Field Density Testing of Peat

[nicka70]

I'm currently working on a project that requires knowing the density of insitu peat and being sure to not apply too much overburden to the native soils. We are installing a drinkwater pipeline roughly 6 feet below grade that traverses an agricultural field. The agricultural field consists mostly of peat. We need to be sure to not apply too much overburden during backfill of the trench to minimize pipe movement. In order to know how much is "too much", we must know the density of the insitu material. I've slung a nuke but am curious if anyone has suggestions of alternatives for peat (moistures of around 200% and dry density of 20 pcf).

The preferred option would be a nuke, however I'm not confident it will be very accurate with such low density and high moisture content material. I could be completely wrong on this; I don't have experience field testing materials with this low of a density.
The alternative would be drives, which may be an issue in more "stringy" peats.

The testing of the peat backfill and insitu will have to be done in the field, so need a quick turnaround time. Thanks!

Any ideas?

 

[Ron]

Shelby tube and/or cone penetrometer......however, you need more than just the density of the material in place to determine settlement characteristics from overburden pressure. You need consolidation testing, moisture content, thickness of the peat, weight of the pipe flowing full and for how long, weight of the overburden, etc.

 

[fattdad]

. . . and the coefficient of secondary compression.

Block sampling also works in conjunction with moisture content, you'd have everything you'd need, or so it would seem. . .

f-d

ípapß gordo ainÆt no madre flaca!

 

[oldestguy]

Have you considered considerably reducing the weight of that backfill? Perhaps bury closed cell poly in the backfill just over the pipe so that even some mistake in backfilling will not overload the pipe. Also consider that there will then be some transfer of that backfill loading into sides of the trench. In the USA years ago an Iowa engineer by the name of
Spangler had a form of trench that would reduce the load on the pipe which he called the "Imperfect Trench". Just over the pipe he placed a compressible fill, corn stalks. The Wisconsin DOT had the need to leave a sanitary sewer outfall in service during and after placing a very high Interstate highway fill over it. They adopted the "Imperfect Trench" and that saved the pipe from the added fill weight over it. It worked.

Keep the trench as narrow as possible to facilitate that transfer of loads to the sides.

Other treatment might be adding of some rigidity materials below the pipe to help minimize differential settlement effects.

Why 6 feet down? Is there any chance of plowing in a coiled plastic pipe to avoid all he grief of trenching? If frost is the reason and this is a usual high water table site, deep frost is unlikely. Also burying a sheet of closed cell poly can protect a pipe at shallow depth.

 

[nicka70]

@Ron, we have enough data for the insitu densities. Our goal is to backfill so the same stresses are felt just below the pipe as before to minimize any movement upward or downward. By applying the same amount of stress to the peat just below the pipe, the worry of settlement is negated. Do you know of any field tests that can be done in a matter of minutes or hours so we can provide feedback to the contractor?

@oldestguy, we are going to be using controlled density fill to reduce the weight of the backfill. That is where this issue comes in. Do you think a nuke would be reasonably accurate in such low density situations? I don't see why it wouldn't be but I was hoping to get the opinions of a few others who are more veteraned.

 

[oldestguy]

As tithe nuc and accuracy, yes there can be a problem. Commonly folks use a calibration technique in the storage container, or similar. What you can do is create some sample containers of known density peat and moisture content. Test that material by a known method and create your own calibration for this special material. Another, possibly more accurate (due to larger volume tested) is to test known locations on the job and run other technique to verify the density of those test locations. Such other technique might be the sand cone or similar technique.

