EM on AM! My homebrew transmitter.
 

Reciently some threads here reminded me that I had some really small tubes tucked away, and got me thinking "what can I do with these?". Well I had been saving up parts to build one of the tube AM transmitters on phill's site ( I had all but the tubes specified). So I built a transmitter based on the 117L7 circuit on phill's site using the PUNY type 6418 I have in my collection of tubes.

I was able to amplify sound and RF from my signal generator, so I set out to take the osc.(or antenna) transformer and tuning cap I got from a pre-war philco I parted out, and make an RF oscilator.

I got it to oscilate this morning, and tuned it to an empty frequency. I then expermented with different ways to connect the portable cassette player I was using as my audio source in an attempt to get amplitude modulation.

I suceeded!:banana: Well sort of...the audio is pretty badly distorted, but I can tune it in on the reciever in the photo if I keep it within a foot or two of of the set up.

It ran well for about 1 hour before the modulation suddenly died. I will need to keep playing with the design to get less audio distortion, but for the time being am over-joyed that it even worked at all.

Here are some photos....
http://i1095.photobucket.com/albums/...r/DSCN1291.jpg
http://i1095.photobucket.com/albums/...r/DSCN1292.jpg
Don't let the disconnected chips fool you....the only active component is the tube in the top right.
http://i1095.photobucket.com/albums/...r/DSCN1293.jpg
http://i1095.photobucket.com/albums/...r/DSCN1294.jpg
http://i1095.photobucket.com/albums/...r/DSCN1295.jpg
http://i1095.photobucket.com/albums/...r/DSCN1296.jpg

If this the wrong section please feel free to move this topic to the right one.

Update: I found out why it died. One of the leads on the secondary of the osc. transformer uncliped it's self from it's terminal. I reconnected the lead and it is working as it did before.

When I come up with a modulation scheme that does not yield an objectionable amount of audio distortion I'll post a schematic.

The small low voltage tubes I'm using should allow for portable battery opperation (and a very small cabinet size), but I'm still probably going to make an AC line supply for home use after working out the bugs in the modulation scheme.

That's the way I'll rats-nest a test setup: got dozens of clip leads. Lots of fun!

If you modulate the oscillator you are bound to get FM as well as AM. The way to do it is to leave the oscillator to its main job of oscillating. Drive an amplifier with it and modulate the amplifier. You will get clean modulation if you do it right.

I actually killed noticable distortion by reducing the input source volume. The signal is not as loud as before, but sounds okay assisde from some buzz (likely from the power supply). I was overdriving the poor thing with the audio signal before.

The amplitude is not modulated in the way I'd have immagined it. Instead of max audio corresponding with max RF, max audio corresponds with min RF.

Reece, I gota agree with you that this is plenty fun.

Bob, I was considering using my other NOS subminiture tube in conjunction with the one i'm using as an osc. in a couple of different arrangements for better modulation. I first tried the easiest thing which was to try to imitate the single tube modulator scheme that the designs on phill's site used. And if I can make my single tube design work a bit better I may possibly keep it limited to one tube as I really want to keep my other subminiture unused as a display item.

Asside from cleaning up the wireing slightly it is still the same electronically.

I drew up a schematic. One mica is unlabled because I don't know how to read the color code on them, and did not have time to setup my cap checker. There is also one resistor that I don't have a measurement for the value of yet. Here is the circuit that I'm currently using.
http://i1095.photobucket.com/albums/...r/DSCN1297.jpg

It is VERY easy to kill the osc presently, and if the tuning cap gets turned so that the plates are more than 1/4 meshed together the amplitude drops from several volts to zero fairly quick. Also the tube and tuned circuit are so sensitive to hand capacitance that it is somewhat tricky to tune (you have to get it perfect and then over-shoot slightly).

That is basically the present state of things.

The audio input is reversed, ground going to hot lead. The antenna is detuning the resonant circuit and so any variation there will change the frequency. You should bypass the B+ supply. Then you can ground one side of the tuning capacitor.

Perhaps the cathode (one side of the heater) should be grounded rather than the grid? :scratch2:

jr

I believe that the audio works ether way, and just happened to be connected that way. I had such a rats nest going early on that I cooked a couple of pots (I had every element voltage except for the filiment supply on potentiometers hooked up as voltage dividers between B+ and ground) in the process of experimenting with a few things.

