Comments for "Tube Regenerative Receiver"

8th December 2014 17:09

Kim VK5FJ wrote...

Hi Alan,

I was wondering about the current through the heater.

I'm building something similar with a 12AT7, but using a large ish 7AH SLA.

As you have no current limiting in the heater circuit I was wondering about power supply options.

The 12AT7 wants 150milliamps, so thinking 80ohms for current limiting.

Did you get around to building that more permanent power supply?

regards and 73,



PS this is that I'b basing mine on;

26th March 2012 01:08

Simo wrote ...

Hi Alan!Nice project!i was looking for something like it for ages,since I found two battery pentodes!I have a question-How many turns should the tickler coil be and how should I space it from the tuning coil and is the ferrite core a must have or I can go without t?Thanks!

21st June 2008 22:31

John Hunter wrote...

Alan, in terms of transconductance the valve is indeed similar to a FET, but in this circuit the grid and cathode behave like a diode which is how it demodulates AM. A FET does not have the same diode characteristics so it is usually used as a class B detector, operating almost at cutoff by means of a source resistor. You certainly don't want any bias with a valve grid leak detector as it will effectively reverse bias the 'diode'. With voltages above about 45, the plate load resistor is the only thing limiting the current so it can't be too low. As far as the plate circuit goes, it functions like any other RC amplifier; higher values of plate resistor give higher gain, but practically speaking you can't go beyond a certain point due to the input impedance of the following stage and/or the output voltage swing being limited by the plate voltage. In practice, for this kind of circuit, plate resistors are usually between 47K and 270K. Incidentally, the grid resistor is higher than usual; normally around 1 to 3.3M is used here. However, it is worth experimenting to find the optimum values for these two resistors. They are by no means critical. You may also have to increase the feedback winding turns to compensate for lower plate current.

To work out the optimum plate resistor theoretically for a grid leak detector is not something I've seen done as curves are normally given for class A operation.

Hope this helps; John.

13th June 2008 11:15

Alan Yates wrote...


How do you calculate the correct value?

I assume it is a lot like a JFET? Normally I'd have a source/cathode resistor in there to set the bias that way, then calculate the drain/plate load needed for the load line I wanted.



12th June 2008 19:54

John Hunter wrote...

The 2.2K plate load resistor is way too low for this circuit. It might be ok for a transistor, but you'll get higher output by increasing it. I would recommend something between 47 and 100K. Of course the higher output resistance means the following audio stage needs a high input impedance.

Leave a comment on this article.