[Elecraft] Noise Generator...idea.

[Ron D'Eau Claire]

It's important to keep in mind the KIND of band pass filter being =
peaked.

Older equipment used tunable designs in which the center frequency =
needed to
be set properly. One of the most common examples were the old
superheterodyne radios with tunable I.F. strips. If you didn't have a =
signal
generator that you could set to the proper I.F. frequency, it was an
exercise in futility to touch any of the adjustments. The same is true =
in a
lot of RF input circuits.=20

Many modern bandpass filters are quite different. They are really "fixed
tuned" but may have one adjustable element - an capacitor inductor - =
used
for "matching" the BPF to the next stage. Adjusting the cap or inductor =
has
no effect on the center frequency, but it is essential for maximum =
energy
transfer to the next stage.

In this case, a noise generator is a good choice, as Dave points out.=20

Still, it is not often a perfect choice, but it is good enough that it's
seldom worthwhile going further. It's not "perfect" because there is =
another
consideration in adjusting such filters - the "noise figure".=20

The inductor in the K2 I.F. is a great example. Elecraft says to adjust =
it
for maximum "noise" in the receiver output. A noise generator is just =
fine
for this. The crystals in the K2 I.F. define the I.F. bandpass filter
frequency, so there is no need to have a calibrated signal generator. =
Still,
the ultimate goal of this adjustment is to provide the greatest noise
figure, not the greatest gain. The problem is that you need a noise =
figure
meter and you need to know how to use it to do that.=20

The setting for optimum noise figure is VERY close to the setting for
optimum gain. It's so close that it simply does not make sense to go to =
the
expense and bother of using a noise figure meter unless one is already
sitting on the bench. The difference simply isn't detectable in normal
operation on the air. So we set it for maximum gain.

That issue becomes more critical as the frequency goes up. At VHF it can
make a significant difference whether a BPF is adjusted for maximum gain =
or
best noise figure. But, even in those situations, it is often true that =
few
operators have a station setup that would allow them to hear the =
difference.


One of the most important things for engineers and technicians to =
understand
- including Hams learning "electronics" as a hobby - is when it makes =
sense
to pursue some measurement or adjustment to a higher degree of accuracy =
and
when it is nonsense to do so.=20

Ron AC7AC
K2 # 1289

-----Original Message-----
Peaking band pass filters with a noise generator is not a good idea, as
others have said it is all too easy to peak them on the wrong frequency.
While this may be unlikely, a proper signal generator (even another
transceiver into a dummy load with a low power carrier) is the better
equipment to use. There are some quite servicable signal generators on =
the
second hand market for very little money, even if they date from the =
1960
era, provided they are of commercial grade (the Marconi TF995 comes to
mind), they are more than adequate to align a K1 or K2.

For aligning the IF filters and BFO of an Elecraft rig, a very basic two
transistor noise source is more than adequate. Mine is not even in a =
box,
has a fixed output and is switched on by connecting the battery "snaps" =
to a
MN1604 battery.

73 Dave, G4AON
K1 #1154, K2 #1892