CBR or not to CBR, that is the question 1

[Ge0dude]

CBR is an undrained/total stress deformation behaviour test on an insufficient depth of ground profile, which provides highway technicians arbritrary values to design their sub-base thicknesses.

The majority of road embankments, specifically in the UK, are founded on clay, in which the undrained CBR is worthless, as the long term drained solution drives the design.

Why do engineers still use CBR, when it doesn't provide an accurate representation of the deformation behaviour of the entire ground profile that will be stressed. Its analogous to using a 0.6m dia plate load test to derive a bearing capacity for a foundation with width of 5m. It doesn't provide sufficient stress/strain data on the required depth of influence. Couple this with not providing long term settlement behaviour. What is the point unless your road is literally built at grade on at least 5m deep granular fill.

Let's not forget that most insitu CBR tests are invalidated during the earthworks, where the Contractor either strips the top formation or raises the levels using engineered fill.

Why are engineers/consultants still wasting their Clients money on these outdated obsolete tests.

 

[Settingsun]

Never seen it used for embankment design, just for the pavement which has short term governing loads (moving vehicles).

Speaking from outside geotechnical engineering, I'd guess the body of knowledge that's been built with reference to CBR is at least one reason to keep using it. There would be hardly any history of pavement design based on other parameters in comparison to CBR. Similar to how geotech finds the jump to limit state design harder: overload conditions haven't been explicitly considered in the past so how to judge the higher loads that come from limit state design is not clear.

 

[dik]

steveh... you pretty much summed it up; there is a lot of history of its use on different soils. For clays, in particular ones with a high PI, there is not a simple alternative test.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[geotechguy1]

I think you'd probably tackle the problem of settlement of embankment fill separately from designing the pavement surface; two different problems. You can do lab CBR tests on material from the borrow source if that's needed.

In general many areas of geotechnical engineering are in need of newer tests and standards. Unfortunately new tests and standards open up many cans of worms. For example, in many areas of North America there is a degree of conservativism in how the results of SPTs are translated into design of pile foundations which is not captured in any code book or manual. According to the manuals, the design might have a factor of safety of 2 or 3 (or a resistance factor of 0.5). In many practical cases if you do a load test on a pile you'll find out the actual factor of safety is much higher, maybe 10 or more - which accounts for a great deal of things like variability in soil conditions, dubious piling contractors, dubious SPT to pile correlations, dubious SPT maintenance etc. If you suddenly switch to designing piles based on CPTs and test piles with PDAs and do the entire design with an FS of 2 on the much higher parameters derived from your load test (or resistance factor of 0.6) without also addressing all of the things implicitly covered in the unspoken factors of safety you're wading into unknown territory.

 

[Ron]

CBR is a drained test, not undrained.