oscillation with subminiature filimentary tube

9/21/2007 11:58:56 AMThomas Dermody

Hello all. Is there some trick to getting a filamentary tube to oscillate? I know that people have trouble with them all the time in battery sets. I am trying to build a superregenerative FM set for my electronics class, and am using all subminiature tubes.

I built a push-pull amplifier, which, after much fooling with the biasing, works quite well. There are four beam pentodes running in push-pull parallel in the output, and two triodes in the input. All runs on 45-50 volts. It actually starts clipping if I use higher voltages. That might be due to a lack of shielding.

As for the superregenerative FM detector, I built it like I have in the past using a 6C4 or other hi-mu triode. The circuit I prefer requires a choke on the cathode, or else the tube won't oscillate. Well, in order to do that with a filament type tube, I put a choke on each side of the filament. I wired the rest of the circuit as I normally would. First I used a triode, and then I switched to a pentode. The screen grid was, of course, fed with a 1 meg resistor, and a .1 MFD capacitor for filtering. Both tubes made a hissing sound, though I think that they are simply behaving as high gain amplifiers, and not as oscillators. I could get the thing to tune in my signal generator, but only with the lead close, and with the generator turned all the way up. With the device operating on its own (no generator hooked up), I could barely hear a couple of FM stations. The circuit is tuning to the FM band, so I know that is okay. It just won't oscillate.

I then changed the circuit to a different one that requires the cathode to be grounded. That particular circuit has the grid and the plate wired to the antenna coil, and has an isolation choke where the grid side of the coil connects to B+ (the grid is connected through a 33 pF capacitor and a 5 meg grid leak). The circuit still wouldn't work. For both circuits I increased the B voltage to 90 volts. The hissing got louder, but the circuit never broke into oscillation (it's usually quite obvious when the circuit starts to oscillate).

A long time ago I tried building one of these using a #30 tube, which is probably a very inappropriate tube for the job. However, I'm starting to wonder if filament type tubes are just poor oscillators.

9/21/2007 1:04:39 PMEdd

Waaaay back in the very early days after I had tried and optimized about everything that I could do with a
crystal set, this was my first step on upward into the tube domain with this superegenerative receiver
article, build as shown...and it was the rage of all who saw it, as working radios just were NOT built that small
in that day.
I was so overly impressed with its performance that I then got away from the batteries and shifted to AC operation and started using a 12AT7 in its place for the det and later using the spare triode for an AF preamp to feed into a 6AQ5 AF output , driving a room filling speaker, vice the weak little phones. (That initial circuit had an RFC choke subbed in place of the phones with the audio being taken away via an AF coupling capacitor.)

Since I was now wanting to move up the freq spectrum from the BC band to the ham bands and the police-iuseses, highway petroleum, firefighters, border patrol, sheriff and ambulance frequencies to see what interesting and covert things that they were up to .

That initial design advanced to a low noise 6BQ7, to then be used for the RF preamp stage and the detector. Eventually , 1 more 12AU7 was incorporated as an amp in taking off the characteristic superregenerative HFaudio "hiss" as was being acquired thru a hf pass filter. That amplified noise then got detected and DC was fed onto the AF preamp stage to bias off that stage. When a signal was received, the audio popped right on through. Bingo...you then had squelch action ....on a minumal parts count, high performing receiver. The other alternative was to have shelled out BIG bux for a full blown dual conversion comm receiver with MYRIADS of tubes being used.

I had found the initial article at that time in Science and Mechanix magazine:

