Analog Video with Full-wave rectifier

I am experimenting with making an analog video signal into an "on or off" state which will basically say, does it have a video signal.

Would using a full-wave rectifier with a capacitor for smoothing be a viable option? unfortunately I can't afford an oscilloscope to test video signals, so I just wanted to ask if I am on the right track. I know I can probably just use my multi-meter to read it once I have the rectifier created...but I also understand very little about video signals and circuits in general (although I am learning a ton and reading a lot)

also, would reading this just be a matter of having the output of it go into a analog Read pin on the arduino since it shouldn't go over 1v?

Thanks,

I would use just a simple single diode feeding into a capacitor with a resistor across it. Yes it will need feeding into an analogue pin, this gives readings in roughly 4mV steps so there is plenty of resoloipution to detect the signal. You are likely to see a voltage of just over 0.3V, you will not see anything like 1V

Why not simply look for vaguely correctly-timed syncs?

Groove:
Why not simply look for vaguely correctly-timed syncs?

Because they are not logic level signals and a sync pulse lasts only 4uS and the analogue read takes 100uS for a conversion.

So, just look for the edges, and don't use analogue.

Groove: So, just look for the edges, and don't use analogue.

Don't be silly.

Spend a couple more quid, and use an LM1881 sync splitter then.

Have you a problem with a diode capacitor and resistor?

No problem, I just prefer a solution less prone to error or misinterpretation

Well you have not got one with your solution. What happens when you try and measure a non existent video sync pulse? At best it times out after a long time.

Thanks for all the responses,

I have looked into the LM1881, but I am more curious of how to do it without that component, plus when I was looking at the spec it seemed like I would have the exact same issue since it doesn't have a digital "this signal is active" pin, instead it has all kinds of information but that info still fluctuates from -v to +v or 0v to Xv etc, which I would have an issue timing, so if I polled it while the sync was low, I would miss that there was a sync right?

Grumpy_Mike: I would use just a simple single diode feeding into a capacitor with a resistor across it. Yes it will need feeding into an analogue pin, this gives readings in roughly 4mV steps so there is plenty of resoloipution to detect the signal. You are likely to see a voltage of just over 0.3V, you will not see anything like 1V

So my lack of knowledge with circuit terminology is not helping, I know what a diode feeding into a capacitor means, but when you say "resistor across it" I am not quite sure how that would look.

I assume you mean something like 6.13 on this page (half wave rectifier?): https://wiki.analog.com/university/courses/electronics/text/chapter-6

Yes figure 6.5 in that link.

but when you say "resistor across it" I am not quite sure how that would look.

How would your car look if you parked it across your drive?

Grumpy_Mike:
Yes figure 6.5 in that link.
How would your car look if you parked it across your drive?

It makes sense with a picture, but I was thinking “Across what?” but seeing the circuit in parallel makes sense when I hear the wording :slight_smile:

when it comes to smoothing, since it is only .3 volts should I make a best effort attempt to get the fluctuation down? also, within the .3 volts I assume I should set my tolerance for detecting the voltage around .1 or .01 or something much lower, although I will have to play around with it.

Would you have a recommended capacitance and resistance when it comes to video signals? I know I saw a formula once, but I am having trouble finding it again.

you have been a great help, thanks!

The formula for a resistor / capacitor is Tc = RC The sync signal is very fast so you only need a small capacitor. Do not bother trying to get it very smooth here is little point. As 0.3 is the black level then setting the threshold at 0.3V is a good start.

I am assuming that your video has the bottom sync level at zero volts, if not it gets a bit more complex with you needing a black level clamp. Do you know what your composite video signal looks like? Let's hope it is not AC coupled.

I have seen the signal images, when it comes down to it I assume I could use the LM1881 as the input to the half wave rectifier.

I will read up on what AC coupled for a video signal is. I assume most devices will hit 0 volts if it isn't making a signal, but what you just said makes it seem like this is not always the case :-|

I will look into what a black level clamp is as well. I plan on using this detection for a multitude of devices, so I can see it having all of those issues at one point or another. So even if I don't use the Clamp or the AC coupled video solutions and instead use the LM1881 with a half wave rectifier, I still want to know how it all works. Knowledge is never a waste of time :-)

Thanks,

Well you could tell us what you have. What is your video source? Do you have an oscilloscope?

AC coupled - connected with a series capacitor. Black level clamp - the black level, the +0.3V level is clamped that is fixed to some point either 0.3V or ground.

when it comes down to it I assume I could use the LM1881 as the input to the half wave rectifier.

No.

I am using a ton of old game consoles to test (nes/snes/n64/ps1/ps2...etc) I am a bit of a collector and very interested in electronics in general. I do not have an oscilloscope, it is outside my price range right now.

I still have a lot to learn as I would need to read more about AC coupled and Black level clamp to start to understand what you said :)

Why can't I use the sync of the LM1881 to determine on or off with a capacitor and resistor, just like the video wave?

again, thank you for taking the time to respond to an idiot like me.

Why can't I use the sync of the LM1881 to determine on or off with a capacitor and resistor, just like the video wave?

Because in the absence of a signal the outputs are not defined. Sometimes they go off by themselves. For example the data sheet says:-

The vertical output is produced on the rising edge of the first serration in the vertical sync period. A default vertical output is produced after a time delay if the rising edge mentioned above does not occur within the externally set delay period, such as might be the case for a non-standard video signal.

Grumpy_Mike: Because in the absence of a signal the outputs are not defined. Sometimes they go off by themselves. For example the data sheet says:-

ok this makes sense.

So if my research has led me down the right road a black level clamp is used on signals that are, in laymen terms "shifted up" right? So the empty or black screen for one signal may be at .5 volts so we need to clamp it down to 0v to make it standard. This makes sense to me.

for the clamp it seems like I would need information like the sync or the back porch to "key" the low point in to clamp down the signal right?

I have been reading and it seems that most items are not AC-coupled with the major exception being items which run on batteries or if it is government mandated. I don't really know the solution to an AC-coupled signal, would this also be fixed by using a clamp, or shift the signal back to the correct black level? what are key words for fixing AC-Coupling?

it really seems like I need an oscilloscope to do this, are there cheap ones accurate enough to analyze video signals?

Thanks!

"shifted up" right?

Yes, or down or indeed float. It fixes the black level at some point.

What you need to do is to try the diode cap and R, with the video signals you have.

Virtually any scope will be good enough to show video.

if it is government mandated

No it is not.