3038 Waveform Generator HELP!

[Don Tillman]

   From: Christopher List <Christopher_List at sonymusic.com>
   Date: 20 Feb 96 11:16:09 

      Date: Mon, 19 Feb 1996 20:36:31 -0500
      From: NegatvNein at aol.com

      I ordered some of those Harris 8038 precision waveform generator VCO
      chips. I built some test circuits and found that it worked quite
      nicely except for the "sine" wave which actually comes out as a
      dememented ramp type wave. 

Sounds like you don't have the resistors on pins 4 and 5 equal in
value. 

				 The datasheets were extremly open ended and
      i have been having trouble designing a circuit that fits my needs
      (i.e. signals from about .01 hz to 20 mhz, duty cycle from 1%/99% to
      99&/1% or there abouts, as well as a standard 1 volt per octave VCO
      operation) I would be grateful if someone could help with precise
      component valuesand circuit design.  Thanks

   Wow! 20Mhz! - that's a 1:200,000,000 sweep range! - I think we know
   what you mean, though :).

No, he said "20 mHz", that's milli, or .02 Hz, for a 2:1 sweep range.
:-)

The data sheet clearly states that the 8038 has a 35:1 speep range,
tweakable to 1000:1.  So this is the completely wrong chip to use for
what you're intending.

   Yes, I too have had some trouble going from data sheet to useful LFO
   with this bugger. I think you will be hard pressed to get an
   exponential converter to drive it accurately. 

Indeed, an exponential-converter-with-a-voltage-output-driving-a-
linear-VCO is truly suboptimal due to the effects of a little offset
voltage.  It's infinitely better to use the classic thermally-frobbed
transistor directly charging the cap.

						 I am just looking for
   linear voltage control in the .01 - 100hz range and haven't been able
   to do it. Seems like the voltage sweep circuit in the data sheet can't
   go that low - even with much bigger caps.

Really?  The data sheet guarantees that it'll go down to 0.001 Hz.

  -- Don