You want to explore your Maslow workspace position accuracy, starting with the horizontal center offset? Write down your left and right chain tolerance parameter (advanced Ground Control settings). Then read on.
1- Compass
I like to think that cathedrals were built with simple tools like a compas 
. So I made a compass with a stick, a screw and a set of holes to squeeze a pen, then drew on my spoil board to find the center line, then started to adjust my chain pitches to fit horizontally on the line when GC reports (x,y) = (0,0). You can try it out too :-).
Actually I realised that it is a matter of care to get a precise result: Using a fine tip pen (0.5mm), making sure the cogs are positionned symetrically, with a notch pointing toward the center, making sure the compass stick does not bow too much, and lightly marking a spoil board that is clean in that center area. Move the sled around to put chains out of the way, yet trace all marks of a cog without moving it twice…
Now you want a lot of X marks to trace with a rather long straight edge a line that best fits through all centers.
Bring back the sled on the center.
Here is what I got:
Looking above the bit on the router: does your router bit center lays on the line?
Here is what I get with default 0 tolerance correction on both chains. Actually, my sled has something like 1.5 mm horizontal offset on the right. (that is a (1/4") 6.35mm bit)
Might be difficult to see on your setup. So let’s talk a moment about observing the router bit while using the maslow.
a) You want to find the center? a pointy v carve bit helps.
b) difficult to look close? Have you though about installing a web cam on the sled to look high resolution images of the router bit from a window besides Ground Control? I found mine used in a flea market. A magnifying glass helps those webcam unable to focus very close. Small flexible usb reading lights are cheap and can light the router bit area too.
Now you can stop here if you want, having a better idea of the possible X offset on your maslow CNC. Just remember to set back the tolerance values you noted at the begining of this post.
This is an indication of the left-right symmetry error in you CNC. This kind of error creates distortions over the workspace in ways that get amplified in some areas. Typically, the top edge of the workspace gets too high on one side and too low on the other side. To the best of my knowledge, the single most important factor causing this symmetry error would be the chain (pitch) tolerance error that is also the only one to create an X offset. Has someone found another one?
Want to know more?
2- Tuning X offsset
Then tune in to bring the sled center on the line: change the chain tolerance on the side the chain is too long.
a) increase that chain tolerance for now. 0,05% at at time. Or adjust your change knowing that this is approximately 0.6mm horizontal shift.
b) Then close the settings pane, click on the stop button repeatedly (the firmware recomputes the x,y coordinates in what is called the forward kinematic calculation) and observe the x and y coordinates move each time to new estimations on the GC screen.
c) Once they stabilize, move the sled back to x = 0. If the bit gets away from the center line, reverse the tolerance change direction.
d) Repeat until your router bit center really looks on the line.
Now look at the difference between your two chain tolerance value. take note. This tells you how different apart your real chains pitches might be. The difference is meaningfull but not the exact value of each: this X offset correction can’t say if the chains are the right length. But this difference between the two tolerance value is likely very close to the right one.
3-What about vertical?
Using GC, send the sled to the bottom center of the workspace (x=0, y=-half work space height). It is likely that GC tells you it is there while visually you can see it is not exactly. If it is, then you might have a good chain tolrance estimation. Or not.
To place the router bit on the workspace bottom edge, you need to set right
a) the motor sprockets distances on the top beam,
b) rotation radius on the triangular link,
c) chain stretch correction,
d) the right common offset on your chain tolerances,
e) and the sprocket height over the workspace.
Too much of one can be compensated by too little of another and the bit may stand on the bottom edge. But elsewhere in the workspace distances would be warped.
Hope this little journey into chain tolerances was insightful to you. To go further, you can learn more about source of errors here.
So now is time to revert your left and right chain tolerance values noted at the beginning of this post. 
