[sdiy] Dumb mixer circuit question

[Seb Francis]

Using a 100K pot and a 100K summing resistor to the opamp doesn't 
introduce too much gain non-linearity.  You can see a graph for yourself 
here:
http://www.mindspring.com/~clist/PotGraph.html
For those having trouble visualising the circuit Harry is describing, 
look at the 2nd one down on this page.

Using 100K pot + 100K resistor the input impedance varies from 50K at 
max gain to 100K at min gain which shouldn't have too much loading 
effect on most signal sources.  If you want more gain linearity at the 
expense of a bit more loading, using a 50K pot looks pretty good.

The way Tom described originally is better (albeit at the expense of an 
extra opamp per input) as it has a completely fixed input impedance, but 
you would still have the gain non-linearity from the interaction of the 
pot and summing resistor in the 2nd mixer opamp stage.  I guess whether 
it's worth using that extra opamp really depends what you're mixing 
together.  For line-level stuff with relatively low output impedance I 
doubt it is worth it.

Another point: if you are using something like 5532 opamps then you 
wouldn't want to be using as high as 100K summing resistor, but for 
TL072 should be fine I think.

Seb




harry bissell wrote:
> Of course there is another way...
>
> Tie the input signal to the top of the pot. Ground the other end...and 
> run the
> wiper to the input of the inverting opamp. If you keep the pot 
> impedance fairly
> high and the wiper resistance even higher... the loading on the input 
> changes very
> little. The impedance to the opamp changes but who cares...
>
> This can save opamps... the linearity of the attenuation is not 
> perfect, of course
>
> H^) harry
>
> Tom Arnold wrote:
>> On Thu, Oct 05, 2006 at 02:27:13PM -0700, Harry Bissell Jr wrote:
>>  
>>> So whats the question ???
>>>
>>> (yes you can do that no problem, its a good approach)
>>>     
>>
>> Yeah. I guess I kinda missed the question part.  I guess the question is
>> putting the pot between the mixer stages and thus increasing the 
>> complexity,
>> worth the increase in complexity?  I suppose my answer is "yes" since 
>> I've
>> proven on he bench that I've got stuff that doesnt otherwise like 
>> being tied
>> to ground across the 10k pot.
>>
>>   
>
>