Now that Ron Black has fully retired, this leaves a bit of a void in the KDX world, especially for the 220 which has a head that leaves a lot to be desired as far as squish clearance.
I'd talked to Ron in March and he said he was still doing 220 heads, but not really any other work. I didn't get a chance to pull the head off my old 220 due to my Mom's battle with cancer, and 2 months later, no replies whatsoever from Ron and Bobby Black... I hope Ron is doing okay, but I took that as him being fully retired.
I've got an old 1959 South Bend Heavy 10L metal lathe that I've been learning to use as needed, and have been curious about the science behind optimizing 2-stroke head squish band widths and dome profiles, as well as how to fixture the KDX heads up in a lathe to perform these operations.
I've been reading up a lot on my lunch breaks and late evenings on this advanced 2-stroke tuning topic, trying to bring more knowledge to the KDX community as our legendary gurus age and retire... Feel free to correct any information which I've misinterpreted or missed, I'm here to learn as much as the rest of you or more. If Ron were responding to emails and still cranking out his work, I would not be posting one bit if this publicly, as you can't ethically take a man's knowledge earned through decades of hard work, and just use it for yourself when the guy is still making a living off of it. But instead now, I'm trying to invoke some of the more active generations to consider learning more about this work so at least they'll know what their 2-stroke engine builder machinist is talking about, and can spot of the builder doesn't really give any concern to optimizing the squish band height and overall head volume/compression ratio... and in rare cases, someone might actually be ambitious enough to learn the finer details and attempt these mods on a metal lathe if they have one at their disposal!
Here's three KDX220R heads that I've got.
Left is a head that someone milled/faced down 0.40mm for significantly tighter squish band clearance and higher compression, but did not touch the squish band or dome, resulting in tight squish height (I'm guessing near the bare minimum of 0.80mm or maybe closer to 1.0mm, I have not confirmed standard KDX deck height or head gasket compressed thicknesses yet). As you can see, although the tightened squish band height surely improved squish velocity and combustion, this head suffered a slight bit of pre-ignition detonation, the pitting marks of melted missing metal scarring the squish band and chamber, which is also very power robbing as well as hard on crank and rod bearings. Too much compression as well as too rapid of squish band air/fuel velocity (from the outer portions of the cylinder and being forced towards the dome/spark plug/flame front and creating mixture turbulence to improve combustion spread as well) can both cause detonation.
Reducing the squish band width while maintaining same squish height / clearance will reduce the squish velocity slightly, and is an important tuning/design tool, but higher octane on the tighter clearances is best advice. You'll get a lot better bottom end power for tight woods riding with the tightest clearance (but top end power suffers on a 2-stroke when you have too high of compression), but high octane is best for the heating of the air/fuel mixture due to the tight clearance as well as if you don't machine the dome for more reasonable compression ratio. Higher octane is also a good thing for slow hot summer tight woods riding with less air moving through the radiators... As is a slightly richer mixture for better piston cooling.
Center in the picture is my Ron Black 220 head... The great KDX tuning master's work

This head has had 0.80mm milled off the head gasket surface, squish band cut and reshaped slightly to around 8mm width from machining the dome wider to cut off the radiused transition to the squish band, and from what I can roughly tell, cutting the chamber about .38mm deeper (material removal, not a measurement off of the new head gasket deck surface height). This head has a taller squish height than the shaved head on the left, despite a lot more height taken off the deck surface of the head, with a squish of about .68mm from the head gasket deck surface (+compressed gasket thickness +/- piston protrusion @ TDC + total squish height). I'm speculating Ron's heads were on the safer side of adequate squish height of closer to 1.20mm total (.048" or so), but still at the wider ranges of the sweet spot to where the squish velocity effect really starts taking place and making a difference.
*note - these are all late night rough measurements with a micrometer, I haven't made a plexiglass blank yet to bolt to the head to seal the chamber off and use a graduated syringe to fill and measure chamber volume with water. YET...
On the right in the first photo is a totally stock and fairly clean 69mm 220 head... much taller squish height, 1.11mm gasket surface to edge of squish band. I'm guessing these heads stock are running maybe 1.7mm or 1.8mm squish height. I need to measure some KDX head gaskets, Wiseco, Athena, and OEM, to get some better data here.
Around 1.25mm is the point where the squish height and squish band start really losing effectiveness, and the engine starts operating less efficiently. Optimizing the squish band height and keeping the compression ratio in check, will make jetting more consistent throughout varied conditions, as well as require a smaller main jet size, as a stock 220 head will use more fuel to get the same combustion, since the combustion is less efficient due to not having the squish velocity causing turbulence to mix up the mixture amongst the flame front, as well as a significant amount of the air fuel mixture lingering on the outskirts of the cylinder and not fully combusting in the effective portions of the power stroke before the exhaust port is exposed as the piston travels downward...