OK, now here is a question. Suppose you are finding that it is not possible to create the density that you want, possibly due to moisture differences, etc. maybe 5% heavier Then what do you do? Let's say this is in the lower half of the backfill, and the question comes up: "How much leeway do we have? Can we make it lighter above? " If water table comes up, what is submerged unit weight? Does it float? What about arching effect? I think the whole idea is rather questionable and you may well break a pipe or find yourself re-digging the trench, etc. I say again, take a look at Spangler's work and you might rethink this. If this were my job, I would not do what you are planning. I've done considerable working with peats and that's a whole different material than earth. Also there are many kinds of peat and they can vary greatly with depth. Many are underlain by organic silt, a whole lot different.

https://www.dot.ny.gov/divisions/en...tech-eng-repository/GDM_Ch-21_Pipe_Design.pdf

 

[oldestguy]

OG back: I see that document didn't fully cover the idea. Here is a paper that is at the
ASCE library but you need to log in and read it then. (Full text).

http://ascelibrary.org/doi/abs/10.1061/(ASCE)1090-0241(2007)133:3(277)?journalCode=jggefk

 

[Ron]

nicka70....I think you might have a logic problem. You are removing material in order to place a pipe that will be flowing with liquid to some level, presumably 2/3 to full. Above that pipe you have to provide fill material to restore the ground level and protect the pipe. You have not considered that the pipe needs reasonable bedding for its support...this will add weight. So now you have the weight of the pipe plus the bedding material, and you have not yet brought the excavation back to grade.

You might consider OG's idea of polystyrene; however, I would put it under the pipe not over. Over the pipe I would use cellular concrete with a dry density in the range of 35 to 40 pcf.

 

[oldestguy]

Ron: Maybe we are not providing the easy OK that is requested, but I hope we get the alternatives thinking process going. Many ways to skin a cat they say, but I've never tried that. I'm pretty good as skinning muskrats however. A close alternative to how it works. Same goes for pipes in the ground.

 

[nicka70]

Ron: I'm not the decision maker in the project, the solution is pretty set in stone and calculations have been done including pipe bedding and the pipe filled, hollow, etc. I'm the field engineer on the site. I think your concerns and recommendations are completely valid and quite well though out and creative solutions. However, my task in this process is QA for the field. Pipe and this specific situation aside, will peat be accurately measured by a nuke?

Oldestguy: Thank you for the literature. As I said to Ron, I'm not the decision maker on the project, just involved in the QA process for the field. I did think that a series of drive samples and associated nuke tests next to them would be a good way to ensure that the nuke is providing the accuracy it typically does or if there is some sort of bias. My concern in this whole process is how accurate the nuke is in such low density and high moisture content materials (afraid the moisture will be much off), and the drive tube and nuke comparison that you suggested would either confirm or deny that suspicion quite quickly. Thank you.

 

[oldestguy]

OG again. I forgot one item that you may know about. The nuc can be affected by material in the sides of the tench above the elevation of the machine base if you are using the surface method, so trench width has to be involved with any calibration. This then may require only using the probe method.

If this were a job I was on and I had been in touch with other engineers with valid ideas, I'd pass that on to the higher ups. If the situation results in a failure, you will have a little ammo.

In my opinion, the proposed plan sounds questionable, to say it nicely.

 

[Ron]

As OG noted, the nuke gauge can be used but must be calibrated to the soil and to the geometry of the trench, otherwise your results could be vastly off.

 

[oldestguy]

OG once more.

I thought more about your project as I was resting from mine and, even though I don't know all the details (pipe size, etc.), you can do the testing job by the nuc and hopefully it will come out OK.

As the trench is dug, or in nearby test pits, take your nuc and use the probe method, taking readings at depth intervals of say every one foot. Don't use only one pit, but several along the alignment. It really doesn't make much difference what identifier you then use for readings, such as soil density, counter readings per minute, etc.. Just be consistent. What you then shoot for in testing the backfill is those same readings with depth. While your readings probably7 mean nothing as to the actual density in #/cf, they do serve as a reference. Then what tolerance can you accept? To answer that go back to the powers that be who thought up this idea and ask them.

Be prepared for a considerable different consistency material as backfill, since it is highly disturbed as compared to in place.

 

[BigH]

No mention that I have seen of what the backfill material will be since it is "set in stone". If silty or clayey you can take field density tests with core cutter. As for undisturbed peat (fibrous or amorphous?) use large diameter core cutter (6" or more).