Pavement Subgrade Question

[swazimatt]

We are doing a new industrial subdivision and the insitu subgrade is quite variable. The natural material is (top down) topsoil - sands (CBR 6) - cohesive silt layer (Cbr 1 and lower) and rock. The depths vary across the site. the pavement has been designed for a CBR6 subgrade with subgrade improvement for lower quality subgrade

Where the pavement is in the cohesive layer we will cut out and replace with subgrade improvememnt layer 600mm thick (CBR10) with a separation geotextile and geogrid at the base but on our circly design we will still not acheive the required 1.2mm benkleman beam deflections. (this is question 1 - any advice to improve this?)

My main question is, where our pavement is still sitting in the upper sandy layer (CBR6 no subgrade improvement layers), what depth should the insitu subgrade be a minimum of CBR6, knowing that there is a weak subgrade layer somewhere underneath it? If we have 400mm of sand and then the cheasive silt under that is that ok? My feeling is that we should have at least 600mm of CBR6 under the pavement, but i do not want to have to remove 400mm of soil if i don't have to. the silt is confined and can't go anywhere (is this a valid point?).

 

[swazimatt]

Are there no pavement engineers here?

 

[CarlB]

What pavement design method are you using?
I'm not familiar with using BB for design of pavements. A quick search indicates it's more for assessing existing pavement, and designing overlays.
I assume this is asphalt pavement? And there will be some depth of aggregate layer beneath asphalt?

Hard to offer an opinion without knowing pavement loading. Assume lots of heavy trucks, for an industrial road?
A typical design procedure which considers the CBRs of various layers should give you answers you seek.

One such pavement design method using CBRs can be found in this UFC:

https://www.wbdg.org/ffc/dod/unified-facilities-criteria-ufc/ufc-3-250-0

 

[SlideRuleEra]

...new industrial subdivision and the insitu subgrade is quite variable. The depths vary across the site.

1) ...what depth should the insitu subgrade be a minimum of CBR6, knowing that there is a weak subgrade layer somewhere underneath it?

2) My feeling is that we should have at least 600mm of CBR6 under the pavement, but i do not want to have to remove 400mm of soil if i don't have to.

3) ...the silt is confined and can't go anywhere (is this a valid?)

1) The answer will depend in large part on the size and type vehicles that will travel on the pavement... fork lifts, heavy trucks, heavier off-road equipment, crawlers (bulldozers, cranes, etc.)

2) Go with your engineering judgement on this one to the extent that you don't skimp on thickness. IMHO, forget about trying to "fine tune" the design, go with a constant thickness everywhere... because:

Cost of this type project is based on more than the design; simple uniform construction (constant thickness subgrade, in this case) allows contractors to work efficiently. If the design calls for differing thickness at various locations the contractor is forced to spend time, effort, and Owner's money coordinating the work.

3) Maybe in theory, but not in practice. If the subgrade is thick enough the entire pavement can settle uniformly as the silt is "pushed" horizontally from under it.

This is a big problem where I work (South Carolina coastal region). In my case, roads at our electric generating stations are sometimes not paved, because settlement continues constantly (for years, and counting). Every couple of years a new layer of suitable material is added on top to maintain drainage.

 

[swazimatt]

Thanks. The more i think about it the more it makes sense that if there is decent soil and then poor soil i should just add the better soil as a subgrade layer and the poorer soil as subgrade extending for infinite depth.

I agree with a simple single pavement approach, the total length is about 300m so any changes in subgrade improvement layers is not going to make a significant impact on the project (and with variability and spacing of scala/dcp testing it makes sense to be a little conservative)

 

[EireChch]

SRE - IF you are saying design then entire pavement with the worst case or near worse case CBR and adopt a thicker pavement, then i would disagree.

With a good geotech engineer on site doing some DCPs and a surveyor you could easily mark out a substantial area that could be designed to a better CBR and thus thinner pavement. This could mean savings for imported material, construction time, traffic on roads etc.

 

[SlideRuleEra]

EireChch - I won't disagree with you for a "large" project.

Many of our in-house projects of this type were relatively "small", like the OP's. With a simple design, we did not have to retain (and pay for), consulting geotechs, surveyors, or full time field construction management.

Note: I'm not saying we did not have professional geotech advice, we would spend a few million dollars for site-wide investigations before design / construction of our electric generating stations.

We would not retain consultants for every follow-on in-house project. Especially since the initial geotech investigation showed that soils at the location of our generating stations were typically "garbage"... just the way it is when you build adjacent to a waterway surrounded by a "swamp" (in order for the operating station to be able go get water easily).