[ARC5] REAL regen receivers RAL/RAK.

[Kenneth G. Gordon]

On 8 Jun 2010 at 7:20, Bill Cromwell wrote:

> Hi,
> 
> I have a RAK-7 and I have been casually looking for a RAL to go with.

Hmmm....maybe we can work out a deal...

> It's definitely NOT a two toob blooper! Early experiences with some
> "beginner kit" regens and an young teenage attempt to homebrew a regen
> based on a 6SN7 made me swear off regens forever - or until I got my
> hands on the RAK. Now I'm a fan, too.

Join the club. :-)

> There are two regens here and
> another two under construction to try out some different arrangements.

Several things you need to be aware of: 1) low mu triodes make the best 
regen detectors, but proper use of tetrodes or pentodes can result in 
good ones too, 2) for maximum selectivity, use as small a grid-coupling 
capacitor, and as large a grid-leak resistance as you can, 3) regulated 
voltage to the detector is a big help for stability, 4) keep the voltage on 
the detector "low", 5) use some sort of low-gain RF amp ahead of the 
detector, even an un-tuned RF amp is better than none, 6) avoid solid-
state detectors, even FETs., 7) use a metal, grounded, panel.

Point 1 is pretty much self-explanatory
Point 2 take a bit of explaining. We have found that the normally 
suggested values of 250 pfd and 100K result in very broad tuning. 
Personally, I think this is so because the resultant load "across" the input 
tuned circuit is fairly low-impedance. We have experimented with values 
of 10pfd and 20 megohms. This resulted in a regen detector stage whose 
selectivity was such that it wouldn't pass a normal phone signal without 
significant side-band cutting.
Point 3 is pretty much self-explanatory, as is point 4.
Point 5 is necessary for several reasons: I) it keeps the antenna from 
effecting the regen point, II) it keeps the output of the oscillating detector 
from being heard as easily by others listening on the same frequency.
Point 6 is interesting: in those regennies I have built, even in identical 
circuits, the FETs provided gains up to 100 times WORSE than even the 
most elementary tube circuits. I have no explanation for this....yet.
Point 7 eliminates "hand capacity" effects.

> The RAK seems to work with converters a la Q5er operation with the
> R-23 (BC-453 - back on topic). Still would like a RAL and I'm sure one
> will show up at a hamfest one of these days.

Using the RAK as a "Q-5er" occurred to me too. Haven't yet tried it, 
though.

I have always had a great fondness for the BC-453/R-23/ARC5. They are 
really superb receivers. On the Glowbugs website, I have reprinted an 
article on a very effective converter to be used with one of those.

I also have posted a completely reversible conversion of the 12K8 stage 
to a dual-triode mixer-oscillator which reduces the internally generated 
noise to a very low level.

http://www.mines.uidaho.edu/~glowbugs/

I have also posted at least one complete AN/ARC-5 manual there in PDF 
format.

The 12K8 is possibly the noisiest "converter" tube ever designed.

One of my pet peeves is noisy mixer stages. I am a great fan of the so-
called "Pullen Mixer". Trouble is, no one uses that, nor has it been 
properly "characterized" yet. I intend to do that asap now that I am 
retired.

Ken W7EKB