The digital scale is the weak link, redo it with a decent spring scale with a slider and I'd be pretty surprised if your results weren't significantly different. It's an effect of the same behavior, the stress is building up faster than the latency of the digital scale, as there's no significant in the system, little vibrations in your hand are fluctuating the strain state, wildly changing the stress state in the material. In fact, if you wanted to do a better test with the digital scale, try adding in a solid ring and tie a piece of mono at equivalent or slightly greater lb test on-line with the fluoro.
This is why I lament most of the homemade line testers, there are a lot of complicated reasons why material testing is done on very expensive stable machines. Now we don't fish under laboratory settings, that's absolutely true, but these kinds of tests introduce errors and other artifacts that don't actually exist in the real problem. In this case, again the bend in the rod, bow in the line, give from your body, etc. change the fixturing so that the real loading condition isn't being replicated accurately. If nothing else, it's an explanation why fish can break fluoro at the boat when there isn't really any stretch left, but we already knew that.
In general, when you get results like that you have to ask if there's another common denominator that's skewing the test.
This is why I lament most of the homemade line testers, there are a lot of complicated reasons why material testing is done on very expensive stable machines. Now we don't fish under laboratory settings, that's absolutely true, but these kinds of tests introduce errors and other artifacts that don't actually exist in the real problem. In this case, again the bend in the rod, bow in the line, give from your body, etc. change the fixturing so that the real loading condition isn't being replicated accurately. If nothing else, it's an explanation why fish can break fluoro at the boat when there isn't really any stretch left, but we already knew that.
In general, when you get results like that you have to ask if there's another common denominator that's skewing the test.