I have run enough tests and sufficiently ordered my thoughts that I can post about my experiences with this TT over the last few months. To summarize where I was at when this thread started: I bought a used original Planar 3 with R200 tonearm and older Grado ZT+ cartridge and set out to do two main things: 1) improve the table's poor wow & flutter (speed stability) performance; 2) restore the tonearm's anti-skate functionality. I'll start with point 1). Bear in mind that some of my results may be specific to my setup and individual parts, and so we may arrive at different conclusions, but I have tried to be as careful and methodical as possible when testing the many variables for speed stability.
Three Clear Improvements to Speed Stability (W&F)
Of all the changes I have attempted via different setup conditions or replacement parts, only three things have definitively improved the audible (and cell phone app measured) wow & flutter of this table: The first, and most obvious thing I should have tried before anything else: change the belt. Mea culpa. When I was given this table to demo, it had a Rega black belt that I didn't know was worn out. When I finally completed the transaction with the owner, he threw in another Rega black belt that had been used previously on one of his other TTs, as I had mentioned to him the speed stability issues I was experiencing. The impact was significant. While there was no apparent difference between the two used belts, but this second one still had some life left in it, and is the belt I am currently using, as it is the only belt I have that gives satisfactory speed stability (including two new opal belts from Tangospinner, either used solo or in tandem).
The second thing that led to a further concrete improvement in speed stability was isolating the motor from the plinth. There are superior looking aftermarket upgrades to this end, but since I had blown my budget on other components, and being naturally inclined to DIY, I whipped up something quick and dirty in my shop. I refined that first design into what I'm using now: a ~ 6" x 9" x 5/8" birch board with three very grippy rubber feet, with a 2-1/2" x 2-1/2" x 3/4" birch motor mount screwed to it. This motor mount has a 2" round recess to support the motor, drilled with a Forstner bit, and the motor i held in place by friction. It would be better if this motor base was made of denser material, but the grippy feet provide plenty of traction to keep the motor immobilized from belt tension while running. When first installing this motor base, or anytime the TT is moved, it is necessary to fine-tune the motor's position relative to the subplatter, as there can be some small variation in both speed stability and speed accuracy (closeness to 33.33 rpm).
The third way I improved speed stability was only an improvement for me because I started out doing it wrong: use the recommended lubricant for the platter bearing. Out of ignorance, I grabbed the first thing I found in my shop, which was 3-in-1 oil, a very light oil that is quite different from what is recommended by Rega. In general, using a heavier lubricant gave me improved speed stability across the board. I'm currently using Nautilus SAE 90 marine gear oil, because I happened to have a quart tucked away in my shop.
With these three principal changes, I generally measure W&F 2sigma errors of between 0.10% and 0.08% at a speed of about 33.40 rpm. This is night and day from my initial experience where the TT averaged about 0.33% error, which was distracting on most LPs, not just those typically prone, such as featuring sustained piano notes. At current W&F levels, I don't notice any speed stability issues during general listening, and it is difficult to impossible for me to discern any wow (depending on music) when doing critical listening or cleaning up needle drops. I have a couple records that are clearly pressed off-centre (the stylus sweeps left and right during play) or are slightly warped, that still exhibit noticeable wow. Without getting into gory specifics in this post, I have made other tests to setup and parts that have not seemed to affect speed stability, have had marginal positive impact, or have made it worse: different motor suspension belt, altering motor position while suspended, different subplatter, different bearing, different motor pulley, different belt / combination of belts, different bearing lubricants, lubricating the motor.
More to come.