Problems making a kill switch with a CD4053. Anyone have any ideas?

[trjones1]

I'm making a kill switch for a friend of mine.  I have it set up so that the guitar signal is passing straight through from the in jack to the out jack and I want to use the switch in the 4053 to dump the signal to ground.  He actually runs three parallel signal lines to three amps, so I'm using all three switches in the chip to connect ground to each of the lines separately.

I have 2 problems:

1. In my plans I assumed that I could choose whether the kill was engaged by a control voltage +9V or by 0V depending on which switching pin I connected to ground.  That is, if pin 4 is the pole of the switch (connected to the signal line), pin 5 is the throw engaged when the control voltage is at 0V and pin 3 is the throw engaged at 9V, I could choose whether the kill was activated at a control voltage of either 0V or 9V by connecting ground to its respective pin.  In practice, no matter whether I connect ground to pin 3 or 5, the kill only occurs when control voltage is at +9V.  This isn't a huge problem, but I want to add an LED to the circuit to show when the signal is "on" (i.e. not grounded), and this only allows the LED to be lit when the signal is, in fact, killed.  Kind of counter-intuitive.

2. the more troubling problem is that there is popping when the 4053 switches the ground in and out.  Because of the nature of the circuit, the biasing schemes outlined at geofex (http://geofex.com/Article_Folders/cd4053/cd4053.htm) aren't applicable.

Is there some solution here that I could try to fix these problems?

[R.G.]

1. In my plans I assumed that I could choose whether the kill was engaged by a control voltage +9V or by 0V depending on which switching pin I connected to ground.  That is, if pin 4 is the pole of the switch (connected to the signal line), pin 5 is the throw engaged when the control voltage is at 0V and pin 3 is the throw engaged at 9V, I could choose whether the kill was activated at a control voltage of either 0V or 9V by connecting ground to its respective pin.  In practice, no matter whether I connect ground to pin 3 or 5, the kill only occurs when control voltage is at +9V.  This isn't a huge problem, but I want to add an LED to the circuit to show when the signal is "on" (i.e. not grounded), and this only allows the LED to be lit when the signal is, in fact, killed.  Kind of counter-intuitive.

I'd have to see the actual schematic. The 4053 is three independent SPDT switches, each switch controlled by a different pin. If you're getting grounding on both 3 and 5 at a control of +9 on the control pin for pin 4's switch, I can only guess that either (a) the chip is damaged or (b) power is not being applied properly to the chip, or possibly ( c) Vee is not grounded.

2. the more troubling problem is that there is popping when the 4053 switches the ground in and out.  Because of the nature of the circuit, the biasing schemes outlined at geofex (http://geofex.com/Article_Folders/cd4053/cd4053.htm) aren't applicable.

Yep, there will be popping on a 4053 unless you can do the biasing schemes. If the circuit forbids that, and you can't use a bias/offset, then the 4053 is out for that application. Without seeing your schematic and hearing why you can't use the biasing schemes, I can't help any more than that.

Is there some solution here that I could try to fix these problems?

I'd use P-channel JFETs if the signals have to be at DC ground all the time.

[trjones1]

Thanks for the reply, RG.  I'll work up a schematic and post it later this evening.

[trjones1]

Ok, disregard this entire post.  I just realized that I can kill the 3 signal lines, plus have LED, with a normal 3pdt switch.  I had previously used a 4pdt switch to do this, which was completely unreliable.  I think the 3pdt will give me decent reliability, so electronic switching won't be necessary.