When a specification calls for a percent compaction eg 95% what would be the effect if you only achieved 90%. In other words how critical is the the asked for figure and what is considered an acceptable variation from that figure (in terms of +/- %)
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Sensitivity of Compaction 15
- Thread starter merryfull
- Start date
-
1
- #2
KAM_imported
Geotechnical
It depends on the reason specified for compaction. Most fill soils are compacted beneath slabs and building foundations because compaction increases soil shear strength and reduces its compressibility, and is usually fairly easily achieved in sandy soils. 95% compaction is a typical standard for soil compaction. Failure to achieve this compaction means that the soil in question would theoretically have a higher degree of compressibility and lower shear strength than a properly compacted soil; however, the difference may be insignificant if the undercompacted zone is fairly thin. If, on the other hand, the poorly-compacted zone is 10 feet thick, this could result in a problem. This is best evaluated by the geotechnical engineer of record. Other soils are compacted to reduce their permeability. In this case, failure to achieve compaction may result in more severe problems (such as in landfill liner situations)due to increased permeability. Failure to achieve uniform compaction is usually an indicator that the contractor's methods are not acceptable, or that the soil is not at the right moisture content.
-
1
- #3
Just some additional information for Merryfull - when I saw your posting, I felt inclined to contribute a little something... as a geotechnical engineer, there is a reason why given compactions are specified. Typically, 100% is specified for fill beneath industrial and institutional buildings, 98% is specified for fill beneath commercial and residential buildings, 95% is specified for fill beneath pavements and pipe bedding. The intention is to specify a compaction that will lead to minimal settlement in the fill material.
Typically, if the specified compaction is not met, it is because of excess moisture in the soil. Drying or blending the soils often solves this problem, although it is not fool-proof. If the specified compaction still can't be met - the safest plan is to bring imported (granular-if possible) material.
As for an acceptable variation on the specified compaction - that depends entirely on the engineer who has to sign off on it. They may be more flexible on road fill, knowing that it will be dug out for servicing, than they would for fill material beneath a new school.
Typically, if the specified compaction is not met, it is because of excess moisture in the soil. Drying or blending the soils often solves this problem, although it is not fool-proof. If the specified compaction still can't be met - the safest plan is to bring imported (granular-if possible) material.
As for an acceptable variation on the specified compaction - that depends entirely on the engineer who has to sign off on it. They may be more flexible on road fill, knowing that it will be dug out for servicing, than they would for fill material beneath a new school.
Compacting soil on the dry side is another and perhaps more common reason for not achieving compaction. Contractors sometimes don't want to rent water trucks. Most specifications require that soil be compacted in a range of -2 to +4 points of optimum moisture as determined by a standard or modified Proctor. My experience is that under slabs and paved areas, the standard is 95-98%. I've seen 100% used a few times, but it is hard to achieve and expensive.
-
1
- #5
How critical compaction is depends on a number of items:
1. what is going on it
2. what material are you compacting
3. what is the thickness of the poorly compacted layer
4. what is the thickness of the overall fill
5. what is the depth to bedrock
Where I work we deal almost exclusively with clays. Compaction and moisture control is very critical with clay because of it's tendency for differencial settlement. Also good compaction will help with shrink/swell problems. For the most part (unless bedrock is extreemly shallow, or the fill is less than a foot thick) the numbers for under the building pad are absoulute. We will make a contractor rework a pad until it passes. We are more giving in pavment and landscape areas, even to the point of allowing some bad fill material (such as topsoil) to be wasted in the fill in thin (less than 2 inch) layers.
Usually when compaction cannot be achieved and the engineer is willing to compromise, the site is proofrolled with a fully loaded dual or triaxal dump truck. At ths point we watch for "pumping and rutting" which is esentially soil movement indicating soft spots which must be undercut and repaired
1. what is going on it
2. what material are you compacting
3. what is the thickness of the poorly compacted layer
4. what is the thickness of the overall fill
5. what is the depth to bedrock
Where I work we deal almost exclusively with clays. Compaction and moisture control is very critical with clay because of it's tendency for differencial settlement. Also good compaction will help with shrink/swell problems. For the most part (unless bedrock is extreemly shallow, or the fill is less than a foot thick) the numbers for under the building pad are absoulute. We will make a contractor rework a pad until it passes. We are more giving in pavment and landscape areas, even to the point of allowing some bad fill material (such as topsoil) to be wasted in the fill in thin (less than 2 inch) layers.
