That Super-Regen rushing Noise is driving me Crazy! In this final video on a tube type superregenerative aircraft receiver, we tame the noise issue with a squelch circuit.
I believe a super regen with separate quench oscillator can be adjusted so it is quiet without a signal. I think the WW2 IFF system used this. Really enjoyed the videos, right up my street!
I am currently working on one of those. And indeed, when operated in linear mode, or below the point of free oscillations forming as well (basically as a sort of supercharged regen, where the name comes from hehe) it is completely noise-free. The problem is that the AGC-function gets lost as well. The regen control is pretty finicky too to remain in this lineair area. I currently use a DK92 heptode (a bit like the 1L6 American pentagrid) as a combined Quench Oscillator and Super-regenerative detector. The Quench oscillator is build between g1 and grid 2, the HF-section is a Hartley between grid 3 and anode/plate. They did use these as IFF-transponder, because basically they just re-emit a signal x1000 or 10.000x larger in amplitude. That's a downside, unless for IFF-transponders ofcourse!
Very interesting. Never built anything working at VHF frequencies. 27Mhz was the highest. Made a two valve two band radio, MW and SW. Looking forward to your next project.
Very cool that you worked for ADI. It was awesome app engineers like you that sent me parts when I was in my teens that had me eventually study electronics in uni. Super cool that you worked with N1TEV too! Sadly my later career has been in banking and software engineering but I get to play with electronics again as my all consuming hobby. 73.
OK. I blame you for the time spent in my building this, I could not resist!! It works like a charm.... except I found the squelch to be insensitive and struggling to cope with weak signals, not opening properly, and the audio would chop on strong modulation. So, after two very enjoyable evenings, I now have it sorted. I dispensed with the diode rectifier on the noise amplifier, instead using the pentode as a simple peak detector. After changing the high pass filter arrangement, and raising the quench freq to 80 kc, it now ignores any modulation, and opens easily on a microvolt of carrier. This was accomplished by using a simple diode switch in series with the audio channel. The pentode now only has to move a few volts to do the switching, it has a full 30 volts of swing from no signal to full quieting, so is very sensitive. . I dispensed also with the regen control, in my case it was not needed, the SR happily tunes from 110 to 144 megs with no issues. The valves I used was the 12AX7 and 2 x ECF80/6BL8, as these were handy. So, thank you once again for the inspiration. If there is some way I can contact you, happy to send you details.
You can send me details sir! Sounds very interesting. Btw, I too always dispensed with Regneration controls on SR-detectors, but now I'm putting them back in because they affect selectivity quite a bit it appears. Did you use a seperate quench oscillaor? Or is a self-quenching affair such as this one? Regards, Thomas
Great work! A DC Rx similar to KK7B maybe a great circuit for Air band AM RX + squelch The inverse circuit with shunt controlled transistor also exists and was used extensively as audio compressor in tape recorders. Great ppl at ADI including Bob Pease and Jim WIlliams.
Some giants were still around when I was there - Barrie Gilbert, Paul Brokaw, Walt Jung, and many of the new generation of SDR folks were in full design mode.
Thanks for this series, I really enjoyed the theory, construction and backstory behind the technology. I'm not at the level of competence to build one but the education is getting me there. Thank-you.
Congrats on an awesome project. Squelching a superregen is difficult to do properly. The noise amp needs to reject any audio modulation, and also to be insensitive to the quench frequency. Usually a bandpass filter is required between the detector and noise amp to adequately reject both, only amplifying the narrow band of frequencies between . Any audio modulation riding on the noise will chop the audio by activating the squelch on modulation peaks, any residual quench will make the squelch gate insensitive. You have done this very well. How anyone can tolerate the constant mind numbing hiss while waiting for a sporodic transmission fascinates me. Squelch is THE most impportant addition to any VHF RX. Excellent project.
Yes this was tricky. Thanks for watching. A proper bandpass filter would allow the circuit to operate more effectively and predictably than the high pass and grid stopping filter approach.
