Introduction

This is a simple one-valve radio for AM reception from approximately 500kHz to 1.8MHz. It should be possible to build it over a weekend and it doesn't require any special tools other than a soldering iron, multimeter and electric drill. Layout of parts is not critical and although I have not tried it with other valves, I suspect that any triode-pentode or double triode may be used. I built it to fill the gap between crystal sets and superhet radios in my collection.

Description

This design uses a PCF801 triode-pentode valve. The only reason for this choice was that my local electronics supplier had a glut of these valves. Since there are really two valves in one glass envelope, the set is really a two-valve receiver. The triode section is used as a regenerative detector, and the pentode section (wired as a triode) is used as a stage of audio amplification. The circuit is based on a design by EN Bradley for a pocket one-valve receiver, which used a 3A5 (DCC90) valve.

Since this set is intended for young persons, I didn't want to have any high voltages, such as one normally associates with valve equipment. Although the set will work off a single 9 volt high tension supply, it seems to work best from 36 volts (four 9 volt 6F22 batteries). Use the cheapest batteries you can find for the 36 volt supply – current drawn by the radio is less than 1 milliamp. The low tension (filament) supply is a nuisance. I found the best solution to be a 4 Amp-hour lead gel battery. Although the PCF801 should have an 8.5 volt filament supply, it seems to operate quite well from 6 volts. (There is an ECF801 with 6-volt filament – but you might have to pay more for it).

The PCF 801 was designed to be used in television tuners.. In this application, it would have run with high voltages of about 170 volts at several milliamps. The present circuit operates the valve at starvation levels with about 2 volts at the anode of the triode, nevertheless, the valve operates in regenerative mode, and with a small aerial will break into oscillation if the reaction control is advanced too far.

Signals arrive at the antenna and cause small voltages to be developed across coil L1. L2 and C1 form a tuned circuit, which separates out the station tuned. A tuned circuit works by swapping energy stored as a magnetic field in the coil and the energy stored as charge in C1. The high frequency voltage across L1 is fed to the grid of V1 via capacitor C2. The voltage is amplified by V1 and a larger version of it appears at the anode (pin 8) of V1. The amplified voltage is applied to the “tickler” coil, L3 and this further excites the voltage in L1. The process repeats until a comparatively large signal appears at the anode. The combination of the reaction control R5 and R4 limit the amount of voltage available, otherwise the circuit would oscillate. The operator should adjust the voltage at R5 so that the valve does not oscillate – the circuit will hiss and then make a plopping sound in the headphones. When tuned to the station, there will be an annoying whistle. Note that although R5 is NOT a volume control, it can be used as such.

The triode portion also acts as a detector and produces an audio signal at the junction of C3 and R4. Capacitor C4 bypasses any remaining high-frequency signals to the ground, so that a clean audio signal appears at the grid of the other half of V1. This is amplified and passed to the primary winding of transformer T1 via capacitor C5. T1 simply converts the “high” (9 Volts!) and low current at the anode to a low voltage and higher current for low resistance phones. T1 is not needed if you are using a crystal ear piece or “high resistance” headphones.

A word of caution – interference or bad connections can cause loud crackling in a crystal ear piece – please operate at the lowest possible volume to avoid ear damage.

In the early days of radio, headphones were made to be sensitive and had a very high resistance of 4000 Ohms. Nowadays, headphones for Hi Fi and iPods have low resistance and very low sensitivity. Although stereo headphones will work in the circuit of Figure 1, the results are not as good as mpre sensitive phones. Best results will be obtained using military surplus headphones or the insides of an old telephone handset. The crystal earpiece works very well. There is some scope for experiment with conventional stereo headphones.

 

Parts List

C1 Tuning Capacitor approx 0.00045 μF – I used one gang of a two gang capacitor.

C2 Blocking Capacitor 100pF silver mica

C3 Coupling capacitor 0.01 μF polyester or mylar

C4 By Pass Capacitor 100pF silver mica

C5 Coupling/Output Cap 0.1 μF polyester or mylar

L1 20 Turns antenna coupling

L2 80 Turns tuning coil

L3 25 Turns tickler

all on 50 mm PVC pipe former. All wire 32 SWG magnet wire (Don't fret if you lose count and miss a turn or two). If making your own, wind all coils in the same direction, leave about a 3mm gap between each winding. Spray with clear lacquer when finished to prevent the coils unravelling. I drill a 1mm hole in the pipe at the start and end of each winding.

