Easy SNES/SFC YUV component video!

[villahed94]

That issue is caused by a weak sync mostly. The TV will take any dark image as its sync, but if the picture goes too white, the TV will lose the sync, thinking the voltage is waay too high(but it's in fact lacking sync).
A 100% white picture, according to NTSC standards should be 1.00 V.
I once read that the absolute admitted value is 1.073V, and anything above that will cause the TV to ignore the input signal.

BTW, I told Guntz to use white screens as reference screens for voltage measuring, because a pure white image is made up of 33.33% of Red, Green and Blue. This applies to Luminance too

[Ace]

... Most home consoles and encoders I know of work on DC current, what do you think the right answer is?

I was inclined to say DC, but this bit from Chilly has me a bit confused:

Most encoders want the RGB inputs less than 1V. In fact, 0.7V P-P is fairly standard for RGB.

The 0.7V P-P bit, in particular, implies AC current. This has caused some confusion on my end, but just to start, I went ahead and measured DC voltage from the RGB inputs on white screens.

Now, here's the thing with DC: I only got a reading if I took the DC voltage from the encoder's RGB inputs, i.e. right after the 1uF caps. On my PC Engine Duo-R, this showed as something which I think is far too strong (I'm not all that familiar with video voltages, so I may be wrong): about 4.3V DC per color on a bright white screen. This is using a voltage divider composed of a 1Kohm resistor in to Ground in parallel with a 470ohm resistor to the 1uF cap. Is the voltage too high?

[villahed94]

Yes, it might be a fake reading....

[Ace]

I wouldn't be surprised if it is. That reading seems far too high.

So now, I've decided to measure the AC voltage between the voltage divider and Ground (I'm not even sure if this is the right way of doing it, but I did it regardless), and I get about 0.13V. From what I know, multimeters measure RMS voltage, which, after having done some calculations, would mean each RGB signal's peak-to-peak voltage is about 0.37V. Seems too low for the encoder, which wants its RGB inputs to be 0.7V P-P. Then again, this might be a false reading, too. I've dismantled the entire circuit as the external circuitry is just giving me hell. Word of note: don't go with C945 transistors to amplify Luminance, they always seem to fail. I'm gonna get a 2N3904 and two new C945s for the colors as well as some new 1uF caps for the RGB inputs and possibly some higher quality surface-mount to DIP adapters than the garbage I got from my local electronics shop.

[evildragon]

I always thought it was DC, but was told it is AC coupled.

But that confuses me, how can it be AC, if it is sharing a DC ground?

[CkRtech]

So you know how standard video output tends to have a resistor on the line (75 ohm) to get it to standard impedance. It also will often use a cap (like a 220uf) as a decoupling component.

Although certainly not a technical resource (ha!), there is some explanation here: http://en.wikipedia.org/wiki/Decoupling_capacitor

[Chilly Willy]

You use AC and read the peak-to-peak voltage. DC won't give squat since it's an AC signal. An oscillascope would be best/easiest for checking the levels since you'd see the p-p voltage on the display.

[villahed94]

I wouldn't be surprised if it is. That reading seems far too high.

So now, I've decided to measure the AC voltage between the voltage divider and Ground (I'm not even sure if this is the right way of doing it, but I did it regardless), and I get about 0.13V. From what I know, multimeters measure RMS voltage, which, after having done some calculations, would mean each RGB signal's peak-to-peak voltage is about 0.37V. Seems too low for the encoder, which wants its RGB inputs to be 0.7V P-P. Then again, this might be a false reading, too. I've dismantled the entire circuit as the external circuitry is just giving me hell. Word of note: don't go with C945 transistors to amplify Luminance, they always seem to fail. I'm gonna get a 2N3904 and two new C945s for the colors as well as some new 1uF caps for the RGB inputs and possibly some higher quality surface-mount to DIP adapters than the garbage I got from my local electronics shop.

I recommend to use the 2SC1741 transistors to amplify Luminance. They never fail.

[Ace]

You use AC and read the peak-to-peak voltage. DC won't give squat since it's an AC signal. An oscillascope would be best/easiest for checking the levels since you'd see the p-p voltage on the display.

Doesn't a multimeter get the voltage reading as RMS voltage rather than peak-to-peak voltage?

[Chilly Willy]

Doesn't a multimeter get the voltage reading as RMS voltage rather than peak-to-peak voltage?

Most, yeah. Some have a peak-to-peak setting as well. I have one with, and one without... but I prefer to use my oscope on AC signals in any case.

[Ace]

I don't have an oscilloscope nor does my multimeter have adjustable RMS and peak-to-peak AC voltage settings. If a multimeter has just one AC voltage setting, does that mean it's only measuring RMS voltage? If that's the case, going by my previous measurements and calculations, the peak-to-peak voltage of my PC Engine Duo-R's RGB outputs using the voltage divider setup I had is considerably lower than the 0.7V P-P the encoder wants. Just to be sure, is this voltage supposed to be the maximum voltage going into the encoder?

[Chilly Willy]

I don't have an oscilloscope nor does my multimeter have adjustable RMS and peak-to-peak AC voltage settings. If a multimeter has just one AC voltage setting, does that mean it's only measuring RMS voltage? If that's the case, going by my previous measurements and calculations, the peak-to-peak voltage of my PC Engine Duo-R's RGB outputs using the voltage divider setup I had is considerably lower than the 0.7V P-P the encoder wants. Just to be sure, is this voltage supposed to be the maximum voltage going into the encoder?

If you look at the datasheet from earlier in the thread - the recommended input voltage for the encoder RGB lines is 0.7V peak-to-peak. If you read page 6, the RGB in are clamped to 3.2V, so if the power supply is 5V, the max for the RGB in would need to be less than 1.8V. The only absolute max is given for the power supply, which is 7V.

[akaviolence]

AWESOME NEWS!!!
I got it working on an rgb modded nintendo (top loader).
Here's the vid, just shows its working.

[Ace]

That is sweet, man. I gotta find a 2C03 and stick it in a spare NES motherboard I have laying around.

By the way, where do you get your SOP to DIP boards? I sure could use some as those that are available locally are complete garbage (they're actually too wide to properly fit a BA6592F or BA6594AF/S-ENC).

[Drakon]

I got my sop to dip adapters from ebay.