Usually when compaction cannot be achieved and the engineer is willing to compromise, the site is proofrolled with a fully loaded dual or triaxal dump truck. At ths point we watch for "pumping and rutting" which is esentially soil movement indicating soft spots which must be undercut and repaired
RESCUENGINEER
Civil/Environmental
I work for a grading contractor. It is not uncommon to see compaction go as high as 110 % . Just because a certain standard is set in a lab, does not mean that different outcomes are possible. At the same time, there are many instances where we can not reach 70 %. From my experience, when the specs say 95%, there is still a lot of variance in what you actually get.
If you consistently see 110% there is something wrong with your proctor. Especially if you are using a modified proctor. Still although it is not uncommon to go above 100% of a standard proctor, equipment is better than when that test was designed, doing so consistently is cause for concern.
If you cannot get compaction, yet moisture is where it should be and the site is proofrolling well then again there is likely something wrong with the proctor.
Often though the cause for too high compaction or too low is the moisture being out of spec.
If you cannot get compaction, yet moisture is where it should be and the site is proofrolling well then again there is likely something wrong with the proctor.
Often though the cause for too high compaction or too low is the moisture being out of spec.
-
6
- #8
If I were a contractor, and a compaction result of 70% was reported, I'd start looking for a new geotech, unless, of course it was April 1st. But seriously.....
The Proctor test is a physical test, brought to the world by, you guessed it, Mr. Proctor (circa 1933). The test proved useful in stabilizing soils for military applications, especially impromptu aircraft landing strips. Proctor (the man) believed that density increased as water, acting as a lubricant, reduced the friction between soil grains, allowing for a more efficient particle arrangement via void filling by the smaller grain sized particles. Additional work by Hogentogler (1937), and Lambe (1958), and the man considered by many to be the master of soil mechanics, Terzaghi, increased the understanding of soil behavior to the point that, under carefully controlled conditions, soil and soil-aggregate composites can be relied on to behave properly under a very wide range of conditions. Over the years, the basic laboratory Proctor test has not changed much. What has changed are the challenges that geotechnical engineers face when evaluating a site. The combination of a good geotechnical engineer and a good structural engineer, working in concert, allow construction to take place at locations that previously were considered 'unbuildable', It would take a detailed understanding to ever allow a recommended compaction requirement to be waived, lowered, or increased. The best course of action is to determine why a spefied relative density is not occuring. Any deviations from the specified relative densities should be brought to the attention of BOTH the geotechnical engineer and the structural engineer for analysis. It may or may not be acceptable. I'll leave you with some examples.
Highly compacted, unyielding soil may be required beneath one type of pipe to avoid shear failure of the pipe, but cause it in another type.
Highly compacted expansive soils, especially when compacted on the dry side of optimum, may heave and cause significant damage when lightly loaded and subjected to moisture infiltration.
Highly compacted fine grained, silty soils, may consolidate when placed and compacted at moisture contents below optimum, This is especially problematic in trenches.
Wide variations in density, even though minimum requirements are met, may lead to a non-uniform soil support system for pavements.
Other, equally important considerations have already been mentioned in this forum. I hope you can see that the question you posed is not nearly as easy to answer as it may have seemed. Whenever I hear it said that the "engineer overdesigned" a project, I always reply that the engineer knew instinctively that someone would try to under build it.
The Proctor test is a physical test, brought to the world by, you guessed it, Mr. Proctor (circa 1933). The test proved useful in stabilizing soils for military applications, especially impromptu aircraft landing strips. Proctor (the man) believed that density increased as water, acting as a lubricant, reduced the friction between soil grains, allowing for a more efficient particle arrangement via void filling by the smaller grain sized particles. Additional work by Hogentogler (1937), and Lambe (1958), and the man considered by many to be the master of soil mechanics, Terzaghi, increased the understanding of soil behavior to the point that, under carefully controlled conditions, soil and soil-aggregate composites can be relied on to behave properly under a very wide range of conditions. Over the years, the basic laboratory Proctor test has not changed much. What has changed are the challenges that geotechnical engineers face when evaluating a site. The combination of a good geotechnical engineer and a good structural engineer, working in concert, allow construction to take place at locations that previously were considered 'unbuildable', It would take a detailed understanding to ever allow a recommended compaction requirement to be waived, lowered, or increased. The best course of action is to determine why a spefied relative density is not occuring. Any deviations from the specified relative densities should be brought to the attention of BOTH the geotechnical engineer and the structural engineer for analysis. It may or may not be acceptable. I'll leave you with some examples.
