29-05-2017, 11:11 PM
Weak signal work on the V U and microwave bands requires a low noise amp on the receiver front end. These are usually homebrewed.
How do we make measurable improvements to our low noise amplifiers .
Typically we can adjust the input inductor coil spacing and tuning capacitors, and the gasfet bias.
The way to make proper measureable adjustments to improve performance comes down to some basic test equipment.
We need a signal source and a measuring device.
The Low noise amplifier to be aligned is connected between a signal source and the radio it is to be used with, and the radio audio output is fed into a laptop or millivolt meter.
Use a noise source which can be turned on and off. These can be bought or preferably homebuilt. The easiest to build is one made from a zener diode with a wideband amp after it. HP calibrated sources are about a thousand dollars but you can make your own for a fraction of that.
You want it to have a output impedance of 50 ohms, so when turned off it presents a nice 50 ohm load. When activated it produces wideband noise 10 or 15 dB up at the frequency you are want to measure, in my case 1296 MHz.
When you connect the noise source to the input of your radio and turn it on you will notice an increase in noise. This is what you want.
Now to set up your test jig.
Set your radio to the desired frequency and upper sideband mode.
Turn off power to the noise source, connect the line output (or speaker output) of the radio to the line input or mic input of the laptop with suitable attenuation as required.
I use a IC910h ICOM as the receiver, a small laptop as the detector and a homebrewed noise source. The software on the laptop is NFM by Owen Duffy.
You can find it here.
http://owenduffy.net/software/nfm/
Install the full package, run it and you will see several coloured clickable boxes.
Click the top left green box marked 1 Noise LO with the noise source off.
Now turn on the noise source and click the green box below it marked 2 Noise HI. you will see a dB reading in the Y(2)dB box.
Look at the S meter display on the radio with the noise source on, if it is moving above zero then you will have to reduce RF gain.
On my IC910h i set the RF gain backwards so the S meter reads S9 on the bargraph. this will give enough attenuation to handle the gain of the preamp which will now be tested.
Connect the LNA between the noise source and the radio. Add a 10 to 20 dB attenuator after the LNA if you have one.
Turn off the noise source, power up the LNA and click on the top left green button 1 Noise LO.
Turn on the noise source and look at the S meter. if it has moved then you will have to reduce RF gain some more. you should hear an increase in noise
Click on 2 Noise HI button and you will see a different reading. the difference between the 2 readings will appear in the Y(2)dB box.
if you have got this far you are ready to make adjustments to the tuning of the LNA. once you get the hang of setting this up the rest is easy.
Turn off the noise source and click Noise LO, make an adjustment to the bias for example, turn on the noise source and click Noise HI.
See if the reading in the box to the right has increased. What you are trying to achieve is an increase in the difference in noise from the off state of the noise source to the on state of the noise source.
Adjust the input coil spacing a tiny bit at a time , measure and remeasure and peak the coils and capacitors(if fitted) for highest reading in the Y(2)dB box.
What I have been doing is peaking up several G4DDK 23CM VLNA2+ preamps after replacing the front end MGF4919 and the ATF54143. Sequencing malfunctions and degradation of the front end over time causes failures or reduced gain.
This adjustment technique is applicable to all sorts of V U and microwave LNA's. If you dont have a noise source you can always use the sun noise vs cold sky just by moving your antenna as you click the boxes in the NFM program.
Added in 1 hour 12 minutes 27 seconds:
How to build a noise source.
Paul Wade has written a great article which you will find here
http://www.w1ghz.org/QEX/Noise_Measureme...ration.pdf
and i recommend the second circuit, it works well.
DJ9BV does one, and there are other designs using the NC305 device. google and ebay are also good places to look.
HP make a good one, but it draws 140 millamps off 28 volts while the others are between 25 and 45 milliamps so are easily battery powered using a small ebay voltage boost converter pcb powered from a couple of lipo cells in series (8 volts).
There is so much more the NFM software can do too, its well worth the download and read of the instructions.
I will post up some pictures to assist with understanding it all.
