[Elecraft] Gain calibration

[Don Brown]

Hi

DeMaw has a circuit in Solid State Design of a low level signal generator=
. It uses double sided circuit board to form a shielded box with separate=
 compartments for the oscillator and each section of the attenuator. It u=
ses feed through capacitors for the power and uses a battery to keep the =
RF from feeding up through a power supply. It has an output of around -10=
0 dbm and he goes into some detail about the shielding requirements to do=
 this without leakage.

He suggests calibrating it against a commercial generator using a receive=
r with the AGC off as an indicator. Again not something a lot of people w=
ill have access to. =20

As I said it can be done but I don't think I would want to try to kit som=
ething like this. =20

Don Brown
KD5NDB =20

----- Original Message -----
From: Bob - AG5Q
Sent: Wednesday, October 30, 2002 2:46 PM
To: elecraft@mailman.qth.net
Subject: [Elecraft] Gain calibration

A signal source, like a crystal oscillator, can be battery operated insid=
e a
metal case to  essentially eliminate stray radiation.  The signal can sta=
rt
out at an amplitude large enough to be measured by a scope for calibratio=
n
purposes (a volt or so) and then accurately attenuated so we know the
amplitude of the signal at the input to the rx.  The attenuator also crea=
tes
an accurate 50 ohm source impedance.  The same box can include a noise
generator too for filter tests.

Even if the rx input impedance isn't exactly 50 ohms, the overall gain wi=
ll
still indicate the performance when it's driven by a 50 ohm source.  The
input filter and mixer loss can be measured by increasing the input signa=
l
amplitude enough to use a scope for measurements in the early stages. 100=
mv
is okay for the first stages when the preamp is turned off.

Then the  amplitude can be reduced to the millivolt range for measurement=
s
in subsequent stages.

A fairly simple addition (NE555) to the signal source is a circuit to tur=
n
it on and off a few times a second. This would allow measurement of the
transient responses of the various filters (for example, ringing and
settling time).  The AGC circuit can be evaluated if the signal is switch=
ed
on and off very slowly.

Bob - AG5Q

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D


----- Original Message -----
From: "Don Brown" <DolfinDon@msn.com>
To: "Paul Jensen" <pj@pcrentals.com>; "Elecraft" <elecraft@mailman.qth.ne=
t>
Sent: Wednesday, October 30, 2002 10:24 AM
Subject: Re: [Elecraft] Re:New Crystals or Not


Hi

A generator to supply 1 microvolt is fairly difficult to design. It is ve=
ry
hard to keep the signal confined in the generator. The small amount of
leakage can be higher than the desired signal. It's not imposable, it's j=
ust
not easy to come up with something with the proper shielding in a kit for=
m.
There is also the problem of calibrating the signal.

Don Brown
KD5NDB


--- StripMime Report -- processed MIME parts ---
multipart/alternative
  text/plain (text body -- kept)
  text/html
The reason this message is shown is because the post was in HTML
or had an attachment. Attachments are not allowed.
Please post in Plain-Text only.---