Use of Proctors 1

[GODsRebel]

Is there a limit to the amount of field compaction tests for any given proctor?

 

[eric1037]

Yes. If the material changes, another Proctor should be performed. If the material doesn't change, it should be good. This a call that needs to be made by the field technician.

 

[emmgjld]

When a significant number of tests have been taken or after a period of time has passed from the beginning of the job and if the technician OR the engineer has any questions about the proctor/soil correlation, It is also prudent to perform 'check points', single point tests at near optimum moisture to confirm near conformance with the original proctor curve.

 

[COEngineeer]

This is really the part of geotechnical aspect I dont like. For road base material, then I think the material is more than likely always the same. So testing it with the densometer is very effective. But for native material compaction, you can put your densometer at where you see a lot of rocks and more than likely youll get your 95% compaction. Taking more sample to the lab takes forever. contractor wants to know whether their compaction is good or not. I hate to admit, but I did let them get away without taking more samples. But I did make sure they put enough water that I think appropriate, compact it very well at one spot, find the density on that spot, then try to achieve that density everywhere. Its all about technician's common sense and experience.

 

[eric1037]

You can do a "one-point" Proctor in the field relatively quickly. An experienced field technician should be able to tell when the material changes.

 

[COEngineeer]

Yeah, but most of the time its not the same technician who took the sample.

 

[GODsRebel]

Well, I don't like using the same proctor over a given stretch of a roadway covering a mile, even if the material "looks" the same. I just wanted to know if there was any standard to using the same proctor for variable/similar in-sitsu soils any number of times.

 

[GODsRebel]

base material is one thing, sub-grade is a diffrent story

 

[civilperson]

A set of proctors should be produced for each job representing the different soil types. These proctors are represented by curves with the peak compaction at optimum moisture. A one point should be pounded in the field and plotted on the curve that is assumed to be representative. If the "one Point" does not land on the curve, (=/- 1/2%), then a new proctor needs to be run for that material. Some jobs go start to finish with one proctor, (15,000 density compaction tests) and others change twice a day as the borrow exposes different material.

 

[fattdad]

Yeah civilperson! Please spread this insight to all technicians throughout the world and it will become a better place. This is the correct answer.

There is no limit to the number of field density tests that are referenced to any given Proctor. There is a compelling need to confirm that you are using the correct Proctor for every field density test; however. The one-point method is great for doing this - providing that the one-point is done BELOW the optimum moisture content. (If a one-point is done above the optimum, you can't properly interpret which unique curve it belongs to.)

Basing your selection of the correct Proctor only on "what it looks like" is a dangerous practice. It takes surprising little change in sand/silt/clay content to affect the maximum dry density. Refer to the 1988 ASTM publication "Compaction Control and the Index Unit Weight" by Steve J. Poulos (my former boss) for more information.

 

[BigH]

Most highway specs that I have seen have, as part of Quality Control, specific tests that have to be done per every so "much" material. Does not the specification indicate frequency of tests? Example - for embankent borrow - do one Proctor for every 3,000 m3 of material from a particular pit/borrow area. Atterbergs might be done every 1,500 m3. For base courses, the Proctor might be as frequent as 1 Proctor for every 500 m3 (1 test for every 150 linear metre of road). I have the Indian Standard Highway specs back at the office and on Monday I'll check them out for you (06.11.06) - the Indians love to spec out stuff like this. In the end, I don't see how you can justify a quality control programme with just one proctor for each type of material or for the whole job - or even each week. This isn't QC. You take concrete cylinders for strength checks almost every pour (or more than one set). Yet, I doubt that the concrete strength for a particular batch is any more variable than for earth borrow - and likely much less.

 

[GODsRebel]

I agree BigH, I hate looking at density reports with a quarter of the results with 101-105% compact. Its ridiculous, especially for roadway work where tests are every couple hundred feet (not many tests for the amount of area).

 

[GeoPaveTraffic]

I agree with civilperson.

GODsRebel, note having a compaction of vover 100% does NOT indicate that the Proctor is incorrect or that the material has changed. Percent compaction is just a ratio of measured density to a laboratory density that was determined in a specific manor.

 

[GODsRebel]

You are right GeoPave, however, another proctor should be performed if a compaction test is > or = 105%. So you are still limited to that ratio.

 

[GeoPaveTraffic]

Why do you say another Proctor is required if the measured compaction is over 105%? There is nothing special about that figure.

 

[TDAA]

Agreed, nothing special with that number, especially if the material is a bit dry. It could just be showing that the level of compaction may be better related to the modified Proctor.

If the "one Point" does not land on the curve, (=/- 1/2%), then a new proctor needs to be run for that material.

Keep in mind that there is variation in the proctor method. Just by the allowable tolerance of the mold volume (+/- 0.0005 cf) there could be a difference of up to about 3 percent. (calced from 4.22lbs in mold sizes of 0.0338 and 0.0328 cf, 15% moisture). Then add in all the other possible variations that may be included. Heck, the 5% variation of moisture listed in the precision section (between labs, i.e. could be lab oven, and field lab oven) of the laboratory moisture (ASTM D 2216) can account for a greater than 0.5% difference in the dry density of the sample.

