Coupling Capacitors on RGB lines with NESRGB 2.0 Necessary?

[erik343]

I learned that there are 220 uf coupling capacitors on the 2600RGB and NESRGB version 1.4 (on the mod board...) as well as on my original unmodified SNES (on the SCART cable that is...). I found out that these components just filter out the DC signal while only allowing the AC RGB signals.

The funny thing is that on the NESRGB 2.0, according to Tim Worthington, these capacitors are not necessary because he uses a THS7374 amp instead of an internal amp from the NES PPU.

My question is, if I have the 220 uf coupling capacitors inside my cable on the RGB lines hooked up to a NESRGB 2.0 (J8 is set to 75 ohms), will this still work or will it have any adverse effects? Is it better to have no capacitors in the cable or is it better or even acceptable to have these caps on the RGB lines inside the cable?

I also have a secondary question:
Does the size of each capacitor make any difference? For example with the 2600RGB for my Atari console, is it better to have the RGB coupling capacitors on the mod board, or instead inside the cable? The capacitors on the mod board are super small, and the ones in the cable are a lot bigger.

[yoshiyukiblade]

There is indeed an adverse effect from having two 220 uF capacitors in series on a video line because it reduces the overall capacitance. Two series capacitors with the same capacitance effectively cut the total in half, so it's like having a single 110 uF capacitor in line. If you're curious, the general formula for equivalent series capacitance is 1/(1/C1+1/C2+...+1/CN). Less capacitance results in more field tilt, or droop in brightness across the screen (left-to-right or top-to-bottom). Even 220 uF is not sufficient for 8-bit video capture purposes (top-to-bottom droop is very measurable).

This somewhat answers your second question: more capacitance gives better field tilt performance. IIRC, professional equipment use 1000 uF or more to keep the droop minimal. I don't think it matters where the capacitor is in line, but theoretically it's safest to have it as close to the source as possible. That way, if there is some freak accident where the cable is cut and short circuited, it wont include the DC part of the signal. Under most conditions, it probably doesn't matter much.

The alternative option is my preferred configuration, which is full DC coupling (no series capacitors at all). There is no droop from the high pass filter formed by the series capacitors. The THS7374 and other modern amps can handle it, and I think that's what Tim meant by saying the capacitors are not necessary. Keep in mind that this means the DC part of the signal will be under a 150 ohm load, which increases power consumption. And depending on the amount of DC offset in the signal line, a short circuit event can draw a lot of current.

[erik343]

Thanks for the information!

Would you recommend keeping the capacitors inside the cable when using the NESRGB 2.0? Remember that this mod board and revision have no 220 uf caps on the RGB lines on the circuit board...

[makar1]

Tim's installation guide for the AV Famicom specifically states to use a Gamecube SCART cable, which has 220uF capacitors on the RGB lines.

https://etim.net.au/nesrgb/installation-famicomav/

https://gamesx.com/wiki/doku.php?id=av:nintendomultiav

[yoshiyukiblade]

Keeping capacitors on the cable is pretty much the "standard" experience now, so it's best to leave them on. I think the onboard capacitors created conflicts with capacitors on cables and were often removed by modders anyway (this is why standards are important!). I think eventually the trends have stuck with keeping capacitors on cables since the old RGB-ready stock consoles did.

[viletim]

Keeping capacitors on the cable is pretty much the "standard" experience now, so it's best to leave them on. I think the onboard capacitors created conflicts with capacitors on cables and were often removed by modders anyway (this is why standards are important!). I think eventually the trends have stuck with keeping capacitors on cables since the old RGB-ready stock consoles did.

This is the way to go for Nintendo SCART cables.

This somewhat answers your second question: more capacitance gives better field tilt performance. IIRC, professional equipment use 1000 uF or more to keep the droop minimal. I don't think it matters where the capacitor is in line, but theoretically it's safest to have it as close to the source as possible. That way, if there is some freak accident where the cable is cut and short circuited, it wont include the DC part of the signal. Under most conditions, it probably doesn't matter much.

Professional equipment is direct coupled. Historically, a video amplifier with a dual voltage rails was required for this, but modern video drivers (like THS...) can operate from a single supply without excess DC offset.

[yoshiyukiblade]

Ah okay, I didn't know they also used to use negative supplies for video equipment in the past. I read an article by Maxim Integrated about AC coupling here: https://www.maximintegrated.com/en/desi ... /3768.html

They mentioned under "Technical Concerns" about broadcast equipment using around 2200 uF capacitors. Maybe broadcast equipment doesn't necessarily mean professional, or AC coupled designs came afterwards.