Radiophile valve transmitter

[Semir_DE]

That's an unusual looking technique, with the output transformer connected between the screen and anode of the EF80.

Hi Jeffrey,

thanks for putting in the google translation. I am quite surprised how well it worked. Apart from some terms like "NF" which is German for Audio Frequency (AF) google did a pretty good job.

The method I used was the method described by Joe Sousa using the 6AS6. That pentode was designed to be driven by both G1 and G3 to facilitate FM demodulation and other functions. I have another project using two 6AS6 as a switching stereo encoder. That project is built as follows:

- ECC82 #1 as audio amp for left an right channel
- ECC82 #2 for 38kc multiplex carrier generation giving two outputs with 0° and 180° phase to drive the two 6AS6
- 2x 6AS6 as multiplexers. 180° and 0° 38kc fed to G1, Audio L and R fed to G3
- PCH200 as a frequency divider /2 to generate the 19kc pilot (yes a single heptode can be used as a flip-flop)
- a third ECC82 to sum up stereo multiplex and pilot.
- a PCF82 as a free running FM modulator and output. This project can provide 250mW in the FM band.

Instead of ECC82 which have become quite expensive I used Russian double triodes 6H3п. Instead of the 6AS6 you can use a 6ж2п the Russian equivalent of the 6AS6 which can be found on Ebay for a reasonable price.

   

With the availability of many small FM TX modules from China certainly nobody really needs such an encoder but I had great fun building it   My main interest was to learn some new things e.g. I found out how I could use a heptode as a frequency divider and I wanted to find out if my idea of using two of the 6AS6 as linear switches would work -and it did.

Here are a few images of the project:

   

   

Below is the output signal with one channel input only with a 1kc audio signal. This is the typical stereo multiplex signal curve...

   

Please excuse the German naming of the image files, I copied them from the German forum where I had originally posted teh project.

If anybody here is into AM Stereo (Motorola CQUAM) here is another thread on RM.org (in English) about yet another project I had built...

https://www.radiomuseum.org/forum/am_ste...lator.html


Cheers

Semir

[Semir_DE]

Hi.
I like making little pantry transmitters and have done a few with Rod Pentodes. Would it be possible to post up the translation and some info on the other ones you've done, or PM the info to me please.

Hi Trevor,

I have several more pantry Tx projects. Please give me some time I have to dig up the circuit diagrams on my other PC. I will be happy to post them here or send you a PM.

Unfortunately I am going to be offline for a few days due to health issues so please be patient I will come back to you in a few days.

Cheers

Semir

[ppppenguin]

Semir, I was surprised at the good quality of the google translation. Not perfect, but easily good enough to understand.

All best wishes regarding your health issues.

[Kalee20]

Interesting approach, and I'm impressed by the linearity shown by the trapezium diagram!

Best wishes for your health, Semir.

[Murphyv310]

I also wish you good health Semir. No hurry for the info.

[Semir_DE]

Hello Everyone,

I just remembered that I had promised to post some more of my AM TX projects here so here goes:

both transmitters use the principle of a balanced modulator with the screen grid and anode as differential outputs. When I first came across this principle I was worried that the output transformer might be difficult to make but this turned out to be relatively uncritical as long as the carrier frequency and the modulating frequency are far apart. I tried using this principle in my stereo encoder project as well but here the carrier is only 38kHz and I could not get a clean modulated output from this circuit as the higher audio frequencies would also appear at the output of the transformer. For AM MW and LW this principle works quite well though and delivers a nice almost completely linear modulation trapezoid.

The output transformer is quite simple to make. I have used a type of litz that can be found on many SMPS power transformers. This litz consists of around 5-7 strands of wire which makes it ideal for winding this transformer. I start out with an empty T50-43 toroid core and put around 40 turns of this litz on it. Then I divide up the individual strands of wire to form the required windings e.g. if the litz consists of 7 strands I divide them up as 2+2 for the primary and 3 for the secondary windings respectively. You can also just use three wires and wind them on the T50 core in one go and then create the primary and secondary windings as needed of course...


One project is a transmitter that uses Russian tubes that can be found on Ebay for reasonable prices.

