[sdiy] Filter problems

[Tim Ressel]

It WORKS! Thank you! Thank you! 2^10 times Thank you!

I would never have guessed the input bias current of the amps would be 
an issue. Live and learn!

I used the OPA1678, which is my new fav amp. It has slightly better 
noise than the 5532 for slightly less cabbage, and 10pA input current. 
Seems to be low enough ;-)

Weee!

Thanks to all!

--Tim (much relieved) Ressel


On 1/27/2018 3:34 PM, Neil Johnson wrote:
> Tim,
>
> Don't use NE5532 for the integrators (U3).  The input bias current
> will integrate over time and push the integrator opamps to the
> positive rail.  This will happen once the gain of the 2164s is low
> enough that the bias current dominates.  Switch to TL072 in that
> position.  I would also suggest a TL072 for the other op-amp as with
> the resistor values you're using and taking worst-case bias current
> you could be seeing 10s of mV (especially U2A).
>
> Also I wouldn't recommend putting the zeners across the integrator capacitor.
>
> Neil
>
>
> On 27 January 2018 at 16:28, Tim Ressel <timr at circuitabbey.com> wrote:
>> Hi,
>>
>> Filters can be a pain. I've got one that is misbehaving. Here is a link to
>> the schematic:
>>
>> https://www.dropbox.com/s/acjvwb17nmuok1n/PrismFilter.pdf?dl=0
>>
>> At low CVs it filters fine, but as the freq cv hits around 2 volts the
>> behavior gets odd. I start seeing a rising DC bias on all the outputs. At
>> about 3 volts all the outputs rail positive.
>>
>> Any help is appreciated. Thanks for playing!
>>
>> --
>> --Tim Ressel
>> Circuit Abbey
>> timr at circuitabbey.com
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
>
>

-- 
--Tim Ressel
Circuit Abbey
timr at circuitabbey.com