Great good news: the thing said its few words and musical notes. I made a rudimentary audio amplifier with a 6BF11 using the pentode section as preamp and a 220/12V as output transformer and could listen the few firsts sounds. Very distorted because the bw seems too low but it is promising. I feel very happy.
Attachments
This page on the math of impedance matching might be enough: https://www.electrical4u.com/impedance-matching/
But a calculator based on (part) of that math is much simpler: https://www.eeweb.com/tools/pi-match/
Keep in mind that apart from matching at the required frequency, a pi filter also is low pass.
But a calculator based on (part) of that math is much simpler: https://www.eeweb.com/tools/pi-match/
Keep in mind that apart from matching at the required frequency, a pi filter also is low pass.
This is an approximate schematic of the entire tuner. Some values may syill be changed. The vast majority of the caps are silver mica except the film units in the cathode follower stage and limiter; and the plate units at cathode decoupling and heaters. Resistor 1/2 watt except two.
If somebody wants the schematic in expresspcb format, it is available; but sincerely I unknow if libraries are also needed.
There is also an undocumented PSU which consists in a 220V transformer to 440V center tapped, an EZ81 rectifier and a 6FD7 wired as series regulator to give about 140DCV well filtered from ripple and an CLC filter between them.
Cx1, Cx2 and Cx3 are the ganged variable tuning capacitor whose exact value is unknown.
If somebody wants the schematic in expresspcb format, it is available; but sincerely I unknow if libraries are also needed.
There is also an undocumented PSU which consists in a 220V transformer to 440V center tapped, an EZ81 rectifier and a 6FD7 wired as series regulator to give about 140DCV well filtered from ripple and an CLC filter between them.
Cx1, Cx2 and Cx3 are the ganged variable tuning capacitor whose exact value is unknown.
Attachments
The Foster Seeley discriminator uses 2 coils like in the other IFTs, probably closer because bandwidth must be higher. The hot side of the primary is coupled to the center of the secondary coil with a capacitor, and the center is connected via an RF choke to the center of the output RC (load) network: see https://www.fmuser.net/wap/content/?6846.html
Bandwidth of the discriminator is determined by coupling primary - secondary, and the load network (lower R ---> higher bandwidth).
The 470K are tentative. I saw there values from 47K to 100K. The use of capacitive coupling from the primary or the use of a terriary winding are the same from the performance point of view, but the cap occupies some space in the shielding cans that I haven't. And the risk with silvered mica cap that a short circuit thanks to silver migration destroy the 6BV8 diodes.
As you can readily deduce, l have the knowledge but not the practice, thus l am learning on the walk.
As you can readily deduce, l have the knowledge but not the practice, thus l am learning on the walk.
This is a general layout of the tuner. Some important items are explained inside it. The general way of the signal is U shaped, say, input is close to output what in general isn't good. But as you can see, the thing grows up a bit chaotically as new stages were added. Obviously more care will be taken in the final artwork.
Attachments
I'm not a mod, but chances are it is related to the EXIF header that was added to the JPEG standard: it features an orientation field and is supported by some programs, but not by others. There is a free program called jhead that can fix these issues.
Last edited:
Again: tertiary is because in the shielding cans (square 14mm side) there isn't enough place to 3 or more silver mica capacitors. As I have lots of winding wires, it is more easy for me to wind a small coil than to add a cap inside the can. Also, some mica capacitors have silver migration that shorts the cap. I saw several of them inside K-Trans IF transformers with embedded capacitors in the plastic base.
Regarding 470K resistors, this value seemed interesting until l found commercial circuits in the Beitman manuals at worldrahistory.com where I found that this value is too large. Seeley in his patent say values for load resistor of between 500K and 1Mohm; look below.
Attachments
Last edited:
Indeed, because otherwise someone could get a patent on a modification: https://patents.google.com/patent/US4254378A/en