Arash Yazdani, Director of Engineering Services at PRI Engineering, is joined by OYA Solar’s Greg Rossetti, Head of Land Origination for a webinar on installing ground-mount foundations for solar.
The hour-long webinar covers the considerations every contractor or developer needs to take into account before construction starts and provides an essential decision-making framework. Focusing on the typical foundation piles used in solar projects, Yazdani and Rossetti detail the possible challenges during construction for each one and provide a risk overview:
- Driven piles
- Ground Screws
- Helical piles
Yazdani follows the discussion of foundations types with four case studies of large solar installations in the United States and Canada, including one currently under construction in Franklin County, New York.
To clarify, when they say “fixed versus trackers” they mean fixed-tilt racking systems and single-axis tracker systems. There are also dual-axis trackers, but I suspect he’s specifically interested in single-axis. So, on that front, the difference is that a fixed-tilt system is set at a fixed angle, meaning that the angle is set to 30 degrees or 28 degrees or 25 degrees or 15 degrees or whatever the designer decides for the angle. A single-axis tracker tracks the sun, so it starts in the east in the morning, and it slowly tracks its way to the west. It follows the sun and allows it to maximize generation.
One thing, that most single-axis trackers systems are equipped with, is wind sensors and temperature and humidity sensors. In a lot of cases, the single-axis tracker will provide you with what they call the top of pile loads. Now in a lot of instances, the top of pile loads for a single-axis tracker is significantly lower than that of a fixed-tilt system, and there are two reasons. One, most traditional fixed-tilt systems are two-up in the portrait system. Most of you probably know what a solar module looks like; it has a rough orientation of about two meters by one meter or roughly three feet four inches by six feet four inches. They are oriented vertically or in a portrait orientation, meaning that they have a much larger area compared to the typical one portrait single-axis tracker. So, you have one up versus two up.
So if you go into a windstorm and you look at a flag at two feet above ground versus 15 feet above ground, the flag, 15 feet above ground, is going to be flopping a lot harder. Essentially, the higher you get, the more wind pressure you get, and the larger the size of what we call the sail means you’re going to generate higher wind loads.
Now back to the sensors, the single-axis trackers are equipped with these sensors that typically will put it in what we call the safe zone when it meets a certain wind speed. Now, I’m sure most of you have driven a vehicle that’s probably about 2015 or newer. Vehicles these days fail because the sensors fail. Electronics fail, and the one thing that I’ve always said to single-axis tracker companies is: you know you really should be considering the loads in the worst-case scenario. You have to assume that those sensors are going to fail and get stuck at that worst-case scenario, and your racking system is going to take the highest load.
So, to answer your question, the loads are reported as a lot lower, but in some cases, they should be reported as a lot higher so it’s a little bit of what I call black magic. I’ve seen racking companies do both. Usually, those that understand the engineering a little bit better will typically give you the loads based on the worst-case scenario. But a lot of racking companies that have single-axis trackers will give you that best-case scenario with the sensors and electronics putting your racking system in the safe zone. Meaning, that you’re going to have much less load and that you’re going to have a smaller foundation. Smaller load means smaller foundation, and typically tracker loads are less than single-axis loads.
What specifics are you looking for in a geotechnical report to help you determine which foundation to use?
Consideration number one is soil type. We classify soils as coarse or fine-grained – that would be the first primary classification. Then, when we’re dealing with fine-grained soils, we’re looking at its consistency, so we want to know about the consistency of a fine grain soil, – or we want to know about the relative density of a coarse grain soil. The other factors are your groundwater condition or bedrock conditions. You want to know what the bedrock type is, is it auger-able, are you able to get a drill hole through it, are you going to need a down the hole hammer, or are you going to need an auger? You also want to know what the strength of that bedrock is and what the depth of that bedrock is. Those are all the key factors of what we’re looking for in a geotechnical investigation report to determine which foundation to use.
I would say that the short answer is no. I would say your best pile for resisting frost heave is a helical pile, but we have to keep in mind that it’s all dependent on soil type. Remember Greg said, “if you need a screwdriver and you show up with a hammer you’re going to be out of luck.” So if you think a ground screw is going to be the best solution to combat frost heave, and therefore, you’re trying to engineer a ground screw to meet the requirements of your site. But that’s not the right foundation type, and it doesn’t mean it’s going to be the best solution to deal with frost heave. That’s because if the pile is not adequately installed, it’s going to probably fail against frost heave.
Are there any significant construction-related challenges (design, cost, time, time of year) associated with installing piles on the wetter regions of a site (wetlands)?
