Going to give the "lil" "Cigar box receiver" in Harry's site a whirl.

Hi Ivan, Harry and all... Thanks for your confirmation of the circuit. I thought that I had found the original 1955 schematic but alas... The May 1955 issue seems to be missing from the archives. The transmitter is there but no rcvr. As far as the present circuit goes it will be interesting to test the difference between the two hookups. It appears that the only part that keeps the RF out of the audio stage is the 100 pf to ground. The project is a real challenge to layout and build with the ganged tuning caps both course and fine tuning with REAL world availible parts. Varactor tuning is sounding better all the time hi hi. 73's, Doug..

KM6OR wrote:Hi gang. While looking over a little more carefully the schematic, sourcing parts and setting up the bread board I noticed what appears to be an error in the circuit as drawn. It looks like the schematic is the original from 1955. It makes me wonder if I am correct (ha ha) maybe Harry might chime in. The audio pick off point from the detector to the first audio stage runs from the plate of the detector to the grid of the audio. I was thinking the audio take off point should be up one part to the junction of the RF and LF chokes. This way the audio is RF filtered by the rf choke. The only way the audio stage is not overloaded with rf is because of the 100 pf bypass to ground. Apparently the receiver works as shown. I would appreciate if others including Harry might chime in on this. Regards, Doug.

As Ivan pointed out, it is more logical to take the AF out from the junction of the two coils. I will check on the original article.

When I built receiver this I did make a few changes to the original article and I thought I had adopted them all in the final circuit. I cannot remember if the AF out tapping was modified. Unfortunately I lost the finished receiver (and the transmitter) after a particularly bad divorce in 1990. I basically left the house with one car-load of stuff and I had to prioritize a little (a lot, actually).

The V2 is a self-oscillating mixer with RF feedback between grid and filament/cathode, so it should not really affect operation noticeably. But it is only a question of moving one wire. When you build it I would be grateful for any comments. Please let me know if that makes any noticeable difference.

I do remember that the final project was re-assembled on an aluminium panel that fitted into the box. This was because a used a strong cardboard box, and without the added mechanical stability there were slight frequency changes when you pressed on the box. But if you use a wooden box then this should not be a problem.

Interestingly I have another similar cigar box that I have been coddling for about the last 25 years with the intention of re-building this project. It was one of those fun projects I wish I had salvaged when I moved. I am sure my ex most probably junked loads of stuff, including these two cigar box projects. If she had moved out instead then I am sure I would have junked loads of her stuff, including the 9 cats  

So please let me know how you get on.

I am working on my all-valve guitar amplifier at the moment and I am about to complete the last of the wiring. The next projects will likely be some more tube projects, unless the winter arrives, in which case it will be back to antennas.

Very best regards from Harry - SM0VPO

HI,
you are right, taking audio from the point between the two chokes makes more sense.
The LF choke may be a tricky part nowadays. Using a resistor only instead of the parralel combination would reduce the receiver gain - but how much? It is worth testing.

VBR from Ivan

Hi gang. While looking over a little more carefully the schematic, sourcing parts and setting up the bread board I noticed what appears to be an error in the circuit as drawn. It looks like the schematic is the original from 1955. It makes me wonder if I am correct (ha ha) maybe Harry might chime in. The audio pick off point from the detector to the first audio stage runs from the plate of the detector to the grid of the audio. I was thinking the audio take off point should be up one part to the junction of the RF and LF chokes. This way the audio is RF filtered by the rf choke. The only way the audio stage is not overloaded with rf is because of the 100 pf bypass to ground. Apparently the receiver works as shown. I would appreciate if others including Harry might chime in on this. Regards, Doug.

Hi Glenn, Ivan,Harry and all! Thats what I love about forums with a bunch of builders that have "been there, done that" commenting. Lots of ideas, keep it up! It makes a great list of ideas with different guys chiming in. To tell you the truth, I had not thought that far ahead about varactors. I dont mind the mechanical complication of dual gang capacitors and they are traditional(hi hi) but I am already going to build it with a I.C driven transistor power supply running off two lithium cells (thats the idea anyway) to get away with 9 volt transistor radio batts series connected for the B+ and lithium ion cells would run the filaments for a long time. I want to keep this particular regen battery tubes. I think I might do some experimenting, breadboarding to see if a fixed tuned preamp might be enough as I want there mainly for isolation and gain control ahead of the detector. This would keep it a one gang tuning arrangement. Anybody out there have experiance with a regen with untuned rf amps? I see quite a few circuits of these in old radio experimenter books. Again, thanks for the replys and comments/ suggestions. 73's, Doug.