One attribute that is seldom mentioned is its use with a superhet receiver. The super regen detector has a radiated signal that can be picked up in the local area and will re-radiate the original signal and have those noise pulses from sewing machines etc quieted. I observed this when listing to a conversion on the CB channels with a Panasonic portable short wave ( HF ) receiver. My mother started doing some sewing and the noise from the sewing machine"s motor seriously conflicted with the CB reception. I turned on an Elcheapo Kiddy Talky. Then like a miracle the motors interference was near ZERO. The broad band nature of the Super regen did not care what channel the CB dudes were on. What ever signal was strongest took control. The Panasonic only responded to the channel it was tuned to. There you have it. Great noise limiting at the front end plus narrow band reception. My challenge is to have a narrow band filter in front of the Super regen. Perhaps an up converter to a variable band pass filter then the Super Regen detector and back down to the origin freq. With all the circuits in front of the Super Regen detector, its radiation signal will be gone.
I would watch out with that final stage, 100 Ohms Rk seems a bit on the small side for an ECC82 as class A amplifier... Might have a too high Anode current there, reducing tube life. Should look at the original datasheet for the right size Rk. And the problem with the 12AU7.ECC82 as a super-regenerative detector tube is that it's gain is much, much lower than all the other triodes you put in. Same with the transconductance. This means it needs much more feedback to get working, and worse: a much higher negative voltage to get the super-regenerating part going (quench voltage needs to be several times higher)
First off, great video and series Mike! I really appreciate your methodical approach to building circuit complexity; you certainly are a classic (junk-box) experimenter ham and I learn a lot from each installment. The radio (if you didn't know it) almost sounds like FM with that squelch circuit. So many squelches are ratty and raspy but yours has a nice sound. My experience with AD was with the AD-590 linear temperature to current sensor; I have always like using currents instead of voltage due to insensitivity to IR-losses. Anyway, I used them to design both temperature controllers and a 16 channel digital scanning thermometer. I have always wondered why other companies don't off current-based sensors; does AD have a patent stranglehold; the AD-590's are SO expensive now? Just getting to this video because YT didn't send the notifications for this and your next video; so I went looking for you! Anyway, love these theory/build videos and hope you stay motivated to make them as they are permanent YT treasures for sure. Best of 73...
I read Charles Kitchen's articles on super-regens and he talks about squelch in Super-regens but I still didn't thin that was possible. Nice demonstration. Also like that it is vacuum tube. BTW, where is the quench circuit and is it fixed or adjustable?
The Self-Quench frequency is primarily set by the grid resistor and capacitor. For example with a 5.6 Meg resistor and a 47 pF cap on the grid, I was happily running around 90 kHz. 5.6M and 70 pF gave 50 kHz and 100 pF and 3.3M gave 40 kHz. But the plate bypass (.001 uF) and Choke value L6 also affect the frequency as does the actual REGEN control voltage. Fortunately the circuit is tolerant. The ridiculous 88 mH L6 choke actually helps in one important way (compared to a small value 1 mH or 2.5 mH choke substitute). It makes the noise squelch work better because of its added filtering action!
@@radiotec76 The 88mH Telephone chokes are getting scarce. I would use some 1 dollar 100 mH radial or axial lead chokes. Bourns RLB0913-104K for example.
There is a lot of controversy about switching from old AM to the NEXTGEN system and digital. The age old defense of AM is safety (of course) and cost to upgrade civilian and vintage aircraft in Borneo.
Yes keep tube technology alive but why not start manufacturing audio tubes in the USA. We need not have to buy them from unfriendly countries. There are musicians that prefer tube amplifiers over solid state. 73
Please search 'radiomuseum df97 fm converter' Leads to a fascinating thread of using a low mu battery triode or pentode as a converter at VHF FM freq. The Triode converter for FM with the 3 point converter - that has a positive fb if loop for more gain ... How they were able to use a triode of 1ma/V or less at VHF. and we find it difficult to design with 12au7 at 2.2maV and unipotential cathode
![Megan Thee Stallion - Neva Play (feat. RM) [Official Video]](http://i.ytimg.com/vi/TpYTyAaTRts/mqdefault.jpg)
I believe a super regen with separate quench oscillator can be adjusted so it is quiet without a signal. I think the WW2 IFF system used this. Really enjoyed the videos, right up my street!
