Alternate RF gain for the Odd To Me Regen

There is always someone finding a problem with whatever is posted. The original circuit used an established method. It has been in use for about 20 years. Here is a little more complicated circuit that may work better. The only problem I had with the original is the RF gain is the volume control so when you turn it up on a weak signal and then get a boomer on channel it may tend to hurt your ears. The diodes used here limit the peaks and help prevent the ear damage. It can be switched from balanced to unbalanced by installing or removing the jumper. This particular circuit has been used in commercial units since the '60s.

The Odd-to-me SW regen update.

The original circuit and the fix. 

I'm not a big fan of the LM386. This is my AF stage. NOTE: Use a 100K pot with the receiver to prevent overloading the detector.

Here is the circuit I built. I'm using slug tuning and diode tuning. I'm using a 9 volt supply. The posted version used a 5 volt circuit.

Here is the modified circuit with the AF amp.

Exploring the world of 3DQ or the mystery of zero bias devices

If you say zero bias in the wrong places they will start collecting fire wood and a stake. Let's don't go into this with a closed mind. After all the manufacturer named these devices. Even if the device is not zero bias the term is being used to define a new class of device. Now that I've set the ground rules here we go.
Hold on it will be a quick run threw without a lot of discussion. I will try to let the datasheets tell the story.

UHF ,high gain, low noise, and dual gate. It's a cascode amplifier. Look at the drain current? Plus or minus 30ma? What's up with that? Can the drain conduct in either directions? That's what it says. Let's turn the page and see what else we see.

Look at the center graph at the top of the page. It conducts at zero volts bias and can respond to positive or negative gate voltage. Ummm! It's and enhancement depletion device! As I said above it's a new family of devices. Let's look at a 'normal' MOSFET for a moment and compare.

We don't have to look at much more than the symbol. The 2N7002 has a protection diode built in to keep the drain current from reversing. If you use a MOSFET in an inductive load it needs a protection diode to keep the inductive kick back from destroying the device.

The depletion MOSFET conducts with no bias and can be biased into higher or lower conduction. This is accomplished by doing an extra step is the doping that enhances the channel. While this is not new and exciting technology it seems when it was new it failed to be exciting enough to become common place to us hobbyist. Now for the good news. These devices are still being made. A quick check on Ebay for 3DQ finds some at $5.50 each. Seems they are obsolete.  But a the depletion MOSFET is alive and well.

This circuit is built using a DN3545 MOSFET. The schematic is for a 'normal' enhancement MOSFET. It would require biasing to conduct. I made the circuit with a Depletion N channel 3545. I did use a 1 meg gate resistor but no forward bias resistor. My plan is to build a ham receiver 80 and 40 meter. I'm sending it a 4 Mhz signal in the test. So it requires no bias and works at 4 Mhz. I think the term zero bias would apply. Now that we know what to look for we can find 3DQ equivalents a lot cheaper. I think I paid 10 cents each for the last ones I bought.
Just look for Depletion in the name.

Making a slug tuned coil with pictures.

My explanation without pictures was a bit confusing. I had a little shop time today and made a new coil. I took the camera along to document the process.

I used this plastic stock for my pads. You can use any insulator. I have used wood but I used an insert in the wood so I could have threads that wouldn't wear out to quickly.

I didn't do any measuring I just cut a piece and used it to size the next piece.

The tube is a sink supply. You can use a plastic pipe. I like the sink drain because the slug fits snugly in it.

I just cut off a piece for the core. I made it a little long and had to trim it. About a hands width is good.

I clamped the two block so they will stay lined up as I drill them.

I just drilled a 1/8" hole about 5/8" from the top. The block being clamped in line will cause the holes to line up in final assembly.

I want to make a snug fitting hole for the former so I select a bit to fit.

I set the drill press to drill half way through.

Another view to show the bit stop is set half way.

The holes are drilled and the fit tested. It is snug.

This is what it will look like when assembled. The reason for drilling both pieces together was to keep the former level.

My rod is a 1/4" brass screw. I need one end threaded to match it.

If you don't have a sink supply and don't want to buy one you could use whatever is on hand. Here I show the core will fit a Bic pin barrel.

I threaded the end cap and have started the core into it.

You could but a brass screw that is threaded. I'm using a screw left over from a plumbing job and had to thread it.

The final assembly except for a knob.

I save whatever I think I might use later. This is a knob from the junk box. The opening is flat on two sides. No problem a few strokes with a file and it fits snugly.

The knob on the shaft.

This is what the finished product looks like. The only thing left is how to mount it.

