Opamp for video component buffer

[wuemura]

Hello there!

I wanto to build a video component selector to use with my monitor (SyncMaster 730MP), my idea is to use simple components basically CD4053 / NE555 / OPAMP (buffer).

So, with opamp should i use or to be more specific what should be my consideration before i select he opamp. I have a bunch of TL072P if i could use it will be nice.

Thanks.

 

[audioguru2]

Hi Wellington,
A TL072 is an audio opamp. A video amplifier IC has a frequency response that is 300 times or more higher. A video amplifier has a gain of 2 and a high current output so it can drive 75 ohm cables that are terminated with the 75 ohms input impedance of the monitor.

Look for Video Amplifier or Video Buffer on the site of a manufacturer such as Texas Instruments.

Years ago I made a video switching system something like what you are doing. It was for a video conferencing system and I used video amplifier ICs from Motorola that are not made anymore. I also used the CD4053 to do the switching.

 

[wuemura]

Hi audioguru,

I was looking around but there is so many options like AD8045/AD8074/OPA693 and the list go on. So i just need to know what is the basic requirement before i can choose to use this one or that one. Now i know about the frequency, for video, high frequency is better? Or with what minimum should i work with.

This will be much more easy to me to search and choose the right opamp that i need.

Thank you.

 

[audioguru2]

Hi Wellington,
An opamp can't drive a load that is lower than about 1k ohms. A video amplifier is designed to drive a 75 ohm video coax cable. To avoid reflections of the signal in the cable then the output impedance of the video amplifier must match the impedance of the cable and the input impedance of the monitir must also match the impedance of the cable. Then the video amplifier must have a gain of 2 to compensate for the matching loss.

The highest frequency depends on the resolution you need.

 

[wuemura]

Ok, thanks.

 

[wuemura]

So i think i will choose the OPA3691:
http://focus.ti.com/lit/ds/symlink/opa3691.pdf

And like you said, i will use a 75R grounded at the input and a 75R as signal output.

One more question, can i make the CD4053 change the internal swich with a CD4017, so i could use a single button to swith the video. I'm thinking in doing a 6x1 switch 6 input and one output.

 

[wuemura]

Correcting my previous post, the better solution will be the OPA3692 and not OPA3691
http://focus.ti.com/lit/ds/symlink/opa3692.pdf

It has a fixed gain of +2. ;D

 

[audioguru2]

The CD4053 is three two-position switches. It doesn't need a CD4017 10-steps sequencer. An alternate-action circuit can toggle all three switches with a single pushbutton:

View attachment 40322

View attachment 40323

 

[wuemura]

Hi audioguru,

The idea to use the CD4017 is to use this logic sequence to trigger different CD4053 on the way, for now is just a concept.

Since i will use 6 inputs:

1) The first block will switch the signals( inputs 1-2, 3-4 and 5-6).
2) Second block will work with the logic of the signals since is only one output i have to disconnect the
output of 1-2 and connect the signal from 3-4 and so on.

The CD4017 will solve in a simple way all that mess.

1) Output CD4017 0 (NC) & 1 (trigger from input 1 to 2).
2) Output CD4017 2, trigger another CD4053 to change from output 1-2 to 3-4

And so on, i think i will need 5 CD4053 and maybe a inverter (74xx) to make all the logic work out.

What do you think?

I will make a basic schematic of this.

 

[wuemura]

So here is the basic structure (is not ready yet), so this could be more easy to explain.

The first output of CD4017 will have a inverter 74xx to make sure that all CD4053 are reseted (LOW), so all CD4053 is in X0/Y0/Z0 position, this will be channel 1

Using a push button to change state of CD4017 output 1 (HIGH) will make IC1 switch from X0/Y0/Z0 to X1/Y1/Z1, all other CD4053 stay at the same position, , this will be channel 2

Changing the state again from CD4017 output 2 (HIGH) will make IC4 switch from X0/Y0/Z0 to X1/Y1/Z1, IC2  hold position, this will be channel 3

CD4017 output 3 (HIGH) will make IC2 switch from X0/Y0/Z0 to X1/Y1/Z1, IC4  hold position, this will be channel 4

CD4017 output 4 (HIGH) will make IC5 switch from X0/Y0/Z0 to X1/Y1/Z1, IC3  hold position, this will be channel 5

CD4017 output 5 (HIGH) will make IC3 switch from X0/Y0/Z0 to X1/Y1/Z1, IC5  hold position, this will be channel 6

When the CD4017 start counting again, it will send a LOW signal to all CD4053 to reset and go back to channel 1 again.

