Hi.
After the great success of the 1ZH42A pantry transmitter I made some years back I thought I'd try a 1ZH37B in a similar circuit.
As usual I have gone for crystal control for the oscillator.
The HT is slightly high at 72v. the filaments are run from a 1.2v Ni-Mh cell and fed to the filament via a 3 ohm resistor. The Crystal is fed between the G2 & one of the G1's.
This valve is much more happy with higher audio drive to the other grid but it needs a little more negative bias to stop clipping of the audio, the bias also needs to applied to the opposite grid, a 3v CR2025 is used for the bias via a suitable potential divider.
A much shorter radiating antenna is now suitable. A proper Pi network will follow to filter out harmonics, there is no FM to the carrier and quality is excellent.
The idea is to make this mains operated in a suitable project box once the initial development work is complete.
Here is a video of the results so far. Note I refer to 1ZH24A instead of 1ZH42A, a human error on my part.
https://youtu.be/yh6w7eEiW_E
Very good project. I must have a go with these valves at some point?
Adrian
Another nice video! I love to see your minimalist circuits Trevor. It will be good to see the end result.
Tracy
Looks good, Trevor!
Absence of FM is certainly going to be assured by using a crystal rather than an LC tank circuit.
Looking forward to the circuit diagram!
Hi.
I've been testing various ways of coupling this circuit to an external antenna. the original lash up had no tuning and produced harmonics. The third harmonic was hard to pick up and there after I couldn't detect it with my SDR Play.
I've attached the circuit so far, It is virtually the same as the the old one with a 1ZH42A. On this setup though I tried a few ideas one is as marked as "A" in the circuit, it is a transistor radio ferrite rod, coil and a tuning cap, this just about fills the shed on its own when tuned. The low impedance winding can be grounded on one side and the other to a 6 foot "long" wire and this easily fills the shed and gives reasonable results in the garden up to the house. I also tried a TOKO oscillator coil wired the same as "A" and it works well with a little more range but it is sharper to tune and the length of the antenna needs retuning if the length is altered. "B" is a Pi tank circuit and this gives the greatest range. This also filters out harmonics as does "A" with 72v HT its actually a bit more than Pantry, reducing the HT tames the output and less drive is required, T1 is a valve output TX in reverse, the low impedance side goes to the headphone socket output from an MP3, DAB radio etc, it can easily overdrive the transmitter.
Personally I dont really think the harmonics are an issue without tuning, the level is low and there is more hash from SMPS, wall warts and Powerline adaptors than a small pantry TX. I did a comparison with my Minimod and I feel it is noisy in comparison.
Tomorrow I'll do the tests with the scope looking at the waveforms and post up the results. Next will be to build it properly and into a box.
PS Tuning caps used were 5-150pf I also decoupled the HT with 100nf
Hi.
I've done another video of the pantry transmitter and did a comparison with the Minimod.
Unfortunately the Minimod produces a lot of generated noise internally, most likely the LM386 and its settings for high gain, it overloads easily and has less range using the same antenna. Audio quality is a bit lower in comparison to the rod pentode pantry tx.
I can upload it if there is interest.
Hi.
Here are some test photos off the scope.
The frequencies used were 100hz, 1khz and 10khz. Response is pretty flat from 60hz to 8khz, slight tail off at 10khz and below 40hz.
The trapezium was at 1khz and with a 1566khz carrier. Also shown is the waveform at 100% modulation and 150% modulation.
It's interesting to note that the oscillator runs with a crystal marked at 42.5 Mhz with excellent sound on a 405 line telly!!
Good stuff!
The trapezium does show some non-linearity, but it's not huge. I'll bet that an RF coupling to a diode detector, to generate a reconstituted AF signal which could be used for negative feedback, would clean it up considerably!
Hi.
NFB is an idea. I did try this on the EH90 tx I did before but the improvement was marginal. A conventional TX I made with anode and screen modulation was no better than the results in this one.
Here is a link to the tests I did a bit later on today.
https://youtu.be/7YtsSWoiWps
There is a lot online re negative feedback. There are advantages and disadvantages with it. It does help with non linearity but not with anything more than slight.
It was way back in the 20s it was added to valve transmitter to cure the "slight`non linearity that all valve AM transmitters suffer from. It is virtually always used in broadcast AM transmitters but not in Amateur Radio as its is frequency dependent so different parameters are required.
One disadvantage is if the TX is over modulated the feedback can swing positive causing it to go into an overmodulated squeal, only shutting it down will stop this.
So there are advantages and disadvantages to NFB. I will give it a go but TBH this sounds extremely good and any improvement will be marginal.