73's de Edd

:Hello all. Is there some trick to getting a filamentary tube to oscillate? I know that people have trouble with them all the time in battery sets. I am trying to build a superregenerative FM set for my electronics class, and am using all subminiature tubes.
:
:I built a push-pull amplifier, which, after much fooling with the biasing, works quite well. There are four beam pentodes running in push-pull parallel in the output, and two triodes in the input. All runs on 45-50 volts. It actually starts clipping if I use higher voltages. That might be due to a lack of shielding.
:
:As for the superregenerative FM detector, I built it like I have in the past using a 6C4 or other hi-mu triode. The circuit I prefer requires a choke on the cathode, or else the tube won't oscillate. Well, in order to do that with a filament type tube, I put a choke on each side of the filament. I wired the rest of the circuit as I normally would. First I used a triode, and then I switched to a pentode. The screen grid was, of course, fed with a 1 meg resistor, and a .1 MFD capacitor for filtering. Both tubes made a hissing sound, though I think that they are simply behaving as high gain amplifiers, and not as oscillators. I could get the thing to tune in my signal generator, but only with the lead close, and with the generator turned all the way up. With the device operating on its own (no generator hooked up), I could barely hear a couple of FM stations. The circuit is tuning to the FM band, so I know that is okay. It just won't oscillate.
:
:I then changed the circuit to a different one that requires the cathode to be grounded. That particular circuit has the grid and the plate wired to the antenna coil, and has an isolation choke where the grid side of the coil connects to B+ (the grid is connected through a 33 pF capacitor and a 5 meg grid leak). The circuit still wouldn't work. For both circuits I increased the B voltage to 90 volts. The hissing got louder, but the circuit never broke into oscillation (it's usually quite obvious when the circuit starts to oscillate).
:
:A long time ago I tried building one of these using a #30 tube, which is probably a very inappropriate tube for the job. However, I'm starting to wonder if filament type tubes are just poor oscillators.
:
:T.

9/21/2007 1:56:49 PMThomas Dermody

Interesting. The circuit you show is almost identical to the one I'm using, though my coil is wound for FM frequencies. However, my design is going off of one that was designed to be run at 180 volts. It has a resistor in both the plate and the screen grid circuit. I should try a choke in the plate circuit instead, and a direct connection to B+ with the screen grid. This will allow me to maximize my 90 volts. A choke is probably more efficient than a resistor, because it only resists when there is a change in voltage, so it will resist for audio and such, but will otherwise feed the tube full 90 volts, allowing it to have enough power to oscillate. ....Hmmm...we'll see. I wonder if a 1T4 performs better than these Raytheon CK6088s that I have. If I have this much trouble with these for an FM radio, I can only imagine what kind of trouble I'd have with them with a television. At some point in time I want to build a really small tube type television. It would be contained in a box no larger than 5 X 5 inches. The CRT would either be one of those 1 inch green CRTs, or I'm thinking about making one from an old C-9 base flood lamp bulb.

:
:
:
:
:
:
:Waaaay back in the very early days after I had tried and optimized about everything that I could do with a
:crystal set, this was my first step on upward into the tube domain with this superegenerative receiver
:article, build as shown...and it was the rage of all who saw it, as working radios just were NOT built that small
:in that day.
: I was so overly impressed with its performance that I then got away from the batteries and shifted to AC operation and started using a 12AT7 in its place for the det and later using the spare triode for an AF preamp to feed into a 6AQ5 AF output , driving a room filling speaker, vice the weak little phones. (That initial circuit had an RFC choke subbed in place of the phones with the audio being taken away via an AF coupling capacitor.)
:
:Since I was now wanting to move up the freq spectrum from the BC band to the ham bands and the police-iuseses, highway petroleum, firefighters, border patrol, sheriff and ambulance frequencies to see what interesting and covert things that they were up to .
:
:That initial design advanced to a low noise 6BQ7, to then be used for the RF preamp stage and the detector. Eventually , 1 more 12AU7 was incorporated as an amp in taking off the characteristic superregenerative HFaudio "hiss" as was being acquired thru a hf pass filter. That amplified noise then got detected and DC was fed onto the AF preamp stage to bias off that stage. When a signal was received, the audio popped right on through. Bingo...you then had squelch action ....on a minumal parts count, high performing receiver. The other alternative was to have shelled out BIG bux for a full blown dual conversion comm receiver with MYRIADS of tubes being used.
:
:
:I had found the initial article at that time in Science and Mechanix magazine:
:
:
:
:
:
:
:
:73's de Edd
:
:
:
:
:
:
:
::Hello all. Is there some trick to getting a filamentary tube to oscillate? I know that people have trouble with them all the time in battery sets. I am trying to build a superregenerative FM set for my electronics class, and am using all subminiature tubes.
::
::I built a push-pull amplifier, which, after much fooling with the biasing, works quite well. There are four beam pentodes running in push-pull parallel in the output, and two triodes in the input. All runs on 45-50 volts. It actually starts clipping if I use higher voltages. That might be due to a lack of shielding.
::
::As for the superregenerative FM detector, I built it like I have in the past using a 6C4 or other hi-mu triode. The circuit I prefer requires a choke on the cathode, or else the tube won't oscillate. Well, in order to do that with a filament type tube, I put a choke on each side of the filament. I wired the rest of the circuit as I normally would. First I used a triode, and then I switched to a pentode. The screen grid was, of course, fed with a 1 meg resistor, and a .1 MFD capacitor for filtering. Both tubes made a hissing sound, though I think that they are simply behaving as high gain amplifiers, and not as oscillators. I could get the thing to tune in my signal generator, but only with the lead close, and with the generator turned all the way up. With the device operating on its own (no generator hooked up), I could barely hear a couple of FM stations. The circuit is tuning to the FM band, so I know that is okay. It just won't oscillate.
::
::I then changed the circuit to a different one that requires the cathode to be grounded. That particular circuit has the grid and the plate wired to the antenna coil, and has an isolation choke where the grid side of the coil connects to B+ (the grid is connected through a 33 pF capacitor and a 5 meg grid leak). The circuit still wouldn't work. For both circuits I increased the B voltage to 90 volts. The hissing got louder, but the circuit never broke into oscillation (it's usually quite obvious when the circuit starts to oscillate).
::
::A long time ago I tried building one of these using a #30 tube, which is probably a very inappropriate tube for the job. However, I'm starting to wonder if filament type tubes are just poor oscillators.
::
::T.

