Telecommunication tower base 2

[phuduhudu]

I am designing a base for a telecommunications tower. It is to be a typical design so the bearing strata need to be checked in each case but this whole country is fairly sandy so it's either deep sand or rock.

The problem with tower bases of course is that any small rotation of the base will cause significant rotation of the 30m tall tower (as they found out at Pisa). I am planning a cruciform base and have tried to do settlement calcs based on assumed minimum SPT values using Burland and Burbridge. However, this gives a large range of results from 3-25mm. The bottom end is acceptable but not the top end. The other complication is that the overturning loads are wind loads so they are very short term loads and I'm sure this has an effect.

Any ideas on what sort of limits to set on a problem like this?

 

[Riz]

Allowable bearing capacity based on SPT values is based on a figure of 25mm. Unless u carry out consolidation test of the soil any lesser or higher prediction of settlement is difficult.
With sand u have inelastic immediate settlement. This normally takes place during the course of construction. To encourage this, u need to compact the existing subgrade and place a hardcore layer to keep it in a some what consolidated state. The weight of base and tower will add into its early consolidation.
If u r founding on an interface of rock and sand, a pdelta analysis based on modulii of subgrade reaction is of utmost important.

 

[emmgjld]

The use of SPT values must always been accomplished with caution. Local experience should be checked. Consolidation Testing in Sand is difficult but, can be done in some circumstances. In my arid to semiarid environment, I have enough 'cementation' due to Sulfate Salts and very small amounts of clay so that dry samples can be prepared from Shelby Tube Samples. You may not be so fortunate.

I consider Riz's suggestion of over excavation and proper compaction of the native sands to be the most obvious course of action. Not only is the inelastic settlement problem immediately addressed but, If the sands have any slight collapse potential (ie metastable) this can also be corrected, assuming the deposits are not too thick.

Soil Improvement is usually superior to playing with the foundation sizing and reinforcement.

 

[phuduhudu]

Thanks Riz and Emmgjld. I realise that allowable bearing capacities gives a potential settlement of 25mm. This would be too much for one side of the foundation but as you say there will be some presettlement from selfweight. I am in Botswana and much of the sand deposits here are potentially collapsing so I would recommend overexcavation and recompaction.

Having done this can you give any idea of the best method for calculating settlements and especially under short term loads. Can I do it from just SPT results or do I need more data (bearing in mind the difficulty of getting reliable results for sand).

 

[Riz]

Carl,
Options for u to consider:
1. At foundation level carry out Plate Loading Test. This will give u both allowable beaing capacity and settlement values.
2. Yes u can predict settlement with SPT test. Look for a suitable book on Soil Mechanics, equation by De Beer and Martens should help.
3. If collapsible soil is giving u a headache, go for a MAT foundation.

 

[Riz]

........further......
From item 1 above u can work out the pad size with respect to the allowable bearing capacity and a limiting settlement value.

 

[emmgjld]

Carlbauer

You appear to have metastable soils and micro climate problems similar to what I experience. My dessicated, metastable soils my be up to 25 - 35 feet (8-10 m)thick and are usually underlain by expansive Shales and Mudstones. I have adopted a scheme of overexcavating, watersoaking (if site conditions allow) and replacing with Structural Fill, usually Native Soil. One of my goals is to change (usually increase) the foundation soil moisture environment to match the later service conditions. The water soaking must be done carefully, in order to avoid problems to nearby structures. There are a lot of subtle and not so subtle factors to consider when soaking.

I prefer to use a low plastic, clayey soil for final, upper backfill around the foundation. I have found this tends to 'shed' the surface runoff and reduces later, deep soil moisture increases.

To put some of the previous suggestions into perspective, we need to consider:

First, the SPT test assumes the soils are reasonably moist to nearly saturated. The Plate Bearing Test assumes the soil moisture conditions of the test are similar to the service conditions. For me, these conditions are usually unrealistic.

Next, what is the actual service conditions of the structure? Such as, will the foundation soils generally gain or lose moisture? What reasonable, total depth of soil moisture change can be expected? Can I postulate 'levels' of potential settlement?

Next, Time for reasearch. Modify a 'test' site by overexcavating, water soaking (if appropriate), densify the subgrade and replace with a fill. Then either drill and get very good sanples or carefully test pit (or completely excavate the improved soils out)to obtain densities and good samples and then laboratory testing. How effective was the soil improvement?

Then modify the soil treatment, as required.

Remeber that the structural fill begins to interact with your foundation and the whole behaves like a MAT Foundation. The total potential settlement can still be fairly high, but the differential settlement is usually dramatically reduced and structure servicability is usually maintained. Reinforcement of the structural fill with layer(s) of a Geotextile or Geogrid Product can also be very effective to creating your 'structural unit'.

At this point, the SPT testing may be corrolated for this soil type/moisture condition. I belive that in-situ soil density will be more helpful, especially if you are able to accomplish consolidation/collapse testing.

Sounds like a lot of work. I just went through this for residential construction design recommendations. Home Owners in the U.S. tend to water a lot and cause enormous problems. The grass must be very green!! We were pleasently surprised at the amount of actual improvement which could occur by our treatment and we have been able to create a corrolation with in-place soil density (Shelby Tubes and possible down hole nuclear densitometer) rather than the SPT, which is always high due to our dessicated soils.

Keep us posted with new information and your suggestions.