   

This project uses one part of a 6N3P double triode as an oscillator and the other part as an AF amplifier. A 6J2P pentode is used as the actual modulator. The 6N3P is a little like the ECC81 but has the filament connections on pins 1 and 9 instead of the usual 4 and 5. The 6J2P-EV is the Russian version of the 6AS6 and more or less a drop in replacement for it, so if you have a 6AS6 somewhere this will work just fine.


The other project I have built is using P-Valves. I had the opportunity to buy a bunch of those for a very fair price so I had to do something useful with them. The circuit posted here is similar to the one I had presented in Radiomuseum.org.

   

This version, however, has an additional output stage with a PCF82 and can deliver up to 250mW -so beware if you do not have a license.... I also have a version where I have used a PCL200 instead of the PCF82 that will output up to 1W. This design also has a power output and modulation level indicator using a PM84 which gives it a neat touch. Here is an image of the version I have built:

   


Apart from these two valve projects I also have built an AM TX sing a TBA520 PAL decoder chip as a balanced modulator and oscillator. Again this was being offered her from a surplus supplier at a cost of a mere 0.20€ so I thought this would be a worthwhile project to build. I still need to get teh data for that from my other PC and change the descriptions to English, so I will post it here in a few weeks when I get around to it.

Please let me know if you have any questions.

I would like to point out that I am posting these circuit diagrams for educational purposes only. By no means is my intention to motivate anybody here to build and run an unauthorized transmitter.

[radioman]

Excellent work Semir - thanks for sharing.
I have some questions about the oscillator : Did you get the crystal specially cut for this frequency or was it just one you already had ?
Also, are modern crystals OK for use in valve circuits like this or are they liable to be overdriven and consequently damaged ?

Cheers,
Andy

[Terry]

i would guess the latter, Andy, as 1580 is not evenly divisible by 9, so does not comply with the MW band plan.

1575kHz and 1584kHz are compliant channel frequencies and transmissions in either of these channels would cause an unwelcome beat. However, with many European countries closing down MW transmitters at an ever increasing rate there must be a fair number of gaps to accommodate such off-channel sources.

Long gone are the days when it was claimed that over 3,000 transmitters occupied 121 MW channels (120 since WARC 79).

Take a look at page 37 of this pdf: https://www.americanradiohistory.com/Arc...966-02.pdf

[ppppenguin]

Apart from these two valve projects I also have built an AM TX sing a TBA520 PAL decoder chip as a balanced modulator and oscillator.

The TBA520 can also be used as a PAL encoder. Labgear made a colour bar generator that used it.

[Semir_DE]

Did you get the crystal specially cut for this frequency or was it just one you already had ?

Also, are modern crystals OK for use in valve circuits like this or are they liable to be overdriven and consequently damaged ?

Hi Andy,

the crystals are in reality ceramic resonators I have a supplier here who had various frequencies in stock ranging from 190kc up to 1600kc that were close to MW band-plan frequencies. Fortunately these resonators can be pulled by about 3-4 kc and are never "spot on" when it comes to the frequency they are specified for, so what I do is I try out several until I find one that is close enough to a desired frequency and then pull it to exactly that frequency. At 20p a piece these are not expensive and work quite well.

I have successfully used this specific resonator rated at 1580kc as a 1584kc reference. I also have used an 800kc resonator for the 801 MW frequency. Once they have been set up to oscillate at a specific frequency they are quite stable but of course not as good as a crystal but far better than an LC tank.

On the subject of RF power handling of the crystals/resonators  I have never had any issues with these resonators, possibly also because I have a variable resistor that sets the amplification or feedback level of the oscillator thus preventing too high an amplitude at the resonator by limiting the RF power fed into it.

I had only one occasion when the high input power of a valve circuit became a problem. That was when I built my FM stereo encoder using valves. Here I used a very delicate 38kc crystal and this would start up and then break with a faint "pling" sound. I killed three of them until I found a solution. This was to place two anti-parallel 4148 diodes in front of the crystal. These diodes effectively limit the amplitude of the RF signal coming from the anode of the oscillator valve. Since adding the amplitude limiter the delicate 38kc crystals work without being destroyed.
I think I posted the circuit of my stereo encoder somewhere higher up in this thread, you can check teh circuit there.
cheers
Semir