R1 1 Meg grid leak

R2 120 k anode load

R3 2.2 Meg grid leak

R4 47k anode load

R5 1Meg potential divider – reaction control

 

Binding posts (4)

Tag Strip

Earth tags

Headphone Jack

TR1 small mains transformer 230 Volt primary, 5 Volt secondary. Primary connected to C5

- or use a crystal ear piece.

S1 – Miniature DPDT switch

B9a Valve Base – Chassis mount

 

Aluminium sheet 175mm * 150mm * 18 s.w.g

Perspex front panel 175 mm * 120 mm * 4mm (Black looks authentic)

2 No. 25mm*15mm* 150mm Supawood offcuts for chassis supports

Small wood screws 3 -3.5 mm x 10mm

Battery HT – 4 x 6F22 wired in series

LT – Best results were from 3 x Enersys Cyclon 2.2 AH (Don't let these dischrge below 1.0 volt per cell) 7 x NiMH 2.5AH NiMH cells wired in series should work even better.

Drilling the Chassis

Ensure that all drill bits are clean and sharp before you begin.

Do not wear a tie or loose clothing. Wear safety goggles. Make sure that all work is securely clamped and that there is no oil or anything that might allow you to slip while using any machine tool.

Make a drilling template on a piece of paper using the approximate layout in the photo of the finished radio. I used a large hole under the coil former with an “International Octal” hole punch. Cut out the surround to the template and glue it to the aluminium chassis plate. Drill a small pilot hole (1.5mm) at each hole location. Remove the drilling template and clean away any traces of adhesive.

Drill each of the small holes to the correct diameter – use a hand drill or drill press and cutting fluid (paraffin is said to work as a cutting fluid for aluminium.

The large holes for the valve base and the coil can be punched out with a chassis punch, or reamed to the correct size using a hand taper reamer.

Alternatively, a device called a “step drill” can be used – take extra care to secure the workpiece!

The hole under the coil can be the same size as the hole for the valve base, or 5 holes of 6mm diameter can be drilled in a circular pattern instead.

If you are really stuck, it is possible to file a neat circular hole.

Place the valve base on the chassis and mark out locations for the two holes to bolt the base to the chassis. Drill pilot holes and finish with a 3.2 mm drill for M3 screws.

Mounti the coil using 3 small right-angle brackets rivetted to the PVC pipe. It might be possible to glue the coil former to the aluminium chassis with epoxy resin adhesive.

Use a countersink bit to countersink the four holes at the corners of the chassis where the screws will be used to attach the supawood supports.

Drilling the Front Panel

Drill the plastic slowly and carefully starting with a 1.5 mm pilot hole. Perspex shatters quite easily and you might like to drill part of the way through, then turn the sheet over and drill from the other side. Work slowly and patiently, allowing the drill to do the work.

Countersink the end holes used for attaching the front panel to the supawood chassis supports.

Assembly

Attach the supawood side supports to the chassis. These run from front to rear at each side of the aluminium sheet.You need to drill 3mm holes slightly shorter than the length of the wood screws being used. Ensure the supports are flush with the sides of the chassis for a neat job. Place a drop of wood glue into each hole before lightly, but firmly screwing the parts together. Attach the front panel to the supawood supports using the same technique.

Leave the glue to dry so that a firm assembly is produced.

Attach the variable condenser C1 to the chassis using two self-tapping screws. Attach a solder tag to the screw nearest the rear of the chassis.

Orient the valve holder so that pin 9 is nearest the capacitor C1. Looking from the underside, the pins are numbered clockwise, starting from the gap. Attach the valve holder to the chassis using two M3 screws. Place a solder tag on each screw before attaching the nut.

Bolt the tag strip to the chassis.

Attach the coil to the chassis using the lugs and three M3 machine screws. (or use epoxy)

Finally, attach the binding posts to the chassis and the switch,reaction control and earphone jack to the front panel.

You might have to shorten the capacitor C1 and reaction control spindles with a small hacksaw – gently does it – let the saw do the work and don't get any metal filings into C1 because its a horrendous job removing them.

Soldering

For this project, you only need an inexpensive 25 to 50 Watt “hobby type” soldering iron. If you intend making electronics a “serious” hobby – then you will need a soldering station with variable temperature control (etcetera.). Avoid using a gas soldering iron – the hot gases can melt everything and spoil your project.

