<<

Accurate 3.7 MHz 0dBm / 50 Ohm signal generator
for calibrating S-meters.
pa0nhc 20180803.

Copyright. The use, copy and modification of all info on this site is only permitted for non-commercial purposes, and
thereby explicitly stating my radio amateur call sign "PA0NHC" as the original writer / designer / photographer / publisher.

Preliminary development information.
My thanks to ON9BOG for all his information and support, who designed the first version,.

The set of seven PCBs for this project will become available, after evaluation of the prototype :
- one PCB for soldering the electronics,
- six more PCBs for soldering the housing.
The panels contain printed construction information.


REM :
Connect 74HC541 pin 1 to ground
.
Connect output pin to R3/4
(via is missing).

This small unit delivers a square wave signal, with a fundamental frequency of 3.686 (or 3.579545) MHz, from a source impedance of 50 Ohms. After inserting a 63dB attenuator, the fundamental frequency output power becomes -73dBm. 

With them, the S-meter of a connected receiver should indicate exactly S9+10dB on the fundamental frequency.

When using standard 1% components, the fundamental frequency output power accuracy is better than +0.5dB / -0.7dB into a 50 Ohms load.
Using accurate components, the fundamental frequency output power error could be less than 0.1dB.
The output connector is a short type BNC chassis bus.
The needed external power supply is 6 Vdc to 20 Vdc @ 14mA into a filtered  5.8mm DC power bus. Use for instance a 9V alkaline battery.

                Critical circuit details :
         The oscillator output duty cycle MUST be as close to 50/50 as possible !
A duty cycle of 55/45 causes a fundamental frequency power error of -0.22dB, and a second harmonic level of -16dBm (in stead of no even harmonics at all). The second harmonic level is an indication for the accuracy of the duty cycle, and the fundamental frequency output error.
The prototype showed a second harmonic suppression of 41.9 dBc.
Use a crystal oscillator unit with "Tight" duty cycle specification. 

         The 5.000 Vdc power MUST be as accurate and stable as possible !
An error of 1% causes an output power error on the fundamental frequency of 0,0864 dB. Measure this 5.000 Vdc voltage to estimate this output error.
The 5.00V in the prototype showed to be 1% to high.

        For R3,4,5,6 use 124 Ohm 0.1% resistors if possible.
Use resistors from the same batch. When using 1% resistors, the fundamental frequency power output error could be + or - 0.2dB.
These resistor values are difficult to measure, as metering cables and plugs easily could have  a resistance of several tenths of an ohm.
In the prototype 0.1% resistors were used.

The calculated fundamental frequency power output of the prototype is + 0.1 dBm.


Principle.

Partslist
Parts locations