TheRadioBoard

Some of you guys may remember my last vintage sig gen...the one with the neon relaxation oscillator for audio? Well, I got it working but it is, shall I say, its really a crappy unit and I have other plans for it.

Being a sucker for airplane dials I landed on this one real cheap.


This one features transformer-less AC powering using a 32 and 30 tube (I guess one is a rectifier?) with series resistance for the filaments (eek!) . Although there's lots of bands on the scale its basically a 100-200 kc oscillator, or so it seems. I haven't quite figured it out. There's no bandswitch, its simply calibrated for the harmonics. Thats what the verbage at the bottom of the dial explains. On the plus side its got a nice Hammarlund brass tuning cap and the output is adjustable.

I suspect it was a low budget kit, doesn't seem to be totally homebrewed. Its small and its cute

-Bill

Be nice to see it when you've 'incorporated' it into a project.

Haw, haw Its the older sig gen that is being converted to a 'project'.

Weeeell,

This is the old Supreme unit that isn't so wonderful.



There was a fellow on one of the other groups unloading a quantity of modern (working) Leader LSG-17 solid state signal generators at $7 a pop. I bought one with the intention of cobbling it into the vintage-style cabinet




We'll soon see how this graft works out


-ex

Here's the schematic for the new little sig gen. Yep, raw AC Anybody care to take a stab at explaining how it oscillates? Beats the heck out of me.

I don't know how well I can explain it, but the part I outlined in green looks like a tank circuit. I can see an RF signal being induced here. Problem might be it having a 60hz signal riding piggyback, but I suspect that the 27mH coil may be just right in this circuit to choke the 60 hz signal.

And the part I outlined in blue looks like a grid leak detector stage. The second grid is being feed by the other half of the 60Hz line voltage through a 14 mH coil, which is suspiciously almost exactly one half the value of the coil that is on the other half of the 60Hz line. If I understand what I just googled about 60Hz chokes, the proper value is going to depend on the rest of the circuit, but these two coils are in the ball park.

And by making the 2nd choke 1/2 the value of the first, it just might push what ever is left of the 60Hz signal out of phase to the point that it cancels out what ever little bit of 60Hz signal snuck thru the first coil.

So if the two coils are stripping both halves of the 60Hz signal, the signal that is left on the plate is going to be what ever the natural resonance of the tank is.

But I could be totally off base. This is about 10% educated guess and 90% wild guess.

But if you added an antenna and the part I outlined in red, it would look like a 3/4's completed 1 tube regen. LOL

_________________
Secret of Electronics - keep the smoke _in_ the wires.

I would guess that it is simply a tuned grid, untuned plate oscillator with the screen grid of the oscillator functioning as the plate. The screen is unbypassed is the clue. With the smaller choke in the screen lead, the "plate" or screen circuit would be tuned to a higher frequency than the control grid circuit, so the phase relationship of the feedback would be correct. I would bet that if you bypassed the screen grid to ground it would refuse to oscillate. That would prove what I am saying.

The raw 60 cycle power simply provides some modulation to the signal so you can spot the signal generator frequency out of all the other signals the set may be picking up.
Curt

_________________
Curt, N7AH
CW forever
Connoisseur of the cold 807

What would be the function of the 4Meg+500pf cap?

I don't know. I wonder if this schematic is for real? There is no coupling capacitor to the grid of the 30 either. The B+ on the plate of the 32 also shows up on the top end of the 20K Ohm pot that feeds the 30 grid. Sure is unorthodox, that's for sure.
Curt

_________________
Curt, N7AH
CW forever
Connoisseur of the cold 807

If Curt is right and the screen grid of the 32 is acting as a plate, is the 4Meg+500pf acting as a gridleak for the 30 tube?

There is another 500pf cap just like it on the output. And (other than a 20K resistor) it's the only thing between 60Hz and the output.

Can a 500pf cap block a 60Hz signal? That doesn't seem right to me.

But maybe I'm thinking about the 60Hz too much. This is a RF generator, so maybe as long as it is kept low the 60Hz signal is just being ignored.

I'm still trying to understand some of the basic circuits so I am probably just babbling.



I've seen schematics that are labled "simplified". Could that be the case here? I suppose the fastest easiest way to tell would be to see if the parts count looks right. Other than the tubes and a switch it looks like 15 parts.

_________________
Secret of Electronics - keep the smoke _in_ the wires.

I thought about the 4meg+500pf cap as the grid circuit of the 30 but it doesn't seem to wash.

This isn't a published schematic, btw. This is an "as-found" rendition. I have no clue to its accuracy and I'm not really fond of plugging things in until I understand what might happen

500pF @60Hz =5.3MOhm ?

I see the logic that you are using, golf. I see that 14mh inductor as a "screen dropper resistance" at 60 Hz.

The whole thing befuddles me having essentially two "B+" sources off the AC line and an equally (un)common return.

One of us is losing 2 decimal places. I come up with 500pf @ 60Hz = 54K ohms, but you are probably correct. At least your answer would make more sense than mine it this circuit.



I know that feeling.

_________________
Secret of Electronics - keep the smoke _in_ the wires.

Micro:
I use this a dozen times a day.

Wack it in your main (first) "Favourites" index - loads instantly. Exray has one on his site too.

enter 0.0005 mF (=500pF - please check me) and 60 in the hertz column.

http://ourworld.compuserve.com/homepage ... en/XLC.htm

PS: If you can do Excell spread sheets, its good to make some up with the more complicated equations in them. Takes the 'alzheimer' factor out of it.

Thanks, I just book marked it.

Yeah, I was using a pencil and a formula from one of my books.

Ohms = 1 / 2(pi)FC

I must of dropped a decimal somewhere. You end up with a string of zeros to keep track of when you have to convert pf into farads.

_________________
Secret of Electronics - keep the smoke _in_ the wires.