9/21/2007 2:11:23 PMThomas Dermody

Same result as before. I should try building the circuit with a 6C4 again, and then see if it will work at those voltages. A 6C4 kind of defeats the purrpose, though, because I want the set to run on a few batteries. As the set is, I can run it for hours and hours with one lithium ion cell for the A battery, and two 45 volt batteries for the B supply. There's also one lithium ion for the C bias, but that'll last forever no matter what you use. I guess a 6C4 should run for a fair amount of time on 4 lithium ion cells. They run a long time. I purchased 4 of them and a charger for a digital camera. They're quite economical because they last a long time and can be recharged over and over again.

9/21/2007 3:53:52 PMThomas Dermody

Wow! I got it to work finally. I realized that the circuit you show is actually running at 30 volts, not 90. I decided to see if I could get my circuit to run at 45 volts. I tried this and that, and finally I got it to bring in some stations a little louder than just barely. I decided that whatever audio I wind up with I'm just going to make due with what I can get, so I built a pentode preamplifier after the detector. Suddenly the thing broke into oscillation, and with certain components I was able to pick up channel 4's audio very loud and clear. It was wonderful. Trying to do the same at FM broadcast frequencies is a bit difficult, though. The thing likes to make all sorts of noises. I don't know how on earth I'm going to attach a tuner to it. If I alter the wiring even slightly, the thing goes out of oscillation. Probably the biggest problem is that I built it on one of those quickie breadboards, where wires are just inserted into holes. There is no shielding. I have everything as spread out as possible so that stages don't interfere with eachother, but it really is poor. I'm about ready to scrap the project because it was only supposed to be something fun, not something permanent. I have been following Abraham Lincoln's (or Thomas Edison's) philosophy, but I feel that I'm wasting valuable time on this.

On a side note, even with 8 tubes, it's amazing how long you can run subminiature tubes on batteries. The batteries last forever. The tubes don't even get warm. I can use 1/8 watt resistors throughout. I even use them for the filaments if I need to (a couple of tubes are actually desgined to run at .75 volts...for the audio I just wired them in series, but to use one requires a resistor).

9/22/2007 9:32:15 PMBrian

Ever try a subminiature with a cathode? Just curious. I have a bunch and wondered about wiring them up.

