🚀 go-pugleaf

In <1992Oct9.203417.246@ole.cdac.com> ssave@ole.cdac.com (The Devil Reincarnate) writes:

>In article <9988@ncrwat.Waterloo.NCR.COM> tjgerman@53iss6.Waterloo.NCR.COM (Trevor German) writes:
>>


>   I have a similar question about op-amps.  This is what I want to do:


>  I have an input sinusoidal signal.  This will be of a real-time variable
>frequency. (That's the problem).  What I want is an inverse of this signal
>that is exactly 180 deg. out of phase with it.  If I pass it through an 
>inverting amplifier, I will introduce a delay in the signal as I pass it
>through the circuitry.  I can't live with this delay. Upto 2 deg. 

>  I need suggestions that will look at the output and input of the inverting 
>amp. and delay the output so that the input and output are *EXACTLY* 180 deg. 
>out of phase.  I can't (don't want to) touch the input. Maybe someone has 
>ideas of reeeeaaaal fast inverting amplifier designs, etc.

>  It has to be instantaneous, some sort of feedback, without any degradation
>of the signals, ie. bucket brigade and charge storage is out.  The freq. range
>is DC (ok, 1Hz) to 20 KHz.  

At 20KHz, the period of oscillation is 50uS.  If there is any time delay on 
an op-amp, it should be much smaller (several orders of magnitude smaller)
than the period in question.  The info on the 741 says something about a    
.3 uS transient rist time delay.  I assume that after the transients have
died off/etc., that any time delay will depend on the phase shift.  You said
you wanted just to invert the signal.  Well that is basically an all pass 
filter with an adjustable phase shift.  If the time delay is real critical,
you could always adjust the phase shift from 180 degrees to compensate.  This
compensation would depend on the particular frequency.  Hope this helps.

-Kip J. Mussatt

hampster@bohr.phys.ksu.edu  ||  hampster@matt.ksu.ksu.edu
--
hampster@phys.ksu.edu