Golborne Vintage Radio

Full Version: Joined the club, Rod pentode superhet radio.
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Pages: 1 2 3
This has kept me occupied for several days, probably over a week, making IF coils and Osc coil, winding the ferrite aerial, making up a spreadsheet so I could work out the values for the Osc to track the aerial tuning, or best I can anyway.  All part of the learning experience.

Looking at the board, it is a bit of a Frankenstein, with trimmer caps strapped to allow for adjustment, wires bridging sections etc.

[attachment=21507][attachment=21508]

But it works and has been a very satisfying little project.

6 x 9 Volt battery for HT 4.4 mA HT current.
3 x 1J18B valves, 1 for local Osc, mixer and RF amp.  1 for IF stage and 1 for driving a speaker, 1 Ge diode for detector.

if anyone knows of a Ki-cad library that includes the rod valves I could draw out a circuit.

Inspiration came from http://www.hi-ho.ne.jp/ux-45/russian.html   now if only I could get things that small?

Adrian
Didnt know what "rod " valves were until now. I have quite a few of them, but with 6 volt heaters. There are some superb twin triodes available, some single triodes, and lots of different pentodes.
I also have the correct silicon rubber sleeving that was used, and some heatsinks for the power pentodes.

Nice little radio Adrian, and layout is nice!!. With Ki-Cad just draw up your own valves by editing an existing valve drawing. Rename it after you modify it ( both for the overlay AND the library ) .

Joe
Rod valves or Rod Pentodes are literally using metal rods rather than the conventional wire grids/screens information can be found here :-
https://www.radiomuseum.org/forum/summar...todes.html

supposedly very robust when it comes to being fired out of big guns when used as proximity fuses etc. But probably not designed for very long life.

I also have a few ECC70/6021 5840 valves that I have used for projects in the passed, these are typical 6.3 Volt heater valves ,Twin triode and Pentode respectively both with conventional grids.

The gm on rod valves seems to be quite low when compared to normal grid valves, but are fun to play with. Will have to get some smaller hearing aid type valves and play with them soon.

Thanks for the comment.

Adrian
Hi Adrian.
I read your post on UKVRRR and the video. In one word..... Brilliant, glad you are now a fully fledged member of the Russian Rod Pentode Club, the RRRC.
These valves are FUN and quite amazing.
I've made up projects using them for RIAA equalisation and pre amp, FM super regen radio, many TRFs, line scan oscillator and output and a Band 1 405 line video modulator, very versatile and I've even had operation over 100mhz. Other projects have included a transmitter for 40 meters AM at 15 watts and pantry transmitters with only 5 components.
Wonderful little valves, buy them now before all disappear due to the crisis in Ukraine.
I have a small stock of them Trevor, mainly 1J24B and 1J29B as well as a few remaining 1J18B's. If doing more radio circuits like this I need a way of holding small formers and counting turns, how the guy on the Japanese web site managed to re-wind the even smaller Toko style coils is beyond me.
He also used a lot of button cells and that is something I was considering, well, I have several packs of old out of date PR48 batteries I would like to try.

Adrian
Well done Adrian.

I have not made a complete superhet with them although I have some modules that I have made. I have a circuit using a 1j37b as a mixer which I have tested. It is a hybrid with transistor output driving a ceramic filter though so not purely valve. Apart from the selectivity it got me round the transformer issue a bit. To get size down a little, and for better stability, I have been experimenting with these tubes (and traditional ones to some extent) with manhattan style layout and surface mount components. Size wise I find I get little difference as I can pack them using conventional tequniques but it should be far better for the higher frequencies.

Tracy
(14-07-2022, 09:22 PM)joebog1 Wrote: [ -> ]Didnt know what "rod " valves were until now. I have quite a few of them, but with 6 volt heaters. There are some superb twin triodes available, some single triodes, and lots of different pentodes.
I also have the correct silicon rubber sleeving that was used, and some heatsinks for the power pentodes.

Nice little radio Adrian, and layout is nice!!. With Ki-Cad just draw up your own valves by editing an existing valve drawing. Rename it after you modify it ( both for the overlay AND the library ) .

