NYD,

Could you post the link to that schematic again?

Thanks,
-E

Not by much!



That's the whole point of the resistor! The signal should be down -3dB in both busses in the center position, to keep it from getting louder as it's switched from L or R to center!

(Why are we using all these exclamation points, anyway?  :wink: ).

The idea for this particular L-C-R switch arrangement did come from me, in response to a request someone made on the old board. It seemed a more usable arrangement than simply connecting the channel to both busses without any level or impedance compensation, as some designs do.
Here's my drawing:
http://electronicdave.myhosting.net/miscimages/lcr_switch.gif

If you do the math, you'll see that the impedances looking in both directions do change as the switch is changed, but not enough to upset the driving source nor significantly alter the loss of the mix buss, outside of the desired additional 3dB of attenutation in the "C" position.

deArry, your diagram looks fine at a glance. Three positions (L, C, R) wouldn't be enough for my style of mixing; but if one insists on a balanced mix buss, it's tough to find 4-pole switches with many positions. If I were building a summing buss box, I would be inclined toward unbalanced internals with transformer I/O. Then there'd be no need to worry about leakage/noise problems from all that circuitry floating above ground, and the pan control could be made much more versatile, either switched with more steps, or continuously variable. But I do understand that some might want a "resistors only" box with no transformers to color the sound, for better or worse.

A friend asked recently about an add-on mixer to implement echo sends on an old ten-channel broadcast board. Here's a sketch I made for him; essentially, it's an unbalanced version of the type of passive mixing network we're talking about here.
http://electronicdave.myhosting.net/miscimages/echosendmxr.gif

Sounds like a good idea to me.

I'll come up with an attenuation value for those "in between" positions, if you like. Or if you prefer to do it yourself, I can give you the equations.



That's for the balanced mix buss. The unbalanced mix buss doesn't need it.

Here's what I came up with. According to the AES Engineering report "Precision Pan Pots" (Richard C. Cabot, April 1978, Volume 26, Number 4), the attenuation values that correspond to displacement of 25 degrees either side of center are 0.5dB and 9dB. This should give just about as smooth a "pan" as you can expect from only five positions.
http://electronicdave.myhosting.net/miscimages/5pospanswitch.gif

The impedance seen by the driving source will vary between 0.7 and 1 * Z. The impedance reflected to the mix buss will vary between R and R + 1.8Z + driving source impedance.

Just to confirm, would these resistor values be correct for that there 5pos pan switch (based on 600R Z, 16 chan)

-0.5dB - 360R
-03dB - 2K4
-09dB - 10K8

ta

EDIT, i'm shit at maths and need to know if i'm calculating the correct values...

600 * 0.06 = 36
600 * 0.4 = 240
600 * 1.8 = 1080

Remember to halve the values if you're building a balanced network.

finally comin together  :thumb:

just need to add another 8 chan ontop of those already there!





(the brd to the right is just a JFP gaintstage - box will remain psv tho)

Looks like 'remote controlled' switches, right ?
Have seen such things been used in older Marantz hi-fi amplifiers (source-select, record-select). Nice system.

Last but not least, that box looks  :thumb:

heh, not that complicated - those are regular switches (4P6Way) for left/center/right control. The cable connects the rotary switch (on one end) to the switch contacts which are housed in that lil ting that looks like a relay...pretty good as the signal is confined to that bak and on that brd



They are alps switches - marked 905E. Got a load surplus but I think theyre pretty old as theyre not in the alps catalog I have. Guess I could let a few go tho :wink:

Cheers!

D