roadbed stabilization

[schreibs]

I am a mech engineer, building a home with 1500 ft lane in from the blacktop over mostly sandy but also clay and silty soil. I have an area on the lane that is soft and spongy and allows the dump truck to sink down and make big ruts. I had already planned to use some Propex separation fabric(200ST) but certain areas are so bad I can stand on them and they act list a tarp laying on water-- almost! Digging down I find clay on top covering a deposit of saturated and mushy sand. I can dig down with my bare hand and remove huge handfulls of the wet snotty sand with little effort, and deeply. There is about 100 yards of lane with this problem. Please offer ideas for remediation. thanks!

 

[oldestguy]

OG here. I Goggled Minn DOT specs and see the class 5 aggregate is similar to what is used in Wisconsin, but I think you need to avoid recycle concrete. The reason is that does not specify much of any binder sizes. Using it for a roadway surface is likely to have much raveling with traffic passes. Otherwise it is a good gradation, well graded meaning sizes work together to make a dense result, strong in shear strength.

 

[schreibs]

OK, I will get a half inch rod and profile the road. Graph forthcoming next time I visit. . .

Keep in mind the 1.5 CBR "rating" reflects only a seat-of-the-pants rating for the worst places on the road, mostly showing up where vehicles wheel travel and in the lowest area near the hayfields low spot. Much (70%?)of the concerning part of the roadbed(farm lane) is better than that by far, no sinking at all under human traffic, hard. The overall farm lane is about 1/4 mile long with only about 100 yards of concern due to likely high water table causing this localized low, apparent CBR.

In the meantime I plan to get a short piece (about 40 ft) of Propex 200ST and apply it thusly: over the "as is" subgrade of silty, wet sand at 20" depth, mixed topsoil and clay above that to the surface, I will spread 2 to 3" of sand from my pit, <<< all this exists already. . . then place the 200ST fabric, then two thicknesses of class 5 aggregate(2" over first 20 ft(A) and (B) 5" for the other 20 ft). Then, I will just start using the road and invite my neighbor to drive his single axle truck over it(no load) a few times. Depending on the response from A and B cross sections I may dump some more sand in his truck to increase load, or just STOP!

Experiment to follow.

 

[oldestguy]

If I gather this right, on top of the fabric you will only use 2" of Class 5 and in another area 5". I envision the use of fabric normally is to have it at the bottom of a thick layer of aggregation. The purpose is to reduce the thickness of gravel that otherwise would be needed to pas over the weak subgrade. This is sort of like say you have a muddy area in your yard path to the garden and you then spread enough gravel to spread out your foot print so it doesn't sink in in a small area, but instead a larger fear, going in less, due to less shearing of the mud. In your case now suppose instead of that gravel you lay down a sheet of canvas instead of the thicker gravel. You probably won't sink in as far, but still plenty. Add a few inches of gravel on the canvas and it helps, but not much.

So for a very weak subgrade, you might need 18 inches of gravel to drive over it. Instead, you have that fabric at or near the bottom and then carry the same traffic, but maybe with 10 inches.

I liken this to reinforcing concrete to carry a load as a beam. For a reinforced concrete beam the steel is needed near the bottom . Put that steel near the top instead, it then has minimal benefit. So a thin layer of gravel on the fabric is pretty much not going to do noticeable good.

For typical highway use of fabric in the bottom of the base course, and for say a CBR of 3, one may need 18 inches of base, instead of 24" without fabric, depending oon the strength of the fabric. I have seen many a failed use of fabric where the covering with base course is too thin.

For your test, I'd start with 6" at one end of the test and end up with 12 or more at the end, a tapered usage. Fabric on the bottom on the subgrade. Test area length probably 10 feet minimum. In a pinch I'd do the test on one wheel path maybe 4 feet wide.

 

[cvg]

well in addition to adding strength to the subgrade, it also acts as a separation layer preventing the gravel from being pushed into the mud.

 

[oldestguy]

CVG: For well graded base course as he is looking at no separation is needed due to the fines filling the voids of the gravel. Fun stuff here.

 

[schreibs]

Thanks guys. I decided to buy the whole dang roll($345, big deal) and go for broke. Based on your input, oldest guy, it appears my A/B test was in the wrong range! Keep in mind my objective was to cause a significant effect, not necessarily to have both sides "work" (i.e. an experiment). Just the same, I will adjust the range upwards. Also, I will try to get you guys the 1/2" rod profile before proceeding if possible.

 

[oldestguy]

The rod probes we use are about 30 inches long with a 1" pipe handle welded on top to make "T".

On that neighbor's truck load test be sure it has a full load and remember the ready mix trucks are much heavier yet.

 

[schreibs]

Here is half inch rod data. Process used: pressed on handle with both hands, if road took the weight easily I stopped. Likely, this applied less than 50# force. However, if the rod sank easily I put my whole 155# across the handle and let it drop till it stopped. I had to upload a pdf file-- pasting into the window here compressed all the numbers, unreadable.

So, my big question for oldest guy: How does the 1.5 CBR estimate square with these numbers? and! HOW bad is this?!
Ha!

 

[oldestguy]

On the basis of what you show, it seems that any test area should be between 170 and 210. That would then be a governing treatment elsewhere. Can you say about how wide the road is and about where the north and south checks were in that width? Could it be that former gravel, if any, was less road width to explain the differences from center to test paths. How much traffic took place and time prior to these rod probes to explain any temporary weakening measured? I'd suppose your test paths were in the wheel paths? Your CBR of 1.5 may well be right for worst areas, meaning the Tensar guy's recommendation may well be too light.