Highly compacted, unyielding soil may be required beneath one type of pipe to avoid shear failure of the pipe, but cause it in another type.
Highly compacted expansive soils, especially when compacted on the dry side of optimum, may heave and cause significant damage when lightly loaded and subjected to moisture infiltration.
Highly compacted fine grained, silty soils, may consolidate when placed and compacted at moisture contents below optimum, This is especially problematic in trenches.
Wide variations in density, even though minimum requirements are met, may lead to a non-uniform soil support system for pavements.
Other, equally important considerations have already been mentioned in this forum. I hope you can see that the question you posed is not nearly as easy to answer as it may have seemed. Whenever I hear it said that the "engineer overdesigned" a project, I always reply that the engineer knew instinctively that someone would try to under build it.
KRSServices
Civil/Environmental
If it was a paved road, and there was frost penetration, the results would be disasterous. The road would not settle uniformly, and numerous ruts and deformations would occur. Besides a costly maintenance headache, if there was a warranty period in the contract, the contractor and/or the testing firm (if recommending approval) could be held liable for the repair costs.
-
2
- #10
Suggest all look, too, at thread 274-5909 on the geotechnical engineering and other related topics forum. We have been having a go at compaction there, too.
One item begging a question in the original query was picked up on by others - what is the underlying theme to the question? For kind of fill and what use of the fill are we discussing? A second item that none of the replies queried was whether the % compaction was modified (heavy) or standard (light). This should have been part of the original query's background too.
There is a good article I saw in Ground Engineering a number of years ago about the "95% Fixation". When I go to the office tomorrow, I will find it and reference it in another reply.
I do want to emphasize a point that I made in the abovementioned thread. For embankment construction, one should, unless the fills are very high, take into consideration too the nature of the foundation soil in design and specifying a level of compaction to achieve. If you have 5 to 10m of soft to firm normally consolidated clay, settlements of the foundation will be much more critical than any self settlement of the compacted embankment fill (say several hundred millimetres or more vs 20mm, perhaps). Further, if properly compacted in a professional manner even to less than 95% standard compaction, most fills <5m or more high wouldn't have any self shear distress either. The foundation in such cases is critical, not the fill.
I do warn the young lions getting their first tastes of the field that regardless of engineering practicality and judgment (i.e., whether or not it REALLY makes a difference), always be up and front about any testing results with respect to the project specfications. Some poor junior field engineer in Ontario was really hauled up on the stick recently for "fudging" a test result of the top soil layer under an approach slab - the test was lower than specified and he willfully made it "good" as the slab poor was already underway. We could all argue whether this really made any difference to the performance of the fill/approach slab and that it was bad judgment to start the pour before really knowing the testing result, but any willful misreprentation of such results . . . NO.
Best regards to all.
One item begging a question in the original query was picked up on by others - what is the underlying theme to the question? For kind of fill and what use of the fill are we discussing? A second item that none of the replies queried was whether the % compaction was modified (heavy) or standard (light). This should have been part of the original query's background too.
There is a good article I saw in Ground Engineering a number of years ago about the "95% Fixation". When I go to the office tomorrow, I will find it and reference it in another reply.
I do want to emphasize a point that I made in the abovementioned thread. For embankment construction, one should, unless the fills are very high, take into consideration too the nature of the foundation soil in design and specifying a level of compaction to achieve. If you have 5 to 10m of soft to firm normally consolidated clay, settlements of the foundation will be much more critical than any self settlement of the compacted embankment fill (say several hundred millimetres or more vs 20mm, perhaps). Further, if properly compacted in a professional manner even to less than 95% standard compaction, most fills <5m or more high wouldn't have any self shear distress either. The foundation in such cases is critical, not the fill.
I do warn the young lions getting their first tastes of the field that regardless of engineering practicality and judgment (i.e., whether or not it REALLY makes a difference), always be up and front about any testing results with respect to the project specfications. Some poor junior field engineer in Ontario was really hauled up on the stick recently for "fudging" a test result of the top soil layer under an approach slab - the test was lower than specified and he willfully made it "good" as the slab poor was already underway. We could all argue whether this really made any difference to the performance of the fill/approach slab and that it was bad judgment to start the pour before really knowing the testing result, but any willful misreprentation of such results . . . NO.
Best regards to all.
geomaritimer
Geotechnical
Good question, as stated by others it is not easy to answer. Keep in mind that if you are using a nuclear densometer (Troxler) that it is a machine, certainly not as smart and/or instinctive than you or other engineers. I have been on jobs where the material seemed bullet proof but failed to reach 95% and down the road using different material (different Proctor) the material passed but still seemed soft and rutted easily. The numbers are for the most part a guide which when combined with experience will allow the engineer to make the right decision(s). Good Luck.