I am currently working on one of those. And indeed, when operated in linear mode, or below the point of free oscillations forming as well (basically as a sort of supercharged regen, where the name comes from hehe) it is completely noise-free. The problem is that the AGC-function gets lost as well. The regen control is pretty finicky too to remain in this lineair area. I currently use a DK92 heptode (a bit like the 1L6 American pentagrid) as a combined Quench Oscillator and Super-regenerative detector. The Quench oscillator is build between g1 and grid 2, the HF-section is a Hartley between grid 3 and anode/plate.
They did use these as IFF-transponder, because basically they just re-emit a signal x1000 or 10.000x larger in amplitude. That's a downside, unless for IFF-transponders ofcourse!
Very interesting. Never built anything working at VHF frequencies. 27Mhz was the highest. Made a two valve two band radio, MW and SW. Looking forward to your next project.
Very cool that you worked for ADI. It was awesome app engineers like you that sent me parts when I was in my teens that had me eventually study electronics in uni. Super cool that you worked with N1TEV too! Sadly my later career has been in banking and software engineering but I get to play with electronics again as my all consuming hobby. 73.
OK. I blame you for the time spent in my building this, I could not resist!! It works like a charm.... except I found the squelch to be insensitive and struggling to cope with weak signals, not opening properly, and the audio would chop on strong modulation. So, after two very enjoyable evenings, I now have it sorted. I dispensed with the diode rectifier on the noise amplifier, instead using the pentode as a simple peak detector. After changing the high pass filter arrangement, and raising the quench freq to 80 kc, it now ignores any modulation, and opens easily on a microvolt of carrier. This was accomplished by using a simple diode switch in series with the audio channel. The pentode now only has to move a few volts to do the switching, it has a full 30 volts of swing from no signal to full quieting, so is very sensitive. . I dispensed also with the regen control, in my case it was not needed, the SR happily tunes from 110 to 144 megs with no issues. The valves I used was the 12AX7 and 2 x ECF80/6BL8, as these were handy. So, thank you once again for the inspiration. If there is some way I can contact you, happy to send you details.
You can send me details sir! Sounds very interesting. Btw, I too always dispensed with Regneration controls on SR-detectors, but now I'm putting them back in because they affect selectivity quite a bit it appears. Did you use a seperate quench oscillaor? Or is a self-quenching affair such as this one?
Regards,
Thomas
@@thomasvandevelde8157 I can email you the schematic if you give me an address?
Great work! A DC Rx similar to KK7B maybe a great circuit for Air band AM RX + squelch The inverse circuit with shunt controlled transistor also exists and was used extensively as audio compressor in tape recorders. Great ppl at ADI including Bob Pease and Jim WIlliams.
Some giants were still around when I was there - Barrie Gilbert, Paul Brokaw, Walt Jung, and many of the new generation of SDR folks were in full design mode.
Hello, superb explanation of receiver design goals. Congrats! Thank you for taking the time to prepare the tutorial.
Thanks for this series, I really enjoyed the theory, construction and backstory behind the technology. I'm not at the level of competence to build one but the education is getting me there. Thank-you.
Thanks for watching this evolve from an idea through a process that ended up in a complete project, even if it is archaic and obsolete tech.
Keep tube technology alive!
Congrats on an awesome project. Squelching a superregen is difficult to do properly. The noise amp needs to reject any audio modulation, and also to be insensitive to the quench frequency. Usually a bandpass filter is required between the detector and noise amp to adequately reject both, only amplifying the narrow band of frequencies between . Any audio modulation riding on the noise will chop the audio by activating the squelch on modulation peaks, any residual quench will make the squelch gate insensitive. You have done this very well. How anyone can tolerate the constant mind numbing hiss while waiting for a sporodic transmission fascinates me. Squelch is THE most impportant addition to any VHF RX. Excellent project.
Yes this was tricky. Thanks for watching. A proper bandpass filter would allow the circuit to operate more effectively and predictably than the high pass and grid stopping filter approach.
