National AN-299 VCO idea (was: Re: [sdiy] Exponential converter information)

[Peter Grenader]

The Plan B uses a servo loop expo as well.  It was critical for me to go
that route in order to manifest the FM timbral response I needed. As far as
temp drift, with a tempco as the shunt (as opposed to in the feedback) if
the tranny pair is supplied adequate current we're talking 8 octaves with
about 2-3 cents drift over a few hours - totally acceptable.  If the tranny
pair is starved for current however it gets bad, as in real bad - no more
than three octaves tracking and then it bends so nonlinear it's laughable -
the forth octave grows to over an octave and a half and then gets worse from
there.  What I'm getting at it the ratio isn't the only concern, it's how
much current it supplies to the servo that I found to be the real deal.

However, my VCO only pulls about 35 mA on a bad day - that's with every
input doing something and all outputs active and patched.  Not all that bad.

- P

James Patchell wrote:

> You are misunderstanding how they are compensating for the drift...they
> actually are using a heater/temp sensor...using one transistor for
> both...in a servo loop to control the temperature of the die.  The tempco
> of the polystyrene cap is -100ppM as I recall...the servo loop itself still
> has a +100ppM drift...this is what they are canceling out...
> 
> The servo temp controller loop does work...although...it consumes a lot of
> power.  If you want good tempco cancelation...you really need to look at
> the expo converter I did....
> 
> This is one I did with an MAT-04
> http://www.oldcrows.net/~patchell/synthmodulesII/200-1007.pdf
> 
> and a 
> CA3046:http://www.oldcrows.net/~patchell/synthmodules/100-1007.pdf  this
> one has an explanation of how the temperature compensation works.
> 
> With the MAT-04 I get about 8 octaves of control to within +/- 2
> cents....the CA3046 gives me +/-2 cents over 7 octaves about...
> 
> And they are rock solid with respect to temperature...with the equipment I
> have...I cannot measure any scale drift.
> 
> 
> At 07:48 AM 6/19/2005 -0700, synth at oldmail.charlielamm.com wrote:
>> Nice links....I esp. like this one:
>> 
>> http://www.national.com/an/AN/AN-299.pdf#page=1
>> 
>> Has anyone actually messed around with this? i.e., can you really use a
>> LM3046 and the temp dependence of a polystyrene cap as they show here, to
>> cancel temp drift of a VCO?
>> 
>> Maybe if you get a bunch of caps and try each one?
>> 
>> Seems like a lot of work.....put in a cap, heat it up, see if it goes out
>> of tune....but it's an interesting idea.....
>> 
>> 
>> On Sun, 19 Jun 2005, Ray Wilson wrote:
>> 
>>> Some links:
>>> 
>>> http://www.national.com/an/AN/AN-311.pdf
>>> http://www.national.com/an/AN/AN-286.pdf#page=4
>>> http://www.national.com/an/AN/AN-299.pdf#page=1
>>> 
>>> I hope you find them useful.
>>> 
>>> Ray
>>> 
>>> 
>>> ----- Original Message -----
>>> From: "Aaron Bader" <sparked at zadzmo.org>
>>> To: <synth-diy at dropmix.xs4all.nl>
>>> Sent: Saturday, June 18, 2005 1:22 PM
>>> Subject: [sdiy] Exponential converter information
>>> 
>>> 
>>>> 
>>>> Can anyone suggest a good tutorial on exponential circuits? I can find
>>>> schematics aplenty, particularly as part of some larger vco/vca circuit,
>>>> but nothing that attempts to explain what exactly is going on in detail.
>>>> --
>>>> 
>>>> "Not a bird not a leaf not a sound/And well after the close of time"
>>>> --Front 242
>>>> 
>>>> Public Key: http://sparked.zadzmo.org/
>>>> 
>>> 
> 
> -Jim
> ***************************************************************
> http://www.oldcrows.net/~patchell
> 
> ***************************************************************
> 
>