KCØPET Pages
Switched Attenuator
The is another project from "Simple Test Equipment for the QRPer " by Graham Firth - G3MFJ and Tony Fishpool - G4WIF. I managed to squeeze it into my favorite enclosure. I used 1/2 watt resistors, so power handling will probably be about 1/2 to 1 watt. Most uses will not involve that much power, though.
The resistor networks are designed to attenuate the power, but still maintain a 50 ohm impedence. Here is the inside:
Accuracy. Al, WB5JNC posed a good question on this design. Some step attenuators use shielding between sections and some do not. Al was curious how much affect the lack of shielding will have on the accuracy of the attenuation. I opted not to shield the sections because the resistor networks are crammed in so tight in the Altoids tin that it was going to be difficult to provide shielding without shorting things to ground. Also I used 5% resistors because that is what I had on hand. So Al's question motivated me to do some testing to see just how accurate the "KC0PET" version of this attenuator is.
I do not have a signal generator (yet), so I used my K2 as a signal source. I applied 1 watt (this is the maximum I am comfortable the attenuator can handle) which is 30 dBm or 7.07V RMS through the attenuator into a 50 ohm dummy load. I used my homebrew Accuprobe to measure RF voltage at the dummy load. I was able to measure up to 48 dB of attenuation. Any higher attenuation resulted in voltages too low for the Accuprobe/dmm combination to provide a good reading.
Here are my results:
|
Attenuation |
dBm |
V Calculated |
V Measured |
Attenuation Measured dB |
Error dB |
|
0 |
30 |
7.07 |
7.0 |
0.0 |
0.0 |
|
1 |
29 |
6.30 |
6.2 |
1.1 |
0.1 |
|
2 |
28 |
5.62 |
5.5 |
2.1 |
0.1 |
|
3 |
27 |
5.01 |
4.9 |
3.2 |
0.2 |
|
6 |
24 |
3.54 |
3.4 |
6.3 |
0.3 |
|
12 |
18 |
1.78 |
0.17 |
12.4 |
0.4 |
|
18 |
12 |
0.890 |
0.89 |
18.0 |
0.0 |
|
24(A) |
6 |
0.446 |
0.45 |
23.9 |
-0.1 |
|
24(B) |
6 |
0.446 |
0.45 |
23.9 |
-0.1 |
|
30 |
0 |
0.224 |
0.22 |
30.1 |
0.1 |
|
36 |
-6 |
0.112 |
0.11 |
36.1 |
0.1 |
|
48 |
-18 |
0.0282 |
0.029 |
47.7 |
-0.3 |
I am pleasantly surprised at the results! Not bad for junk box test equipment (even to buy all the parts would be less than $10).