Joe

All rod valves are Pentode-like and all have 1.2V filaments. Some have two filaments for two beams in parallel.

They are actually a kind of elongated electron gun with rods as focus elements. A plate on either side of the filament and beam provides the control grid action. The Screen grid (g2) and Suppressor grids are actually an electrostatic lens, so in fact the g1, g2, and g3 are not "real" grids but pairs or plates or rods on either side of an electron beam. Some higher power versions are actually two electron sources in parallel, so the two middle "g1" plates are one plate.
The dual control grid types (1j37b, 1j42a etc) are not dual grids at all, but the two plates either side of the filament are not shorted together, so a differential voltage on g1a or g1b will steer the beam. This means that EITHER polarity of differential voltage reduces the gain and maximum gain is when there is no differential signal.

The g2 has most effect on beam focus, so at 80V HT on anode, that pair of rods on each beam has an optimal voltage for maximum gain, thus on a 1j24b, about 20V is very low gain and 55V is about maximum. But it varies with HT.

The structure is doubled, so all rod tubes have two sets of anodes, g3 rods and g2 rods. This is because the larger width g1 plates either side of the filament result in two beams.

There are very many USA, European and Russian sub-miniature tubes/valves. Some are the same shape and size as rod tubes and some are flatter. They are diodes, triodes, pentodes, triode-pentodes etc and work the same way as regular valves (wire coil grids) and are not rod tubes.

The beam-tetrodes that are like pentodes replace the g3 wire grid with beam focussing plates either side of the flow to the anode.
Some heptode (or maybe octode) valves actually don't have a regular wire grid for g2, but rods. On the Heptode/Octode mixer-oscillators the g2 is basically used as an anode for the oscillator, so you can use an adaptor base and wire a pentode-triode or hexode-triode into an heptode (or octode) socket and vice versa.

You can triodise a pentode by connecting g2, g3 and anode together. But a Rod pentode is using the rods to focus, so to use it like a triode only g2 is connected to anode. The g3 has to remain near 0V (usually f- or f+). Because a Rod tube g3 is really part of an electrostatic focus lens and not a grid, then you can vary the gain a bit by varying g3 by about + or - 5V.

There can be no remote cut off/ variable mu because on a regular valve that's achieved by varying the spacing on the turns of the control grid. Instead, for AGC, you need to vary g2 driven by and AGC amp (which could be the magic eye deflection pins or anode of a DM70 (DM71).

The filament has to be very thin to be a point source electron beam for the g1 plates either side. So the tubes are very low power. The filaments are spring loaded for longer life and if power is derated on output types and filament is 0.8V to 1.25V, then life exceeds the regular B7G 1.4V valves.
The short life was only at excessive power for radio transmission (primitive radar) for a proximity fuze of a shell or bomb as operational use would be very short! The designs used in the regular military and aerospace gave long life:
MiG fighters
Satellites inc Sputnik (used older types).
Vehicle military radio
Man-pack military radio.
One model of man-pack for forestry workers known
No consumer applications known.

LT was usually rechargeable cells, NiCd or Silver based. HT was often via pnp germanium transistor push-pull inverter, sometimes batteries.

The 1j17b is an early type replaced soon by 1j18b, it was made obsolete by the 1j24b.
The 1j29b and 1p24b replaced earlier power types. The 1p24b is the least efficient of later types as it was for short life proximity fuses, MiG and vehicle sets. A 1j29b is quite high power and a pair in push pull can use a 110-0-110 centre tap mains transformer for maybe 2 watts class B. A more negative than 9V g1 bias needed. A pair of the low power 1j24b with a centre tap mini-mains transformer will give maybe 500mW class b.

The 1j37b was the most used one for mixers and AGC. You connect a centre tap transformer to g1a and g1b. The two RF signals feed the centre tap (common mode) and primary (differential). It acts simply as an RF rectifier or frequency doubler is fed matching anti-phase to the g1a and g1b.

The 1j42a was one of the last, optimised for about 9V HT.

