Wind farm civil works projects: typical errors

Here you have my collection of errors I frequently see when I check wind farm project developed by external companies.

Being a quite peculiar sector is no surprise that I normally found several mistakes: here you have the most commons.

 

Error #1: two levels crane pad / foundation area designed without considering the slopes of the foundation pit.

Here the problem, as you can see in the transversal section, is that the slope of the foundation pit “enters” inside the crane pad and the road nearby, reducing the available space.

The only way to build something similar is with a "2 steps" constructive approach, i.e. to build the foundation, to close the hole and after to build the crane pad.

An example of this 2 steps approach can be seen in the following drawings, taken from a real wind farm. The ground was very steep, so first of all we calculated the elevation of the bottom of the foundation pit to ensure the necessary soil covering that help balancing the overturning moment. In several cases the result was that the center of the foundation was too far away from the border of the crane pad (around 16 meters), making difficult the work of the main crane (a standard distance is around 10 meters).

So we had to approximate the border of the platform to the foundation filling the area in between until the required distance is reached:

 

Error #2: crane pad and foundation on an embankment.

The problem here is that the foundation must be realized below the natural ground for stability reasons. Normally the depth of the bottom of the foundation pit is around 3 meters, calculated from the lowest point of the terrain around the circumference of the foundation.

But if in the project the WTG is shown on an embankment, probably the stair or even the tower door will be below ground. In the transversal section you will see clearly the problem.

 

 

Error #3: road and crane pad at different levels

Here you have a top view of a platform and the access road. Everything looks fine:

But then, when we check the longitudinal profile, we discover that the road is going down (while obviously the platform stays at the same level, 646 meters). The only way to build something like that is to use a wall to retain the earth inside the crane pad, but this is obviously not the case.

At the end of the platform, the height difference is almost 5 meters:

In the longitudinal profile above you can see the platform in yellow and the road in dark blue.

 

Error #4: insufficient vertical transition curve

Sometimes the Kv parameter for the vertical transition curve, defined as  is not adequate.

This happens when it’s lower than 400 – 500: in these cases, the truck can remain “stuck” because it touches below.

 

Floating roads on peat

Floating roads are a solution developed in Scotland during the construction of several Km of internal wind farms roads on very poor, organic materials such as peat.

It is an interesting constructive technology, developed on a very compressible, decayed material normally considered as the worst possible subgrade for road construction. In a normal highway project, peat would be removed

Usually floating roads are stabilised haul roads with one or two layers of geogrid. This approach leads to several advantages:

 

  • Reduced road thickness (and subsequently less weight)
  • Better distribution of pressure
  • Untouched surface layer, lesser impact on vegetation
  • Less construction material needed
  • Usually cheaper

 

The suggested approach to floating roads construction is based on six steps:

 

  1. Carrying out a detailed survey to define the hydrology of the area and the paet type (according to the “Von Post” system, there are 10 possible classification for this material depending on the decomposition level)
  2. Identifying the value for in situ peat strength
  3. Estimate the expected traffic loading
  4. Design the road
  5. Monitoring the construction
  6. Recording action and outcomes for future projects

 

Among the design assumptions, it’s relevant to remark that a rut with a depth up to 10 cm is to be expected.

The first step is an in depth site investigation, where several parameters are recorded: peat depth and classification, side slope angle, hydrology and permeability. Many in situ test are available, from probing and sampling to more sophisticated techniques such as ball penetrometer or Mexecone penetrometer and ground penetrating radar. After, in laboratory, water and organic content can be defined, together with vane testing and direct simple shear tests.

When all the data is collected, normally is the geogrid manufacturer who will design the cross section of the road, deciding the type of grid (there are several dimensions possible – they must match with the aggregate size), the number of layers and the height of the compacted stones.

The designer will use semi empirical rules, using as inputs shear strength of the peat, weight of the road and expected number of equivalent axes.

It is standard practice to use at least 2 layers of geogrid: a lower one directly on the existing material, and an upper layer on the top of the embankment, used as a support for the controlled granulometry crushed gravel.

Good practice during windfarm construction

“Good practice during windfarm construction” is a document produced by several Scottish agencies (Scottish Renewables, Scottish Natural Heritage, and the local Environmental Protection Agency and Forestry Commission).

It provides several useful advices on wind farm construction, based on the experience they gained with the development of several pretty big projects (for instance Whitelee W.F., with 140 WTGs, 86 Km of roads and 940 Km of cables).

The document is based in a northern Europe environment, where many roads are on peat and heavy precipitations are expected. Here the main points of the documents:

 

  • They suggest designing the drain for a 1:200 year event and the use of pre-earthworks drainage (a solution seldom used in dry countries).
  • Regarding the shape of the ditches, they explain the differences between “V” shaped ditches (they maintain more the vegetation, but they have more erosion) and “U”shaped ditches (they allow easier access and egress to wildlife).
  • Considering that many times internal road are built without camber to allow the use a narrow truck crane, they propose the use of a series of cross-drain to divert the flow to the side ditches.
  • Several techniques are suggested as protection measures: silt traps, silt fencing, straw bales, settlement lagoons or even the use of flocculant.
  • They explain that cable trenches can act as a water drainage route, so in case of strong gradients clay plugs and impermeable barriers should be used to limit water flow.
  • They recommend not to install MV cables in soft peat, because they sink adding tension and potentially breaking.

