Instead of the LF356s on the ASR I used CA3140AEs. Most important
parameters for a S&H buffer op amp are settling time and input bias current.
The LF356 has better settling time but as the switch remains open for the
entire clock cycle this parameter is not important in this circuit. For
typical slow rates as note shift registers the op amp input bias current is
the most important factor along with switch (4052) and capacitor leakage.
TL071 will be worse in the application than the LF356. Other factors are
board layout and parasitic capacitance from use of IC sockets. I do not
recommend sockets in this application, though I can't quantify the
contribution to the droop.
I used 47N caps, Mouser part number 1429-2473. Droop did not seem to be
significant, but I must admit I did not measure this.
Op Amp Parameters
Chip settling time input bias input offset
To 0.01% current (pA) current (pA)
Typ max typ max
LF356 1.5 us 30 200 3 50
TL071 ??? 65 200 5 100
CA3140AE 4.5 us 10 40 0.5 20
For a built in clock I recommend rates typical for note or arpeggio
generation, roughly 0.1 Hz to 30 Hz.
John Loffink
The Microtonal Synthesis Web Site
http://www.microtonal-synthesis.com
The Wavemakers Synthesizer Web Site
http://www.wavemakers-synth.com
parameters for a S&H buffer op amp are settling time and input bias current.
The LF356 has better settling time but as the switch remains open for the
entire clock cycle this parameter is not important in this circuit. For
typical slow rates as note shift registers the op amp input bias current is
the most important factor along with switch (4052) and capacitor leakage.
TL071 will be worse in the application than the LF356. Other factors are
board layout and parasitic capacitance from use of IC sockets. I do not
recommend sockets in this application, though I can't quantify the
contribution to the droop.
I used 47N caps, Mouser part number 1429-2473. Droop did not seem to be
significant, but I must admit I did not measure this.
Op Amp Parameters
Chip settling time input bias input offset
To 0.01% current (pA) current (pA)
Typ max typ max
LF356 1.5 us 30 200 3 50
TL071 ??? 65 200 5 100
CA3140AE 4.5 us 10 40 0.5 20
For a built in clock I recommend rates typical for note or arpeggio
generation, roughly 0.1 Hz to 30 Hz.
John Loffink
The Microtonal Synthesis Web Site
http://www.microtonal-synthesis.com
The Wavemakers Synthesizer Web Site
http://www.wavemakers-synth.com
> -----Original Message-----
> From: Larry Hendry [mailto:hendrysr@...]
> Sent: Saturday, May 22, 2004 9:08 PM
> To: cgs_synth@yahoogroups.com
> Subject: Re: [cgs_synth] Sample & Hold capacitor, CX, for Analog Shift
> Register
>
> Well, I plan to leave the original 2 I built stock. I am building 2
> more. I think I'll use the cap you have listed below (thanks). Like
> you, on this one I intend to socket the chips so I can compare. Once
> I know what I want, I'll go back and change the originals.
>
> Now, here is another question. I plan to add internal clocks for my
> 2nd dual ASR. I see no reason in many cases to tie up an external
> osc or clock for the job. I do intend to normal that to the jacks so
> I can still jack in an external clock.
>
> I am trying to decide what I think the useful range of the internal
> clock shoudl be. I want to scale the clock frequency control to only
> include the useful range. Since I am not an experianced ASR user,
> any input would be apprecited.
>
> Larry
>
>
> --- Richard Brewster <pugix@...> wrote:
> > Hi Larry,
> >
> > Yes, I want to use the 100n to lessen the droop. Isn't the LF356
> > better
> > here than a TL071? I am looking at the LF356 data sheet, and one of
> > the
> > features is extremely fast settling time. Must be a reason Ken
> > picked it
> > over the TL071. I socket all my CGS boards, because they are
> > experimental, so I can play around with different chips.
> >
> > -Richard
> >
> > Larry Hendry wrote:
> >
> > >The stardard recommended cap is the 47nF IIRC. I used that one,
> > and
> > >the stock op amp and I am less than satisfied with the droop. I
> > plan
> > >to change the S&H op amp to a FET input amp TL071 and probably
> > change
> > >the cap to a 100n. So, I think this you list below may be a good
> > >choice.
> > >Larry
> > >
> > >--- Richard Brewster <pugix@...> wrote:
> > >
> > >
> > >>I'm ordering parts for the ASR. Found this mouser part:
> > >>505-MKP20.1/250/5. WIMA metallized polypropylene film. It even
> > >>says,
> > >>"suited for sample & hold circuits." I think this is the right
> > >>type of
> > >>capacitor, yes? Now this is a 100nf part. There is a discussion
> > >>on
> > >>Ken's web page about this capacitor, which is specified to be
> > >>anywhere
> > >>from 47nf to 470nf. You may need to add series resistors from
> > the
> > >>4052
> > >>outputs into these caps to protect the chip. So my second
> > question
> > >>is
> > >>what values have folks used here and did they need to add
> > >>resistors?
> > >>Details appreciated.
> > >>
> > >>Thanks,
> > >>
> > >>-Richard Brewster
> > >>
> > >>