The solder is important. Use a good quality resin core solder for electronics work. Normally, an 18 or 22 s.w.g. Solder made by Kester or Multicore solders will be fine. If using lead free solder – be sure your soldering iron is suitable.

Its also a good idea to have a solder sponge and a small tin of “tip tinning solution” handy.

Since solder contains lead – remember to wash your hands after soldering and before handling food. Try to avoid breathing solder fumes.

Once the soldering iron is up to temperature, melt a little solder on the tip (or dip the tip in tinning solution), then wipe the tip on a wet sponge. Heat the part to be soldered with the iron and apply a little solder to tin the part. The solder will flow onto the part and wet it. If it forms a blob, then the part needs to be cleaned by scraping it with a sharp blade. Once two parts have been tinned they can be easily joined by heating and applying a small amount of solder.

When joining parts together with solder, it is no longer necessary to wrap wires round solder tags three or four times before soldering as was done in World War II military projects. If you do that and make a mistake, it will be a devil of a job unsoldering the whole mess. (Tektronix never did that with its famous oscilloscopes – the component was simply placed on the tagboard without wrapping.). All you need do is place the component wire through the hole in the tagboard/valve base/component tag and ensure a good solder joint with both surfaces wetted. (Remember the solder will flow round the parts – not form a blob.)

Clean off any flux residue with a small stiff paint brush and lacquer thinners.

Nearly all the parts will be “pre-tinned” making the soldering job very easy.

The exception is the enamel wire used on the coil. The enamel used on this wire is supposed to act as a “flux” - but you will have to slowly melt it away with the soldering iron and tin each lead quite carefully.

Wiring the Set

If possible, work from the circuit diagram and skip this section. In early texts on radio construction, the wiring instructions and a wiring diagram were always provided.

 

Its always nice to begin by wiring the LT circuit. Wire from the LT binding post to the switch and then from the switch to pin 4. Connect Pin 5 to ground. Connect the “Ground” binding post to ground. Important! - check your connections – be really, really sure that you are not going to short the LT battery to ground. Connect a charged 6 Volt “Lead Gel” (or whatever battery you are using) to LT+ and Ground. Insert the valve in its base and switch on. After a few seconds you should be able to see the filament glow cherry red.

If at this point the valve doesn't glow red – re-check your connections. Check the battery voltage is above 5 volts. If the wiring heats up and the insulation melts, then you have shorted the LT battery, and will have to re-wire everything with new wire and a new switch (most likely). You will also have to recharge the battery. Always check your wiring most carefully.

Unplug the valve and disconnect the battery.

Next wire the tuned circuits. L1 is the antenna coil situated nearest the mounting lugs. L2 is the tuning coil – the one with the most turns in the middle. L3 is the top coil. Connect the bottom-most winding of L1 and L2 together and connect to ground.

Connect the top end of L1 to the antenna binding post. Connect the top end of L2 to the fixed vanes of C1.

Connect one end of the tickler winding to the anode – Pin 8 and the other end to the first free terminal of the tag strip. (You may have to reverse these connections later).

Connect C2 between the stator of C1 and pin 9. Connect R1 between Pin 9 and ground.

The following is a photo of the underside of the set with the component names noted.

 

Continue to wire the triode section wiring R4 to the slider of R5 via the 4^th terminal on the binding post and wire the other terminals of R5 to ground and HT+ (I used pin 3 of the tag strip). At this point wire pin 3 of the tag strip to the switch. Wire the switch HT+ connection to the binding post.

Wire C4 and R3 between pin 2 and ground. Wire C3 from tag strip terminal 1 to pin 2.

Wire the anode resistor R2 between pin 6 and HT+ tag terminal 3

Finally, wire C5 to T1. Wire the headphone jack. Make sure connections 1and 3 of the valve are grounded.

Now double-check your wiring against the circuit diagram (pins 4 and 5 wiring may be interchanged) – and I may have omitted something or made a misconnection in the above list of connections. Check again

Make a 36-Volt HT battery by wiring 4 9-Volt 6F22 (PP3) batteries in series – just solder directly to the tags.

Checking the Operation

Turn the reaction control to mid-way. Attach the LT battery between ground and LT+ binding post. Attach the HT battery pack between the HT+ and ground – observe polarity!.

