TheRadioBoard

Hi people!

I'm a beginner to homebrewing and I just built my first shortwave receiver - what a blast! It's a 2-transistor regenerative design from a Japanese webpage. I have some problems though, so I hope maybe you can help.

The design is from here:

http://www.excite-webtl.jp/world/englis ... N&wb_dis=2
(original Japanese webpage: http://www.puni.net/~aniki/school/radio/3.htm)

I'm living in Japan now so I was able to get all the parts exactly as listed. Instead of winding my own coil I used a pre-wound 7mhz FCZ coil, which is explained here:

http://qrpjapan.blogspot.com/2010/01/fcz-coil.html

The circuit diagram shows coil pin "2" going nowhere, but the wiring diagram shows pin 2 connected to pin 3, so I connected them.

Now, the set kind of works, but it only seems to go into regeneration when the tuning cap (one of the compact plastic types, not an air gap type - I think it's called a polyvaricon) is in the higher half of its tuning range (turned to the right). In the lower half of its tuning, I can't get it to oscillate, though the potentiometer does seem to be able to increase the gain.

Any ideas what could cause this? As a note I have pretty long leads going to the potentiometer and the tuning cap (10cm or so), and the set has hand capacitance issues. Could the long leads be causing the problem?

I built the set almost exactly as shown on the web page, using a tag board. Not sure if that's the best layout or not though.

Anyway, it's kind of working, and it was quite a thrill to be able to hear some stations. However the performance isn't what I expected from the legendary regenerative design, so I figure my layout or implementation is suboptimal.

Any hints on how to get the set to be able to go into regeneration over the whole tuning range?

Thanks!

Try increasing the value of the resistor connected from the 2SK241 FET source to ground. You can add a fixed resistor in series with the pot. Try starting with a 1k. That might get it oscillating farther down the band. If it still won't oscillate at the bottom, make the resistor larger.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Hi.

I'll take a guess - but I'm a novice at this too.

A. -- It MAY be that the set is being overloaded.

1.) I'd try putting a variable capacitor in the antenna line between the antenna and point 4. If you don't have another variable capacitor, try putting various small (10pF,or 47pF, or 100pF, etc) capacitors in the antenna line.

B. -- OR, It MAY be that not enough energy is being fed back between from point 8 to the coil 1-2.

2.) If the set still won't oscillate at some portion of the tuning range, I'd:

a. - leave a small capacitor (say 100pF or so) in the antenna line, then

b. - try varying the 47pF capacitor value between points 8 & 1. Increasing it to a higher value (like maybe 150pF or more) may help. If not, try decreasing it.

Hopefully a more knowledgeable person will jump in with better advice. Then we'll BOTH learn something.

Good luck - let us know how you make out.

73, Dan

why is this a regen?

Doesn't look like a regen to me, so it won't oscillate.

Looks like an RF amp, diode detector, and Audio amp.

Bottom of antenna coil should go to ground.

Rich

_________________
Homebrew Radio ex-Silicon Valley

It's a regen. It oscillates at the high end of the band. This is very common with regens. The tank reactance increases with frequency, and so does the Q and so does the gain. Some regens become very difficult to control at the high end, but are smooth at the bottom of the band. His FET doesn't have enough gain as biased to oscillate at the low end of the band. I had exactly the same problem with my MW regen when I swapped in a lower-gain switching FET in place of a J310, to try to make it more controllable. That's how I know to increase the source resistor. In my writeup for that receiver, I said:

http://qrp.kearman.com/html/mwregen01.html Page 2.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

If it's a regen, where is the feedback path?

Why is there a diode detector?

Rich

_________________
Homebrew Radio ex-Silicon Valley

From the original Japanese website:



Rich

_________________
Homebrew Radio ex-Silicon Valley

That Japanese circuit is kind of the same idea as this regenerative receiver:

http://home.pacbell.net/lengal/ip/amregen.pdf

By varying the source or emitter bias, you can make it go into regeneration. These circuits use a variable resistor to vary the source or emitter bias for that purpose. In the Japanese circuit, the diode detector between the regenerative detector and the amplifier stage doesn't seen to make sense to me either. It seems like that diode would cause the circuit to not operate correctly. Am I missing something here? There is a 100pf cap and a 1M resistor before the FET, which makes it a detector.

Norman
KA1GUK

Varying the source resistor changes the gain of the RF amplifier.

I suppose if you adjust for very hi gain, you can get oscillation due to stray coupling from the tank to the output of the first stage.

There was an article in Electric Radio some years ago about a "q-multiplier" front end for a regen, but even that had some kind of feedback (tapped coil, tickler, coupling cap, etc.).

Rich

_________________
Homebrew Radio ex-Silicon Valley

The referenced circuit has a feedback capacitor, C7, from base to collector.

The Japanese circuit has no feedback that I can see.

Rich

_________________
Homebrew Radio ex-Silicon Valley

Sorry, but while I was editting my post, you were sending yours. I was just adding that when you have that combination of a 100pf cap and 1M resistor before the FET, you form a classic 'gate leak' detector. This is basically the same as a 'grid leak' detector with a tube/valve.

Norman
KA1GUK

Check the layout. Plenty of feedback. It's not uncommon to use a diode detector in a regen. There are several examples, including at least one by Kitchin. I used an infinite-impedance detector in one of mine to see how it affected detector loading.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

We are jumping all over each other aren't we?

Norman

Kitchin did use a diode in his regen, but if you notice, the regen stage has a tapped tank coil for feedback.



Rich

_________________
Homebrew Radio ex-Silicon Valley