30-07-2020, 10:36 AM
(This post was last modified: 30-07-2020, 10:40 AM by Mike Watterson.)
As well as adding about 6pF to 9pF on the anode going to RF, you might need 2pf to 3pF on the RF grid, the pentode mixer. It's a crazy simple circuit and handy that no external g2/screen grid supply is needed for either valve. The g1 of the Heptodes tends to be low anyway and g1 of the Rod tubes has the highest capacitance being plates very near the cathode filament. I think I'll make up a socket with 6MHz to 15 MHz and 15 MHz to 30MHz oscillator for a Heptode/Octode, an IFT and a tuned RF in. Also a FET buffer to drive a frequency counter from the g2.
Then I can see exactly the difference between the DK92, 1L6, DK96 and my Triode-Pentode in terms of:
1) Pin capacitances. I'll add a trimmer each on the LO, RF and IF with a scale in pF.
2) Mixer conversion gain
3) LO leakage on RF in
4) FM of the LO by a strong signal outside the IF pass band. May need an FM detector on the FET buffer or my Spectrum analyser.
5) Maybe add a low noise IC amp on the IF to amplify noise. The Rod Tubes are MUCH lower noise than any Pentode, Hexode, Heptode, Octode or Nonode, in theory. Joe Sousa has an article on it.
The simple Triode-Pentode mixer using two 1j24b Rod Tubes isn't testable on a regular valve tester; at least maybe it's possible if you add a series resistor to the triodised anode that also feeds Pentode g2.
I have some NOS 1V6 https://www.radiomuseum.org/tubes/tube_1v6.html
The 1V6 is a Triode - Pentode and should also work in any Heptode circuit.
Read Joe Sousa's interesting article on the 1V6, which is no ordinary Triode - Pentode.
Another idea is a Noval socket for an ECH81/ECH83 and have 2 x 1j24b on a Noval base with series filament and 4.1V dropper resistor,
I've found too that 10 Ohms is more ideal than the higher value series resistors, giving 1.25V from 1.39V cells.
I also have some NOS 2G21 valves; the only radio I find using it are 1945 and 1948 Belmont Boulevard, an early truly pocket radio.
https://www.radiomuseum.org/r/belmont_bo...5p113.html
It can't have had much battery life!
https://www.radiomuseum.org/tubes/tube_2g21.html
It's a Triode - Heptode! It will be interesting to see how it performs.
The Hallicrafters S72 does actually use the 1R5 (DK91) as a mixer only and 1U4 as the Local Oscillator, no doubt because it gives better strong signal performance and also works to a higher frequency. The 1U4 is maybe better than the 1T4 for higher frequencies. The original Transoceanics used non-B7G mixer/osc (some maybe 1LC6) till the 1L6 came out. European sets used the DK40 instead of the DK91 for better shortwave till the DK92 came out.
Then I can see exactly the difference between the DK92, 1L6, DK96 and my Triode-Pentode in terms of:
1) Pin capacitances. I'll add a trimmer each on the LO, RF and IF with a scale in pF.
2) Mixer conversion gain
3) LO leakage on RF in
4) FM of the LO by a strong signal outside the IF pass band. May need an FM detector on the FET buffer or my Spectrum analyser.
5) Maybe add a low noise IC amp on the IF to amplify noise. The Rod Tubes are MUCH lower noise than any Pentode, Hexode, Heptode, Octode or Nonode, in theory. Joe Sousa has an article on it.
The simple Triode-Pentode mixer using two 1j24b Rod Tubes isn't testable on a regular valve tester; at least maybe it's possible if you add a series resistor to the triodised anode that also feeds Pentode g2.
I have some NOS 1V6 https://www.radiomuseum.org/tubes/tube_1v6.html
The 1V6 is a Triode - Pentode and should also work in any Heptode circuit.
Read Joe Sousa's interesting article on the 1V6, which is no ordinary Triode - Pentode.
Another idea is a Noval socket for an ECH81/ECH83 and have 2 x 1j24b on a Noval base with series filament and 4.1V dropper resistor,
I've found too that 10 Ohms is more ideal than the higher value series resistors, giving 1.25V from 1.39V cells.
I also have some NOS 2G21 valves; the only radio I find using it are 1945 and 1948 Belmont Boulevard, an early truly pocket radio.
https://www.radiomuseum.org/r/belmont_bo...5p113.html
It can't have had much battery life!
https://www.radiomuseum.org/tubes/tube_2g21.html
It's a Triode - Heptode! It will be interesting to see how it performs.
The Hallicrafters S72 does actually use the 1R5 (DK91) as a mixer only and 1U4 as the Local Oscillator, no doubt because it gives better strong signal performance and also works to a higher frequency. The 1U4 is maybe better than the 1T4 for higher frequencies. The original Transoceanics used non-B7G mixer/osc (some maybe 1LC6) till the 1L6 came out. European sets used the DK40 instead of the DK91 for better shortwave till the DK92 came out.