As indicated, you might all wish to get a copy of paper by Charles, J.A., H.D. Skinner and K.S. Watts "The specification of fills to support buildings on shallow foundaions: the "95% fixation"." Ground Engineering, January 1998.
If you don't have access advise and I can scan and forward copy.
Best Regards.
If you don't have access advise and I can scan and forward copy.
Best Regards.
KRSServices
Civil/Environmental
Geomaritimer hit it on the head. Many municpalities and contract documents specify 95%, 97%, 98% SPD or modified due to the fact, and in my experience, that a specified density is the only measurable method to evidence re-compaction of disturbed material, embankment fill, and trench fill. AS everyone has sort of pointed out, once native soils are disturbed, various soils become mixed (any organics and unsuitables removed) such as clays, sand and maybe silt lens. Unless a proctor is taken for each excavation bucket, some other method has to be employed. A combination of expertise and sound judgement is used to enable the contractor to achieve the required densities. Many of my contracts allow for remove of unsuitables and the importing of borrow material. Moisture content is also important as well as a good working knowledge of the material. I always specify test strips be placed prior to placement of base course on a road project, and encourage test areas in trenching to ensure the protor being used is actually representative of the material. I have seen variations in gravel stockpiles. In those cases, I have erred on the side of caution and obtained a new proctor. However that being stated, in most cases, the geotech has always been very knowledgeable and readily able to recommend the proper course of action. Again, nothing substitutes for experience, therefore if in doubt or uncertain, ask someone. I hope this helps. KRS Services
Personally I have never understood the procter test. Someone has finally hit on one of my pet peeves and waken the sleeping beast.
First off, The proctor test, if done by an experienced technician is fairly reproducable, but not perfectly. I would guess it is reproducible around 3%.
It is not a perfect test. However, based on the number of completely screwed up procters I have seen, good technicians are apparently in inceasingly short supply.
But my biggest gripe is that the proctor does not give us a number to which to design. If I were testing native material for suitability to support load by testing weather the insitu dry density excceded 95% of the modified procter you would call me a nut (or worse). Yet if I pick up a pile of soil and move it 50 feet across the site, this becomes the only criteria for acceptance. We do not know the bearing capacity friction angle, cohesion or any other properties. We do not konw what the capacity of the soil is verses the load we apply to it, yet if we compact it to 95% of an test standard that we do not have any hope of correlating to our field compaction equipment, we trust everything will be fine, and if it is only at 94% our work will collapse. I guess my big question is now that we have the fill at 95% of modified proctor, how do we know from a design standpoint that the fill has sufficent strength to support the load?
First off, The proctor test, if done by an experienced technician is fairly reproducable, but not perfectly. I would guess it is reproducible around 3%.
It is not a perfect test. However, based on the number of completely screwed up procters I have seen, good technicians are apparently in inceasingly short supply.
But my biggest gripe is that the proctor does not give us a number to which to design. If I were testing native material for suitability to support load by testing weather the insitu dry density excceded 95% of the modified procter you would call me a nut (or worse). Yet if I pick up a pile of soil and move it 50 feet across the site, this becomes the only criteria for acceptance. We do not know the bearing capacity friction angle, cohesion or any other properties. We do not konw what the capacity of the soil is verses the load we apply to it, yet if we compact it to 95% of an test standard that we do not have any hope of correlating to our field compaction equipment, we trust everything will be fine, and if it is only at 94% our work will collapse. I guess my big question is now that we have the fill at 95% of modified proctor, how do we know from a design standpoint that the fill has sufficent strength to support the load?
-
3
- #17
That's a complicated subject; have you read all the other threads on this site?
It appears to me that much of what you complain about comes from a basic misunderstanding of soils in general. You can't specify them the way you can steel or concrete - they're variable and quite non-linear. You may not be able to separate different soils solely on the basis of color or texture. And they don't come with a yield stress stamped on them...
It's unrealistic to expect a simpler specification than the common relative compaction standard - in the design and specification process, you have to look for a range of compactive effort that provides suitable performance. Your statement that ... my biggest gripe is that the proctor does not give us a number to which to design suggests that your expectations may not be realistic.
You can correlate field compaction with the lab test, but it only works for a given borrow source, site, compaction equipment and compaction procedures. Change one or more of these and the correlation breaks down.