Hi Glenn,
vatactor diodes in a tube receiver? OH NO!!!   

BR fron Ivan

Ivan wrote:
...
1- The need for two ganged variable capacitors.
...
VBR from Ivan OK1SIP

Hi Ivan

n-ganged varactor diodes could be used - or is that not vintage enough :-) :

1SV149 Toshiba ca.= BB112 ca.= KV1236

BB112 17-560 PF 12V. 8V 20pF; 1V 500pF
KV1236 26-490PF
BB212 22-560pF (= 2*BB112)
BB130 ca. 15-500pF 1-28V = KV1225, KV1226
1SV102 ca.= BB130 high Q

( https://sm0vpo.forumotion.com/t484-sw-receiver#2933 )

CIGAR BOX RECEIVER:
http://sm0vpo.altervista.org/rx/cigarrx.htm

-

This might be interesting:

From:
https://sm0vpo.forumotion.com/t484-sw-receiver#2933

See paper page 26:

Frequency compensated LC networks for oscillators with the wide tuning range:
http://www.kearman.com/vladn/hybrid_feedback.pdf
https://web.archive.org/web/20161022044737/http://www.kearman.com/vladn/hybrid_feedback.pdf
Quote: "...
Notes for the Rs+Rp mixed loss model plot (middle):
Nearly perfect zero-tilt for a mixed Rs+Rp model. The LC network values can be tuned to achieve exactly zero tilt for any given combination of Rs and Rp (not possible for classes 1, 2 and 3 alone).
Note nearly constant Q for the mixed load model.
...
All three hybrid feedback oscillator prototypes built in this study confirm that after properly balancing the two feedback paths the oscillation onset happens at exactly the same gain settings at the two opposite frequencies of the tuning range (or any two frequencies of choice) and that the remaining non-linear term of the threshold dependency is quite small.
...
The end result was a surprisingly well behaved regenerative receiver. It has a very good frequency stability and does not require adjusting regeneration level as you tune from 3.3MHz to 9.1MHz which is highly unusual for a conventional regen. The oscillation threshold level is so flat that it is completely feasible to have just two fixed regeneration presets - one for AM and a higher one for the SSB/CW modulation types. To illustrate these somewhat unconventional results I put several videos on youtube.
The two most relevant to the subject of this discussion are:
..."
http://www.kearman.com/vladn/hybrid_theory.pdf
https://web.archive.org/web/20210211011108/http://www.kearman.com/vladn/hybrid_theory.pdf

Hi,
I did not pay much attention to this project, but it really has potential. I see three drawbacks:
1- The need for two ganged variable capacitors. IMHO the input circuit could be fix tuned to the center of each band (changeable coils) and the tuning would be accomplished at the regen detector only.
2- The cathode of V2 has RF potential. I would move the reaction coil to the g2 and connect the cathode to the ground. It would be better for directly heated tubes.
3- I am a non-smoker, so I have no cigar box    This is the least problem, hihi.
I would like to try it with Russian military tubes 1Zh29B (1J29B) as V1, V2 and V3 or 1P24B-V as V3 instead. I have not used them yet, so my experiments may take "a bit" of time.

VBR from Ivan OK1SIP

I really think the cigar box regen has possibilities . The screen grid control of regen in the detector (properly ranged of voltage control) along with the somewhat unusual "electron coupled" in a filament cathode is really cool! I plan a RF amp tube with variable gain using a pot to vary minus bias to the bottom of the grid return. This could cause the B+ voltage to vary with gain control setting to a zener on the top of the RF gain pot (say 27V). I think after the receiver is built (or maybe first!) I will make a lithium ion cell driven filament LDO regulater and isolated transistor batt (11V) to 90v, 45v and bias dc to dc power supply. The receiver should only draw between 3 to 4 watts. Could make a real learning experiance. Any help and suggestions welcome... New to forum, Howdy to All! Doug