:Wow! I got it to work finally. I realized that the circuit you show is actually running at 30 volts, not 90. I decided to see if I could get my circuit to run at 45 volts. I tried this and that, and finally I got it to bring in some stations a little louder than just barely. I decided that whatever audio I wind up with I'm just going to make due with what I can get, so I built a pentode preamplifier after the detector. Suddenly the thing broke into oscillation, and with certain components I was able to pick up channel 4's audio very loud and clear. It was wonderful. Trying to do the same at FM broadcast frequencies is a bit difficult, though. The thing likes to make all sorts of noises. I don't know how on earth I'm going to attach a tuner to it. If I alter the wiring even slightly, the thing goes out of oscillation. Probably the biggest problem is that I built it on one of those quickie breadboards, where wires are just inserted into holes. There is no shielding. I have everything as spread out as possible so that stages don't interfere with eachother, but it really is poor. I'm about ready to scrap the project because it was only supposed to be something fun, not something permanent. I have been following Abraham Lincoln's (or Thomas Edison's) philosophy, but I feel that I'm wasting valuable time on this.
:
:On a side note, even with 8 tubes, it's amazing how long you can run subminiature tubes on batteries. The batteries last forever. The tubes don't even get warm. I can use 1/8 watt resistors throughout. I even use them for the filaments if I need to (a couple of tubes are actually desgined to run at .75 volts...for the audio I just wired them in series, but to use one requires a resistor).
:
:T.

9/22/2007 9:59:54 PMNorm Leal

Brian

I find sub miniature tubes just about the same as larger equilavents. Some seem better as they were disigned later. 6111 is similar to 6SN7. 6112, 6SL7.

Norm

:Ever try a subminiature with a cathode? Just curious. I have a bunch and wondered about wiring them up.
:
::Wow! I got it to work finally. I realized that the circuit you show is actually running at 30 volts, not 90. I decided to see if I could get my circuit to run at 45 volts. I tried this and that, and finally I got it to bring in some stations a little louder than just barely. I decided that whatever audio I wind up with I'm just going to make due with what I can get, so I built a pentode preamplifier after the detector. Suddenly the thing broke into oscillation, and with certain components I was able to pick up channel 4's audio very loud and clear. It was wonderful. Trying to do the same at FM broadcast frequencies is a bit difficult, though. The thing likes to make all sorts of noises. I don't know how on earth I'm going to attach a tuner to it. If I alter the wiring even slightly, the thing goes out of oscillation. Probably the biggest problem is that I built it on one of those quickie breadboards, where wires are just inserted into holes. There is no shielding. I have everything as spread out as possible so that stages don't interfere with eachother, but it really is poor. I'm about ready to scrap the project because it was only supposed to be something fun, not something permanent. I have been following Abraham Lincoln's (or Thomas Edison's) philosophy, but I feel that I'm wasting valuable time on this.
::
::On a side note, even with 8 tubes, it's amazing how long you can run subminiature tubes on batteries. The batteries last forever. The tubes don't even get warm. I can use 1/8 watt resistors throughout. I even use them for the filaments if I need to (a couple of tubes are actually desgined to run at .75 volts...for the audio I just wired them in series, but to use one requires a resistor).
::
::T.

9/23/2007 1:41:29 AMThomas Dermody

A subminiature tube with a cathode will perform like any other tube with a cathode, except, like Norm said, they are often improved. They might not necessarily be able to handle as much power as a larger tube, but they certainly have less internal capacitance, which makes hitting the higher frequencies a bit easier. Filamentary tubes in general are pains in the rear to work with. They are trickier to bias. If strung in parallel, voltage dividers must be arranged in the B circuit, since all cathodes will be at the same potential, and not all tubes require the same bias. C batteries can be used. If wired in series, the filament string can be taken advantage of to some degree, to acquire negative or positive bias for grids. These tubes are extremely picky with bias, too. Being off by just 1/2 a volt can make a huge difference in performance. Check a Zenith portable for example. Check a 5-G-500 or a 6-G-601. Look, for instance, at how the volume control is grounded. It has a resistor to each side of the 1LH4 filament, which affects how the AVC circuit performs. Place the AVC at B- and the circuit distorts. Notice where the 1st IF secondary is placed on the 1LN5 filament. Move it to the other side of the filament and the circuit won't even work. A 1LN5 is, of course, a sharp cut-off tube, so it's extra sensitive to bias. This can take advantage of the relatively small AVC that's generated in these radios.