Only a fool - with no understanding of the test and its' history - would arbitrarily say that a project would collapse at 94% relative compaction, yet is acceptable at 95%. I don't know who told you that, or why, but I suggest that you go back to that person and ask him/her to explain the logic of that statement.
Keep in mind that we have two issues to deal with on most jobs: design/performance needs and contractual obligations. It's terribly important not to confuse the two -
![[pacman]](/data/assets/smilies/pacman.gif "[pacman] [pacman]")
Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.
It appears to me that much of what you complain about comes from a basic misunderstanding of soils in general. You can't specify them the way you can steel or concrete - they're variable and quite non-linear. You may not be able to separate different soils solely on the basis of color or texture. And they don't come with a yield stress stamped on them...
It's unrealistic to expect a simpler specification than the common relative compaction standard - in the design and specification process, you have to look for a range of compactive effort that provides suitable performance. Your statement that ... my biggest gripe is that the proctor does not give us a number to which to design suggests that your expectations may not be realistic.
You can correlate field compaction with the lab test, but it only works for a given borrow source, site, compaction equipment and compaction procedures. Change one or more of these and the correlation breaks down.
Only a fool - with no understanding of the test and its' history - would arbitrarily say that a project would collapse at 94% relative compaction, yet is acceptable at 95%. I don't know who told you that, or why, but I suggest that you go back to that person and ask him/her to explain the logic of that statement.
Keep in mind that we have two issues to deal with on most jobs: design/performance needs and contractual obligations. It's terribly important not to confuse the two -
Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora.
DRC1:
You make some interesting comments. The Proctor test is valuable for geotechnical engineers in conveying the idea that a particular soil that will be placed as fill needs to be compacted to a state where it will have the appropriate density, strength and stiffness to meet the expectations of the design engineer. It is not a perfect method (as with most of what is done for geotechnical engineering) but it does help the field people to focus on the issues that are most critical to achieving reasonable performance. The soils need to be at a moisture content that would lend itself to an appropriate degree of compaction. The soils need to be placed in lifts of proper thickness so that the compactive effort is adequate and the right equipment needs to be selected to accomplish the compaction.
In various parts of the world with various soils, local geotechnical practice is developed over time with anecdotal performance information that generally achieves the objectives. What we all need to look out for is a geotechnical engineer going into a new area with new soils that specifies compaction criteria without an understanding of what is achievable.
Field compaction control is really somewhat of an art where we have to keep in mind our objectives as others in the thread have pointed out.
I hope this helps.
Glen
You make some interesting comments. The Proctor test is valuable for geotechnical engineers in conveying the idea that a particular soil that will be placed as fill needs to be compacted to a state where it will have the appropriate density, strength and stiffness to meet the expectations of the design engineer. It is not a perfect method (as with most of what is done for geotechnical engineering) but it does help the field people to focus on the issues that are most critical to achieving reasonable performance. The soils need to be at a moisture content that would lend itself to an appropriate degree of compaction. The soils need to be placed in lifts of proper thickness so that the compactive effort is adequate and the right equipment needs to be selected to accomplish the compaction.
In various parts of the world with various soils, local geotechnical practice is developed over time with anecdotal performance information that generally achieves the objectives. What we all need to look out for is a geotechnical engineer going into a new area with new soils that specifies compaction criteria without an understanding of what is achievable.
Field compaction control is really somewhat of an art where we have to keep in mind our objectives as others in the thread have pointed out.
I hope this helps.
Glen
I would like to say in my own defense that I do have an understanding of soils and soil mechanics, and I feel Focht3 and ganderson make good points.My point is which gets back to the orginal question of the thread - How do you know, without additional testing, that a particular percentage of a procter value will give you the performance you want, especially when the percentage is specified prior to the selection of the material?
Focht3 also raises an interesting point -
"Keep in mind that we have two issues to deal with on most jobs: design/performance needs and contractual obligations. It's terribly important not to confuse the two - "
This is true of construction but should it be so? If the owner is paying for work that is specified, but does not advance the design/perfomance needs of the project, what benifit does the owner recieve?
An interesting digression.
Focht3 also raises an interesting point -
"Keep in mind that we have two issues to deal with on most jobs: design/performance needs and contractual obligations. It's terribly important not to confuse the two - "
This is true of construction but should it be so? If the owner is paying for work that is specified, but does not advance the design/perfomance needs of the project, what benifit does the owner recieve?
An interesting digression.
- Status
Similar threads
- Locked
- Question
- Locked
- Question
