Do you still use Panasonic's F15 cameras from 1991, and want to get colour bars out of the camera when you don't have one of the accessories that has the colour bars switch? While that's possible, you might not want to bother. This page explains why.
In the normal run of things, colour videos cameras produce colour bars as a means of showing that the camera's colour encoder is working correctly, as a test pattern for aligning the colour encoder in the workshop, as a test pattern for phasing up the camera's synchronisation with other equipment, and as a test pattern for setting up the monitors showing the camera's picture (including the camera's viewfinder). The colour bar test pattern is generated in the camera head, and fed into the RGB inputs of the video circuitry instead of the picture signal from image sensor. This tests almost the entire path of the video signal through the camera.
Unfortunately, the F15 camera doesn't have a colour bar generator in it, at all. It has an input for feeding an already (PAL or NTSC) encoded into the camera, and switching it through some of the final video stages of the camera. This doesn't test the camera's RGB encoder. And since the video is external, it can be out-of-phase with the camera's colour encoded video, so you can't rely on switching the camera to bars to phase the camera up with analogue video equipment. And, as well as that, its signal levels may be different from the cameras, as at least one camera I tested was. So it can't be used for setting video levels.
So, if you want useful colour bars, you have to re-align the bar generator, properly (I had to tweak trimpots, and change some surface mounted resistors to fix up incorrect sync and white levels). Or, build your own for a camera that doesn't have the WV-AD37 back end.
All-in-all, this is a pretty stupid design decision/flaw, on Panasonic's behalf. Colour bars aren't for making a pretty picture, they're an alignment tool, they need to be accurate.
So, to properly phase up a F15 camera, using the colour bars: First you turn on the bars, phase up the system to the bars in the way that you expect phase alignment to be done (with the horizontal and sub-carrier phase controls on the RCU or one of the genlock adaptors on the back of the camera). Then you turn off the bars, and align the camera video signal to match up with the colour bars. Either by using a vectorscope locked to an external sync signal, and lining up the burst vectors. Or aiming the camera at a primary coloured object, and using a vectorscope to line up the image vectors.
Phasing up the camera image on a vectorscope may be difficult with NTSC cameras, as few cameras have a text-book video output. i.e. If you aimed a camera at images with 100% primary or secondary colours (like colour bars on a test card), they won't land on the vector targets the same as electronically generated colour bars. The same for PAL encoded video, but we have an advantage when the scope has the automatic reversal mode of the V vector, to make the scope look like a NTSC display: You rotate the phase until the reversed vector lands on top of the unreversed vector (one set will rotate clockwise, while the other rotates anti-clockwise). It won't matter if the dot for a red image lands in the red target square, all you care about is the two red vectors merging into one dot somewhere near the red targe. Likewise for if you picked another colour to adjust by.
Once you've lined the camera up with its back end, you shouldn't have to re-adjust the phasing between the camera head and its colour bars any more, unless you change back ends on the camera (each back end may, and probably will have, different characteristics from each other), or if you have unstable camera that keeps going out of alignment. The next time you have to phase up the camera to other equipment (because you've changed cables, or other equipment), you only need to do the phase alignments that you expect to do with most video cameras (horizontal and subcarrier adjustments of the genlock controls). And you can do that with the colour bars from the camera back end, or the camera video.
Since two of our cameras didn't have the backs with a colour bar generator in them, we're used to phasing up by the video image vectors. And we tend to do that more than look at the burst vectors (as we're going through equipment that re-inserts house sync, burst, and blanking), and it's the image part of the video signal that's more important, to us, than the camera's colour burst.
The way many people in the PAL television world would handle that, as a quick and dirty method, was to adjust the phase while looking at a colour video monitor, and tune for maximum colour saturation. So long as the monitor doesn't have an alignment fault, that will get the sub-carrier very close to being in phase (good enough for most people, though not for broadcasting). But many monitors do have colour errors, so an improved variation on that method would be to view two or more monitors on the same signal, as you adjusted it, and tune up for best results between the lot of them.
If you have an ordinary CRO and a DVD recorder with component outputs, you can almost make your own vectorscope. While the burst vectors won't be shown, and the dots won't rotate as they go out of phase, you can watch the dots move away from the centre as you get closer to being in phase, and towards the centre as you go out of phase. This is easier to look for than the colours brightening up on a monitor. Put your CRO into X/Y mode, and connect your X & Y inputs to the B−Y and R−Y component outputs. Now feed encoded video into the composite or S-Video inputs of the recorder.
This crude approximation of a vectorscope can, also, help with manual white balancing. You'd adjust the white balance controls to make the vector dots converge into one central dot. Some people find it easier to white balance using a vectorscope, others prefer to use a waveform monitor, tuning for minimum fuzz (chroma signal) on top of the luminance waveform.