Hi all.

I have a FET preamp, tuned by means of two fixed capacitors and one variable
inductor that serves as an impedance transformer.

Am I correct in thinking that if there is an impedance mismatch (say the
input circuit was tuned into a 50 ohm load, then was finally presented with
say a 105 ohm load), then the preamp's noise figure would increase and the
FET would start to become artificially noisy?

Then if the FET was pushed into overload by an adjacent strong signal, the
noise would then start to reduce.


I have this odd situation on a repeater, and I can only put it down to
mis-tuned cavities. I've never come across it before though, where the
repeater actually IMPROVES in sensitivity when the transmitter keys up
(?!?!), so I would like to know what's going on before I fix it.


Thanks all,

Sam
M1FJB

In article , "Samuel
Hunt" wrote:

Am I correct in thinking that if there is an impedance mismatch (say the
input circuit was tuned into a 50 ohm load, then was finally presented with
say a 105 ohm load), then the preamp's noise figure would increase and the
FET would start to become artificially noisy?

Then if the FET was pushed into overload by an adjacent strong signal, the
noise would then start to reduce.


I have this odd situation on a repeater, and I can only put it down to
mis-tuned cavities. I've never come across it before though, where the
repeater actually IMPROVES in sensitivity when the transmitter keys up
(?!?!), so I would like to know what's going on before I fix it.

Sam-

There may be some measurable increase in a preamp's noise output, but a 2
to 1 mismatch will probably not cause such a serious degradation in its
noise figure.

Perhaps I misunderstand you. When the noise reduces due to overload, that
is NOT the same as an improvement in sensitivity. That is commonly
refered to as desensitization (desense). An amplifier's gain is reduced
to zero when it is in saturation (or oscillating).

Inadequate or mistuned cavities can cause desense. It can also be caused
by poor shielding, often from poor co-axial cable. You might benefit if
you switched to double-shielded cable. Also look for poorly attached
connectors.

73, Fred, K4DII

Hi Fred and all....

It's not due to desense. When I tune the cavities up properly, there's about
10dB noise figure on the preamp (reciever is 0.8uV sensitive). Then when it
transmits, the noise figure doesn't change.

When I tune the notches off slightly so the preamp starts to desense, the
noise figure goes down as it starts to desense, then the noise figure goes
up high as it pushes the preamp well into desense.


I'm going to retune the cavities for minimum SWR at some point anyway (we've
got problems with the frequency and a deaf repeater is actually quite a good
idea at the moment, so it's not an issue), and see what that does.


I just wanted to try to understand why it's doing this, does the FET noise
figure increase with an impedance mismatch on the input, and so pushing it
into desense reduces the amount of noise on the preamp?


It's just an interesting thing I've never come across before and would be
interested to find out a technical explanation to it all.


Thanks all,

Sam


"Fred McKenzie" wrote in message
...
In article , "Samuel
Hunt" wrote:

Am I correct in thinking that if there is an impedance mismatch (say the
input circuit was tuned into a 50 ohm load, then was finally presented
with
say a 105 ohm load), then the preamp's noise figure would increase and
the
FET would start to become artificially noisy?

Then if the FET was pushed into overload by an adjacent strong signal,
the
noise would then start to reduce.


I have this odd situation on a repeater, and I can only put it down to
mis-tuned cavities. I've never come across it before though, where the
repeater actually IMPROVES in sensitivity when the transmitter keys up
(?!?!), so I would like to know what's going on before I fix it.

Sam-

There may be some measurable increase in a preamp's noise output, but a 2
to 1 mismatch will probably not cause such a serious degradation in its
noise figure.

Perhaps I misunderstand you. When the noise reduces due to overload, that
is NOT the same as an improvement in sensitivity. That is commonly
refered to as desensitization (desense). An amplifier's gain is reduced
to zero when it is in saturation (or oscillating).

Inadequate or mistuned cavities can cause desense. It can also be caused
by poor shielding, often from poor co-axial cable. You might benefit if
you switched to double-shielded cable. Also look for poorly attached
connectors.

73, Fred, K4DII

On Sun, 26 Feb 2006 09:54:49 +0000 (UTC), "Samuel Hunt"
wrote:

Hi all.

I have a FET preamp, tuned by means of two fixed capacitors and one variable
inductor that serves as an impedance transformer.

Am I correct in thinking that if there is an impedance mismatch (say the
input circuit was tuned into a 50 ohm load, then was finally presented with
say a 105 ohm load), then the preamp's noise figure would increase and the
FET would start to become artificially noisy?

Then if the FET was pushed into overload by an adjacent strong signal, the
noise would then start to reduce.


I have this odd situation on a repeater, and I can only put it down to
mis-tuned cavities. I've never come across it before though, where the
repeater actually IMPROVES in sensitivity when the transmitter keys up
(?!?!), so I would like to know what's going on before I fix it.


Thanks all,

Sam
M1FJB


How are you measuring this? How do you hook the signal generator to
the system when you check the sensitivity with and with out the
transmitter on?

It sounds like a mis tuned duplexer but by what you are saying it is
hard to tell what you are seeing.

73
Gary K4FMX

In article , "Samuel
Hunt" wrote:

It's not due to desense. When I tune the cavities up properly, there's about
10dB noise figure on the preamp (reciever is 0.8uV sensitive). Then when it
transmits, the noise figure doesn't change.

Sam-

What frequency is this repeater on? A sensitivity of 0.8 uV doesn't sound
very good to me. On 146 MHz I'd expect that without any external preamp,
although it might be good at 1200 MHz???

73, Fred, K4DII

I think that basically is the problem.

Because the impedance is not set up properly on the cavities (the pass
plunger is a little off the correct impedance), then the preamp is operating
with a non-optimal impedance.

Desensing the preamp reduces the gain so reduces the amount of noise the
preamp is generating.


Sounds like a little lesson well learnt to me. Now on to returning the
cavities up properly.....;


Thanks all,

Sam

"K7ITM" wrote in message
oups.com...
It's important to realize that the optimum match for noise is in
general NOT the same as matching the impedance seen at the FET input.
The optimum impedance to match to is the FET's input-referred noise
voltage at the frequency of interest, divided by the corresponding
noise current. RF transistor data sheets will often list the impedance
for optimum noise figure. The difference between impedance matching
for maximum power transfer and lowest noise figure can be significant.
This is true with both bipolars and FETs.

Not sure what's going on with the change in noise as the repeater keys
up. Any chance some of the xmt RF is being rectified and changing the
bias on the preamp?

Cheers,
Tom