By "bypass the B+ supply" do you mean connect the B+ directly to the plate, and then hook one side of the transformer primary to the plate, the other to one side of the tuning cap and ground the other side of the tuning cap?

This is my first stab at tube design so given my limited knowledge of the design end I'm pleased that I got the thing to oscilate in the first place....Let alone got recievable, listenable modulation.

The Reason the unmarked cap has an "x" is that it came with the coil and the osc. doesn't seem to be significantly affected by it's presence or absence (by absense I mean one lead cut). Just thought I'd mention it if it makes a difference to anyone.

I would recommend plate modulating or even using the second grid rather than using the first grid. That would reduce FMing and also give you much cleaner audio. You should only drive the tube at 1/4 of it's maximum output power, that way your modulation will swing your carrier both negative and positive with your audio which is the proper way to run AM.
If you run the tube to full power and then modulate it then there is nowhere the audio can push the carrier into the positive territory, thus resulting in only dropping carrier power with modulation peaks.
A brief brush up on AM transmitting basics will help you out on this.
Also beware of over-modulation as that will cause bad distortion.

On monday night I reworked the audio slightly so the audio ground was correct, and redrew the schematic. This is the working update.
http://i1095.photobucket.com/albums/...r/DSCN1298.jpg

This design has always had bad capacitive pickup from hands, rediculous lead lengths connecting the tube to the osc. cap and coil, and refused to oscilate if not connected to ground through my scope (which has always had it's ground connected to the cathode and probe connected to the plate).

And so after doing the audio rework I pulled the parts for the osc. and modulation, and experimented again with osc. design using shorter lead lengths. My interpretation of bob's statement "bypass the B+ supply" did not work, but I came up with something that does work, and has killed the hand capacitance effect on the tuning cap and reduced it on the osc coil. It can now oscilate with the scope probe disconnected form the scope, but not with the probe disconnected from the circuit. Modulation is about the same, and is likely going to be experimented with more in the future. I intend to test the capcitance,resistance, and inductance of my probe in soon to see what parts will mimic it's effect on the circuit.

this is the present design...
http://i1095.photobucket.com/albums/...r/DSCN1299.jpg

I will try to implement modulation of that sort. I did initially try second grid modulation on the first design, but it killed the osc. if I rember right. I will try it again with the new osc. design I'm running.

I want to keep this design limited to one tube if I can, and this is just so I have something good to listen to on my old tube radios.

I don't have one bit of Tube or RF design knowledge. My forte is trouble shooting and repair, and being a sophmore EE student my only design knowledge relates to OP AMPs and is not very detailed (especially in the RF frequency range). I was thinking before you posted that the 0 modulation RF power should be around 50% so the output is high with modulation of a fairly high percent.

I've used my non-equational knowledge of tube circuit design to get this far, and plan to keep playing with my design in hopes of improving it. This design came about from a "get enough monkeys with enough typewriters together and you get the work Shakesphere" type of scenario. I'm basically comeing up with things to try, testing them, and accumulating/refining/building upon the stuff that works or seems to work.

You might be able to build a decent AM transmitter using a 12CS6 or less so 12BE6. (the 12CS6 is more linear than the 12BE6).

http://www.wa2ise.com/amtxmit.jpg
http://www.wa2ise.com/radios/amxmit.html If you have enough input audio amplitude you can take away the triode audio driver.

Or solid state:http://www.wa2ise.com/amxmit.gif Here I used an open collector gate to chop the analog audio waveform, making AM.

Avioding hum can be troublesome, using small chokes like 100mH on the audio lines and power supply lines, as well as bypassing the rectifier diodes in the power supply (like 0.01uF) can help. The hum can come from the RF finding a stronger path thru the diodes when they are conducting 60Hz than when they are not conducting, thus making a hum buzz that looks like AM modulation be seen by the receiving radios. The chokes are to break this RF path.

Thanks for that circuit. I may build that exact design in the future, and will likely try to emulate the design with the components I have at my disposal.

My goal in using hearing aid sub miniture tubes is to build a VERY small transmitter that can be run on batteries. It has also become a good learning experience for me.

It will likely be around a week before I can work on it again.

Looks like you guys would have no trouble fixing my radio L-660 !