Attach a 20 metre aerial and attach the ground terminal to a good earth – metal water pipe for example – otherwise you are going to have to drive an earthing spike into the ground.

If you are a flat-dweller – try using 6 metres of aerial wire wrapped round the room. Attach the earth wire to a large metallic object. The set will operate without an earth – but the sound quality will be poor and less stations will be heard.

With the tickler coil correctly wired, the set will tune to medium wave stations quite sharply and they will be at a comfortable listening volume. If the tickler is incorrectly wired (as mine was at first) – stations will be quite weak and you need to reverse the connection to the tickler. With the tickler correctly wired, as you advance the reaction control, the set will begin to make a hissing sound in the phones as the valve breaks into oscillation. This is particularly noticeable on a short aerial. The long aerial may prevent the valve from oscillating altogether. Please don't operate the set with the reaction control so far advanced that the set oscillates, as it will interfere with other radios.

Results

This radio was able to receive all medium-wave stations that were audible on a normal modern medium wave radio. Tuning is very sharp, so a reduction gear drive or dial-cord drive is essential. The reduction gear on the tuning capacitor is perfect for medium wave, but if you are contemplating a short-wave version, you may have to find a slow-motion drive.

Using a 20m aerial with a poor earth in the Northern Suburbs of Johannesburg, it was possible to receive Metro FM (rebroadcast on medium wave), Radio Pulpit, Zimbabwe radio, Voice of America, Hellenic Radio, 1485 AM, Radio Islam and many others after dark. The use of a good Earth caused 1485 AM to overload the receiver.

Its worthwhile experimenting with different antennas and earthing arrangements to get best reception.

To thank the Author because you find the post helpful or well done.

 Hi Bryce,

Thats a good circuit you came up with and you got some good DX with that radio! 

In order to improve reception, I would recommend changing the way you control your Reaction(regeneration). I would remove the 1meg potentiometer that you used, connect R4 strait to your B+, install a 10K(linear taper works best) potentiometer across your tickler coil and connect a .005uf capacitor to the wiper and the other end of the capacitor to ground.

What this will do is increase the B+ voltage on your detector(making for better DX reception) but still allowing you to easily control your reaction.

Good Luck!!!

To thank the Author because you find the post helpful or well done.

Bryce and Wyatt,

Thanks for the very well thought out construction article, with very good detailed performance results.

I would like to share a few thoughts.

The modification that Wyatt suggested helps to improve the triode gain by running it at a higher plate voltage. Distortion should be reduced for loud stations because the cutoff voltage is more remote with higher plate voltages.

But higher plate voltage also directly impacts the rectification behaviour of the grid. You can think of the grid as a diode with a controlling back plate. One way to measure this effect is to note the DC voltage at the control grid as the plate voltage is raised. You should note that a higher plate voltage makes the grid voltage less negative.

The voltage that is normally present at the grid should be around -1V to -0.5V, depending on tube type, age and plate voltage. This voltage is essential to establish how much current the grid leak resistor will draw, thus setting the impedance of the grid, itself. The grid detection efficiency will be highest when the conducting grid impedance matches the impedance of the tank circuit driving it.

If you increase the plate voltage greatly, you could bring grid conduction to zero, thus eliminating all detection. Under these conditions, you should see very little if any negative voltage at the grid.

One way to solve this problem without giving up the nice high voltage at the plate that provides high gain, is to add a diode in parallel with the grid. If the grid voltage is already at Zero volts, then a 1N34a germanium diode may work well, to provide the detection that the plate back biased grid can't. You could experiment with biazing the diode, perhaps with 1uA of current for a diode impedance around 26k. Some 1N34a diodes already have thi much saturation current and operated near this impedance.

A tube that has a spare detector diode, like the 12AV6 or some pentodes, like the EAF42, could be separately optimized for maximum amplifier gain, and maximum detection efficiency in the grid circuit, with the spare detector diode in parallel with the grid.

---

A further comment on the saturation problem with the local station at 1485kHz, and a good ground.

The antenna coupling coil L1 should be loosely coupled to the main tuning coil L2. This tends to increase selectivity and Q of the tuning coil circuit, which helps reject the strong signal. Loose coupling means that a good portion of the input coild flux is not coupled to the secondary coil.

One of the most common methods of input antenna coupling is to have a primary with many turns that is far from the secondary. The high number of turns tends to increase the flux from what are usually high impedance antennas in the AM band, while the long distance between coils keeps the Q high in the main tuning coil.