One attribute that is seldom mentioned is its use with a superhet receiver. The super regen detector has a radiated signal that can be picked up in the local area and will re-radiate the original signal and have those noise pulses from sewing machines etc quieted. I observed this when listing to a conversion on the CB channels with a Panasonic portable short wave ( HF ) receiver. My mother started doing some sewing and the noise from the sewing machine"s motor seriously conflicted with the CB reception. I turned on an Elcheapo Kiddy Talky. Then like a miracle the motors interference was near ZERO. The broad band nature of the Super regen did not care what channel the CB dudes were on. What ever signal was strongest took control. The Panasonic only responded to the channel it was tuned to. There you have it. Great noise limiting at the front end plus narrow band reception. My challenge is to have a narrow band filter in front of the Super regen. Perhaps an up converter to a variable band pass filter then the Super Regen detector and back down to the origin freq. With all the circuits in front of the Super Regen detector, its radiation signal will be gone.
Should be a great project!
👍 Awesome.
I would watch out with that final stage, 100 Ohms Rk seems a bit on the small side for an ECC82 as class A amplifier... Might have a too high Anode current there, reducing tube life. Should look at the original datasheet for the right size Rk.
And the problem with the 12AU7.ECC82 as a super-regenerative detector tube is that it's gain is much, much lower than all the other triodes you put in. Same with the transconductance. This means it needs much more feedback to get working, and worse: a much higher negative voltage to get the super-regenerating part going (quench voltage needs to be several times higher)
First off, great video and series Mike! I really appreciate your methodical approach to building circuit complexity; you certainly are a classic (junk-box) experimenter ham and I learn a lot from each installment. The radio (if you didn't know it) almost sounds like FM with that squelch circuit. So many squelches are ratty and raspy but yours has a nice sound.
My experience with AD was with the AD-590 linear temperature to current sensor; I have always like using currents instead of voltage due to insensitivity to IR-losses. Anyway, I used them to design both temperature controllers and a 16 channel digital scanning thermometer. I have always wondered why other companies don't off current-based sensors; does AD have a patent stranglehold; the AD-590's are SO expensive now?
Just getting to this video because YT didn't send the notifications for this and your next video; so I went looking for you! Anyway, love these theory/build videos and hope you stay motivated to make them as they are permanent YT treasures for sure. Best of 73...
What about KORGs NUTUBE 6P1 as a push pull amplifier or as a balanced mixer.
Standby - KORG soon.
I read Charles Kitchen's articles on super-regens and he talks about squelch in Super-regens but I still didn't thin that was possible. Nice demonstration. Also like that it is vacuum tube. BTW, where is the quench circuit and is it fixed or adjustable?
The Self-Quench frequency is primarily set by the grid resistor and capacitor. For example with a 5.6 Meg resistor and a 47 pF cap on the grid, I was happily running around 90 kHz. 5.6M and 70 pF gave 50 kHz and 100 pF and 3.3M gave 40 kHz. But the plate bypass (.001 uF) and Choke value L6 also affect the frequency as does the actual REGEN control voltage. Fortunately the circuit is tolerant. The ridiculous 88 mH L6 choke actually helps in one important way (compared to a small value 1 mH or 2.5 mH choke substitute). It makes the noise squelch work better because of its added filtering action!
@@MIKROWAVE1, where do you find those 88 mH chokes? They were common in the 1970s in unregulated power supplies.
@@radiotec76 The 88mH Telephone chokes are getting scarce. I would use some 1 dollar 100 mH radial or axial lead chokes. Bourns RLB0913-104K for example.
@@MIKROWAVE1 82c on Digikey. I happened to purchase an 82mH choke DigiKey: RL181S-823J-RC-ND which is rated for 10ma.
@@vincei4252 more than adequate Vince.
Air Band is DSB doesn't have a carrier.
There is a lot of controversy about switching from old AM to the NEXTGEN system and digital. The age old defense of AM is safety (of course) and cost to upgrade civilian and vintage aircraft in Borneo.
Yes keep tube technology alive but why not start manufacturing audio tubes in the USA. We need not have to buy them from unfriendly countries. There are musicians that prefer tube amplifiers over solid state. 73
Please search 'radiomuseum df97 fm converter' Leads to a fascinating thread of using a low mu battery triode or pentode as a converter at VHF FM freq. The Triode converter for FM with the 3 point converter - that has a positive fb if loop for more gain ... How they were able to use a triode of 1ma/V or less at VHF. and we find it difficult to design with 12au7 at 2.2maV and unipotential cathode