[sdiy] Auto Center waveforms

[Tom Wiltshire]

Sorry to repeat Eric's question, but what *is* wrong with a DC  
blocking cap?

I also don't really see a problem with Bill Forbes' suggestion of  
integrating over a number of wavecycles to get the DC level. If you  
integrate (say) 10 cycles, you'll have a delay of 0.1 sec for 100Hz  
before the DC level settles down. 100Hz is pretty low, so consider  
that close to the worst case - and a tenth of a second isn't very  
long. Or am I wrong in assuming these are audio waveforms? Is this an  
LFO?

Why does it matter so much? What have you got following this circuit  
that is so sensitive to DC offsets?

T.




On 14 Jul 2009, at 16:56, Jerry Gray-Eskue wrote:

>
> I know there are some guys out there with a heavy math background  
> (including
> at least one PHD)
>
> How about this approach, I use a Differentiating Amplifier to  
> calculate the
> Derivative of the input signal and take that output (the  
> Derivative) into an
> Integrating Amplifier to recreate the signal without the DC component.
>
> It is in the simplest form just 2 op amps 2 caps and 2 resistors  
> and runs
> Real Time with no significant delays.
>
> Is it mathematically sound?
>
>
>
>
> -----Original Message-----
> From: synth-diy-bounces at dropmix.xs4all.nl
> [mailto:synth-diy-bounces at dropmix.xs4all.nl]On Behalf Of Jerry
> Gray-Eskue
> Sent: Tuesday, July 14, 2009 8:46 AM
> To: Synth-diy at dropmix.xs4all.nl
> Subject: [sdiy] Auto Center waveforms
>
>
>
> I have been thinking about centering arbitrary wave shapes on 0 volts.
>
> Wave shapers and some VCO waveforms tend to create waveforms with a dc
> offset and it would be nice to have a circuit that would "balance"  
> these so
> that the sum of the + voltage = the sum of the - voltage.
>
> To do this it would seem that if you took the sum of one waveform  
> cycle
> using a simple integrator you would have a voltage that represents  
> the DC
> offset. Using the proper scaling you could then subtract this  
> voltage from
> the waveform to remove the offset. A pair of voltage hold circuits  
> would
> store the voltage of the previous cycle and remove the offset from the
> current cycle. The integrator (s) would reset each cycle after the  
> voltage
> hold circuits were updated.
>
> The problem that I am having with this scheme is that it appears  
> the period
> of the waveform must be used to scale the resulting voltage. Now it is
> simple enough to use a constant current into a second integrator to  
> derive a
> voltage representing the time period, and it would appear that  
> using this
> voltage and a multiplier in the feed back of an op amp will allow  
> division
> of the total voltage / period voltage.
>
> The issue with this approach is cost and complexity. It seems like  
> there may
> be a "Slick" way to do this without resorting to expensive ICs and  
> still
> give good offset removal performance.
>
> Any thoughts, or ideas on a better method ?
>
> - Jerry
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>