I am slightly lost here… what is the whole point of this build. All I see is low print quality and crazy amount of failure points.Not quite sure what you are after.
@Matt_Hallett This is an experiment - an attempt to build printer with a slightly-larger build volume, that can do fast and accurate prints. Also aiming for simplest feasible fabrication.
You might be right about the quality and failure points. Or not.
So is this just a mock up and once everything works you are going to cnc all the parts?
I am not trying to be rude just trying to understand…
@Matt_Hallett No problem!
This is meant to end in a working printer. The complex specialized bits are meant to be 3D printed, much in the RepRap theme.
That said, I use mock-ups along the way to try out ideas. The hunk of plywood pretending to be a print-bed is not meant to be the final result. Earlier I used sticks of pine for my XY gantry (until the carbon fiber tubes arrived).
Also I am trying some pretty dubious things, to see if they might work. Getting some potentially interesting results, but the final test will be when/if this is a functioning printer. Got a ways to go.
That sketch is certainly inventive.
Never seen that before.
But I have tried and tested similar approaches.
Now, your mileage may vary, but I’m afraid not that much.
The problems start with the stiffness of your entire bed construction. As I look at yours, I see a lot of failing points. I believe you, when you state that the bed moves smoothly by hand, so did mine. But as soon as weight was added in the form of a real bed, and real heating, it still seemed okay. ( and don’t forget, you would want to be able to print big things, right? Think about the added weight of that)
But. There is always a but, isn’t it? But printing with the damned contraption didn’t go so well.
The layers came out in varied thickness, but since a fdm printer does not know the nozzle is 0.05 of a mm too close, it makes them wider. Took me some time to relate it all back to the power of the two motors I used. Even when I couldn’t hear them skipping, they did. Not much, but enough.
I tried several mechanical solutions, one of them similar to the sketch from your father.
Used pulleys and wire to do it. Switched to kevlar, because braided fishing line stretched just a little too much.
Oh, yes, before I forget, that kevlar is mean. You can cut through wood, even hard wood, with ease with that stuff.
All I ever accomplished with all those pulleys and wire contraptions was a funny kind of see-saw. ( Not so funny at the time)
I ended up with three motors with lead screws, one on each side of the bed, and one in the back.
All driven by one big external stepper driver.
This has the advantage that I can now move the bed accurately on all corners within 1/100 of a millimetre.
It has one disadvantage though, if something mechanical goes wrong with one of the lead screws, and that motor stalls, the whole bed gets skewed, and it’s recalibration time. It’s been more than a year ago that such shit happened.
But, you being who you are, feel free to ignore my and other experiences, and surprise us. Working good so far.
@Rien_Stouten Oddly, I am glad to hear you had the difficulty you describe (though that sounds wrong).
I am very wary of the print-bed wobbling, as 0.1mm seems likely to throw off the print, given the larger bed and taller volume. That is why I much prefer locking down motion on all corners (if I can come up with a feasible scheme).
But … there do not seem to be a lot of stories in the community about problems of this sort. There are at least a few CoreXY designs with cantilevered beds, and single screws … maybe this is a one of those cases where problems tend to cancel out. Gravity works. Maybe I am wasting time with more complex solutions.
Or maybe this is often a problem, but not well-diagnosed.
So I am glad to hear you had the problem described, and were able to diagnose. Helps me to feel my wariness is not misplaced. 
Also, what was on my desk, this morning.
@Preston_Bannister You are very right to be wary. The problems I had probably occur with a lot of printers out there, but as long as the bed is small, and lightweight, the effect is minimal.
And cantilevered beds? Not for anything bigger than 20X20 cm.
And even then. It isn’t all that strange a lot of people are using auto-bedlevelling. They are compensating for a lot of mechanical shortcomings, in a way I personally dislike very much.
I always learned the best way to solve a problem, is to attack the root cause, not the symptoms.
In my opinion, the best option for a larger bed is 3 lead-screws on 3 0.9 degree stepper motors. (3 point leveling is very manageable, where 4 point leveling makes your life hell)
Direct drive. No gearing. And if you do feel the need for gearing, use a belt.
That photo of herringbone gear-wheels brings back bad memories from the time I used those for my extruder.
The backlash of those things is small, but not zero.
And the noise it made retracting…
Errors don’t cancel each other out, only seem to do that, and tend to create bigger problems in the long run.
You say you are a software guy, right? Go thinker with that, and not with hardware, which is very unforgiving and often costly.
Not to mention all the misery coming from wrong choices and building on those errors.There is very little room for error when building big printers, and everything matters.
@Rien_Stouten Well, we are in agreement with all except the last, and you might be proven right on that also, in time.
That said, I will continue the current experiment.
@Preston_Bannister
You would dissapoint me if you said anything else.