dual gang slide pot supplier (uk) advice

[synergy]

i'm about to install a soundsystem in a bar in town using a behringer dcx2496 and ep2500 amplifiers

in an attempt to offer them zone control i'd like to be able to insert a balanced volume control for each channel between the controller and amps (the amp rack will be out of reach and i'll be using a multicore mic cable to give 'send and return' to the control box by the bar)

i through my research on the net have been led to believe that you should match the value of the pot to the input impedance of the amplifier in this case 20kohm for balanced (is this theory correct?)

here's the rub for neatness and ease of use i'd like to use sliders and to maintain the balanced signal i'll need dual gang

can i find 20k log sliders anywhere let alone dual gang ones not a ******* chance! and i can forget about less than 60mm travel

10k is the only value in a dual gang i can find

so my question is if anyone knows of a place i can get a 20k version or at worst something above can they please shout out otherwise i'll be forced to use rotary pots


cheers Dave

[richie00boy]

Originally posted by synergy
i through my research on the net have been led to believe that you should match the value of the pot to the input impedance of the amplifier in this case 20kohm for balanced (is this theory correct?)

No not really. Passive level controls are a real pain. Ideally any source should be loaded at 10x it's impedance or more. So working backwards, if your amp has 20k input impedance, you need a 2k pot. But therein lies the rub, your mixer might well struggle to drive 2k.

The above is compounded by the fact that each amp as above will need it's own 2k pot, so if you have 4 amps the load on your mixer will be 500 ohms. That will definitely kill the sound, if not the mixer output stage eventually.

This is one of the many reasons why active/buffered level controls are just so much better.

A last resort may be to improvise a law-faked control, using 100k LIN (yes lin not log) pots assuming a 20k input impedance on the amp.

Suppliers to try:

RS
Rapid Electronics
CPC
Farnell

[synergy]

ok now i'm confused (too many years since i did electronics at college)

the dcx2496 has an output impedance of 160ohm the ep2500 has an input impedance of 20k

that's 125 times the difference

that works though because the output stage of the dcx will not be driven very hard into such a high impedance and allows for chaining of amps in parallel off one output

following this theory through if two amps were paralleled off one output the total load seen would be 10kohm

which would logically suggest i should be able to get away with a 10kohm pot which in series with the input of the ep2500 would be over 180 times the output impedance of the dcx

does this make sense or am i farting in the wind?

[richie00boy]

No read my post again, there is a specific reason why I started from the end of the chain with 20k.

The point is that if you use a 10k pot, the load on the pot must be 10x which means your amp must be 100k.

Also, for example, the DCX might have 1 ohm output impedance, that doesn't mean it can drive 10 ohms happily, due to current constraints.

[dhaen]

If you want a quality product, try:
Penny and Giles
The PGFX3000 series is available in 50K, but lin only.
They're not cheap however.

[synergy]

ok now i'm confused (too many years since i did electronics at college)

the dcx2496 has an output impedance of 160ohm the ep2500 has an input impedance of 20k

that's 125 times the difference

that works though because the output stage of the dcx will not be driven very hard into such a high impedance and allows for chaining of amps in parallel off one output

following this theory through if two amps were paralleled off one output the total load seen would be 10kohm

which would logically suggest i should be able to get away with a 10kohm pot which in series with the input of the ep2500 would be over 180 times the output impedance of the dcx

does this make sense or am i farting in the wind?

[richie00boy]

You're not farting in the wind but you are repeating yourself.

Take a read of my posts again, particularly the second one.

[synergy]

dunno how that happened at all

anyways thanks for your help (and patience, i'm being particularly stupid today) i think i get you now...

the dcx with its 160ohm ouput impedance needs to drive at least 1.6kohm impedance

a 1.6kohm pot would need to drive at least 16kohm

correct?

so in theory it would work with a 2kohm pot but the dcx driving such a hard load would probably give up the ghost sooner rather than later

back to the drawing board methinks...

the fake-law looks interesting my trouble is it has to be a passive design as i cannot power an active circuit where it will be situated

any ideas?

[richie00boy]

Yes you're more on the right track now. Although it is possible the DCX could struggle to drive 1.6k still due to current capacity of its output stage.

A fake law pot does not have to be active, in fact if you see my first post I suggested a way to do it.