73 Dave
How do we make measurable improvements to our low noise amplifiers .
Typically we can adjust the input inductor coil spacing and tuning capacitors, and the gasfet bias.
The way to make proper measureable adjustments to improve performance comes down to some basic test equipment.
We need a signal source and a measuring device.
The Low noise amplifier to be aligned is connected between a signal source and the radio it is to be used with, and the radio audio output is fed into a laptop or millivolt meter.
Use a noise source which can be turned on and off. These can be bought or preferably homebuilt. The easiest to build is one made from a zener diode with a wideband amp after it. HP calibrated sources are about a thousand dollars but you can make your own for a fraction of that.
You want it to have a output impedance of 50 ohms, so when turned off it presents a nice 50 ohm load. When activated it produces wideband noise 10 or 15 dB up at the frequency you are want to measure, in my case 1296 MHz.
When you connect the noise source to the input of your radio and turn it on you will notice an increase in noise. This is what you want.
Now to set up your test jig.
Set your radio to the desired frequency and upper sideband mode.
Turn off power to the noise source, connect the line output (or speaker output) of the radio to the line input or mic input of the laptop with suitable attenuation as required.
I use a IC910h ICOM as the receiver, a small laptop as the detector and a homebrewed noise source. The software on the laptop is NFM by Owen Duffy.
You can find it here.
http://owenduffy.net/software/nfm/
Install the full package, run it and you will see several coloured clickable boxes.
Click the top left green box marked 1 Noise LO with the noise source off.
Now turn on the noise source and click the green box below it marked 2 Noise HI. you will see a dB reading in the Y(2)dB box.
Look at the S meter display on the radio with the noise source on, if it is moving above zero then you will have to reduce RF gain.
On my IC910h i set the RF gain backwards so the S meter reads S9 on the bargraph. this will give enough attenuation to handle the gain of the preamp which will now be tested.
Connect the LNA between the noise source and the radio. Add a 10 to 20 dB attenuator after the LNA if you have one.
Turn off the noise source, power up the LNA and click on the top left green button 1 Noise LO.
Turn on the noise source and look at the S meter. if it has moved then you will have to reduce RF gain some more. you should hear an increase in noise
Click on 2 Noise HI button and you will see a different reading. the difference between the 2 readings will appear in the Y(2)dB box.
if you have got this far you are ready to make adjustments to the tuning of the LNA. once you get the hang of setting this up the rest is easy.
Turn off the noise source and click Noise LO, make an adjustment to the bias for example, turn on the noise source and click Noise HI.
See if the reading in the box to the right has increased. What you are trying to achieve is an increase in the difference in noise from the off state of the noise source to the on state of the noise source.
Adjust the input coil spacing a tiny bit at a time , measure and remeasure and peak the coils and capacitors(if fitted) for highest reading in the Y(2)dB box.
What I have been doing is peaking up several G4DDK 23CM VLNA2+ preamps after replacing the front end MGF4919 and the ATF54143. Sequencing malfunctions and degradation of the front end over time causes failures or reduced gain.
This adjustment technique is applicable to all sorts of V U and microwave LNA's. If you dont have a noise source you can always use the sun noise vs cold sky just by moving your antenna as you click the boxes in the NFM program.
Added in 1 hour 12 minutes 27 seconds:
How to build a noise source.
Paul Wade has written a great article which you will find here
http://www.w1ghz.org/QEX/Noise_Measureme...ration.pdf
and i recommend the second circuit, it works well.
DJ9BV does one, and there are other designs using the NC305 device. google and ebay are also good places to look.
HP make a good one, but it draws 140 millamps off 28 volts while the others are between 25 and 45 milliamps so are easily battery powered using a small ebay voltage boost converter pcb powered from a couple of lipo cells in series (8 volts).
There is so much more the NFM software can do too, its well worth the download and read of the instructions.
I will post up some pictures to assist with understanding it all.
73 Dave
. Yeh its a bit dodgy, 28 volts comes in from the right and noise comes out of the sma on the left. But it works and its a plug and pray replacement for the nice HP noise source in the first photo - lower right.