<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=us-ascii" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 8.00.6001.23562"></HEAD>
<BODY>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>Hello
Friends,</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020></SPAN></FONT> </DIV>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>You may recall a few
years ago that I built an interpolating scanner with a trapezoid generator for
controlling linear VCAs. The trapezoids were generated with a kind of
folding circuit that uses 5.1 zener diodes to limit the voltage in a couple of
places. As the incoming CV rises linearly, the circuit output rises
sharply from a little less than 0V to 5V, stays there briefly, and
then falls sharply back to a little less than 0V as the CV continues to
rise.</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020></SPAN></FONT> </DIV>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>So, I built the
original circuit with 1N4733 5.1V 1W zener diodes, and it worked
perfectly. Just the other day, I built a very similar circuit for someone,
and again used 1N4733s. However, in this circuit, instead of getting nice
well defined trapezoids that peak and ride at 5V, I'm getting narrow peaks that
don't quite rise to 5V before falling again.</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020></SPAN></FONT> </DIV>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>I have checked and
re-checked (and re-checked, and re-checked) the layout against the schematic,
and it is correct. I went to my simulation, and nothing I did changed the
general robustness of the circuit. However, then I tried replacing the
simulated 1N4733 diodes with simulated 1N5338, a 5.1V 5W zener, and -- lo
and behold -- the simulation generated wimpy little narrow peaks up to about 4V
instead of trapezoids up to 5V. So, I can only conclude that the 1N4733s
that I bought (from Small Bear) are actually 1N5338s in disguise. I
haven't yet desoldered the old zeners and tried them in the new circuit, but I'm
going to shortly. I'd bet that this will fix my circuit. (And,
before you ask, I decreased the current-limiting resistors from 4.7k in the
original circuit to 3.3k in the new circuit, to no avail -- also, the simulation
shows no effect of decreasing these resistors.)</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020></SPAN></FONT> </DIV>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>Has anyone else
experienced similar inconsistent behavior with zener diodes? How
likely is it that these 4733s are actually mislabeled 5W
zeners?</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020></SPAN></FONT> </DIV>
<DIV><FONT size=2 face=Arial><SPAN
class=311575304-30112020>Cheers,</SPAN></FONT></DIV>
<DIV><FONT size=2 face=Arial><SPAN class=311575304-30112020>Dave
Dixon</SPAN></FONT></DIV></BODY></HTML>