I want to get a tri-level video sync to to act as a rising edge trigger on an Atmel AVR. The Tri-level signal is a square wave that goes from 0V->-300mV->300mv->0mV that is typically running at 24-60hz. The actual sync point is where the -300mV->300mV transition occurs. What I would like to do is shift and amplify this wave so that it falls into a 0-5V range (2.5V->0->5V->2.5V) thus falling into the range that would trigger an interrupt on the AVR. I am a software guy and am just beginning to get my head wrapped around the hardware side of things. From what I have seen it looks like an OpAmp circuit would be the way to go but I am having trouble working out the details. Any help would be appreciated.

 

First you need to set up the opamp as an inverting output amplifier and then feed the inverted and amplified signal to the AVR.The voltage gain for the inverting configuration is set by the ratio of the feedback resistors. If your video sync is not a true square wave you may false trigger the input on the AVR. Also I don't know how the AVR input will take the negative part of the wave. You will have to experiment with it.
I am sure Google has a lot of examples of opamps.

 

You don't really have to invert the signal first.

See the attached for one possible solution.

C1 isolates the incoming signal's DC level. R1/R2/R5 allow adjusting the DC level on the noninverting input to accomodate offset. R3/R4 set the opamp's gain.

 

The problem that I see is that you are assuming only the tri-sync (which implies an HDTV signal) signal level. With a general op-amp circuit you will amplify the sync pulse right along with the active video signal and possibly exceed the voltage rating on the I/O port.
I would suggest a sync separator;e.g., LMH1981. There are separate horizontal and vertical sync outputs that you can use to take into your micro-controller.

 

Distort10n's idea is far better than what I came up with.
Here's National Semiconductors' page on the LMH1981:
http://www.national.com/pf/LM/LMH1981.html

 

Thanks guys. I had looked at the LMH1981 and will probably end up going that route. Was just wondering if there was a more straightforward approach if I knew the signal I was getting was directly out of a Denecke SB-T (http://www.denecke.com/Products/tcgenerators/tcgenerators.htm#SBT). Thanks again. I will probably be back for more...

 

I want to get a tri-level video sync to to act as a rising edge trigger on an Atmel AVR. The Tri-level signal is a square wave that goes from 0V->-300mV->300mv->0mV that is typically running at 24-60hz. The actual sync point is where the -300mV->300mV transition occurs. What I would like to do is shift and amplify this wave so that it falls into a 0-5V range (2.5V->0->5V->2.5V) thus falling into the range that would trigger an interrupt on the AVR. I am a software guy and am just beginning to get my head wrapped around the hardware side of things. From what I have seen it looks like an OpAmp circuit would be the way to go but I am having trouble working out the details. Any help would be appreciated.

How about using the LT1399?
Take a look at the datasheet on the first page they have a circuit that could generate a Tri-level sync.
Joe

 

Last post was 12.5 years ago. I doubt the TS is still around.