Getting the most from your G4DDK VLNA - NTMS
Getting the most from your G4DDK VLNA - NTMS
Getting the most from your G4DDK VLNA - NTMS
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
<strong>Getting</strong> <strong>the</strong> <strong>most</strong> <strong>from</strong> <strong>your</strong><br />
<strong>G4DDK</strong> <strong>VLNA</strong><br />
Sam Jewell, <strong>G4DDK</strong>/W5DDK<br />
14th International EME Conference<br />
Dallas 2010
Origins<br />
• WD5AGO 23cm LNA <strong>from</strong> Microwave<br />
Update 1999<br />
– NE32584C > ATF10136<br />
• Typically 0.33dB noise figure and 28dB gain<br />
– Self supporting input inductors and capacitor<br />
to reduce loss<br />
– Stable design with typical input return loss of<br />
around 7dB
WD5AGO 23cm pre-amp
Development of <strong>the</strong> <strong>VLNA</strong><br />
Difficulty obtaining <strong>the</strong> ATF10135/6 GaAs<br />
MESFET 2nd stage device<br />
and<br />
Requirement for a lower noise figure &<br />
more gain<br />
Fuelled by <strong>the</strong> needs for better performance<br />
<strong>from</strong> my small (2.3m) EME dish
23cm <strong>VLNA</strong>
Noise figure results for 23cm<br />
<strong>VLNA</strong> 1<br />
Noise figure spread<br />
Noise figure (dB)<br />
0.4<br />
0.35<br />
0.3<br />
0.25<br />
0.2<br />
0.15<br />
0.1<br />
0.05<br />
0<br />
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21<br />
Unit<br />
Built by <strong>G4DDK</strong><br />
Measured on <strong>the</strong> same test gear over a<br />
three year period
Gain results for 23cm <strong>VLNA</strong> 1<br />
Gain spread<br />
Gain (dB)<br />
39<br />
37<br />
35<br />
33<br />
31<br />
29<br />
27<br />
25<br />
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21<br />
Unit<br />
The same batch of preamplifiers shown in <strong>the</strong> previous slide<br />
Unit 20 was a single stage pre-amplifier with ~17dB gain<br />
21 uses a different second stage.
23cm <strong>VLNA</strong> 2<br />
• Work by Sergie, RW3BP<br />
– 0.14dB noise figure based on Skobelev feed<br />
cold sky source<br />
– Unconditional stability<br />
– Good input return loss
Initial attempts to improve<br />
performance<br />
• Use MGF4919<br />
• Remove drain resistor<br />
• Increase source lead inductance
More source inductance<br />
G<br />
S<br />
S<br />
D<br />
R2<br />
0 Ohm<br />
MGF4919G ‘raised up’
How’s s that<br />
Unfortunately, not unconditionally stable<br />
Open circuit input = strong oscillation<br />
around 1100MHz<br />
RW3BP suggested more source inductance
Just about right!
Source, drain and L1/L2 for 23cm<br />
<strong>VLNA</strong> 2<br />
5mm<br />
6mm<br />
5mm<br />
L1<br />
L2
Curing any unwanted oscillation<br />
Note <strong>the</strong> two pieces<br />
of absorber material
Tune up<br />
• Set bias as described in <strong>the</strong> <strong>VLNA</strong><br />
document<br />
• Connect 50 ohm SMA termination to input<br />
• Connect output to a spectrum analyser set<br />
to cover 100MHz to 2GHz<br />
• Reference to 0dBm<br />
• Power up <strong>the</strong> preamp
Tuning without a noise figure<br />
meter<br />
1296MHz
Improvement in performance –<br />
Noise figure<br />
Noise figure spread<br />
Noise figure (dB)<br />
0.3<br />
0.25<br />
0.2<br />
0.15<br />
0.1<br />
0.05<br />
<strong>VLNA</strong>1<br />
3 different preamps<br />
MGF4919 MGF4919 MGF4919<br />
MGF4919<br />
+10R<br />
MGF4919<br />
+0R<br />
0<br />
1 2 3 4 5 6<br />
Unit<br />
1 is <strong>the</strong> reference ‘golden’ unit with NE32584C<br />
2, 3 and 4 modified with MGF4919 and 22R drain resistor<br />
5 as 4 but drain resistor = 10R<br />
6 as 4 but drain resistor = 0R
But what about gain<br />
• Applying negative feedback reduces gain<br />
• Reducing <strong>the</strong> value of <strong>the</strong> drain resistor<br />
increases gain<br />
Gain spread<br />
Gain (dB)<br />
39<br />
37<br />
35<br />
33<br />
31<br />
29<br />
27<br />
25<br />
<strong>VLNA</strong>2 with 22R 10R 0R<br />
1 2 3 4 5 6<br />
<strong>VLNA</strong>1<br />
Unit
And<br />
• Replacing R3 (51R) with a 3n3H choke<br />
also helps increase <strong>the</strong> gain
And <strong>the</strong> results<br />
At least for 23cm
Noise figure bandwidth<br />
Noise figure bandwidth<br />
Noise figure (dB)<br />
0.5<br />
0.45<br />
0.4<br />
0.35<br />
0.3<br />
0.25<br />
0.2<br />
0.15<br />
0.1<br />
1000 1100 1200 1300 1400 1500<br />
Frequency (MHz)<br />
<strong>VLNA</strong>2<br />
<strong>VLNA</strong>1
23cm <strong>VLNA</strong>2 After modifications
13cm <strong>VLNA</strong>2<br />
• Improved noise figure<br />
– Down <strong>from</strong> average 0.35 to 0.25dB (or better)<br />
• Gain is unchanged<br />
– Approximately 26 to 27dB
13cm <strong>VLNA</strong>2<br />
• MGF4919 front end<br />
– Same length source leads as <strong>the</strong> 23cm <strong>VLNA</strong>2<br />
• C1, L1 and L2 as 13cm <strong>VLNA</strong>1<br />
• T2 absorber strip not required, although<br />
T3 is required<br />
• L9 changed to 5.6nH
13cm <strong>VLNA</strong>2<br />
Position of L1 and L2 for lowest noise figure<br />
Absorber tile<br />
Too large!
13cm <strong>VLNA</strong>2 Performance spread<br />
13cm <strong>VLNA</strong>2 Noise figure<br />
Noise figure (dB)<br />
0.4<br />
0.35<br />
0.3<br />
0.25<br />
0.2<br />
0.15<br />
0.1<br />
<strong>VLNA</strong>1 <strong>VLNA</strong>2 <strong>VLNA</strong>2 <strong>VLNA</strong>2 <strong>VLNA</strong>2 <strong>VLNA</strong>2<br />
1 2 3 4 5 6<br />
Unit
13cm <strong>VLNA</strong>2 Gain spread<br />
13cm <strong>VLNA</strong>2 gain spread<br />
Gain (dB)<br />
29<br />
28<br />
27<br />
26<br />
25<br />
24<br />
23<br />
22<br />
<strong>VLNA</strong>2<br />
<strong>VLNA</strong>2<br />
<strong>VLNA</strong>2 <strong>VLNA</strong>2 <strong>VLNA</strong>2<br />
<strong>VLNA</strong>1<br />
1 2 3 4 5 6<br />
Unit
13cm <strong>VLNA</strong>2
Thank you!<br />
More information @<br />
WWW.<strong>G4DDK</strong>.COM