I was afraid to connect anything else to the power suppllies feeding the deflection board or I would have gone exactly that route with a 24VDC fan!
Been terrified to blow something up here all along really.
And it isn't even so much the $$ but the fact that these kits are still kind of in the unobtanium kind of realm still currently.
I waited over a year for a USB-DVG unit (even after I had paid for it) and I have been on the wait list for an Amplifone repro kit for longer than that at this point.
Not to fault any of the crew involved in making these things mind you, they all have busy (and sometimes rather complicated it appears?!) lives of their own to lead, but when you are obsessed with this stuff like I am it is unnerving to wait so long- and to sometimes think it is never going to happen anyway.
I had experimented with the fan positioning and orientation before I decided on the current layout.
Literally just waved it around in the cabinet with it plugged in and running to see how it affected the image on the monitor.
Even tried to Faraday cage the stupid thing in its original spot to no avail.
I found how close I could get it to the tube without any drama with the hot and floating technique and then mounted it another inch further away than that even.
And so we have composition #3.

Hopefully the only thing that moves from here is the audio amp, so it can be played with through the eventual coin door.
With the thousand bpm shake managed it was time to address the broken vector weirdness which randomly appeared once I had nearly everything else sorted out.
Full disclosure:
I have tried three different rewound yokes on this thing so far.
They have very similar values in terms of resistance but produce different quality images for sure.
One was the original to the tube.
Sadly I pulled a bunch of correction strips off of it before I realized that maybe they should stay and I have had major convergence issues with that one since.
Maybe some day I can get it back to good, but dialing in a yoke is pretty heady stuff that I still know almost nothing about yet.
Many of the numerous documents I have found on purity and convergence tuning speak as if it is a very scientific and systematic process that is not difficult.
Ha-
Of all the esoteric themes in this hobby yoke gymnastics are easily the most arcane.
Another fellow CRT maniac here (who is far beyond me in knowledge and skill) has a bucket of old yokes from tubes he has scavenged and last I was at his house we dug through it with an inductance meter and found a few with promising values- so I rewound one of those.
The mounting collar is kinda beat up so it is hard to get the thing to stay seated really well but I tried it anyway.
It too produces a fairly un-converged image.
The kit that I bought second hand from a gent in Canada (although I don't believe it was used- maybe not even tested out) came with a rewound yoke- and that is what is currently (haha) on the tube.
The convergence still gets a bit wonky as you get to the outer perimeter of the screen, but during game play there is really no discernable issue thankfully.
And so even as good as I could get this last yoke to look there was still some weird stuff going on with the image on screen.
Such as....

What was really weird to me at first about these bizarre tails to some of the images is that they don't all head in the same direction or have any consistency to them at all really, but after I thought about it this makes sense.
The image is being drawn by the guns which are aimed at the center of the screen and then the beam is deflected to the proper position and then lit up when it is in the right spot, but it is not as if the beams need to draw one image, go back to center, and then get bent around again to draw the next.
If you crank up the screen adjustment on the HV unit (lower dial on the 'flyback transformer'- which it isn't in this case- even if it looks like one to you) you can actually see the path of the beams as they are flying around the screen- and then these weird tails on images make sense.
It is a pretty impressive sight really.

What I was looking at is the beam getting lit up ('Z' axis voltage in Atari-speak) before it actually got to the location it was supposed to be drawing something.
So the deflection board is (very!) slightly slower in bending the beam than it is in lighting it up- and hence the timing being off just that little bit makes for a line where you don't want one.
Enter Slew Rate.
The USB-DVG has a menu all on its own, accessed by a button on the board itself.
The first thing you see when you access it is this.

As my world expands here I am starting to figure out what all this crap is now.
The first line is an obvious one, as it lists all of the original vector monitors here (G-08, WG6100, Amplifone, etc...) and selecting one automatically loads the settings below for what works best with a healthy original monitor.
I selected the Amplifone one for this to begin with since that is what Barry cloned here with this kit but this is where the nuances of this hole thing begin to get tricky.
Note that I said that those default settings are for a "healthy original monitor."
My boards and tube are fine BUT, this is NOT an original yoke- and this is where it starts getting sticky and this is all finally making sense to me now.
My rewound yoke does not act EXACTLY like an original Amplifone one because the outer yoke coils, although having been rewound to get close, are still a slightly different inductance/resistance than original (so are the inner ones for that matter) and therefore they DO NOT deflect at exactly the same rate as original ones do.
Cranking up the Slew Rate allows the beam to get knocked around faster (it allows a faster rate of voltage change to the deflection to more precise) and therefore get to where it is supposed to be in time for the 'Z' axis to get lit up at the right time and draw the images properly.
Neat, yeah?!
Even just a few ticks of bumping up the rates yielded great results.
Behold!

I'm still trying to figure out exactly how the "Jump Rate" and "Beam Delay" settings affect what I am seeing.
Anyone who knows more about electronics here please chime in!