A Measurement of Drake2880 MMDS Downconverter
by Shiro Yoshida JI5MFZ
JAMSAT Newsletter Vol.29-No.3
Translated by JI1IZR, JH7BZR, JF6BCC
1. At The First
Recently the Drake2880 MMDS converter has become popular, especially since the launch of AO-40 satellite. We can find many articles and documents about it. But I thought that we could not make suitable modification on it while we had not known the detail about it. So I tried to measure the performance of all sections in 2880 board separately by cutting its PCB pattern and attaching measureing instruments on its 50-ohms lines.
Additionally I tried to measure the performance after I did the well-known modification. Now I think I have got the characteristics of 2880 and best design for it. Here I am introducing the result of measuring this 2880.
I measured on the condition of; (1) Making the case opened, (2) Removing the shield cover, (3) Detaching the IF-DC biasing choke coil and supplying power of +5/+10V.
2. Measuring all sections of 2880.
2.1 The performance of the first FET stage of RF sections.
I removed the first stage FET and then connected a ERA-3 matched series-L type preamp(LNA) and measured its NF and Gain.
| NF: 0.9dB | Gain: 32.5dB | @2400MHz 20degree centigrade. |
I have already known the performance of ERA-3 amp as 3dB/NF and 17dB/Gain, so I can calculate the first stage FET performance as;
I have already known the avarage loss of matching-network. So I can guess the FET performance;
2.2 RF stage.
MHz |
Gain(dB) |
Return loss(dB) |
VSWR |
NF(dB) |
2300 |
10.9 |
2.7 |
6.35 |
- |
2350 |
13.4 |
3.8 |
4.68 |
- |
2400 |
15.3 |
7.0 |
2.62 |
1.40 |
2450 |
16.2 |
11.5 |
1.72 |
- |
2500 |
16.2 |
15.1 |
1.41 |
- |
2550 |
15.8 |
14.5 |
1.46 |
- |
2600 |
15.4 |
13.0 |
1.58 |
- |
2650 |
15.0 |
12.2 |
1.65 |
- |
2700 |
14.7 |
11.9 |
1.68 |
- |
* Without shield-cover. Cutting a part of input matching-L for measuring.
(From Translator: See this! The original RF amp has a very good performance. We should not destroy this with any alternative devices! :-p)
2.3 BPF (3-pole interdigital type) section.
Pass band and insertion loss (dB).
MHz |
Original |
Extension(1) |
Extension(2) |
2300 |
- |
-4.8 |
-6.6 |
2350 |
-18.4 |
-3.9 |
-5.8 |
2400 |
-5.7 |
-4.1 |
-5.1 |
2450 |
-5.1 |
-5.4 |
-5.3 |
2500 |
-4.9 |
-8.2 |
-6.2 |
2550 |
-4.7 |
-14.7 |
-7.3 |
2600 |
-5.1 |
- |
-10.0 |
2650 |
-6.2 |
- |
-14.5 |
2700 |
-14.0 |
- |
- |
Extension(1): Adding +2mm at 1st/2nd element and +1.5mm at 3rd element.
Extension(2): Adding +1mm at 1st, +2mm at 2nd, +1.5mm at 3rd element.
This extension of elements was suggested by
G0MRF. Measuring without shield-cover.
2.4 IF stage (Gain and Return loss)
Freq.
(MHz) |
Original |
Mod(1) |
Mod(2) |
Mod(3) |
Gain(dB) |
R/L(dB) |
Gain(dB) |
R/L(dB) |
Gain(dB) |
R/L(dB) |
Gain(dB) |
R/L(dB) |
50 |
- |
- |
6.5 |
- |
16.5 |
0.5 |
16.7 |
0.5 |
100 |
- |
- |
16.3 |
- |
21.4 |
3.7 |
22.1 |
5.5 |
140 |
- |
- |
20.1 |
0 |
21.4 |
6.6 |
22.0 |
11.5 |
144 |
4.1 |
- |
20.3 |
0 |
21.4 |
7.0 |
22.0 |
12.6 |
145 |
4.8 |
- |
20.3 |
0 |
21.4 |
7.0 |
22.0 |
12.6 |
146 |
5.5 |
- |
20.4 |
0 |
21.4 |
7.0 |
22.0 |
12.6 |
150 |
- |
- |
20.7 |
0 |
21.4 |
7.1 |
22.0 |
13.5 |
200 |
21.9 |
- |
22.5 |
3.0 |
21.6 |
9.4 |
22.0 |
20.0 |
250 |
21.5 |
- |
23.3 |
9.4 |
22.1 |
11.0 |
22.1 |
12.2 |
300 |
24.1 |
- |
23.6 |
20 |
22.7 |
11.0 |
21.9 |
7.7 |
| Mod (1) - | Removed 2 pairs of red coil and chip-capacitors. |
| Mod (2) - |
Removed 2 pairs of red coil and chip-capacitors, then replaced the series
chip-capacitor (8pF) to 30pF. This is the way suggested by G0MRF. |
| Mod (3)- | Mod (2) + add a 12pF chip-capacitor between the IF output and GND. |
Consideration: Modification (1) had too much bad Return loos of 0dB! This means that there might be so much loss of IF port. Modification (2) is better but the peak is out of 2m Band. So I will recommend the style of Modification (3).
2.5 The total performance of Drake2880 with IF-Mod (2) only.
Total performance |
IF output |
MHz |
Gain (dB) |
Input Return loss (dB) |
NF (dB) |
2400 |
22.8 |
8.7 |
5.20 @20c |
2410 |
23.6 |
11.0 |
- |
2420 |
23.6 |
12.2 |
- |
2430 |
23.6 |
12.2 |
- |
2440 |
23.6 |
12.2 |
- |
2450 |
23.4 |
11.0 |
- |
2500 |
24.1 |
9.6 |
- |
|
MHz |
Return loss (dB) |
144 |
8.2 |
145 |
8.5 |
146 |
8.5 |
|
2.6 My Consideration
For the valuable use of this 2880, using a preamp before 2880 is the simple and best way I think. But in this case, the total gain will be too much so we should try a way how to reduce the gain of IF stage.
If you have any questions, please e-mail to Yoshihiro Imaishi JF6BCC in English or Japanese. Please note that Mr. Yoshida JI5MFZ will not accept any direct request, and will not have any responsibility about this English version article.
