XPost: sci.electronics.design   
   From: donnyduck@gmail.com   
      
   On 2024-10-20 5:24 a.m., albert@spenarnc.xs4all.nl wrote:   
   > In article ,   
   > Cursitor Doom   wrote:   
   >> Gentlemen,   
   >>   
   >> Last week I got an old (1968) pulse generator out of mothballs and managed   
   >> to get it fully functional again. However, before replacing the case, I   
   >> (true to form) dropped it on the bench and something on the PCB must have   
   >> shorted out against the metal tools it fell on, because it no longer works   
   >> properly.   
   >> I've found an issue with the principal oscillator. It's generating   
   >> distorted sine waves. It's a wien bridge type using BJTs as the gain   
   >> element and fine tungsten filaments as thermistors, so should produce near   
   >> perfect sine waves before they're chopped and shaped by subsequent   
   >> circuitry, but since the fall, it's not.   
      
      
   The whole idea of making a sine oscillator is positive feedback with an   
   AC gain  =1  while DC biased with negative feedback and not boosting to   
   a pulse with high gain. So something must limit the gain smoothly like a   
   hot bulb with lower impedance driven by a lower voltage.   
      
      
   Below the inverting gain can be  R1/(R2 + (R(Q1)) =2 maximum unless Q2   
   is slowly turned off. and quickly turned on ;) otherwise known as fast   
   attack , slow decay.   
      
      The non-inverting side is unity gain for AC signals. So you got an   
   oscillator and the diode voltage turns off the PFET or Pch JFET. The   
   output amplitude is controlled by the gate control voltage V(AGC) which   
   can be attenuated to boost output voltage to 10Vpp with a series R   
   around 4 Meg to the 1 Meg shunt.  The FET threshold of 0.5V and the   
   diode voltage with 1Meg is only 0.4V so slightly less than 1Vp is achieved.   
      
     With a -voltage below ground must meet the FET threshold  to control   
   gain with a ground reference.   This was copied directly from LTSpice   
   examples > education.  If you understand any of what I said then you   
   recognize the differences with Hewlett Packard's old design.   
      
   Whatever is boosting the gain of your circuit or NOT cutting the gain   
   with high R must be fixed.   
      
   Cheers   
      
      
   Tony Stewart, near Toronto   
      
   EE since 1975   
      
   Learning how to retire since age 54.   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)