The multiple itself will not introduce signal loss. However, if you
distribute a signal that uses the typical 1K series resistor on its
output and patch that to a lot of other inputs, there will be a small
signal attenuation, due to the voltage divider formed between that 1K
and the other input impedances. Examples:
1K output to a single 100K input. The signal is 100/101 or about 99% of
what's on the op amp output pin.
1K output to 4 100K inputs, which is equivalent to 25K. The signal is
25/26 or 96%. This is still a small attenuation, that you might not
notice unless you were driving 1V/oct VCO inputs, where the pitch
difference would be apparent. Many controllers that drive 1V/oct for
pitch have something around 100 ohm output impedance in order to
minimize this problem when driving multiple VCOs.
In short, signal loss isn't much of a problem with using multiples. If
you know the output and input impedances in your patch, you can do the math.
I never fully understood the choice of 1K as a standard output series
resistor. Electronotes (Bernie Hutchins) used 1K. I think one reason
is that you can hard patch 1K outputs together in a multiple and not
draw excessive current. Most MOTM modules use 1K. Serge uses 330 ohm
resistors here. I've seen 100 ohms used on Blacet module outputs.
Richard Brewster
http://www.pugix.comStephen Drake wrote:
> Wire all the tips together, wire all the grounds together. You can
> also do all sorts of fancy switching things using the switched jack if
> you're so inclined, like the motm mult panel.
>
> I use the bottom 8 jacks in my blacet-motm conversion panels, and make
> 2 sets of 4 multis. You can never have enough...
>
> I've never noticed any signal loss. No more so than any other
> connection in the system.
>
> On 7/17/06, ithaib <ithaib@...> wrote:
>
>> Hi all,
>> I want to make a Mult Panel. Is the correct way of doing it connecting the
>> Jack's Tip & Ground
>> to one another in a chain? Is this how the MOTM Mult is built?
>>
>> Is it true that I lose a certain amount of dB with each connection? (like
>> 3dB?) is there a way
>> around this?
>>
>> I'll appreciate any insights on this.
>>
>> Thanks,
>> Ithai B.
>>
>>