So, this is what i was thinking, i'm open to suggestions.

To make this more easy:

Channel 1
IC1 X0/Y0/Z0
IC4 X0/Y0/Z0
IC5 X0/Y0/Z0

Channel 2
IC1 X1/Y1/Z1
IC4 X0/Y0/Z0
IC5 X0/Y0/Z0

Channel 3
IC2 X0/Y0/Z0
IC4 X1/Y1/Z1
IC5 X0/Y0/Z0

Channel 4
IC2 X1/Y1/Z1
IC4 X1/Y1/Z1
IC5 X0/Y0/Z0

Channel 5
IC3 X0/Y0/Z0
IC5 X1/Y1/Z1

Channel 6
IC3 X1/Y1/Z1
IC5 X1/Y1/Z1

View attachment 40324

 

[audioguru2]

Hi Wellington,
Your inputs and outputs are L channel audio, R channel audio and video. Your monitor has RGB inputs, not just a single composite video wire. I hope the video standards for the signals of your sources and monitor are all the same. There are many different types of video signals around the world.

Look at the CD4067 16:1 multiplexer and the CD4097 dual 8:1 multiplexer ICs.

Your pushbutton must have a de-bounce circuit or each time it is pushed and its contacts bounce then the 4017 will be clocked many counts.

 

[wuemura]

Hi audioguru,

No, no, i will not work with RGB just component video (Y'PbPr). My monitor can handle component video (Y'PbPr) and 1080i signals.

This structure is just the basic connection to work with video, the audio block still missing in there. But before i build the design to the audio block i have to solve the video block first, that is why i ask for a good opamp video buffer because of the loss of signal when it pass by the CD4053 internal switches.

Maybe i have some where a schematic for the circuit of the pushbutton using the CD4017.

But, the basic idea, do you think it will have too much signal loss?

 

[audioguru2]

Where is the loss in the video signals?
The video source is made to produce a certain level of video in the 75 ohm input of your switching circuit.
The CD4053 has a switched-on internal resistance of only about 150 ohms and it feeds the 1M or higher input impedance of the video amplifier so the loss is nearly nothing. Then the video amplifier has a gain of 2 to make up for the loss in its 75 ohm series output resistor.

Look in Google for Debounce Circuit for the clock input to the CD4017.

 

[wuemura]

I see signal loss at input 1-2 and 3-4 because it has to pass by 3 CD4053 (150R*3=450R), the input 5-6 has only a total resistence of 300, so i fear that this could mess up with the signal even if it is buffered at the output, like diference in signal at input 1-2/3-4 and less at input 5-6, but i can compensate this by adding 3  150R resistors at the output of the IC3 (pin 4, 14, 15).

Any way, i need to build this first and test latter.

I will search for this debounce circuit, thanks!

 

[audioguru2]

The opamp has a typical input impedance of 100k ohms in parallel with only 2pF. So the series 450 ohms resistance of three CD4053 switches won't cause any measurable loss.

 

[wuemura]

Good to hear (read) that!
Thanks!

 

[Shahriar]

Hi

I have Uploaded a Video Aid Project which I have downloaded from SILICON Chip Website.
I think it will help you with what you are going to do.

Download it from here
http://www.4shared.com/file/9112625/a494a2d5/Video_Aid.html

Shahriar

 

[audioguru2]

Hi Shahriar,
Thanks for the article. But it is for low quality black and white video for a TV. Wellington needs a swicher for high quality colour for a computer monitor.