There's a lot of science and math behind 2 stroke squish bands, porting, crankcase volume and crankcase compression etc... I'm just starting to grasp this, but the head work is the most critical part, and also the most advanced knowledge that an amateur home hobbyist (rookie) machinist and novice bike builder/tuner will likely get into if even touching the head themselves.
To reiterate, the squish band, the more flat outside portion of the combustion chamber area, is there to promote more efficiently mixed and properly timed combustion, but this doesn't really have much of an effect unless the squish clearance is 1.20mm or less (some say 0.050" which is about 0.48mm). Running this tighter around 0.80mm works very well, but the actual squish band width may need to be narrower in order not to exceed the maximum recommended squish velocity. Too high of velocity will compress the air too quickly in this area as opposed to out in the dome of the combustion chamber area, and can cause pre-ignition detonation which is always to be avoided... 1.0mm is a safer bet with that in mind, and also allows for a bit more wear and tear on the crank and rod bearings, which will change the clearances slightly, as well as piston rock in the worn cylinder bores with a worn piston, and also figuring in rod stretch at high temperatures and high RPM... this is a real thing, the rods actually do stretch a very miniscule amount. Everything adds up.
The squish band width and height both have to be calculated to figure squish band total area, in order to achieve the targeted optimum squish velocity without exceeding it. The minimum desired squish velocity would basically force the mixture out from the edges of the cylinder towards the flame front (as well as causing turbulence to scatter the flame front amongst this newly introduced air/fuel mix from the outskirts for even better combustion) at a rate that most of it has combusted before the piston has been pushed down to the point where the exhaust port is exposed or "opened." One thing to consider here, if running higher compression for added low end power and response, higher octane fuel is required, and high octane fuel actually burns SLOWER than lower octane fuel... so the squish velocity for high octane has slightly different optimum requirements than for standard 87 pump gas... running race gas in a 2-stroke really doesn't help you much, unless somehow it also contains more BTU (any non-ethanol gas contains more BTU's of potential energy than 93 octane 10% ethanol gas blends, FYI...
Basically the biggest factory here is to make sure the squish velocity is timed such that the squish velocity event is completed by the time the exhaust port is exposed by the piston. And again, squish velocity is dependent on adequately slim squish band height as well as squish band area (height vs width of the squish profile in the head).
As you can imagine, a ported cylinder with raised port heights will also come into play, and will require pushing the squish velocity up to the maximum allowable in order to get the same complete burn to take place before the exhaust port is exposed.
The squish band angle generally needs to roughly match the angle at the edge of the piston crown, but say if the piston were at a 10 degree angle at the edge, it's commonplace to make the squish band at an 11 degree angle just to make sure there are no variances due to cast piston irregularities or piston rock which would cause the squish to be taller at the edge of the piston vs shorter at the inner edge of the squish band. This is a bad thing, as it will heat up the mixture even more and risk detonation, since the mixture would be ever so slightly compressed further when funneling it down.
Another thing to note is that due to larger bores increasing the compression ratio due to more available volume of air to compress, often times the piston manufacturers will give the oversized pistons a slightly more mellow dome than the standard bore. This would work on the safer direction as far as squish band angle, but also will make it less optimal if going say +1.00mm oversized piston and only just machining 0.50mm of material off the perimeter to get the extra 1.00mm of bore diameter. So when going oversized piston, both the angle and the width of the squish band need to be re-evaluated vs standard bore. Most people don't even consider this. The 223cc "225 kits" are somewhat popular due to Eric Gorr's well known big bore work, so boring and re-plating a 200 or 220 to 223cc is a common and popular mod. No sleeving required of the plated cylinder, cast iron or steel sleeves are generally a huge downgrade, and should always be avoided even if a cylinder requires welding repair work.
Lastly for now, reiterating that too high of compression ratio on a 2-stroke actually hampers high RPM power, and takes more away from that than the gains that higher compression gives the low/mid power and throttle response. There is a fine line. optimized squish band height promotes a lot better combustion, but when shaving just the deck surface of the head, you'll also be increasing the compression ratio quite a bit, which then in turn needs to be taken back out by enlarging the dome portion of the combustion chamber. As we see here by the work of the wizard himself, it looks like widening the dome to cut away the radiused transition between the dome and the squish band is the ticket... Ron has more decades of experience building and tuning and machining engines, and his work speaks for itself, he really has all of these variables completely dialed in as he's done up the perfect head for stock 220 port heights and stock bore size.
My big quest here is also an even greater quest, as I have a ported 223cc cylinder that I have no matching head for, and I need to do more than just copy my RB head here, I actually need to recalculate everything slightly with any altered port timing (I think the port heights are still stock at the higher end) and the larger area/volume of the overbore. The squish band widths are said to be a loose guideline of somewhere around 50% of the entire bore area's square centimeters etc... But this figure is just a rough starting point, and is dependent on all the factors I've outlined above.