95% vs 100% MDD - calculating difference in self weight settlement? 1

[jth88]

Situation - I have been recommending 95% compaction for controlled fills and using the rule of thumb that you may see as much as 1% of settlement from the controlled fill due to self weight (assuming a silty sand). Additionally, I understood that this settlement predominately occured during the construction of the fill.

I am with a new firm and they want to recommend 100% MDD with the belief that this will eliminate the self weight settlement. They also do not believe the self weight settlement occurs during construction.

I have spent a few hours trying to calcualte this behavior without success. Has anyone been through this exercise?

Thanks,
jth88

 

[GeoPaveTraffic]

I would not expect a significant difference in self weight settlement with 95 vs. 100% compaction. Also, settlement will start as soon as the material is placed, period.

One other thought, if the fill is being placed on a compressible foundation, the extra fill weight will create additional foundation settlement.

 

[jth88]

GeoPaveTraffic - Thanks for the sanity check. Is their a way of mathematically proving our beliefs? I will have to have something to change supervision's beliefs.

Settlement of the foundation soils is an additional source of settlement that was calculated, both quantity and time rate.

Thanks,
jth88

 

[Ron]

jth88...I agree with GPT....little difference. It takes very little compactive effort to get from 85 to 90 percent compaction in most materials, a fair amount of energy to get from 90 percent to 95 percent, and a lot of energy to get from 95 to 100 percent. Once you achieve 95 percent compaction, it takes something relative strong, not just self-weight to cause that last 5 percent of compaction...for instance, the additional compaction can be gained in a roadway under heavy traffic, but most likely would not occur under static conditions. Keep in mind that soils compacted near the surface are usually of select grading and usually not subject to much further consolidation, since about the only further consolidation they will undergo is elastic compression, which will be done by the time construction is done.

If you have clayey or silty materials, more movement will take place.

 

[kslee1000]

I would agree with Ron, further consolidation "within" well compacted fill will require much more than just selfweight, however, depending on site conditions, the settlement is likely to occur within the underlay soil layers immediately upon placement of the fill, and may continue long after the construction.

 

[BigH]

If your fill is placed in thin lifts - say 250mm max loose lift, and you are using a heavy vibratory roller, you will be compacting lower lifts while the upper lift is being compacted. Therefore, settlement of the first, second and even third previous lift will occur as it gains compaction energy from the upper lift. See D'Appalonia's work on the subject (as seen and referenced in Terzaghi Peck & Mesri, page 311-312). In India, we placed sand fill in 1000mm lifts and did compaction checks at three different levels in the lift and the compaction was fine (note that this was very clean river sand). I usually test the surface of a sand lift - if passes - then okay but if it is a bit low, I scrape out 75 to 100 mm and retest. Usually it will pass if the work was done properly. In my view, in compacting sands, you should be checking the previous lift, not the lift being placed.

 

[msucog]

it sounds like you're talking about fill induced settlement? (settlement of the underlying soils due to the new fill weight). usually for well compacted sandy/silty fill (95%+ standard), i typically do not see significant settlement within the fill itself...but "substantial" settlement of underlying soils due to geology (silty sands/sandy silts). for large/thick fill areas, i typically see (and calculate from laboratory testing) 2%+ of the soil profile thickness (in rough, round figures--keep in mind it is dirt). begins immediately after fill placement and is pretty "quick"...but quick could mean weeks or many months (again silty sands/sandy silts--not clays)...so confirm with field monitoring.

 

[BigH]

msucog is right - but I took the OP to be "of the fill itself" - given that 95% would be less stressful to the foundation soil than 100%.

 

[jth88]

All - thanks for the responses. Sounds like most feel that the difference in self weight settlement between silty sand soil placed at 95 and 100% MMD is insignificant. I could use the math to prove what we collectively know is true or a suggestion on how to mathematially prove it and I will do the math.

BigH - you are correct, I am interested in the self weight settlement of the fill. Or how can I mathematicaly show that an incresing from 95% to 100% MDD does or does not reduce self weight settlement. For this exercise I am not worried about the settlement of the underlying foundation soil.

BigH - not sure what you mean by "given that 95% would be less stressful to the foundation soil than 100%."