I drilled a small hole in the bottom of each block for mounting screws. While I was at it I drilled a hole in the top of each block. I decided to terminate the wires on the blocks.

I installed the pins for the wire termination.

I put a couple of self tapping screws in the bottoms.

The final product ready to wind coil and mount.

I looked in the bench drawer and found a roll of wire to wind the coil with. Wound the coil and soldered it to the terminal pins.

I'm ready to mount the coil. You could measure and mark but I just laid the coil on its side and marked the holes in line with the screws.

I have drilled the holes and placed the board over the coil to check the alignment.

I screwed the screws into the block threw the board.

The coil mounted on the board and ready for a build.

Just another view. The meter reads 16uh to 23uh.

This is my audio amp.

This is the radio complete with pilot light. There is a LED in the hole under the switch.

The back side of the board. It isn't pretty but it work

Another version of variable inductor

Varactor tuning circuit.

Here is the basic circuit with some things to consider when sizing parts. I have seen 10k pots in tuning circuits. There is really nothing wrong with this but the 100k will use 1/10th the current thus conserving battery. The 1 meg resistor should be fine with a good diode but if it is leaky it could drop voltage and cause problems. This can also be referred back to the pot. If the diode is leaky it will draw current threw the portion of the pot being tapped and cause excess voltage drop thus limiting the tuning range. I will build it as shown and see how it functions.

I built the basic circuit (RF, Detector, and oscillator) it is very strong. I used jumpers to connect some test coils and a tuning capacitor to it. I received some broadcast band and short wave. The J112s work very well. It has four knobs to turn and takes a little adjusting. More on that after I get the tuning section built.
EDIT 1: The four knobs have become 6 knobs. Now with a volume control on the AF amp and the slug tuning knob.
EDIT 2:  The slug tuning gives about 100KHz for two revolutions. With a one inch or larger knob it gives a sensitive adjustment.

The tuning system starts here with a slug tuned coil.

You need two blocks about 1" square and 1-1/2" long. I used plastic. Stack the blocks and drill a hole where the green line is. This hole needs to be sized to match your core material. I'm using a 1/4" brass screw. I drilled the hole and tapped one block. My former is a sink supply line. The line is 3/8". Drill each block about 1/2 way through with a 3/8" drill using the holes you have drilled as a guide. Stand the blocks on end and drill a hole in the center of each block.(where the red rectangles are). These holes will mount the blocks to your chassis. Tap the holes for mounting screws.(I used 1/4" can be whatever you have on hand.) The core needs to be cut to length.(about 4" should do.) I just marked the core about 1" from one end and another mark 2" away. Wind the coil between the two marks. The coil value will vary with the size wire used. I used #26 wire and the 2" length coil tunes around 3 Mhz with a 120pfd capacitor. I mounted a three terminal block on one mounting block to terminate the wires . (this is optional otherwise you need leads long enough to reach your circuit.) Now you put the coil in the blocks and measure the distance between the mounting holes. Drill two holes in your chassis where the coil is to be mounted and lay the coil on the chassis and use two screws to secure it. Screw your core into the mounting block and the coil is done. Next we assemble and mount the tuning diodes. Another way of doing the coil is found here. Tuning a coil with a brass screw

The tuner with pictures

JFET regen short wave receiver

I finally decided on short wave receiver to build. This is it.

This the RF amp, detector and AF amp. I will add a tuning circuit and it's a done deal. I'm planning diode and slug tuned coil tuning. I pruned the circuit and am piecing it back together with my modifications. Q3 is the RF amp and it receives power through the tuning coil which I cut off. (Don't want to create confusion, I'm piecing together stages I hope will perform well together.)

This one uses J310s. I will be using J112s. I made tested circuits using 2SK193 and J176 JFETs. I expect the performance to be better with the J112s. The SPICE circuits seems to show better results with the J112s than the J176. I don't have a SPICE model for the 2SK193 so I will have to test them side by side. The J112 is a 'modern' transistor made for the auto industry which is readily available. My first hit on Ebay was 20 for $5.25. There may be better offers on Ebay I looked no further. Arrow Electronics sells them for 5.6 cents each. The catch is you must order in 1000 increments. The 2SK669 are 2 cents each. Still $20 a thousand isn't bad. My plan is to build with parts which are readily available so I'll be using diode and\or slug tuning.

Russian rod tube circuits and data

Here is a link to a regen build
I have a couple of booklets which are in PDF format. The program doesn't allow PDF so I converted them to image files. Hope they are readable.