The AM band air core antenna coils I often seen in radios have the primary and secondary space by a distance that is about the same as the diameter of the coil, or perhaps a little more. The primary inductance should be a few hundred uH, perhaps up to 1mH with a Universal-wound technique. Too many turns beyond what is needed for 1mH end up loosing the high end of the AM band.

The impedance of a 20m wire antenna is going to be high in the AM band because the wavelenght at 1MHz is still 300m. So 20m is still a short high impedance antenna.

Summarising, a very loosely coupled antenna primary, with many more turns than the secondary, is preferable to the low turns primary for at least two reasons:

1-More energy can be extracted from the high impedance short antenna,

2-The Q of the secondary main tuning circuit is not compromised by primary coil loading. Note that this selectivity that is unaided by regeneration is essential to avoid saturation. Trying to obtain the same selectivity with regeneration is much more likely to cause amplifier distortion.

---

The following is a survey of early regenerative circuits. I got this from Ron Roscoe a few years ago, but I don't know the origin otherwise.

 Note that none of these circuits includes a spare detection diode in parallel with the grid, that would be unaffected by grid voltage. 1920's era radios could have benefited greatly by the extra diode of the type 55 triode from 1932 or in the 1H5G from 1935. What a difference a little detector diode platelet at the negative end of the filament of a 1920's 201a would have made for detection efficiency.

Thanks again for the very interesting, real-world experience thread.

-Joe

To thank the Author because you find the post helpful or well done.

Joe and Wyatt,

Many thanks for your suggestions.  I am certainly going to try some of them out on my next project, and I would suggest to anyone that builds this set that they do the same.  I imagine that the original design by EN Bradley was centered around minimising battery expense. The original HT bateries have been in the set for about a year.

Joe's points about High Q LC circuits and antenna coupling are well taken. I imagine the coils would benefit from using Litz wire and I think the antenna coil is too tightly coupled, leading to problems with 1485 (Which can be received with no aerial at all -just).

Kind regards - Bryce

PS I lost my reel of Litz wire ...  does anyone know where I can get some more?

 

To thank the Author because you find the post helpful or well done.

Adding a diode to the grid of an audion has been suggested by Nestel in 1935.
[Nestel, W.: Rückkopplungsaudion mit verringertem Klirrfaktor. ETZ 1935, p. 1021; 'regenerative audion with reduced distortion']

Audions are very sensitive but the drawback of an audion is its low dynamic range. This can be seen from its characteristic. [Pitsch, H.: Lehrbuch der Funkempfangstechnik, vol.1, VAG, 1963]

When the lefthand side oft the grid voltage exceeds the bent of the tubes characteristic resp. the lower envelope of the anode current touches the zero line, distortion occurs. Depending on the amplitude of the carrier, typical distortion of an audion is shown in the next figure.

After Pitsch part of the distortion is due to the fact that the system grid  - cathode is not totally equivalent to a diode. Therefore he suggests to add a diode.

In a Lorenz 100W receiver the Nestel audion has been realized. Here one system of an AB2 ist used as the "Nestel-diode". The AF7 has a cathode resistor of 1kΩ, so it rather operates as an amplifier for AF and RF.

This receiver was designed for quality local reception as can also be seen from the triode AD1 in the output stage.

Regards,

Dietmar

A German text about AM demodulation see here.

To thank the Author because you find the post helpful or well done.

The Lorenz 100W with the extra diode in parallell with the AF7 grid has more overall gain than without the diode, for several reasons.

The following curve tracer photo shows grid conduction for a 3A5 triode from the 1950's. Note how enve just +20V at the plate of this triode pushes the conduction slope to the right of the 0V line. If the rated 90V operational plate voltage were applied, the grid conduction slope would be so far to the right that no grid rectification would be possible. (This photo came from an earlier thread exploring tube substituitions)

Some of the reasons for the benefit of the diode can be enumerated as follows:

1- Diode conduction is independent from the screen voltage, which has similar effect to plate voltage in a triode. This improves detection efficiency and helps keep the distortion on the left side of the distortion plot (Klirrfaktor) in the previous post to a minimum.

