[sdiy] LM13700 single supply?

[Tim Stinchcombe]

Upon further study, I am now quite sure this won't work with an LM13700, not
due to any considerations of single-sided supply/whatever, but because of
the different current mirror arrangement in the LM13700 at the Iabc pins. (I
now understand the bit the eluded me earlier - the constant current source
of 10nA - but that is a red-herring as far as the expo conversion is
concerned.)

Simulating the key components using the 3080 - the input divider, Q1, 10nA
source, ideal buffer, 3080 Iabc input Q's - gives a lovely expo response up
to 80uA for Vin up to 8V. Doing the same with the LM13700 Iabc (Wilson-type)
current mirror gives a paltry current of 5nA max! The diode-connected
transistor in the emitter of the input transistor is messing things up big
time!

If you could work out how to drive the LM13700 from an expo current source
such that there is no conflict with the voltage at the Iabc pin, I see no
problem using an LM13700: however even a standard
two-emitter-connected-PNP-type expo converter (assuming split supplies too)
will give grief because the collector will barely be a diode drop above
ground, which won't be enough!

Tim
__________________________________________________________
Tim Stinchcombe 

Cheltenham, Glos, UK
email: tim102 at tstinchcombe.freeserve.co.uk


> -----Original Message-----
> From: synth-diy-bounces at dropmix.xs4all.nl 
> [mailto:synth-diy-bounces at dropmix.xs4all.nl] On Behalf Of Tim 
> Stinchcombe
> Sent: 15 June 2007 20:40
> To: flightofharmony at comcast.net; 'synth DIY'
> Subject: RE: [sdiy] LM13700 single supply?
> 
> 
> > I am experimenting with the exponential VCO from Walter
> > Jung's IC Timer Cookbook 2nd ed., p. 239.  I want to use an 
> > LM13700 in place of the CA3080, but he has this circuit set 
> > up for single-supply. Can the LM13700 be used this way? 
> 
> Having been staring at the book for half an hour, my initial 
> thought would be no, it won't work with an LM13700. I don't 
> fully understand the circuit operation, but op amp A4 is 
> actually supplying a *voltage* to pin 5 of the 3080, not a 
> current as is usual - this is backed up by the sentence 
> "Since Q1 and the internal transistor at pin 5 of the 3080A 
> have closely matched Vbe/Ic characteristics...", so he seems 
> to be relying on the interaction of these two transistors to 
> give the exponential response (and I'm also guessing that in 
> saying that transistors in the 3046 and 3080 are 'matched' 
> that he is implying that the chips are fabricated using the 
> same process...). The arrangement of transistors for the 
> equivalent pins on an LM13700 are quite different, so subbing 
> that chip I think will mess up the whole thing (generally the 
> 3080 voltage at the Iabc pin is a diode drop above ground; 
> for the LM13700 it is *two* diode drops).
> 
> A quick simulation will probably clinch it one way or the 
> other, but before going to that extent I shall wait for input 
> from other more experienced members who maybe *can* fathom 
> the circuit quicker than me, and if they disagree I might 
> then run a sim as a double-check!!
> 
> Tim
> __________________________________________________________
> Tim Stinchcombe 
> 
> Cheltenham, Glos, UK
> email: tim102 at tstinchcombe.freeserve.co.uk
> 
> 
> 
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl 
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
> 
>