Amplifying and buffering chroma and luma signals to S-Video Standard Levels on a VCR?

Hello all, I have what is probably a rather simple problem for an electrical engineer or someone who knows a bit about video signal buffering or op amp selection for analog video applications, but these are not my areas of expertise unfortunately.

Basically what I’m trying to do is convert a video tape player that originally only has composite video output into having S-Video output. I’d like to be able to do this with betamax and VHS players and possibly some others. These formats store their luma and chroma data signals separately on the tape and these signals are usually kept separate within the player until they are joined into composite right at the output.

It hasn’t been too hard to actually locate valid chroma and luma signals on video boards, however, they are not at standard S-Video levels - say luma and chroma might be half of what their peak to peak values would need to be for standard S-Video and my guess is that they’ll drop even lower without some type of buffer or op amp once they are put under the usual 75 ohm load.

My question is whether there’s a (hopefully) simple circuit that I can can try to boost these existing chroma and luma signals to levels that are usable for a standard S-Video out that will have a 75 ohm termination. Ideally it’d be some type of Op Amp that buffers the signal so that loading the output down doesn’t affect the existing circuitry or input signal. I would imagine it would contain a couple of variable resistors to fine tune to the desired output levels as well.

I also understand that different op amps have different noise and bandwidth profiles and I do not know what would be ideal for these analog video formats, so I’d like specific part numbers suggested if possible and a crude drawing of how it might be wired up to try. It’d also be ideal if this can be accomplished with a single input voltage to the ICs as opposed to say BOTH plus 5V and minus 5V as some op amp applications require.

Any help is greatly appreciated!

Hi,
this could do the trick:

You can use the bias pots to set the correct DC offset for amplifying. One problem with Luma signals is that the DC point after coupling through the capacitor varies with the line content. to get around this you could replace the 1 meg resistor with a schottky diode (for low forward voltage).
I only tested this on a breadboard with a 1 meg resistor though. For testing I used a color bar image so there was no varying luminance…
As the chroma signal is a sine wave there is no problem with ac coupling at this point.
Use the potentiometers in the opamp feedback to set the gain for each signal. The gain formula is the standard one for noninverting opamp amplifiers
gain = 1 + (R_feedback/R_ground)
The capacitors going to ground in the feedback path are there to block DC voltage being amplified. They are needed because you set a DC bias at the opamp noninverting input. This forms a highpass filter so you should use larger caps here (10µf seems to be fine).
The big capacitor in the Luminance output should be this big because it also forms a highpass filter with the 75 Ohm + 75 Ohm Resistors like this:


The Cutoff frequency is: fc = 1/(2pirc) , with r = 2*75 Ohm
With the 1000µF cap you will get around 1 Hz which is low enough.

The LMH6643 is a low cost rail-to-rail, highspeed opamp. Michael Egger from anyma suggested them. He uses these for his circuits and they are really good for video stuff. However they are only available in SMD SOIC-8 format. If you want to build this on veroboard you can deadbug them.

Hopefully this does the trick :slight_smile:

Edit:
One more suggestion: the LMH6643 can take up to 12 Volts on the power supply. To get more headroom for amplification you can power it from 9 Volts. So everything thats labeled +5V in the schematic will then be +9 Volts.

3 Likes