100 percent. For starters, I would urge you to try to avoid wetlands if you are going to be building a solar farm. Remember those risks I talked about at the start or about halfway through the presentation? We have connections here between biology, financing, and geotech risks. Wetlands to me are a biology risk so ultimately try to avoid those where you can. I do know there are cases where certain authorities are allowing under certain conditions or if there’s a reward system. So if you go and damage one acre of wetland here, you need to build two acres of wetlands across the road to make up for the fact that you’re building in that area. At times, I know that it’s unavoidable.
However, wetlands are, as you could probably imagine, wet, and when you take heavy construction equipment into wet areas, there’s a really good possibility that they’re going to sink. So, the first consideration isn’t actually even an engineering consideration, it’s a construction management consideration. If your excavator goes into the wetland, and it sinks out of sight, you have some bigger problems like – how am I going to get my pile in the ground? And how am I going to get the excavator out that’s stuck in the wetland? One way around that is to do your construction in the winter when the crust of the wetland is frozen.
My recommendation to anybody that is looking at building in a wetland or a swampier area is to assume that you need to swamp map the whole site. For those of you that don’t know what a swamp map is, it’s essentially a large skid (it is around six feet by six feet and in wetlands, I’ve seen it up to two rows thick), and basically, you’re building a temporary road in the wetland with timbers, but that’s just to access the site. We haven’t even started building those foundations yet, and in most cases, we have low resistance soil – we’ve got poor skin friction. If we drill a micro-pile and pull the tooling out, that hole is going to fill with water, and all that soil is going to collapse.
I mean, there are 101 technical issues with building in a wetland. However, if you’ve got to do it, I can’t even explain how important it is to do a very well-thought-out pre-production testing program ahead of time. Doing a pre-production testing program lets you know exactly what you’re going to run into from a construction standpoint and exactly what type of resistance you’re going to get with your foundation because you’re going to have a challenge to get the uplift resistance.
Have you ever seen a driven anchor solution lying attached to an anchor driven and pulled up to achieve the final position?
So, not a big fan. I’ll start there. This foundation type does not penetrate a foundation into the ground; it penetrates a grounded anchor system into the ground. Essentially the racking system sits at grade, and then you have some guy anchors that attach to the racking system and are embedded below grade (it’s like aircraft cables that are pounded into the ground). Then from the tension, it’s able to achieve a little bit of lateral resistance and to some degree uplift resistance.
I don’t have any proof or anything to support what I’m about to say – but I think it’s a great system for California, maybe Texas, but I think it’s a terrible idea for Northern environments. Those legs are going to sit on the ground with no consideration for frost uplift design. I believe that this system is only designed for wind uplift, and there’s an operation and maintenance/longevity of the product concern when you’ve just got all that steel sitting in the ground, and it’s going to be absorbing freeze-thaw, which is going to potentially cause fatigue failure of that steel. We would also be concerned about how the racking is going to flex in the freeze-thaw scenarios because the reason you go through all that effort in resisting frost is to avoid having that movement in the racking and inducing potential microcracks into the panels through that movement.
You have to keep in mind that the reason we go through this effort of putting in a secure or solid foundation is that frost-action is not going to be consistent across the site. One square foot of earth can have different actions, and you can have sizable differences where you’ll have frost heave, and the earth will push four inches – six inches in one area and not at all in another. You start doing that to your tables, and you might find that the panels don’t like that. The point is about differential heave, and if you have a rigid deep foundation, then that’s going to be a much lesser of a concern than if you’ve got a surface ballast system.
If I were to be using that system, I would want it to be a free-moving system and what I mean by that is it would just have two posts – one on either end of the table. If you introduce any middle posts and let’s say, this pile shifts up, it’s going to be anchored by these two other posts. But if you’ve only got two posts, it’s going to be a little bit freer to flex, which isn’t going to cause as much stress to the steel frame and most importantly as Greg noted to the panels.
At the end of the day, the whole point of solar racking and designing your foundation properly is so that the stress from your external loading conditions is not being transferred to your module. As long as the modules are safe, everybody’s happy; that’s the most important thing. So if you’ve got a 2-post system, and it’s able to move a little freely, I’d be less concerned because there’s not going to be as much stress transferred to the modules. However, over time, there might be some operation or maintenance issues with that actual steel frame structure, and it’s going to go through fatigue or other types of stresses that are going to probably affect its design life. One last thing to add is that typically guy anchors are used for providing lateral resistance to transmission poles, so it’s not exactly the right application. I personally designed a remediation design for piles that didn’t have the adequate lateral capacity using guy anchors. But we have to keep in mind that there was already a deep foundation there. We didn’t have the adequate lateral capacity so we anchored it so that it wouldn’t you know it stiffens it up a bit.
*Edited for clarity