a result of this, the only way to include sub 500Khz in the Speaky would be by down-conversion or even up-conversion to another band. With the PLL values and mixing scheme it's simply not possible.
Without conversion to another band it's still possible but without using the internal PLL; making a fixed frequency oscillator or a VFO controlled by the bandpass control voltage (T2), anyhow, loosing the PLL control.
I might go that way...




My experimental band module:




I used 1N4007 rectifier diodes as PIN diodes substitutes and they seen to work nice, also tested 1N4148 without significant changes in signal for the other modules sharing the same signal bus.
Initialy the main oscillator wasn't working and I traced to a faulty BF981 so I used an BF961 instead.
T1 was a 2n3904 instead of the original BF199

A side by side comparation with the 40m module:




Have fun!

No comments: