Compensating technics for long ribbon driven axes

hsverre

Hi

I have built a 3d printer with 80x60 cm buildplate z is about 80 cm. I did this only for test purpose, i had very low expectations of a printer this size and ribbon driven.

The ribbons have limits when beeing this long but i wonder if it is possible to make programmable tunings to compensate for mechanicle limitations.

A ribbon driven printer is by far the easiest compared to big heavy lead screws. It think it would have been a really great idea to have electroniccly compensation for ribbon cables, aka ressonance compensation/accelerometer. Is there possible other methods to compensate?

Anyone have thought about this? This printer has a Duet Wifi and Duex 5

Regards

dc42

@hsverre resonance compensation is available in RRF 3.4.40, it's called input shaping. There's a summary at https://docs.duet3d.com/en/User_manual/Tuning/Input_shaping.

hsverre

@dc42 said in Compensating technics for long ribbon driven axes:

@hsverre resonance compensation is available in RRF 3.4.40, it's called input shaping. There's a summary at https://docs.duet3d.com/en/User_manual/Tuning/Input_shaping.

Thanks for your reply.

I know about the input shaping but thats for resonance. I guess the input shaping does not compensate for the lack of rigidity in ribbon belts?

I see that the the tolerance is acceptable when the print head is in the center of the printer. I guess its because there is aqual amounts of belts retracted from the stepper motor and the lack of rigidity is equal on each side of the print head.
But as soon as the print head is at the most extreme side there is a long ribbon retracted on one side (x-) and very short on the other side (x+). And therefore there is to little precision.
The Voron 2.4 fex does not recommend to have builds exceeding 35 cm xyz beacyse of belts limitations.

I was thinking about this also to make bigger 3d printers more accessiple to people who does not have access to costly big industrial precision printers.

mrehorstdmd

@hsverre Are you concerned about the belt stretch? I can't say much about it in the X and Y axes, but there shouldn't be any problem with a belt lifted Z axis longer than 35cm. My printer has 69 cm belt lifted Z axis and I often make prints that are 600 mm tall without any issues. I have measured both steel and glass core belt stretch in the Z axis and found it small enough to be inconsequential.

I measured stretch under multiple loads at about the center of the Z axis](https://drmrehorst.blogspot.com/2018/12/comparing-steel-core-and-glass-core.html), and yes, the numbers probably vary a bit depending on the bed position- less stretch near the top and a little more when the bed is near the bottom, but we're talking about a few microns. The roughness of the print's top layer is probably going to be worse than the error caused by Z axis belt stretch for most prints.

The ultimate stretch of the Z axis belts should be easily predictable for any given print -the slicer tells you the print mass, and a simple test tells the amount of stretch to expect for a given mass, so simply scaling the Z axis in the slicer should be able to compensate for stretch of a super heavy print and/or a super long Z axis, if you're really worried about it.

If you're worried about belt stretch in the X and Y axes, you can select a mechanism that minimizes the belt length- maybe drive the Y axis with two belts and mount the X axis motor on the X axis to keep belts as short as possible. Corexy might not be ideal if you're worried about belt stretch in X and Y in a very large printer.