[sdiy] CEM3372 passband gain trick?

Magnus Danielson magnus at rubidium.dyndns.org
Tue Aug 14 08:34:48 CEST 2012

On 08/14/2012 12:54 AM, Dave Manley wrote:
> If I keep asking innocent questions wiil you draw up the schematic, prototype and test it and send me a pcb???
> :o)

Iiiiiiin yoooooooouuurrrr dreeeeeeeaaaaammmmmssss :)

I rarely get things done is the bitter fact of life. I might provide 
ideas, on a rare occasion I do a sketch of a concept idea and might even 
put it on my webpage. Even simple stuff I came up with 10 years ago I 
haven't tried and there is obvious mistakes which other figure out when 
trying it. Ah well.


> -Dave
> -------- Original Message --------
>   From: Magnus Danielson<magnus at rubidium.dyndns.org>
>   Sent: Mon, Aug 13, 2012 01:57 PM
>   To: Dave Manley<dlmanley at sonic.net>
>   CC: Tom Wiltshire<tom at electricdruid.net>; Synth-diy at synth-diy.org
>   Subject: Re: [sdiy] CEM3372 passband gain trick?
>> On 08/13/2012 10:13 PM, Dave Manley wrote:
>>> The mini filter is discrete, so adding an opamp is easiest,  although it would interesting (perverse?) to keep it discrete.
>> Keep your perversities to yourself and your MiniMoog. :)
>> It's a two-tree transistors away, maybe four.
>> A diff-pair (like the bottom of your Moog chain), with a resistor at the
>> top of the collectors on both sides, then a current mirror on the common
>> emitter side, and then a resistor from +15V to give the current for the
>> current mirror. The bases is the + and - inputs. There, done it.
>> Counting in 4 NPN transistors and three resistors. Keep the current to
>> around 1 mA to start with, so the current resistor should be 15 kOhm.
>> Start with 6k8 resistors on each arm and see how that goes. The current
>> mirror has the collector and base connected to the 15k resistor while
>> the emittor hits the ground rail, the other transistor has it's emitter
>> on the ground rail, the base to the first transistors base (and
>> collector) and then the collector hit's the transistor pairs common
>> emitters.
>> This is a febrish head's back of the envelope sketch (without the
>> envelope!), so it hasn't been proven in battle. Using a current mirror
>> will improve performance to "just" using a resistor, for the cost of two
>> transistors. One of them may actually be replaced by a
>> Cheers,
>> Magnus

More information about the Synth-diy mailing list