Ron- do you have any references on the energy increase required to go from 85 to 90, 90 to 95 and 95 to 100 percent compaction.

Another problem I have with requiring 100% MDD is the difficulty it will cause the contractor. He will have to watch the moisture content very closely or he will have to use larger equipment to make sure he is above 100% MMD based on Standard Proctor.

Thanks,
jth88

 

[fattdad]

If you are dealing with silty sand, settlement of thick fills is not likely a problem. If you are placing and compacting a sandy lean clay, lean clay with sand or such, then compaction will result in increased pore pressures in the soil fill. This is especially the case for thick fills (i.e., greater than 10 or 15 ft). For the case that you are placing a thick fill of fine-grained materials on a completly rigid soil (theoretically), then some measure of settlement after compaction can occur.

In Central Virginia (i.e., in the Piedmont), there are many projects that use fine-grained earth fill. When the earth fill exceeds 10 or 15 ft (thicker?), I recommend settement plates.

Regarding 100 percent Modified v. 95 percent Standard. I always reference 95 percent Standard - never had a problem. I'm also very critical of the reference Proctors being used and insist on confirmation one-points and full curves on a regular basis to make sure we gauge that the reference density that we're using is accurate.

Regarding 100 MDD v. 95 SDD, yeah it may be "better", but why not require 105 MDD, that would beat 100! Somewhere there has to be reason included into this discussion. I mean what would be the contract value of 100 MDD - ALOT!

f-d

¡papá gordo ain’t no madre flaca!

 

[BigH]

muscog was worried about settlement of foundation. If you have a fill placed at 95% MDD (say, for example) equivalent to 95pcf and if it were placed at 100% MDD, say 100 pcf, and a fill of, say, 20 ft in height, which compaction level would impose a greater stress at the surface of the foundation soil? 1900 vs 2000 psf. Not really significant I know . . .
As for calculating the effects, start with a layer of x01 thickness and assume an MDD of, say 110 pcf and a moisture content of 10% and assume an E value of E01. Put another layer on top of x thickness. Calculate the stress in the first layer x01. Estimate the settlement due to x02 as delta p x 0.7/E (assuming poisson's ratio of 0.3). Then put on another layer x03. Compute the stress increase in x01 and x02. (neglect, for now the increase in E). compuate the settlements of x01 and x02 due to x03. Continue. Then do the same for 100% MDD.
The fallacy of the exercise is that (1) you are assuming uniform loading in an elastic medium (not a particulate medium as put forth by M.E. Harr), (2), the compaction is exactly 95% MDD and that in placing the layer and compactin, you do not get any increase in compaction of the underlying layers, (3) E remains the same throughout the compaction process where, in fact, it would slightly increase due to additional confining pressures, etc. . . .
I suppose you could try it for 10 layers or so and see what difference you get - for academic exercise.

 

[aeoliantexan]

I have never seen much post-construction self-weight compression of lean clay fill so long as the moisture content remained nearly constant. If a lean clay or clayey sand fill becomes saturated or nearly so at some time after construction, it may well compress under its own weight. I use a rule of thumb of 1% to 2% of thickness for this situation if the fill was compacted to 95% of Standard Proctor density and was more than about 15 feet thick.

Poorly compacted fill or fat clay may compress more. Fat clay, of course, may swell, especially under small overburden pressure.

Some research on "Hydrocompression Setttlement of Deep Fills" was reported by Brandon, Duncan, and Gardner in the Journal of Geotechnical Engineering Vol 116, No. 10, October, 1990. See also "Collapse of Compacted Clayey Sand" by Lawton, Fragaszy, and Hardcastle in the same Journal, Vol 115, No. 9, Sept, 1989.

Some earth dam designers use a 1%-2% rule of thumb for settlement of the embankment, expecting half of this during and soon after construction and the other half as seepage develops. I have generally found this to be conservative if 95% compaction was specified and enforced.

 

[darthsoilsguy2]

to avoid confusion reference whether the Proctor is standard or modified.... like "let's compact 95% Standard... or 100% Modified" The MDD is throwing some people off becase Max Dry Density is a given. i understand from your second post that they use 100% Standard Proctor Maximum Dry Density (SPMDD or STD) as their general spec.

