4066 Problems

[john-mike]

I hate to follow up FP's somber post with my tiny problem but it's dirving me mad.

I'm using a square wave osc to control a switch on the 4066. It works BUT when the sitch is off (the osc goes low) you can still hear the signal. I have tried caps going in and out but I really don't understand the biasing of it even after search and searching here.

Is biasing the problem?

The ground is currently conected directly from the signal source to the amp while just the + is getting switched.

Thanks

[David]

I hate to follow up FP's somber post with my tiny problem but it's dirving me mad.

I'm using a square wave osc to control a switch on the 4066. It works BUT when the sitch is off (the osc goes low) you can still hear the signal. I have tried caps going in and out but I really don't understand the biasing of it even after search and searching here.

Is biasing the problem?

The ground is currently conected directly from the signal source to the amp while just the + is getting switched.

Thanks

John-Mike:

Find the site of Andrew, "the Tone God".  Look up an article there called "Wicked Switches".  It describes how to use the 4066/4016 in electronic footswitching.  He likes the 4016.  I breadboarded the unit with a 4066.  I thought it was weird to wire up (but for me, everything is weird to wire up on a breadboard), but it worked just fine.  No noise or biasing problems that I could detect.  I just couldn't figure out how to wire the bloody thing on perf!

[john-mike]

That was the first thing I read on the whole 4066 thing.
If I wire my switch exactly like his example (I don't need any of the logic) the problem is even WORSE!
You still hear the source when the switch is off and it's distorted and louder.
This is why I tried placing caps and resistors inline with the input and output like http://members.shaw.ca/roma/switching-2.html did, but I still hear the signal.

Sould I use <9v?
Buffer the infoming with a jfet or even opamp?

Help!

[David]

That was the first thing I read on the whole 4066 thing.
If I wire my switch exactly like his example (I don't need any of the logic) the problem is even WORSE!
You still hear the source when the switch is off and it's distorted and louder.
This is why I tried placing caps and resistors inline with the input and output like http://members.shaw.ca/roma/switching-2.html did, but I still hear the signal.

Sould I use <9v?
Buffer the infoming with a jfet or even opamp?

Help!

I had high hopes for Laurier's stuff, too.  Unfortunately, I've never been able to get any of the stuff of his I tried to work.  Especially his switching circuit.  As for the logic in Andrew's circuit, yeah, you absolutely DO need it.  If you wire it properly (I used the SPST switch and 4049 example), I assure you that it will work.  You need the 4049 to produce the correct inputs for the 4066 / 4016.  Without the buffers, you won't get a bounce-free state change to trigger the 4066.

[john-mike]

Well it's good to see other peope are having problems with Laurier's stuff but I don't think my problem is one of bounce...

My Circuit is a 556, One half an astable which triggers a oneshot on the other half witch is ment to control the 4066.
Since a one shot inherintaly has no bouce do I need his logic or is it doing something else that i dont understand

Thanks

[David]

All I know is that if you follow the recipe as written, the casserole will come out fine.  I believe you're introducing unnecessary variables with the astable multivibrator thing, but suit yourself.

[john-mike]

Unfortunatley I have to have the osc's. The whole thing is meant to open an audio channel at set intervals for set amounts of time for and gallery instalation for my noise makers. http://johnmikereed.com/

[puretube]

There might be capacitive crosstalk happening....
use a second switch of the 4066, which is driven by an inverted control voltage, and which shorts the unused (off-switched) signal to ground.
(maybe wire a series-resistor of 2k2 to the output of the un-used item,
for safety-reasons, beforehand).

[mikeb]

Just a thought - have you tried measuring the resistance of the paths through the IC when it's in the different on and off states (disconnect the current audio connections first)? When I'm trying to solve problems and I'm feeling dumb (happens often ) I often just try and collect a lot of information about the issue at hand, and usually sooner or later the resolution 'bubbles' out of my subconcious. Of course, applying logic and understanding the engineering is the best approach, but if you have a 'weird situation' then my simple and dumb approach might be worth a shot.

Good luck!

Mike

[The Tone God]

Well theres a couple of different things going on here.

Lets deal with the 4066 (which is what I prefer for analog applications). I don't think you have a biasing problem when you have signal coming through in the off state. Its sound more like the switch is not closing properly or the signal is leaking through to the output in some other way.

Just to get alittle more information from a few questions could be helpful. What is the code on the IC ? Is a 74, HCT, CD, etc. 4066 ?

Seperate the 4066 from the 556 for now so we can figure out which could be causing the problem. Disconnect the 556's power supply so it doesn't even run. Have you tied all unused inputs/outputs to either ground or V+ ? Have you tried every switch in the package ? I have zapped individual switches inside a 4066. Have you tried to drive the switch control pin directly from the power rail ? Have you tried driving just a LED with the switch. If you still have problems then you might have a bummed 4066.

As for the 556 are you using a real 556 which is really two 555s in one package. If they are then you might be having problems driving CMOS with TTL logic. In which case you can can try to over come the differences using something as simple as a pull up resistor or a transistor to drive the inputs or use another logic IC (i.e. buffer, inverter, level translator, etc.) to drive the CMOS logic from the 556.

Another issue could be depending on how you have the 556 configured the output might not be going fully to 0 volts therefore not fully shutting off the 4066.

Yet another issue with 556s is that they throw noise spikes into the power supply so even if you get it working you'll probably have noise issues that are nearly impossible to get rid of.

I prefer to keep my logic stuff within the same family so use a CMOS 555 to drive the 4066. Even then you might need to put something in between to clean up the pulses.

Hope this helps.

Andrew

[john-mike]

Hey! Just the guy I wanted to talk to!

Im  using The cd4066 and the 556 is a cmos.

Even with the 4066 completley naked, just the leads from the device and the amp to pins 1&2 and 13 juct conected to ground, I still get bad distirtion.

[john-mike]

I have had the rest of the inputs grounded and the resistance in the off state is apparnently <<20M, btw.

[The Tone God]

Hey! Just the guy I wanted to talk to!

Funny most people try to run away when they see me. Oh well you'll learn.

Even with the 4066 completley naked, just the leads from the device and the amp to pins 1&2 and 13 juct conected to ground, I still get bad distirtion.

Naked! Is that allowed ?!?!? Distortion ? Is it loud or quite ?

I have had the rest of the inputs grounded and the resistance in the off state is apparnently <<20M

Just ground everything but the switch you are using.

What kind of signal are you putting through the switch ? Is the a plain guitar signal ? Boasted ? Each switch still has a resistance of about 100ohms per switch at 9v so you may not be getting enough ground. Try your signal raw and take a 100ohm resistor and tap it to gound. Does it gound out the way you want to ?

Andrew

[john-mike]

Fixed!
Their were apparently two things wrong.

The pot I was using to bring down the incoming signal was broken! The  volume was coming down but was not getting grounded.
AND!
I'm an idiot.

Seriously, thanks everybody for helping a long time lurker.
And big big thanks to Andrew for the original 4066 schematics!

[The Tone God]

Fixed!

YEAY!!!

I'm an idiot.

I just spent 45 minutes screwing with a Morely ABY pedal only to figure out I needed to reverse two wires so I'm not only a member but I'm also the president.

Seriously, thanks everybody for helping a long time lurker.
And big big thanks to Andrew for the original 4066 schematics!

Aw shucks. Glad you enjoyed them and you got your project to work. Also Welcome to the board.

Andrew