2-Now the screen voltage can be optimized for higher transcoductance, wider ouput voltage swing, and lower distortion. The schematic notes for the Lorenz 100w indicate that the screen operates at 60V, which is much higher than the typical 20-30V for detector operation. Higher mu tubes can run with higher plate or screen voltage because of the reduced effect of the plate or screen on the grid. The higher screen voltage moves the steep worsening of distortion on the right side of the plot, where cutoff is approached by a factor of three, as compared to typical detector operation.

The unregenerated audio gain of the AF7 pentode is still high despite a relatively low 100k resistive plate load. The plate draws 0.75mA at 180V and the HV supply has approximately 260V, according to the schematic notes. The load resistor is 100k, so a gain toward 100 would be expected from 1mS of transconductance.

Usually, running a resistively loaded pentode with a low screen voltage gives the highest gain because the transconductance drops with the square root of the plate current while the availabe voltage drop for the load resistor increases linearly with reduced plate current. The higher the available supply voltage the higher the gain that can be realized.

------------------------

Triodes also operate under these constraints, but their much lower plate impedance makes transformer use practical to extract additional voltage step up ratios of 3x to 10x between the triode plate and the grid of the next stage.

In a 1920's set, the shunt diode would have allowed operation of the transformer loaded detector triode type 201 at a full 90V instead of the usual 22V. At 22V, and with a mu=7, it only takes 3V of negative grid bias to turn off the triode. This means that with a 5V filament is cutoff where it's DC drop is between +3V and +5V. The operation in this regions is one of remote cutoff, further complicating the plate cut-off distortion.

With the full 90V at the detector plate, a tickler coil could still have been employed for regeneration, and a higher step up ratio for the output transformer would have been practical because the plate impedance would be much lower at 90V.

It still remains amazing to me that the small plate detector diode as part of a triode or pentode that became ubiquitous after the introduction of the 55 triode/diode combination in 1932, was not invented or commercialized earlier.

---

On a different regeneration related topic: Bryce, have you tried homodyne/synchronous reception with your set? You need a steady hand by the results are impressive. It makes a regenerative set sound like Hi-Fi. Use your nicest earphones to enjoy this. Look for details in this thread about my Crosley 51.

Prof Rudolph illustrates the perfectly linear characteristic of synchronous detection at page 12 in AM_Demodulation_Teil_1 I have read many of Prof Rudolph's writings with Google-Translate, and have always found it to be a very worthwhile effort. The German language thread indicated above is well worth reading with Google-translate.

 

Regards,

-Joe

To thank the Author because you find the post helpful or well done.

Bryce,

Winding the coils with Litz wire would make almost no difference. The reaction is what gives the circuit its Q so Litz wouldnt help too much., sells Litz. Hes a good guy to do buisiness with.

The problem with 1485 is mostly caused by the tight antenna coupling but it can also be fixed bu using an antenna tuner, simply another tuned circuit loosely coupled in series with your antenna.

Dave Schmarder, 1n34a.com sells Litz. Hes a good guy to do buisness with.

 

Joe,

Using a 12AV6 would be a great Idea, but some sort of RF amplification would probably be best for using it though.

I have found that using very low plate voltages is best(less than 20V). I have had great success using a 1625 as a regenerative detector with 12V on the plate!

To thank the Author because you find the post helpful or well done.

Wyatt

I think you're right about the Litz wire. Thanks for the info on where I can get some for my non-regenerative  projects.

 

Joe

If I tune sligtly off the station and advance the reaction until I get a weak beat note, then I can get the oscillations to "lock" on to the stations frequency. This is only possible with fairly strong stations. You get a fairly gritty beat note until the oscillation suddenly disappears. The sound is then fairly good - akin to the sound when there is almost no reaction.I assume there will be no interference, since the oscillator is now locked in frequency (and presumably phase?) to the tuned station. I don't recall being able to do this very easily with other super-regens. I used normal stereo eadphones for this test. Does it prove that one triode can do the job of a PLL i.c ?  ;)

 

To thank the Author because you find the post helpful or well done.

I've been thinking about this circuit for a while now.  Looking around for a tube to use, I'm currently considering the 6AS8. This tube has a pentode section and a separate diode section that can be used a high perfomance detector. I think this tube would work well in the Nestel Audion circuit. Thank you Professor Rudolph for bringing the Nestel Audion circuit to my attention! um.org/tubes/tube_6as8.html 

Norman Bourassa

KA1GUK

To thank the Author because you find the post helpful or well done.