ground switch control

[bent]

hello,

i have a switch that simply make on/off to ground...

I would like to cut the ground line when my LED is ON

-----------  switch  ----------x-----------  GND
                                      LED
so if LED is ON- the ground is cut and the switch is not working,
but when the LED is OFF-then the switch can work...
I could use a relay with the gnd activated by a NPN transistor, but it will use too much current...
Can it be done by using only transistor or optoisolator?

thanks

bent

[GibsonGM]

More of an idea what the heck you're doing would be good, Bent    

What kind of switch you talking about?  Bypass?    Don't keep us in the dark here! 

[bent]

Ho boy... my problem is to explain.....

the switch is a on/off for a effect, i want to use the LED ( ON ) to prevent the switch to be reactivated by another device (long story)....

so  how could i use the 2.5v or the LED ON to cut the signal from touching ground, and when the lED is OFF the signal can touch the ground and switch is working....

[GibsonGM]

Can you draw what you want to do?   Is it a push button switch with FETs that do the actual switching, or a mechanical type that latches?     A switch shouldn't be wired in such a way that another device CAN trigger it....just can't picture what you want to do.   There are ways around this sort of thing, yes, but a clear idea of what the whole thing has to do is pretty much required

If I can't see what you want to do, I can't give you a GOOD way to do it.  I could tell you how to add reverse bias to an LED, and control that, and a bunch of other things that may not do anything like what you want!

Draw your idea out, a simple block diagram will probably do...

[bent]

First: Thank's GibsonGM for trying to help me !  

There's 4 switch on the DM-4 (line 6). Each switch have +/-3 volt on one side , and ground on the other... so when it makes contact, the voltage is sunk to ground momentary , that's how the pedal know wich switch is activated....So if you activate switch #3, LED #3 is ON....
I want to use each LED ( or the 2.5 volt on the anode) to cut the signal that goes from one side of the switch to the ground.

+/-3V  ------> switch ---------X------------> ground
                                        LED
                        or the 2.5 volt from anode
                          cut the signal when ON

Why? Because i have a memory system that can replace the switch , but if it send the switch again, the pedal go OFF.... welll this part is hard to explain...anyway...
i try to use a PNP transistor 2n3904, but when the 2.5 volt (from the anode LED)  is apply to B , the ground pass from C to E.
I would like to make the opposite: when volt is apply to B, ground doesn't pass from C to E ... and when no voltage is apply , the ground pass from C to E..
Is there somesort of transistor or photocoupler that can do that? when power is apply , it close the signal?

bent

[GibsonGM]

OK, well, that is a little more clear, but not as much as I had hoped for.  See - I don't know what is happening to your SIGNAL when you use the switch.    We are only talking about switching voltages here, they are control voltages and not audio, you see.    It must be turning on and off FETs to allow signal to pass or not pass.  Like in a Boss pedal, perhaps.     I'm no expert on those systems, being an old mechanical type builder

Are you asking how to hold the 'switch' (and LED) HIGH when you apply another signal from some other device directly to the switch?   That means more design info is needed: is the memory going to be in use all the time, or only some time?  What voltage level does it output - how does it function?
It's not enough to just connect them together, as you found out...

If you can find a schematic of the Line 6 unit, I think you'll find that the switches are not a major factor here - they send CONTROL voltages to FETs and so on, which do the actual switching electronically.    Now - you CAN hold them high, with some creativity - but that may not be the best way to do what I suspect you might be trying to do.   That might be more involved, requiring you to work right on the PCB with those switching FETs (if you even can, as they are probably SMD - tiny devices). 

.......>>>   2N3904 is NPN......2N3906 is PNP and will do the opposite.   Depending on how you have it wired, and to what device, you will find that taking the base H results in the transistor turning on (collector current flows).      A PNP will do the opposite....base H will take the collector path low.   There are ways to use each type in its opposite fashion, though....

Again, I can't see if you are trying to use this as a common emitter amplifier to control something, or a basic switch IN the collector path or not.

Need more info, Bent!!  Sorry.   This is the hard part - thinking of your design, EXACTLY how something is operating....   

[bent]

exactly this:   

from this page: http://hk.renesas.com/products/opto/technology/laboratory/index.jsp 
the "Break" type ("b" contact:Normally closed)

[bent]

exactly this:   

from this page: http://hk.renesas.com/products/opto/technology/laboratory/index.jsp 
the "Break" type ("b" contact:Normally closed)

what would be the part number? searching and can find it.... 

[GibsonGM]

They look like a supplier...have you written to them??    This is probably a package, like a relay, that you buy!    You could try a work-around with MOSFETs if you wanted, but it's probably easiest to contact them, in my opinion.

[GGBB]

I won't pretend that I understand everything here, but what I hear is that you want to prevent a connection to ground from being made while current is flowing through an LED. If you direct the current through the LED so that it also controls a normally closed relay, and that relay is between the switch and the ground, wouldn't that solve the problem?