Thoughts on the usefulness of the USCS and it's derivatives 1

[geotechguy1]

I'm curious what everyone's thoughts are on the usefulness of the USCS and it's derivatives to classify soils for engineering purposes (or more broadly the joys of soil classification)

The advantages I see: Has been used extensively, large availability of correlations

Some points I've been contemplating:

[ul]
[li]Is the A line on the Atterberg chart actually useful? I've found the usefulness of it to largely be: more plasticity = more clay content (or a higher fraction of more active clay minerals) The silt/clay choice being based on being above or below the line has never made any sense to me. The golder version of the USCS calls plastic soils below the A-line 'Clayey silt' rather than 'silt'. I've also worked for employers that just ignore the A-line and call all plastic soils 'clay' with modifiers for varying silt and sand content.[/li]
[li]Is the 50-50 split between 'fine grained' and 'coarse grained' actually useful? I've never found calling a soil that's 55% sand, 20% silt, 25% clay 'clayey sand' to be a useful classification - and contractors and developers often see the word 'sand' and interpret it to mean 'beach sand' and then accuse the geotech of being a moron when it looks like plasticine. [/li]
[li]Does a USCS classification actually tell us anything about engineering properties? Eg. Strength, stiffness, permeability [/li]
[/ul]

I' find completely ignoring the soil classifications and solely looking at shear vanes, SPTs, lab results and CPT results to be substantially more useful.

 

[EireChch]

I havent worked that much with USCS classification, however Eurocode classifies soil with more than 35% fines as cohesive, which I find to me more realistic as it only tanks a small portion (15-35%+)of fine grained soil to make to whole material perform as a fine grained soil

 

[geotechguy1]

Yes, I agree with the 35% as an improvement over 50%. NZGS uses 35% as well. NZGS has it's own flaws (namely that I haven't found any consultancies that do atterbergs, sieves, hydrometers and moisture contents often enough to train field staff to 'calibrate' their logging and thus all the field classifications are absolute total bullshit).

The 15-35% criteria - perhaps even 10% if it's bentonite - is how I have always understood it, and in Canada at consulting firms I worked out the majority of engineers largely ignored the MUSCS/USCS as it suited them and classified soil according to engineering behavior - which sometimes means a soil can be 20% fine grained but it may be preferable to classify it as a fine grained soil. I'm not sure how things are in Europe or Britain - I have heard in Europe things are much more code-heavy which may mean they take the classification system literally.

 

[geotechguy1]

I wonder if anyone's ever thought of using the 'Activity' of the clay fraction - or better yet maybe activity of the clay fraction normalized by by the % passing number 200 sieve) as a criteria to make the fine grained - coarse grained decision. I haven't been able to dig up much research on soil classification though.

 

[jdonville]

I vastly prefer USCS to the textural classification system mandated by the DOT in Indiana (and I think also maybe Illinois?) where it seems everything is some kind of 'loam'. Only useful IMO if you're a farmer.

The 35% rule on fines is new to me, but seems to make some sense, unless you have overwhelmingly silty fines.

 

[geotechengineer1]

According to the USCS a high plastic clay constitutes 'silty fines' as long as it plots below the A-line.

 

[fattdad]

VDOT (Virginia Department of Transportation) uses ASTM D2487 (laboratory) or D2488 (visual). These represent standards that are available to all. I really don't like Burmister or USDA. I don't want folks to make claims on weight percentage of clay or silt. The ASTM methods use weight percentages for sand and gravel, but behavior (i.e., LL, PI) to reference the nature of fines. I don't really care whether the fines are clay or silt sized (ala USDA) as I'm not in the pore-size/root growth business!

f-d

ípapß gordo ainÆt no madre flaca!

 

[geotechengineer1]

Fatdad, the USCS makes the distinction between gravel and clay based on weight percentages and between silt and clay based on nonsense.

 

[fattdad]

geotechengineer1, Huh? Gravel and sand, based on weight. Nature of fines (i.e., whether clay or silt sized) based on behavior. It makes sense to me? Then again, maybe that's a reflection on me!?!

f-d

ípapß gordo ainÆt no madre flaca!

 

[geotechguy1]

Gravel is a soil with more than 50.01% coarse grained material and more than half the coarse grained material is gravel. 49.99% clay, 25.1% gravel and 24.9% sand is a "gravel" according to the uscs. If the fines plotted slightly below the A line you would call it a silty gravel...which says nothing useful about the engineering properties, which supposedly is the entire point.

 

[MTNClimber]

I've used modified versions of the Burmister and USCS systems on the east coast and Colorado, respectively. It seems like each company has their own take on how to classify soil.

The modified Burmister is relatively simple to me. Visual classification is split up by gravel, sand, and fines. Proportions are "trace" 0-10%, "little" 10-20%, "some" 25-35%, "and" 35-50%, and the major component is just listed on all caps. It also breaks out sand and gravel by fine to course particles sizes. Fines are visually ID'ed by field thread diameters, i.e. Clayey SILT can roll 1/4" dia, SILT & CLAY can roll 1/8" dia, CLAY & SILT rolls 1/16" Silty CLAY 1/32"... ect. Example: fine to medium SAND, some Clayey Silt, little fine Gravel paints a relatively complete picture of the gradation of the soil to me.

The modified USCS version that I've used in Colorado is lacking in my opinion. Visual classification is as follows:
0-4% Not described
5-12% "slightly" - silty, clayey sandy, gravelly
13-35% silty, clayey, sandy, gravelly
36-45% "very" - silty clayey, sandy, gravelly
45-55% and/or

example would be "Slightly Silty Clayey Sand" which paints a vague picture in my mind and requires an actual lab gradation for me to fully understand. My other gripe is that Silty Sand could have 13% fines or 30% fines which varies A LOT from a material use perspective.

Those modified USCS rules were not hard lined because if it was 34% gravel, 33% sand, 33% silt then it would be classified as Silty Sand and Gravel (SM-GM). One company I worked with also didn't use "gravelly" instead would use "with Gravel" at the end (Silty Sand with Gravel).

 

[GeoEnvGuy]

Similar to FD i use ASTM D2487 (laboratory) or D2488 (visual). Also I find value in knowing the difference between a very loose silty sand (SM) and a very soft lean clay (CL) in terms of shear strength.

 

[geotechguy1]

When you think of silty sand is your mental model a cohesive soil or a non cohesive soil

 

[BigH]

MTNClimer - in our practice in Canada (the original Geocon and then Golders as all those whos started Golders were originall with Geocon) was to use 0-10: trace; 10-20: some; 20 to 35: -ty; 35 to 50: and. Of couse, now, thing might have changed but I haven't been practicing in Canada since the mid-90s.

 

[geotechguy1]

https://www.barrie.ca/City Hall/Pla...echnical & Hydrogeological Investigations.pdf

This is the one I've seen recently from Golder

 

[GeoEnvGuy]

When I think the field engineer writes down silty sand, I think a non cohesive material that shows a rapid dilatency and contains primarily fine grained sand. Both the description of the grain size and dilatency should be described by the field engineer.

 

[fattdad]

when I see, "Silty Sand" I immediately want to know the plasticity (elasticity) of the silt. I just don't have a rebuttable presumption that it's non-plastic.

f-d

ípapß gordo ainÆt no madre flaca!

 

[MTNClimber]

I guess that's what lab testing is for. If it's important information, visual classification shouldn't be the only tool used.

 

[geotechguy1]

Which silt particles cause plasticity, fattdad

 

[fattdad]

the elastic ones. Unclear question?

f-d

ípapß gordo ainÆt no madre flaca!