Inexpensive function generator kits have historically been based on 8038 and XR-2206 chips. For around $20 US you could get a PC board and components to get started. I choose to build a kit # 20685 ). This kit is limited to < 100 kHz and has an output impedance of 600 ohms. Lately kits have come out based on the MAX 038 chip which is good to about 20Mhz. I noticed one for $55 on the web Kit #101 which includes a case, voltage regulators, and buffer amp. The kit I bought didn't include these items. If I was starting over, I'd give kits based on this newer chip some serious thought.
The contents of the kit are displayed to the right. The catalog I ordered from indicated that they now (fall 1998) included the external pots needed for the frequency and amplitude control. One of these turned out to be a trim pot however!
For power I found an old wall wart that put out about 20 VDC open circuit, This was feed into +12 volt and +6 volt three pin regulators. Since these regulators are not meant to sink current, I connected a resistor in series a LED to the +6 volt output to keep the current flowing in the right direction and provide a power on indicator.
A few modifications I made: The frequency is meant to be controlled by a one meg pot, I added a 50K ohm pot in series with it to make it easier to set it to a particular frequency. Added a .1uF decoupling capacitor across the power pins on the IC. I don't know if this makes a difference but it's easy enough to do. The accuracy of any frequency dial you make up depends on the values of the four timing capacitors that are used. For best results, they should be in exactly one order of magnitude steps. I used a capacitor tester as best I could to make sure that none of the capacitors stood out as higher than expected . I then left the leads long on the timing caps when I soldered them in. With the long leads, it's easy to add small caps to the bottom of the board so that the range switch steps in close to 10X increments. A six pole switch is supplied to change frequency ranges. Using one position as "off" and four positions for the supplied timing caps, there is one empty position left. It's easy to add a fifth timing capacitor consisting of a small cap and a trimmer capacitor. In theory this cap should be 100 pF, but I found that about 80 pF worked best. Stray capacitance probably accounts for the difference.
The kit will actually oscillate at several megahertz, but don't expect much of a waveform above 100 kHz,
Copyright © 2003 by Stephen M. Powell
Comments on this page? E-mail smpowell@taxspam.usa.com
(remove the obvious extra seven character political statement from e-mail address)