Hello,
I'm working on an audio/video switch for my video game consoles. I want it to have 16 inputs and 2 outputs. It must be able to have up to 2 systems going at once.
I made a prototype which was a modified version of this:
How-To: Make a solid-state A/V switcher
It performed perfectly when I only had one system turned on but as soon as I turned on the second one the audio on both consoles became slightly distorted. The best word I can think of to describe it is "crunchy". I think the audio from one is interfering with the other and vice verse.
Then I designed a similar setup with these chips instead:
Invalid Request
It had the same problem but nowhere near as bad. If the first version was a 10 for interference, this would be a 1. Still unacceptable though.
The prototypes were all built on a breadboard with 22 gauge wire.
In both cases the video worked perfectly.
Does anyone know of better chips for A/V switching? Or have any ideas why these dont work? I checked the outputs on a scope and the voltages are within the limits of the second chip.
I'm working on an audio/video switch for my video game consoles. I want it to have 16 inputs and 2 outputs. It must be able to have up to 2 systems going at once.
I made a prototype which was a modified version of this:
How-To: Make a solid-state A/V switcher
It performed perfectly when I only had one system turned on but as soon as I turned on the second one the audio on both consoles became slightly distorted. The best word I can think of to describe it is "crunchy". I think the audio from one is interfering with the other and vice verse.
Then I designed a similar setup with these chips instead:
Invalid Request
It had the same problem but nowhere near as bad. If the first version was a 10 for interference, this would be a 1. Still unacceptable though.
The prototypes were all built on a breadboard with 22 gauge wire.
In both cases the video worked perfectly.
Does anyone know of better chips for A/V switching? Or have any ideas why these dont work? I checked the outputs on a scope and the voltages are within the limits of the second chip.
what do you mean be decoupling them? I dont have a whole lot of background or experience in electronics.
Decoupling seems irrelevant - I can't see any amplifiers anywhere.
In the absence of the actual circuit diagram you're using, and by looking at that very primitive engadget article, I think the problem is that you're using a switch IC that is powered by +5V and 0V to switch signals that span somewhere from -1V to 1V. When a signal voltage goes below the negative supply of a 4000 series CMOS switch, all sorts of bad and odd things happen. You can do several things to fix this, all of which will take something more than a pile of wire and chips, probably some sort of perfboard at the minimum.
The basic idea is that the DC level of each signal switched by the IC either need to be raised to say 2.5V by using coupling caps and resistors, or the power supply to the switch chips needs to be changed to a dual bipolar supply, +2.5V and -2.5V, so that the signals themselves will sit (at idle) between the two supply voltages.
The second method is by far simplest, especially if you built it like the engadget article with a pile of ribbon cable soldered directly to ICs. All you'd have to do is change the supplies to the audio switching chips. This will work basically OK, but still, switching audio with 4000 series chips is pretty nasty. It can 'work' though, but the switches will change the way things sound. Not sure if it matters to you though!
There are better CMOS switches made for audio these days that have a charge pump built in. This means that they can handle +- signals with a single +3.3V and 0V supply since the chip generates its own negative supply. They are also far far cleaner sounding. The Intersil ISL54405 is the nicest of the bunch, but it's only available in tiny packages. Few or none of them are available in 0.1" leaded DIPs though, so you'll need at least some sort of surface mount prototype PCB and make a proper circuit out of it.
Best of luck...
In the absence of the actual circuit diagram you're using, and by looking at that very primitive engadget article, I think the problem is that you're using a switch IC that is powered by +5V and 0V to switch signals that span somewhere from -1V to 1V. When a signal voltage goes below the negative supply of a 4000 series CMOS switch, all sorts of bad and odd things happen. You can do several things to fix this, all of which will take something more than a pile of wire and chips, probably some sort of perfboard at the minimum.
The basic idea is that the DC level of each signal switched by the IC either need to be raised to say 2.5V by using coupling caps and resistors, or the power supply to the switch chips needs to be changed to a dual bipolar supply, +2.5V and -2.5V, so that the signals themselves will sit (at idle) between the two supply voltages.
The second method is by far simplest, especially if you built it like the engadget article with a pile of ribbon cable soldered directly to ICs. All you'd have to do is change the supplies to the audio switching chips. This will work basically OK, but still, switching audio with 4000 series chips is pretty nasty. It can 'work' though, but the switches will change the way things sound. Not sure if it matters to you though!
There are better CMOS switches made for audio these days that have a charge pump built in. This means that they can handle +- signals with a single +3.3V and 0V supply since the chip generates its own negative supply. They are also far far cleaner sounding. The Intersil ISL54405 is the nicest of the bunch, but it's only available in tiny packages. Few or none of them are available in 0.1" leaded DIPs though, so you'll need at least some sort of surface mount prototype PCB and make a proper circuit out of it.
Best of luck...
- Status
- Not open for further replies.