LM13700 based filter

danial stocks diode at hotmail.com
Mon Mar 6 15:01:21 CET 2000

>The filter works fine and I sweep the control voltage to the OTA's from +15 
>v to -15v (I am not really sure if I
>can sweep the CV up to 15v but it works ok.)
these chips are really current controlled.. with a res on the bias pin this 
makes it vtg controlled.. the max i/p current = 2ma - so if you are using a 
15K [or higher] res on bias pin then you could go to + 15V [it references 
the bias current to v- so +15 in a +/-15V really appears as 30 V on the bias 

>The thing is that the lowpass filter works fine but the high pass gives 
>some cracks when I turn the CV almost to
>-15V (the bias current is almost zero uA then).
Resonance high? these chips are not too good with high levels of audio drive 
sometimes either... they can bleed through etc.. try reducing the res, or 
the audio input lvl

>And when I turn the CV even lower both filters stop working because the 
>outputs of the OTA's run into one of the
>supply rails. I have checked if this is because of a small voltage 
>difference betweent the inputs of the OTA and
>this indeed is the case.
If you look at the internal cct diagram of the LM13700 you'll note that the 
bias pin attaches to a transistor base.. I think this trans may be part of a 
darlington [not sure, haven't looked at this in a while] at any rate, this 
pin is always 1 or 2 Vbe's above V- when any current flows.. as such, you 
will always encounter a dead zone if driving it from simple cv into resistor 
arrangement and lower the CV right down close to V-... To get around this 
problem if necessary, I usually drive the bias pin with a voltage controlled 
current source - that way, the CV in translates directly into a current 
without need to consider the Vbe offset problem.. otherwise, offsetting with 
an opamp is just as complex.. of course you can just live with it if you 
dont want to use the extra parts, it isn't really a prob for many 

   And I was sondering what the expected tuning range is for such a filter. 
The datasheets say it's possible to tune over a
   few decades and I have designed the filter to go from 20 to 20000 Hz. But 
it doesn't sound like it is reaching the 20000 Hz.
Yes, it is possible to get this sort of range, would mean 2ma to 2ua bias 
current not a problem @all Ive made vco's with these going from 5hz - 5K 
[same range..] bear in mind that 20K is pretty high for a filteer.. either 
everything you can hear gets thru or nothing does.. again, reduce the aud 
lvls and see if that helps..
5% res are usually fine for this sort of app..

   When I have a 808 snaredrum across the filter the snaredrum stays a 
little muddy.
Get Your Private, Free Email at http://www.hotmail.com

More information about the Synth-diy mailing list