 

Last but not least, there is a detailed explanation of floating tracks (road construction on peat using geotextile or geogrid). The suggestions are:

 

  • To build them only in flat areas, to avoid the risk of a slip or circular failure.
  • To leave vegetation and tree roots in place, and lay the geogrid directly on the existing terrain.
  • To avoid leaving open trenches or drainage ditches in the proximity, because the cyclic loads due to truck passage can lead to a soft materials migration and collapse of the road.

 

If you are interested you can download the entire document here: Good practice during windfarm construction

Tensar Triax geogrid use in wind farms

Here you have a real world example of geogrid use.

We are in a wind farm in southern Spain, and thanks to previous experience with this technology the client decided to use a Tensar Triax geogrid TX160.

This is a triangular geogrid (the “old school” version was square). It seems that the triangular geometry guarantee a better distribution of the loads

The same geotextile  has been used in many wind farms (more than 100) around the world, with several big project in Germany and UK where in some cases more than half a million square meters have been used. Often the soil was peat (turf), with a very low CBR (even less than 1).

The saving in crushed stone using a geogrid can be around 30% to 40%, so with a price around 2-3 €/sqm it is normally a cost effective solution.

When we opt for this solution we’ve had helpful feedback and hints from Tensar, as they can study the existing info and provide a design based on the soil conditions and the available materials. There are several models available with a different triangular size, so it is not easy to choose the right model.

We haven't had any problem during the first half of the civil works circulating with machinery and trucks, but when we started moving the narrow track crane on the internal roads between the pads significant damages appeared:

Maximum wind farm internal road gradient

This is another standard problem I found in the wind farm I’m working with: mountainous areas, with difficult access and very strong inclination.

The standard maximum slope imposed by several manufacturers (for instance Repower and Gamesa) for safe transport on gravel roads is about 6% to 7%. Above 7% other technical solution may be necessary, depending on the trailer used to pull the T1 and the nacelle.

With an average quality track surface a 6x6 tractor unit can pull the nacelle approximately up to a 9%-10% slopes.

Than it is necessary to pull the truck with a bulldozer, using a steel bar or a steel cable (see pictures below). In this case the inclination was around 15%, and we used a D8R Caterpillar and a steel cable), without particular problems.

It must be noticed that in steep sections long horizontal alignment are preferable respect to closed bends.

What we normally do in extreme cases (above 15%) is to build a ramp using concrete slabs or an asphalted road: this solution is not only more expensive, but can also introduce additional problems, such as the need for environmental authorizations or from other authorities.

Here you have a 17% asphalted curve:

Concrete slabs are normally cast on site on a layer of aggregate stone, and they have dowel bars for load transfer and sealed contraction joints. An initial texturing is made with a burlap drag or a broom device, while final texturing is made with a spring steel tine.

You can find more information on the subject in the JPCP Design and Construction Guide

Jointed Plain Concrete Pavements Design Construction Guide

 

Geotextile vs Geogrid: which is the best solution?

This is a debate we are living each time we have to build internal roads on soils with a low (<5) to very low (<2) CBR.

Basically when the others alternatives (mainly soil substitution and soil improvement) are not feasible we are adopting two different solution, either a strong geotextile with reinforcement and separation properties or a geogrid (coupled with a thin geotextile used as a filter if necessary).

Presently both of them are working well, but only after many years we will know which one works best. I'm hearing very different opinion on the subject, so there is not an universal consensus.

As geotextile, both woven and nonwoven alternatives seem reasonable. Both of them provides separation of the aggregate from the subgrade and have high tensile strength and modulus, adding reinforcement to the foundation soil. Right now the woven solution is widely preferred.

As woven geotextile we have used the US250 from US Fabric, with the following properties:

PROPERTYTEST METHODENGLISHMETRIC
Tensile StrengthASTM D-4632250 lbs1,112 N
Elongation @ BreakASTM D-463215%15%
Mullen BurstASTM D-3786450 psi3,102 kPa
Puncture StrengthASTM D-4833100 lbs445 N
CBR PunctureASTM D-6241900 lbs4,005 N

And as nonwoven, something like the US Fabric US 160NW  looks like the best option:

PROPERTYTEST METHODENGLISHMETRIC
Weight - TypicalASTM D-52616.0 oz/sy203 g/sm
Tensile StrengthASTM D-4632160 lbs711 N
Elongation @ BreakASTM D-463250%50%
Mullen BurstASTM D-3786305 psi2,103 kPa
Puncture StrengthASTM D-483390 lbs400 N
CBR PunctureASTM D-6241410 lbs1,824 N

Regarding geogrid, it has been used in several wind farms all around Europe. I had a meeting with the representative from Tensar, and their product looks interesting.

It is a triangular net, providing support to the stone aggregate. It works equally well in every direction.

We have used it in Spain, and it have been used in several other projects in UK and Romania.