Given the above discussions - I have a granular subbase (GSB) material. It is actually crushed stone. The specfications say that the GSB must have a Liquid Limit and Plastic Index of less than 25 and 6 respectively. The limits at 0.425mm is a range of 10 to 25% passing. The material is classified as non-plastic but the lab did do a liquid limit test and got a value of 27. Question: Does the material pass the specifications (in strict compliance)? The GSB is non-plastic - does it fail due to the "liquid limit" being greater than 25? and, is this a case that there really is no liquid limit to a non-plastic material? This is a real life situation.
As the guy out in the field, presented with the data, I would not accept the material as it does not meet the 25% LL Max. As an advisor to the engineer, I would recommend that, at a minimum the lab run the test again or review the labs procedures... they should not be outside of the SDev of similar labs.
My experience of looking at the liquid limit and plasticity index of sub-base materials is it's there to provide an indication of how the material will behave (similar materials act in similar ways), and is the material likely to be frost susceptible. Experience has shown that where a sub-base has fines that are plastic, the material is more likely to be frost susceptible, even to the point of 'suggesting' that if the fines are plastic, its not worth spending the money on a frost heave test.
Going back to the point about the result of the LL being 27, and the limit being 25, I think this depends on whether you want to approve the material or not. If your experience suggests that this material is suitable, and there is evidence that this source has been succesfully used in the past, I don't think it should be a problem - subject to all other tests complying. There is an increasing interest in the 'degree of certainty' of measurement, what are the statistical confidence limits on the result? - not easy to determine but even taking a rough +/- 5% certainty on the result you get a rough range of 25.6 to 28.4. If there is statisitcal data available you should get a much better 'feel' for the range of the result. Also its worth looking at the repeatability and reproduciblity of the testing lab - again if its available. I am not sure of the QA system in India, but in the UK under the UKAS system, you have to be able to provide evidence of repeatability, reproduciblity and tracability of testing and equipment. Its not called in to question often, but on 'big' jobs where the cost implication can be massive it can be a very powerful tool to allow the identification of erroneous results.
If it was me, I would not worry about a single result, but would want to see the spread of results on a selection of samples, or at least the 'typical' range of results on the material over a suitable period of time. Similar to the way in which the new EN standards look at the typical property as opposed to singular values.
If on the other hand you don't think the material will be suitable - you can always argue that its failed so its no good, but somehow I get the feel you value your engineering judgment above conflicting arguements.
I would also like to make the point (I think I've posted this before) that from my own experience of doing the test, it is VERY difficult to get repeatable liquid limits on samples which are non-plastic - hence my suggestion of getting a spread of values and making an engineering decision, a single reading on this type of material would not, in my opinion, be sufficient to pass/fail the material source.
Big H,
I was just looking at your real life situation. What did they end up doing? Based on your specs data of <25LL and <6PI, the required material would be fine grained. Since crushed stone hardly meets fine grained criteria, I'd say the GSB material should be rejected. What's the GSB supporting, anyway? Regards.
Just saw this thread. BigH-you haven't said what the %passing for #200 sieve is. If it is less than 50% passing, you have coarse grained soils and the LL and PI specs would not apply. As Iandig suggested run multiple tests and look at the range.
If the lab can run a sand equivalent test, have them try it and that will help us get better feel of these soils.
Aw - there's the rub. The material is a crushed stone GSB with less than 8% passing #200. Surely, it is not a fine grained soil so it must be classified as G or S if you go that way on crushed aggregates. The point is the nature of the fines - are they plastic or not? Most highway specs have a clause that say that the GSB must be non-plastic or, as in our case, having a PI<6 and LL<25 (meaning the fines portion, e.g., <0.425mm). This is so the fines that are there don't cause "harm" to the subbase (read that expansion).
Our PI was less than 2 or non-plastic. The "determined" liquid limit was really slightly less than 25% so it was passed. The point to my original post was a what if? Having 24% in the tests, we were close. We didn't get 26% - here, it would have been rejected by nearly all; for me, I would have passed it as the "liquid limit" (in this context) has no meaning without plasticity, in my view.
Thanks for your interest -
![[cheers]](/data/assets/smilies/cheers.gif "[cheers] [cheers]")
