3 Vacuum Tube Superhet Design

[Daniel Bingamon]

This is something I like to do, I'm 61 yo and mostly self-taught in electronics.  I still like to work with tube circuit designs.
Anyway IF transformers for such circuits are hard to come by so I drawn this up and I'm working on a PC board as well.
This design will use the LTM455EW resonator, just one stage so I don't have to worry about matching two resonators.
I know that the resistors in the IF will have to be adjusted to get voltages right.   I'm open to any comments on this:



And Circuit board:

[TimFox]

R2 = R4 = R3 = 470k seem much too large.
Is 470k merely a placeholder?
Consider the plate voltage on V4.

[schmitt trigger]

Just curious…Why are you connecting T2’s primary to T1’s secondary? Wouldn’t it be better to connect it to the 120 VAC line directly?

Otherwise congratulations on your retro circuit.

[Xena E]

Good project!
Regarding the resonator. That may be problematic as they do have sporadic resonances away from their design frequencies. That is why they are normally operated in conjuction with traditional transformers...

One way to incorporate a traditional tuned transformer in the circuit is to make them! A short cut is to use unshielded axial chokes that couple nicely when placed in parallel and the coupling can be adjusted by varying this spacing, the centre frequencies can be trimmed by use of parallel ceramic variable capacitor/fixed capacitors.

They can be incorporated into a style of construction that mimics that used in the original vintage circuits.

Good luck.
Xena E

Edit: I think this subject has been discussed as a source of replacement components in vintage tube radios on the Antique radio forum, may be worth looking there for inspiration too...

[rsjsouza]

Interesting circuit; I haven't seen the resonator used with tubes before, thus I can't comment on that.

In addition to the other comments I have a few more:
- Capacitors of the "5" series (C2, C4, C5, C6, C8) are usually hard to find these days. Perhaps consider the much more accessible "4.7" values?
- Z2 is 1A, which is typically quite excessive for +B. A slow blow fuse rated for less current would not be stressed by the inrush current during power on, and the protection would be more effective.
 

[szoftveres]

I've once tried this concept of "building" a coupled IF transformer from stock 330uH inductors, 330pf fixed caps and 10-60pf trimmer caps from DigiKey, it worked out quite well:
https://www.antiqueradios.com/forums/viewtopic.php?f=12&t=320861



You can also use values stated in the above forum topic (1200uH, 68pf + trimmer) for higher impedance.

I'd recommend trying to build an IF transformer instead of the resonator; the resonator being a fairly low impedance device is really badly matched to the high-impedance tube circuit and consequently you'll lose a lot of gain and will hardly get anything at the output.

[fant]

At power ON, on the input side of the resonator will be applied a voltage pulse that will destroy it; better couple it through a transformer.
Mandi

[schmitt trigger]

Now that you bring the power up sequence, a similar fate could happen to the germanium diode.

[RoV]

Besides comments on transformers and too high plate resistors, I would add that the cathode of D1 needs to be DC-coupled to ground with a suitable resistor (e.g. 100 k), otherwise the diode would desensitize itself by self-bias; besides, it wouldn't be able to provide AGC voltage on the anode.

[Daniel Bingamon]

Good point, being on the secondary of the main transformer is kind of redundant and creates additional load for it.

[Daniel Bingamon]

R2 = R4 = R3 = 470k seem much too large.
Is 470k merely a placeholder?
Consider the plate voltage on V4.

Yes, sort of.

[TimFox]

R2 = R4 = R3 = 470k seem much too large.
Is 470k merely a placeholder?
Consider the plate voltage on V4.

Yes, sort of.

If you need a high source impedance out of the pentagrid, you might use a simple L-C tuned circuit in place of the plate resistor, capacitively coupled to the resonator.

[Daniel Bingamon]

I think I might just use your discrete components IF circuit.  I'm going to build it right on the PCB board and then put solder slots for a shield with a hole for the trimmer.

I've once tried this concept of "building" a coupled IF transformer from stock 330uH inductors, 330pf fixed caps and 10-60pf trimmer caps from DigiKey, it worked out quite well:
https://www.antiqueradios.com/forums/viewtopic.php?f=12&t=320861



You can also use values stated in the above forum topic (1200uH, 68pf + trimmer) for higher impedance.

