[sdiy] CA3280 VCA Design

[James Patchell]

Sorry about that last message...forgot to type in the content.... :-)

for those interested...take a look at :
http://www.oldcrows.net/~patchell/synthmodulesII/200-1007.pdf

I tried to FTP the VCA I did up to the web sight, but I can't seem to get 
an FTP link this morning for some odd reason.  I will try again later.

At 08:22 AM 11/21/2005 -0600, Grant Richter wrote:
>Terry Michaels told me about a design he saw in a trade magazine. It
>uses a second section of OTA to predistort the input and fully
>compensate for the input non-linearity.
>As I recall, you run the voltage straight in through a resistor, tie
>the first (predistortion) section back to it's own input and into the
>input of the gain control cell.
>The first section will generate the compensation curve needed to
>produce a full level linear output after the gain control block,
>without input attenuation.
>This allows you to run full +/- 10 volt signals through the OTA. The
>monolithic matching of the LM13700 sections should allow the
>compensation to be perfect.
>
>Anybody else see that article?
>Maybe someone said Jim had already posted this somewhere as a
>subcircuit in a drawing?
>
>It seems like you could use a dual LM13700 like a single CA3280 and
>get maximum S/N from the LM13700 (get your 60 cents worth).
>
>Otherwise the input diodes on the LM13700 allow you to input +/- 500
>mv at around the same THD as +/- 20 mv (10K/500 vs. 100K/220)
>
>Also, who still sells CA3280s? I should buy some before they
>disappear altogether. Thank you.
>
>On Nov 20, 2005, at 1:32 PM, James Patchell wrote:
>
>>On the CA3280E, you don't have a choice...you have to use the
>>current sources.  So...it would be difficult to compare.  The
>>LM13700, you have choices, at the expense of adding more parts.
>>However, with the LM13700, you are starting out with an inferior
>>part (oh...I shouldn't call it that...it is the last remaining DUAL
>>OTA :-).....personally, a few dB isn't going to really make a whole
>>lot of difference...although, soon, I guess I will be exploring
>>this since I will be looking into ways to get the most of what I
>>want out of those parts.
>>
>>One of the main reasons you get better S/N with the linearizing
>>diodes is because you can drive the input harder (more signal), and
>>still get the same distortion level that you would get with a
>>standard resistor attenuator on the inputs (which has to knock the
>>signal down quite a bit, depending on how much distortion you are
>>willing to live with).
>>
>>At 11:10 AM 11/20/2005 -0800, Jay Schwichtenberg wrote:
>>>One question I have is about using the linearization diodes.
>>>Obviously
>>>you'll get some SNR improvement by using the diodes. But, I've
>>>gotten the
>>>impression (I can't remember if it was from the data sheet or the
>>>Electronotes discussion) that if you drive the diodes input with a
>>>current
>>>source vs. a resistor that you can get a few extra db on the SNR.
>>>I think
>>>that might be what Paul does with the MOTM VCAs.
>>>
>>>Any comments?
>>>
>>>Thanks.
>>>Jay S.
>>>
>>>
>>>-----Original Message-----
>>>From: owner-synth-diy at dropmix.xs4all.nl
>>>[mailto:owner-synth-diy at dropmix.xs4all.nl]On Behalf Of James Patchell
>>>Sent: Sunday, November 20, 2005 10:26 AM
>>>To: Rob Keeble; synth-diy at dropmix.xs4all.nl
>>>Subject: Re: [sdiy] CA3280 VCA Design
>>>
>>>
>>>You might try looking here:
>>>http://www.oldcrows.net/~patchell/archives/archives.html
>>>
>>>and here:
>>>
>>>http://www.oldcrows.net/~patchell/smb1/smb.html
>>>
>>>lots of other crap on my website as well....
>>>
>>>At 05:20 PM 11/20/2005 +0000, Rob Keeble wrote:
>>> >Hi,
>>> >I have a tube of CA3280's which I plan to turn into dual VCA's,
>>>however I
>>> >can't find a circuit design on the web...so I have cobbled one
>>>together -
>>> >which is here (37k GIF) - mainly using ideas from the SCI Prophet
>>>5..
>>> >
>>> >www.emulatorarchive.com/assets/vca.gif
>>> >
>>> >Before I put this on the breadboard, any changes I should make?
>>> >Audio and CV ins are 0 to +10V standard. VCA is linear, unity
>>>gain, and
>>> >should't invert signals. Only one VCA shown...
>>> >
>>> >Many thanks
>>> >Rob
>>> >www.emulatorarchive.com
>>>
>>>          -Jim
>>>***************************************************************
>>>http://www.oldcrows.net/~patchell
>>>
>>>***************************************************************
>>
>>         -Jim
>>***************************************************************
>>http://www.oldcrows.net/~patchell
>>
>>***************************************************************
>

         -Jim
***************************************************************
http://www.oldcrows.net/~patchell

***************************************************************