QUESTION: Hello sir. I am currently working on a thesis about mixing silt with high plasticity with shredded rubber, 10, 20, 30, 40 and 50 percent by weight. My hypothesis is that in a certain percent by weight the shear strength will reach its optimum, say 20 percent. I am just wondering, if i would only determine the trend of the behavior of soil, should it be valid to consider only one trial, considering also that the increment is 10 percent by weight? Is it considerable enough? If it is considerable, what is the basis? Or else, should i pursue three trials in order not to be biased with the result?
ANSWER: Dear Erwin
This is kind of problem I like to think about. I have had similar problem several times before.I need a couple of day to study more about and I will let you know.
---------- FOLLOW-UP ----------
QUESTION: Thank you sir. I appreciate that. It would really be helpful because it will save time and money. :)
I suggest you a try and error strategy. For example try first with 3 samples and for a controlling process try 2 more ( between 1 and 2 and between 2 and 3 ) if the modification of the average and standard deviation is not more than 10% it is not necessary to continue and 5 samples is ok. If not you should continue with 4 more samples. This is an example . You can use your own strategy to satisfy your project requirements.