LF: accurate signalsource]
Thu, 24 Feb 2000 23:11:10 -0500
Rik Strobbe wrote:
> For some while I was looking for an easy way to calibrate the s-meter of my
> Sometimes the solution is that simple that it takes some time to find it :
> with a handfull of common parts it is possible to make a 5mV into 50 Ohms
> (S9+40dB) source that needs no calibration. I did built it 5 times with
> random parts and alway came to 5mV within 0.5dB (checked with a Rohde &
> Schwartz spectrumanalyzer at work) !
> The principle is rather easy : with a 74HCT4060 a 5V square wave close to
> 136kHz is generated. The advantage of the HCT types is that they are very
> good 'rail to rail', so if you feed them with 5V you have an almost perfect
> 5V square wave. With the knowledge Mr. Fourier left us we know that a 5V
> square wave has a fundamental content of 2.25Veff. With a 2 resistor
> divider (of 1% metalfilm resistors) the signal is reduced to 5mVeff.
> (S9+40dB) into 50 Ohm.
> With the more classic 50 Ohms attenuators the signal can be further reduced
> to any wanted level.
> More informantion can be found at http://www.qsl.net/on7yd/ref136.htm
> 73, Rik ON7YD