Re: Desert Ratt


From: Paul Harden, NA5N (
Date: Tue Jun 01 1999 - 22:27:43 EDT

On Tue, 1 Jun 1999, Norm Melick wrote:
> Now that my desert rat is up and running, I'm curious. What
> would be some nice transistors to replace Q1, Q2, and Q4 in
> the desert rat and get some additional performance out of
> it?

Norm and others,
The Desert Ratt regen was intended to be a VERY simple, but workable
circuit for a short wave receiver. That is, minimum of components.
To do this, you want as bloody much gain as you can get out of Q1 and
Q2 particularly. In most properly designed amplifiers, you bias the
base voltage with a voltage divider (resistor to ground, another one to
+V), add some emitter resistance and a bypass cap. This standard biasing
scheme precisely sets the gain of the transistor, such that the gain is
the ratio of the collector resistor divided by the emitter resistor
(discounting the load of the next stage). Thus, you set the gain with the
EXTERNAL components, and the gain of the transistor itself (the forward
current transfer ratio or HFE) plays virtually no role.

In the Desert Ratt, the RF amplifier Q1 for example, this biasing
scheme is not used. Instead, a single resistor from the collector to
the base (from 220-470K) is used to bias the base just a hair above the
0.7V needed to make Q1 conduct. The gain of this amplifier is determined
SOLEY by the value of the transistors HFE. And for a 2N2222, 2N3904
or equivalent, HFE will be somewhere between 50 and 200! Usually 100
is estimated for most bias calculations. But if you take 5-10 different
transistors, they will all have slightly different HFE values, with one
or two of them being exceptionally poor (low HFE) or pretty hot (a high
HFE). This is why most good amplifier stages (which I'll admit, Q1 is
not) base the gain on external components, not the transistors HFE.

So the moral of the story is to try different transistors for the Q1
RF amplifier and the Q2 regen stage. After only a few, you will find
a couple that are especially "hot." In fact, you'll be amazed at the
difference in the "feel" of the regen action from one transistor to the
next. You are finding one that just happens to have a high value of
HFE. You can even play around with different NPN's if you wish.

In my original Desert Ratt, I am using a hand-picked 2N3904 for Q1,
and a hand-picked MPS918 for Q2. And like Mike Gipe, my MDS is around
-120dBm. Although, you will beable to receive several of the major
shortwave broadcasters with a MDS as lousy as -80dBm or even worse.

You can do the same for Q4, find one with a higher HFE for more audio
gain. The HFE used for Q5 doesn't matter, since it is biased properly
with the gain split between the emitter and collector resistors, and
Q3 is an emitter follower, which is used as a hi-Z in/lo-Z out converter,
and has a voltage gain of 1 (unity) - so put your lousy 2N2222 with the
crudiest HFE there :-)

Kinda like matching crystals!

GL, Paul NA5N

Search QRP-L Archives

[ QRP-L Archive | ]
[ 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 ]


This archive was generated by hypermail 2b29 on Fri Jun 02 2000 - 11:41:19 EDT