 

[COEngineeer]

If found it its very difficult to achieve 100% compaction if you have the right proctor. Unless you are compacting it againts hard surface like concrete, I never get 100%.

 

[emmgjld]

The discussion of the last 4 posts indicate that one-point checks need to be accomplished when these questions come up. This was the point of my first post. Supervision and quick checks to identify changes is in order.

On the flip side.
About 15 years ago, My father sent me a copy of a review of a large project (A water filtration plant, if I recall) in Colorado Springs which was in litigation. Not Ours!
Hundreds of density tests had been taken.
The soils were complicated.
I believe there were over 60 proctors. Very few one-point checks.
Many proctors did not have a sieve analysis run and recorded with the test.

Apparently, every time the soil looked different, a proctor was run. The supervision was not as frequent.
Many of the proctors differed from previous by only 1-1/2 to 2-1/2 pcf.
Several of the Proctor Curves were significantly over a reasonable Zero Air Voids Curve.
Many of the proctors were obviously very similar, to almost identical with earlier proctors. But, without sieves, who knew?

The reviewing Engineer for litigation concluded that:
The tester had lost track of what he had.
Too many Proctors and how to sort them out???
Many Proctors differing from previous by only 1-1/2 to 2-1/2 pcf may very well be within the normal range of field sampling and laboratory testing. [See Precision & Bias Section in ASTM]
[Over 1/2 of the proctors should have been one-point checks]
A significant number of the Proctor tests were not valid (over Zero Air Voids Curve).
Supervision was worse than poor.

!!!Take it to heart and don't let it happen to you!!!

 

[BigH]

The point I made which seems to be lost is that a particular project needs to have an Inspection and Testing Plan. This plan will lay out for each material the type of test (laboratory or field), the frequency of each test, the criteria on which the test is to be judged (it may be that a particular day's work is based on the average being greater than minimum value in a statistical fashion). You don't use a proctor from a month ago some 5 miles away with material from a differnt borrow pit - this is why you have frequency of testing per X m3 of material. In addition to the spelled out QC, there should be a QA procedure for verifying that the testing results are "correct". The QA would likely be performed by a different agency but on the same sample (or sample location). As much as I hate to say it, these procedures are set up so that variational mistakes, etc., in a fashion, are not made. Of course if the source of material changes, new testing is required.
Years ago, Onatario had a comparison method for relative compaction - you did your in situ density test (rubber balloon or sand cone). The material that comes out of the hole is then "adjusted" to be near optimum moisture. You then recompact it into the proctor mould and measure the volume. Compare its volume to the hole's volume and you get the relative compaction. There were adjustments to be made on the blow counts per each layer of the recompaction in the mould depending on the volume of the field hole.
As much as everyone (generally clients, contractors and even engineers) think that they are experts on compaction (it is SO simple), it is amazing that in many handbooks (Fang for instance) the largest chapter is on compaction. And Monihan (NJNIT) has written a pretty good book just on compaction.

 

[DRC1]

I really like Monihan's book. It is a good combination of theory and practcality. I am also partial to Fang as I was fortunate enough to study under him. So I recomend both books.
Big H has probably the best point. It's not just matching a compaction result to a proctor. It is an overal program. Inspecting the borrow pits and sampling, Doing laboratory multiple point proctors. doing sieves moisture content and limit tests, having the engineer review and accept the material and coordinating the work with the contractor. I have seen jobs where several pits, each with approved proctors, have been pulled out of simultaniously to go in the same fill. Note that Proctor A + Proctor B does not equal Proctor A+B. Also I have jobs where stanard proctor is required and proctors in the lab were modified because nobody comminicated (or vice versa) Note that it is very easy to get in excess of 100% standard proctor.
Also rember that if you are testing with a nuclear gauge, the toal density is the average density of the lift (up to 12 inches)but the moisture is a backscatter measurement in the top 2-3 inches so unexpected high or low readings should be checked prior to acceptance or rejection.
Proctors are multi point tests that require 3 days to perform. One point proctors are performed in the lab to cross check the test. One points, even done by experineced technicians can be way off. I would not recomend using one pointers for feild inspection.
If the project does not have a testing plan for eathwork, use the states DOT's practices. A.) They are probably the largest owner contracting for earthwork B.) They have refined a system for that area over the years and C.) Most earthwork contractors are familar with it.
That said one of my pet peeves in civil engineering is compaction testing. I have yet to figure out quanataively what compaction tests say about bearing capacity or settlement pediction. So if we arbitarily accept (and the acceptance value seems rather arbitary to me)of 95% and we have 92%, how do we know the fill won't perform as intendend? On the other hand if we have 98% how do we know it is enough?