Belt tension vs. temperature

[ReXT3D]

I am curious if anyone has investigated the magnitude of change in belt tension vs. chamber / frame / component temperature?

I noticed on my recently built 2.4 that there is a very significant increase in belt tension as the chamber heats up. While this is to be expected, the magnitude of this increase was a bit surprising to me. For example, the AB belts on my 2.4 are nominally adjusted to approximately 1.7 lbs (~100 Hz @ 150 mm) at room temperature (21-22 C). When the chamber is heated to just above 50 C, the tension increases to nearly 3.5 lbs (~145 Hz).

Note that I reduced the initial recommended target tension of 2 lbs because I was troubleshooting some very faint trace indications of belt teeth resonance in the Y axis.

 

[NoGuru]

No, but I was curious about the same thing. So would it be better to tune the belts lower and have them come up to 110Hz, or spit the difference? Hard to say but I would think somewhere in the middle and as long as the tension is equal.
I don't particularly like tuning by Hz because each pluck is different. There has to be a better way. The Prusa belt gauge seems to be somewhat decent in doing this based on belt tension.

Printables

www.printables.com

 

[ReXT3D]

The reason I posted this in the first place is because I see something vaguely resembling belt artifacts in my prints, but I don't think they are belt artifacts. Using genuine Gates pulleys and belts...