[sdiy] Some thoughts about ladder filter circuit

[Yves Usson]

Befor going to THAT which are quite expensive, the CA3146 is the  
modern ROHS and
non-obsolete version of the CA3046. therwise with a minor redesign of  
PCB one may use
CA3083 (I am currently redesigning some of my yusynth modules with  
these) which are
non obsolete, available in ROHS and easily sourceable.

Cheers

Yves, yusynth

Le 12 déc. 08 à 10:02, Tom Wiltshire a écrit :

> Simon,
>
> Perhaps you could use the THAT Corp matched transistors in your  
> filter instead?
>
> http://www.thatcorp.com/300-series_Matched_Transistor_Array_ICs.html
>
> A couple of these chips would deal with the ladder quite nicely.
> I've also read of people using the THAT340 matched NPN/PNP chip in  
> the discrete SSM2040 designs that are around (Jürgen Haible's/ René  
> Schmitz).
>
> Regards,
> Tom
>
>
> On 11 Dec 2008, at 20:48, Simon Brouwer wrote:
>
>> Hi,
>>
>> I have some thoughts about this ladder filter implementation:
>> http://www.elby-designs.com/asm-2/vcf2/vcf2-filter-asm2-cct.pdf
>>
>> I have been matching transistors for this filter using a method  
>> described here: http://simonbr.xs4all.nl/wiki/index.php/ 
>> Matching_transistors
>>
>> I get the impression that CA3046 is not being made anymore,  
>> Intersil lists the part as inactive, and in any case Farnell  
>> doesn't have it. So I am considering to use the hand-matched  
>> transistors for all the pairs in the ladder filter. As the CA3046  
>> specified a maximum VBE mismatch of 5mV, using transistors matched  
>> to 1mV could actually be an improvement.
>>
>> However, I'd like to keep the dissipation in the expo converter  
>> transistor (Tb in the schematic) as low as possible to minimize it  
>> heating up compared to the reference transistor Ta.
>>
>> I'm a bit puzzled by R835. Its only function that I can think of  
>> is to limit the current into the ladder to about 2 mA. But  
>> couldn't you have this as well by choosing a value of ca. 5.3k for  
>> R836?
>>
>> If  R835 can be removed (i.e. shorted) then by lowering the  
>> voltage on C808 to 0.7V (change R814 to 56R and R808 to 510R) the  
>> collector voltage of Tb would become around 0V, and the maximum  
>> dissipation in Tb would be more than halved.
>>
>> I may have to look at keeping C 809 sufficiently biased.
>>
>> Apart from that, I think it would not change the operation of the  
>> filter.
>>
>> Any comments?
>>
>> -- 
>> Vriendelijke groet, Simon Brouwer.
>> | http://nl.openoffice.org | http://www.opentaal.org |
>>
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at dropmix.xs4all.nl
>> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy
>>
>
>
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy



-------------------------------------------------
Yves Usson
email :       yves.usson at yusynth.net
web site :    http://yusynth.net