Failed Crown-test

So all you need to do is make sure your machine is square before you hit start.

Each roller should be the same distance from the corner as it’s partner. Both X rollers should be equal distance from the corner as well as both Y rollers should be the same distance as each other. If you start it square it will make a square. Easy.

At the same time…0.8mm should be close enough for most projects, that is 1/32" I would suggest designing with slightly looser tolerances and things will be much more enjoyable to make.

Yes. I check these distances before the start. But, the diagonals still do not converge.
I noticed that the pulley has 16 teeth.
The repetier-host stubbornly calculates the course of the belt for 18 teeth. Could this be the cause of curved diagonals?

It does not effect the steps per mm, the firmware does. Even if the steps were wrong it would still draw equal diagonals.

What are the diagonal measurements?

56x55.5 cm

Two squares. The deviation is the same in both cases.

My details are larger than this square and there are more divergences of diagonals.

 

UPD:

Accuracy disappears in the process of drawing a squares.

I carefully adjusted the distance between the rollers before the start.

Then the machine drew a familiar black square. I carefully checked the distance between the rollers “after all”.

I found a gap.

Take a look /

UPD:

On the short side, the accuracy is maintained (even though less quality rollers).

The loss of accuracy after the “square” is observed only on the long side.

 

Are you sure all of the pulleys have the same number of teeth?

Yes, my friend.

The problem was in the pulley.

One of the setscrews fell out. Another one was barely able to hold on.

As a result, the pulley slipped out on the sheave.

I fixed the pulley properly on the sheave and got a fine square.

 

Thank you very much.

Nice!!!