95% Standard has worked for a long time... i'm hesitant to believe that any "raising the bar" has been based on 95% STD actually failing in the field (as opposed to be perceived as failing). seems to me like the bar is just being raised to get a Safety Factor on getting plain old 95% STD.

 

[iandig]

Just to throw my 2 pence into the discussion.
Based on my own experience and the soils I normally come across, then there is a case for distinguishing between 95% and 100% of the 'standard' [2.5kg] proctor. Most of the earthworks I get involved with have a high proportion of cohesive soils. Following current standards [SHW & BS EN] National Guidance [BRE] and best practice, we do distinguish between the two levels of compactive effort. This is of note because we find that in many instances, 100% of the 2.5kg rammer is close to 95% of the 4.5kg. Please note this reflects the cohesive soils from my region and would not want this to be a 'global' generalisation. Where we are placing cohesive soils as 'general fill' then typically we would look at the 95% of the 2.5kg rammer, which in turn 'typically' returns shear strength of around 50 kPa and/or CBR in the order of 2.5% on the wet side of the curve. 2.5% CBR is the minimum allowed for sub-grade to pavement foundations [IAN 73/06], below this you "have to do something else..." and this corresponds well with the min shear strength for trafficability. Where the fill is placed below more sensitive structures, i.e. buildings, we are reccomended to use the 4.5 kg rammer, as this should have a higher bearing capacity and less suceptible to long-term settlement. Tyically we would find CBR's in the order of 5% /shear strengths of around 80-100 kPa at 95% MDD of the 4.5kg compaction on the wet side, which in turn fits well with the expected performance of the material at 100% of the 2.5kg rammer. We do, however, also place great emphasis on the air void content, with a maximum of 5% specified for cohesive fill.
Now going back to the original post:-
in my experience for cohesive soils, 100% MDD of the 2.5 kg rammer will settle less than 95% of the same MDD, but will not remove it completely. Cohesive soils tend to settle less during construction and more long-term due to consolidation settlement. Granular soils will probably experience most if not all of their settlement during construction. The general split between granular and cohesive, based on the SHW, is the amount passing the 0.063mm sieve, more than 15% its cohesive, less than 15% its granular. This is NOT the same assessment we make when describing the soils geologically as part of the SI.
So in summary it depends on the soil type and nature of the structure being constructed. I have tested fine grained 'cohesive' soils [sandy SILTS] where the MDD using the 2.5 and 4.5 were very similar, but were unable to achieve less than 8% air voids.

 

[jth88]

Thanks for the responses.

BigH - I hope to have time this weekend to try the academic exercise.

I was also going to look at the change in air voids from the 95% and 100% MDD Std Proctor states.

I will review the paper listed. I fear that these papers are for larger >> 20 foot. Not sure of the applacability for smaller fills.


Thanks,
jth88

 

[muuddfun]

1 Percent settlement seems to work on a general basis from my experience with running a few remolded consols everyonce in a while on some prospective fill soils. It is deependent on the particulars of each soil, and how much compaction you have done. Remold a sample to 100 percent of the standard proctor and do a consol, then remold to 95 percent of the modified and check again on another consol and see what difference you have. That is the only true way to find out if there is any real difference. A lot of the settlement in sandy or silty sand fills that have been well compacted would be from hydrocollapse after the fact, the majority of any settlement would take place during placement so the only thing left would be the collapse potential.

 

[BigH]

Again, I think that we must also remember that no soil placement is to 95% MDD (whatever standard you want to use). Some layers get compacted to 97% average, some to 96%, some to 99% - there is a non-uniformity both between layers and within layers and also neglects lower layers being compacted somewhat higher during compaction of the upper layers - so, in my view, this is typically an academic exercise for our jth88 to see the difference if "all" were at 95% and "all" were at 100% to show his boss that there is very little or little or somewhat differences in settlement. One difference is that the higher compaction, will cause "more" settlement - thereby requiring the contractor to bring in more fill to reach final grade which, in the end, might reflect in higher fill prices . . .

 

[civilperson]

A better use of the construction cost used to compact the soil from 95 to 100% would be a temporary surcharge above the finish grade, (perhaps with some wick drains).