Core One - Belts tension adjustment - Your opinion, please
Hello to all Core One owners.
I was fortunate enough to be able to assemble my Core One myself. Initially, I only adjusted the belt tension using the app, and everything was great.
However, after a few test prints, I noticed that the VFAs were still present and visible at certain angles (especially at 45 degrees). An additional VFA test print from OrcaSlicer confirmed this.
The manual belt adjustment in the Core One menu was well described and worked fine, but somehow the results were worse or differed from the app version.
I was confused by the instruction at the end to “turn 2/8 of a turn to the left.” There is no scale or reference point where you can turn exactly “2/8 of a turn” from point A to point B. No matter how hard you try, it remains an “approximate setting” (in my opinion). I was about to print out a scale and hold it up when I had an idea.
I thought I'd reverse the order and save myself the step of first finding out what frequency I currently have on the belt and adjust it right away.
Using the control dial on the xLCD, I first set the frequency on the upper belt that Prusa specifies as optimal. Then I simply tightened the belt with the adjustment screw until the maximum deflection was visible on the LED control on the belt. I repeated the same process for the lower belt.
And lo and behold, it worked immediately. The deviations from the acoustic “Belt Tuner” in the app were significantly smaller with this method.
I created a GitHub ticket for this.
https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/5095
If anyone would like to test this, feedback or criticism is welcome.
RE: Core One - Belts tension adjustment - Your opinion, please
IMO, if things are very close this might work, but because of the interaction between the adjustments and the overriding need to keep the X-axis square, it can't be recommended. You can only control two of three things when adjusting the tension. You have the upper belt frequency, the lower belt frequency and the squareness of the X-axis. Pick any two. Hint- it better include the last!
I set up squareness with the belts somewhat loose, as is well described. After that I only move the adjustments by small amounts and by the same amount, to get the upper frequency set near the top of the range. In theory, that's it, but I then check the squareness and tweak one or the other adjustment to get it perfect. Then I run the utility again. It will usually say no adjustment is needed, or a very small one. The squareness is a very good indicator of equal belt tension, if it was square to begin with.
You could print a slip-on scale for the Torx wrench, but I just go by the different color flats. It's easy to work in eights of a turn, as you're just splitting quarters.
RE: Core One - Belts tension adjustment - Your opinion, please
It's usually not a good idea to tighten the belt screws asymmetrically. You could end up introducing skew in the gantry. Also the 98/92 reference points are just that. Reference points. The belts don't need to be exactly those frequencies. As long as the top belts is 4-6hz higher than the bottom belt and you are inside of 90hz then you are fine. VFA's are a fact of life with rigid frame Core XY printers. They all have it. The easiest way to get "rid" of them (i.e. hide them) is to use matte filaments. Or to not care.
As for 2/8 (1/4) of a turn, that is 90°. I have a Wera hex driver and put a mark on it with a Sharpie.
RE: Core One - Belts tension adjustment - Your opinion, please
To be honest, at first I also thought, “Damn, how much is 2/8?” 🤣
It's usually not a good idea to tighten the belt screws asymmetrically.
Yes, okay, that's what I did at first. Adjust evenly on the left and right to get the right frequency.
That's why I wrote that the belt needs a certain amount of tension, which may be visible from the outside via an indicator.
IMO, if things are very close this might work, but because of the interaction between the adjustments and the overriding need to keep the X-axis square, it can't be recommended.
I will work through it step by step several times over the weekend and test it.
I would argue that if these metal brackets were more solid, you could also adjust them individually without running the risk of deforming the frame.
RE: Core One - Belts tension adjustment - Your opinion, please
You might, but they aren't and the interaction between the adjustments would drive you crazy. The belt tuning utility works the way it does for a reason, and it works well. I went through some similar thoughts when first starting out, enough that I dug fairly deeply into the system. I'll eventually set up a Core One page on my site and edit and move this but see if you can get anything useful from it. I'm new enough at this that I still make mistakes! https://conradhoffman.com/CoreOne_assy.htm