 

[oldestguy]

Shreibs: A rough idea of how I evaluate conditions using fabric follows. Commonly the legal truck axle load limit is 18,000#" (18kips). That means a set of duals carries 9,000 pounds and likely the air pressure is about 100 psi, sometimes a little lower. That gives an area loaded of 90 square inches, probably on either one tire or duals. I then figure the pressure on the fabric by computing the fabric loaded area using a load spread on an area computed by enlarging the dimensions of the area (9 by 10) by adding the stickiness of cover to these numbers. Thus for a 10 inch layer the fabric loaded area is 19 by 20 or 380 square inches making the pressure there 9000/380 = 24 psi. Once I find a suitable thickness by the tapered test area and compute the max pressure I want on the fabric, that then is used for other axle loadings and a "pavement thickness" design. It may not be precisely right, but gives an idea of how one might do this. So, if you run a test with a lighter axle load and find a cover thickness that just holds and does not rip the fabric, that would then be used for those heavier trucks, which likely won't exceed the 18kip axle load. Hope this helps. Pressure limits on the fabric are assumed to be related to shearing and the tearing of fabric. Complicating this is stretching or even moving the fabric, leaving a hammock situation and a rut. So for a rutting situation I use one inch as maximum rut depth accepted. That may well control if side dimensions off from the traveled path are not much.

 

[schreibs]

additional info you requested: . . . and MORE!
-- virgin land for the most part. No gravel ever saw this road until I dropped some class 5 on top of the concrete pieces last week.
-- the "road" is really just a farm lane that until I bought the property was used only by the farmer who made hay off the grasses and we drove through 3' high grasses on this "lane" when viewing the property first time!
-- the farm lane is variable in width, but approx 10 ft wide. I will HAVE to make it 12.5 ft wide to accept the fabric. The North and South lanes are about 5 ft from the center data, not measured. I just plopped the rod down in the likely track location to within 14" or so.
-- I think the important thing here is to consider if the fabric will be affected enough to fail early when there is SO much variation across the width as these data indicate. I figure the fabric will be severely depressed on one side and be deformed there while the other might hold up OK.
-- the lane lays at the bottom of what I believe is likely a hill of sand, mostly and glacial till at the bottom. The hill probably exists because the original river eroded away the till eons ago and left silt in this narrow river valley up to the river edge 100 yards away. This is backed up by soil borings from my Septic designer and borings from he gravel pit north, and by my observations of the subgrade I unearthed last week, and by geology info found on my property on the net.
-- I get my sand from what really looks like an esker on my land. It is a long, narrow ridge of sand that runs through my land and north of it, north and south.
-- We have not heard much about the potential the (supposed) raised water table may have on this road stability. . . Wondering if I waited a year would it be gone?
-- The Tensar fellow gave me a cute little nomograph he used to calculate the CBR that is able to use seat of the pants criteria OR calculated numbers. We used seat of the pants criteria column to lead us to the 1.5 CBR because we had no CBR measurement tester OR the neat 1/2" rod data! Anyway, it is my belief without trying to analyze his little slide rule / nomograph thing that it takes into account the patch dimensions, pressure, number of passes a truck makes, and it has two sides to it. One side is to get a CBR related to a one inch rut and the other is for 3" ruts.
-- I agree with your conclusion about the location for a test area and will do "something" there. . . !
-- Remember the Tensar guy's recommendation were NOT for fabric originally. He wants me to use his nice, expensive $1660/ roll plastic triangular mesh-- TX160
-- I will run a test patch of Propex Geotex 200ST in the test area and if that fails look to his high buck product.
-- I have a local paving company looking to see if they have any left over Tx160 they could sell me. Brock White(dealer) will only sell full rolls.

 

[oldestguy]

It looks to be a job patty well thought out. With fabric there any future excavations, as for buried electric lines or drainage pipe installation will "mess it up". Be aware that if the edges of the fabric become exposed, snow plowing also will do a real job on it. This must be kept in mind when placing the covering gravel also.

 

[schreibs]

Well guys I put in about 20 dump truck loads(7 yd)in all-- much of that over just the basic fabric(not "GEOGRID" plastic) and it appears to pretty much have solved the problem. Of course, time needs to take its course to really prove this out.

I did not use the geogrid materials on this mushy area because it just cost too much(from Tensar). However, on a tough area where vehicles must make a hard turn ON a hill slope the grade was getting beaten up pretty badly and I was able to buy some material from a local paving company that was not even as strong as Tensar's TX140 and placed about 45x 13 ft of that on the turn (two pieces, $120) under some class 5 (clay, sand, gravel mix). I plan to top that with some ground blacktop to prevent erosion.

There is no way to tell the relative worth of the fabric (mushy area) in this test probably but I did leave some areas on that stretch untouched with a plan to use the knowledge gained after some "running" time to determine whether it is worthwhile. Two things appear to be major contributors better road so far: 1) got the road higher off the water table and 2) the total thickness enabled the forces to spread out the effect of weight and shear. Of course, fabric helped to accentuate that effect. I am hopeful the grid material used does help with the shear being generated as big trucks take the turn on the grade. For now, I am a lot smarter for having visiting you folks and have a neat 1/2" road bed tester to boot.

Thanks for everyone's help!