I'd recommend trying to build an IF transformer instead of the resonator; the resonator being a fairly low impedance device is really badly matched to the high-impedance tube circuit and consequently you'll lose a lot of gain and will hardly get anything at the output.

[Daniel Bingamon]

I've dropped the Resonator and will add the LC circuit on the PC board with provisions for a solder-on shield.
Might as well make it a two stage IF at this point.
Provision for adding coupling caps if needed.
LED's are being added as power indicators.

Here is the revised schematic with some of the improvements made per various suggestions:

[szoftveres]

One more thing - triode as IF amplifier with tuned circuits on both the input and output will not work. Use a pentode, like 6BA6.

[TimFox]

Important point:  without neutralization, that triode is bound to oscillate.
The 6BA6/12BA6 allows gain control (manual or automatic):  if you don't want AGC, you could use a 6AU6/12AU6 sharp cut-off pentode.

[Daniel Bingamon]

Thanks everyone for their suggestions.  The 12BZ7 has been replaced with a 12BE6 and 12BA6.  As a result it is now a four tube design.
Making a good reference to WA2ISE Hot Rodding AA5s.
Being only 3 miles by air from a 50KW AM radio station it's probably a good idea to have well working AGC.

[james_s]

Looks like a cool project and I don't want to just sound critical but the PCB layout looks like it could use a good tidying up. I generally try to keep all traces a 0, 45 and 90 degree angles as it looks a lot more neat and clean. All the weird random angles kind of triggers my OCD. Looks like it wouldn't be too hard to rearrange a few things such that almost all of the traces could be on one side of the board leaving most of the other side available as a ground plane. That should improve performance, I have an educational kit AM radio on a single sided PCB and putting my hand near the board alters the tuned frequency. The old tube radios I have work remarkably well for the rat's nest construction style they use but they do have a grounded metal chassis.

[Daniel Bingamon]

I might just change that front end 12BE6 to a 12BE7 based on comments from this site:  https://www.wa2ise.com/radios/12ba7.html.  Any thoughts?

[Daniel Bingamon]

At this point I'm going to redo the whole PCB.

Looks like a cool project and I don't want to just sound critical but the PCB layout looks like it could use a good tidying up. I generally try to keep all traces a 0, 45 and 90 degree angles as it looks a lot more neat and clean. All the weird random angles kind of triggers my OCD.

[james_s]

Check the clearance of the mounting holes too, that has bit me more times than I care to admit, screw heads are larger than they seem like they'll be and it's easy to forget that they're usually made of metal so clearance to high voltages is something to pay attention to. Consider some screw holes near the tube sockets if practical too, it can take quite a bit of force to insert and remove tubes so it helps if the board is well supported. Other than that it looks pretty good.

[TimFox]

I might just change that front end 12BE6 to a 12BE7 based on comments from this site:  https://www.wa2ise.com/radios/12ba7.html.  Any thoughts?

The 6BA7/12BA7/6SB7Y pentagrids have a higher conversion transconductance than the 6BE6/6SA7 tubes, but is otherwise very similar.

[Daniel Bingamon]

Yes, I read that the 12BA7 is 950 Micro Mhos transconductance and that is almost twice as much as the 12BE6.

[TimFox]

At V_bb = 100 V , 6BE6 is 455 uS, 6BA7 is 900 uS.
The metal-octal 6SB7Y was introduced for FM service, after the 6SA7 was introduced:  the 6BA7 and 6BE6 are "miniature" versions of their octal predecessors.

[Daniel Bingamon]

Hey James,

Looks like a cool project and I don't want to just sound critical but the PCB layout looks like it could use a good tidying up. I generally try to keep all traces a 0, 45 and 90 degree angles as it looks a lot more neat and clean. All the weird random angles kind of triggers my OCD. Looks like it wouldn't be too hard to rearrange a few things such that almost all of the traces could be on one side of the board leaving most of the other side available as a ground plane. That should improve performance, I have an educational kit AM radio on a single sided PCB and putting my hand near the board alters the tuned frequency. The old tube radios I have work remarkably well for the rat's nest construction style they use but they do have a grounded metal chassis.

Is this any better?