attenuate signal levels on video signals

I often modify video output on gaming consoles. Often I find myself confused on the value of capacitor and resistor I have to put after the video chip before the video connector. Different users and manuals do the mods extremely different.

What I understand is that sync should have around 0.3 Vpp and R G and B around 0.7 Vpp. When I measure them with an oscilloscope they tend to bee too high. So I want to lower them. Here kicks my problem in:

How do I actually measure the signal level? I can't see any difference in signal level, no matter what I do.
Please see my example: I measure the CS# (c sync near TTL level) output of my Famicom AV over its RGB output Board (made by Tim from Australia).

I get Max 55 mV, Min -516 mV, mean 5.4 mV and RMS 131.9 mV. Vpp seems to be 571 mV. Please the see attached picture.

So now I put a 460 uF capacitor with positive leg facing the chip to the chip. After the capacitor I put a 75 Ohm resistor. Then I measure again with my oscilloscope. What stuns me: I get the same result!!! The difference is below the limit of reproducibility.

What am I doing wrong? How is the proper way to measure signal level? Why doesn't the value change?
I experience this "phenomenon" (I am not an electrician, I am a chemist) my whole life. It is really frustrating.

 

Are you measuring while the console is connected to the a TV?

The resistor on the output forms a voltage divider with the resistor in the TV. If the connection isn't connected to the TV, there's no voltage divider and the resistor is just limiting current only.

 

Oh dear... that might be the problem! Yes, it's connected to a TV but the C Sync line I am currently measuring not... Is this the problem? Am I just allowed to measure when the very line I am using is connected to the TV?

 

Yes, that's the problem .

Yes you can measure it connected

 

Thank you so much, Bad_Ad84. You can't imagine how much you helped me! I would have been unable to continue and would have been stranded with an oscilloscope I wondered about "why it does't work"

 

Now I definitely measure something.

But the mystery continues as:
A wire directly connected to the CS# Output to TV yields me: Vpp 0.083 V, Max 9.4 mV, Min -73.4 mV, Mean -0.76 mV and RMS 17.77 mV

The before mentioned set up with cap and resistor:
Vpp 0.152 V, Max 14 mV, Min -128 mV, Mean -0.6 mV and RMS 31.8 mV

So with my attenuator the signal even seems to got bigger. Which shouldn't be. As the capacitor does take away the DC part the signal should lower. Furthermore the capacitance of the capacitor should just mess around as being a filter and doesn't affect much of the signal level. Then there is just the resistor (the value of which I chose for matching impedance) which can't be an amplifier.

I imagine this: Without an attenuator, the signal level is too high. Therefore my TV pops in something like a stepped attenuator. And thus it happens that the signal level is shown as being really small instead of way over 0.3 V. Is that plausible?

So would that mean that to get the start point I would have to measure sync disconnected to get kind of "the worst case" and then, when adapted, measuring again with a TV to see what it actually gets as an input as a cause of my alternations?