Adapt, improvise and overcome. After all, 'engineer' and 'ingenuity' come from the same root word.
"...students of traffic are beginning to realize the false economy of mechanically controlled traffic, and hand work by trained officers will again prevail." - Wm. Phelps Eno, ca. 1928
BA, with all due respect, what kind of advice is that? Perhaps he could do a test or two or ten, getting some sort of value for the friction coefficient. But what sane engineer would use that value for design? What does he do if there's a problem down the road and he's asked to defend his design? "I stacked some weights on these two pieces of material in my garage and pulled them real slow until they moved." Or "I lifted the plate a little until the weight slid, and I grabbed my kid's plastic protractor to measure the angle." Rather, the prudent move is to look for documentation of values already determined, under more rigorous testing conditions.
It is not a jest. It is a very simple experiment which we performed in class during first year engineering. The coefficient of friction between vulcanized fiber (or a similar material) and glass is likely dependent upon who manufactured the fiber and the glass, and the surface roughness of each.
And nutte, there are better ways of measuring slope than your kid's plastic protractor.
I too have done a number of "back yard" tests to get a feel for what a likely answer might be. If they were very consistent - I might go with them - with appropriate SF.
That said - it would not cost all that much to take your ideas and results to a professional lab and have them redo and/or certify the results.
Coefficient of friction for fibre on fibre is 0.16, for fibre on smooth cast iron is 0.21. Unfortunately, they don't provide figures for fibre on glass.
I'm a bit surprised that people think that using manufacturers' data from the Internet (which may well not be exactly the same material under the same conditions) is more reliable than doing a test fore yourself.
In any case if the friction value is critical to the design you should do analyses with upper and lower bound values, and it would be a good idea to do your own tests in addition to looking at published values to establish reasonable values for those.