Any of the dual grid models behave the same as the rest if g1a and g1b are shorted

The g2 and anode can act like a current mirror, so gain varies horribly if g2 is fed by a single resistor from HT, so in the Russian designs they had 3 options:
An HT tap on battery (or secondary HT on invertor)
A pair of resistors as a voltage divider (some regular B7G battery sets use 1T4/DF91 g2 on a pot for manual RF or IF gain).
Direct connection to HT (if about 55V on 1j24, or 80V on 1j29b)

A pair of 1j24 can replace an Octode, Heptode, triode-pentode or triode hexode DIRECTLY (assuming filament voltage & curreent is sorted with series & parallel resistors). You wire one as the triode (or g1 & g2 oscillator) by short g2 and anode, and g3 to f-. The other one is wired as a pentode, but original Screen Grid connection is ignored, the g1 is RF in and g2 connects direct to anode of the one as an oscillator. This works as DK9x, 1L6 etc, no change to set.

The capacitance of the electrodes is much lower than all the regular Dx9x battery types, so a small cap and trimmer from anode to f- will mean the radio IF or oscillator doesn't need retuned. The LO leakage is lower than an Octode or Heptode (which are dreadful) because RF at g1 and LO at g2.

The 1j24b can thus replace all battery valves except the output. Just add an 1N60 for any of the DAF9x or similar.
You only need 1/2 the filament of 1j29b to replace any DL9x valve. The 1j24b can replace the DL91(=1S4) or the DL92 (=3S4) in 1.4V LT sets if HT is 45 to 67V, but the 1j29b is a better choice.
Thanks for the superb, in depth explanation Mike.
I seriously had never heard of these valves untill this post, which is also on UKVRRR forum.
So, after a lifetime ( im almost 70 ) of playing with valves I learned something again.
The small wire ended types I have are all of US decscent I guess, 
because they are just minaturised versions of conventional valves. ( or nearly so )

In reading this post, I have understood that rod valves are of russian only make ?. is that correct ?.
I "suppose " that being in Australia has sheilded me from russian stuff. 
I have bought capacitors and the like from Ukraine that was all ex mil russian.
Big Bulky, usually pretty ugly, but test superb and seem to last without leakage.
For instance, .1% capacitors, which if made in the west would be at LEAST 20 or 30 $ each cost a few cents.
Soldered together tin plate cans with large ugly solder terminals on mini ceramic supports, and bolt down tags
to secure said capacitor to the chassis. They are about 1 1/2" long, 1" wide and 3/8" thick.
The same solver mica capacitor of western design and manufacture is about a 1/2" square,
perhaps 1/16" thick and dipped in some for of epoxy, and costs about $10 each.


Once again, thanks for the instruction.

Joe
Yes, the Rod types are Russian Military only. The idea was mooted generally in the 1930s and as I wrote some octode/heptodes and all Pentode characteristic Beam Tetrodes (or Kinkless Tetrodes, hence KT66 etc) use electrostatic lens principles. Development & production was from early 1950s to the early 1980s. Production only ended with the collapse of the USSR. Soviet factories, especially for military production, had almost automatic production quotas. So in 1991 there was excess stock of millions, hence various sellers in Ukraine, Bulgaria, Russia and some other ex-Soviet influence countries.

I have quite a few NOS and ex-equipment Western sub-miniature types. Also developed for proximity fuzes, but earlier. There is even a pentode-triode for mixer-oscillator. Quickly used in handheld two way radios, some larger PRC series US military, hearing aids (some with 0.7V filament for a two valve amp, 1.4V LT and 22.5V HT) also some US and quite a few Japan pocket radios that usually had a regular B7G battery valve for speaker unless a earphone only model. Using 45V or 67V layer pack like an elongated 6F23/PP3/9V pack.

The last Western miniaturisation was the nuvister, used in some USA mains FM radios. Really only Europe did domestic portable FM battery valve sets. None at all in USA except one novelty set and the only USSR battery valve VHF portable were for military or space use. The one Forestry model is maybe just HF, not sure. Could be band 1.
Can I ask if the 1J37b valves would work around the 88 to 108 MHz band or higher, I have ordered a few and would like to use them as rf input and mixer if possible down to 10.7 MHz IF. I am guessing they would be better than just a normal 1J18b or 1J24b for this?

It could be a while before they arrive so at this stage just planning.

Adrian
Pages: 1 2 3