It is definitely an awkward way of doing it. I had the original Snapmaker when there was limited software support (and personal knowledge) so awkward was the norm. While cumbersome, it worked well for my flight sim controllers; which required printing, lasing, and CNC’ing. Made enough money to cover the purchase cost of both machines.
I used F360 for the printing and have since switched to Lighburn for lasing. I will probably spend today (day off) learning the machining function of F360.
I did a test cut on a piece of cast aluminum at a cut depth of .05mm and a feed rate of 30mm/min. Worked great. No slowdown of the spindle during the cut. No indications of added strain on the modules. I do need to slow down the spindle speed as 12,000 RPM causes the whole machine to vibrate. The vibration goes away below 65% of spindle speed.
Good call on the air test. As it turns out, the x-axis will only move to a maximum of 136mm from center this is about 5mm short of being able to mill the entire bosses on the right side of the carriage. Bummer. I guess I could machine the left side and center most bosses. Remove the carriage and turn it 180 degrees; then machine the remaining bosses. If I maintain the work origin I can re-order the CNC file and split it into 2 files to do this. It would be nice to have a plate that would allow me to move the head over those few mm’s. I may see about doing that.
Aside from riding on the assumption the mount and everything is perfectly square, a long shot unless verified, that sounds like more work than checking high spots with dykem and a flat plate and flattening accordingly like mentioned by @eh9 here
Sorry I misread the screen. the maximum x-axis movement is 136.5. The module will move further by hand but maybe the limit switch position won’t allow it to move further under power.
I had considered having the carriage machined. But while the carriage would be parallel top to bottom; the possibility that the module sleighs are not entirely square and on plane would not be being taken into consideration. So by machining the carriage in place, all the factors that could contribute to an uneven surface, would be.
We used to machine vehicle brake rotors on a lathe. But that didn’t account for variations in the hub mounting surface. So, often times, the vehicle would return after a few thousand miles with a brake vibration that was attributed to the rotor surface not being true. Several year ago we started machining the rotors while mounted to the vehicle. This eliminated that issue. I’m thinking the same would hold true for this application.
I had 2 small brackets that allowed me to move the head over to the next hole. I’m going to run a dry test again.
Wow, this thing is .7mm out of level. highest in the left front and lowest n the left rear. The original owner buried the head into the left rear so it makes sense.
I’m taking it down .05mm per pass at 65% which is about 7700 RPM… I had to rearrange the g-code so it would cut them from highest to lowest in each pass. Each pass is taking about 25 minutes. I’ll let you know how it turns out.
I’ll just repeat this again:
And then do it once more 
What am I checking for?
In general, the squareness between all axes, and that everything is parallel.
I was assuming that you measured the difference with everything mounted and did so by measuring it by moving the tool head around?
If you measured using a flat worksurface and the carriage unmounted you can disregard my comment.
Just want to avoid you milling it parallel and later come to the conclusion that one of your linear modules was mounted a bit tilted.
Everything @brvdboss said. Several people have milled their wasteboard at an angle because they incorrectly assumed that the X axis is always parallel. It is not, and can easily be tilted by raising and lowering the Z tower mounts where the X axis connects to each tower. If you are measuring the right side consistently lower than the left that that’s a very good indication your X axis is not tram to the linear modules that make up the Y axis.
MS-paint-cad exaggerated view here:
One of those people was me 
In my case the only damage was 1mm of wasteboard. Not that big of an issue and easy and cheap to replace.
Experience is the best teacher of all.
I also ran into that issue by screwing something up 
I t seems that you are saying that the baseplate itself is not even; causing the towers to be of different heights. SO I would need to put my square across the y-axis modules and check that the z-axis modules are square to them and that they are square and parallel to each other. Is there flex in the baseplate or is it rigid enough not to flex?
Not necessarily. Could be though. But yes, I did measure the 3 rotations of each Z tower and they turned out to be pretty good, within .2 degrees or so of perfect.
Moreso is what I mentioned with the x axis linear module being rotated. If you turn the machine off and push down or lift up on the right side mount for the X axis you will find you can move it up and down probably 1/4", introducing serious tilt. That needs to be somewhat regularly fixed, I reference off of the surface of the Y linear modules.
If your machine isn’t square you’re going to mill an angle into your platform. It’ll be flat, but angled, and the Z up direction the Z towers move won’t be perfectly normal to the platform, resulting in some weird geometry in 3DP and milling.
That’s just left-right angle. Front back is equally an issue, but requires shimming to fix. The platform’s XY plane needs to be normal to Z, and the toolhead needs to be parallel with the Z axis. That is “tram”.
@Thick8
You do have everything correctly assembled too?
Nuts down on bed frame. Linear modules centered and not hanging off front or back.
Don’t think I saw that asked of you.
You’d be amazed how many times that’s the case, so I figured I’d just make sure.
-S
I assembled it 3 times. The first time by the book. Then with some modifications to the procedure. Then 1 final time with some more modification and a torque wrench. All to make sure everything was tightened and secured in a logical process. I’m a mechanic (of 40 years) by trade so I recognized the errors in the book’s assembly/tightening sequence. I did use a straight edge but not a square. I assumed that the baseplate was properly true. I removed the rubber feet and have it sitting on memory foam to eliminate vibration and allow for even support. I will have to pull it out of the enclosure to put a square on it.
This dual linear module idea seems to have some inherent issues that were not present on the original model.
a support connecting the tops of the z-axis modules would probably be a good idea.
y-axis: parallel, level across the tops (on the same plane), centered within the depressions/pockets.
z-axis: parallel, square to the y-axis, to each other, and baseplate, parallel to the y-axis during its full range of motion.
x-axis: square to the z-axis, parallel to the y-axis during it’s full range of motion.
So that just leaves the carriage being bent as the cause of the difference in height. Right?
Sounds right. I’m having trouble picturing how the x axis is parallel to the y axis. If that was instead ‘square’ sounds perfect.
The plane of the x-axis is parallel to the plane of the y-axis through the entire range of the z-axis.
EDIT: So I went over the whole thing again with a sliding square and a .001" feeler gauge. I did note a discrepancy in the x-axis being slightly lower on the right side. I was able to reposition it on the z-axis sleighs to correct for it. Something I found odd though is that both z-axis modules are tilted slightly inward at the top. The degree to which they are tilted does not change with the height of the x-axis module though. So it seems that everything is <.001" . Now to verify the carriage.
I put 2 sliding squares on top of the x-axis module with the rule touching the highest bosses at the front comers of the carriage. I then ran the y-axis back, stopping at each boss. I also did this with the middle rows of bosses. The carriage is bent towards the left rear. I also placed my machinist’s rule across each row of bosses and was able to rock it. Indicating the bend. So now it’s time to machine it. Will report back.
Thanks for all your help with this guys. I feel a lot better about machining the carriage now…
EDIT: Woohoo! It worked. I went from a z height of 55.9 to 55.4. I was unable to take more because I didn’t want to take the boss height below .2mm. The left rear boss will still have to be shimmed .17mm. Straight edged them and they’re all < .03mm ( that’s my thinnest feeler gauge) from each other and the x-axis.