A thousand things have to go right for a wind turbine to run, but only one thing has to go wrong to bring it down. This might paint a grim picture for some, but not for me. You see, I work on a wind farm with 167 wind turbines. If I wake up from a night’s sleep and find only two turbines are faulted, that means only a couple of things went wrong and at least 100,000 things went right.
The work is fun, but the days can be long. We’re talking 14-hour butt-kickers. I’ve been trapped in the hub of a wind turbine 300 feet off the ground – during single-digit temperatures and rated winds – chilled to my core chasing a pitch fault that seemed impossible to fix, and I had to go right back to it the next day. Anybody who has ever spent the day up-tower knows the feeling.
I will outline some general practices and troubleshooting tips and some helpful advice to use on those troubled turbines so that you can come home happy and relaxed because you fixed the turbine and saved the day – there is nothing better than that feeling you get after fixing the faulted turbine that has been down for too long.
Training. One of the key building blocks for a good technician is training. Understanding basic mechanical and electrical theory in combination with specific original equipment manufacturers’ (OEMs) turbine systems will set you up for success when assigned to the unthankful tasks of identifying those hard (or hard to find) faults.
To date, I have been trained on three different turbine platforms by OEMs, as well as the Danish Windpower Academy. Outside of that, I have brought in specialty trainers for turbine controllers, such as Bachmann, to help my team completely understand all of the components in a turbine.
Establish a pregame ritual. Before you hit the field and make your ascent up to your office for the day, you should be doing some pregaming – identifying the primary fault and the sympathetic faults that follow using your SCADA system; researching those faults in whatever fault library you have available; analyzing the event data to determine what occurred first; digging further into any other systems that could give you insight, such as Toolbox for you GE techs; tracing out the schematics with your team for the system you will be working on; and then gathering up all of the essentials for the job.
We have all heard the saying, “Don’t bring a knife to a gun fight”; well, the same thing holds true for technicians. Don’t bring a Fluke 87 Digital Multimeter if you need to test insulation breakdown; bring a Megger MIT410 Insulation Resistance Tester. The point is that there is a proper tool for every job – if you want good results, bring the right equipment.
Schematics should be another critical element of your pregame ritual. If you have a set of pristine schematics – without any voltage readings, highlighted circuits or resistance measurements that are written down – you are not going to be as successful. Mark them up all you need to. Trust me, it will help. You can always print out a new set of schematics.
Troubleshooting. You have reached the turbine site and are ready to start digging into the fault. Whether you are looking into a communications failure, a safety chain, pitch issues, a mechanical failure, a grid sync issue or a converter fault, there are some basics that you have to always keep in mind.
The first troubleshooting method that I like to use is the sensory method. Take the time to look, listen, feel and smell what is around you. Most of the time, if an electrical component failed, you can smell it. If you have a mechanical problem with the drivetrain, you can hear it. Some of the more obvious faults are the ones you can see, such as a set of step-down transformers that have caused a small arc blast due to their inability to stay bolted down to the back of an axis cabinet in the hub. This method is simple and might seem silly to even mention, but it’s useful and practical, as long as you follow the proper lockout/tagout procedures to protect yourself.
Navigating electrical faults. Some of the most difficult faults that can occur in a wind turbine revolve around electrical circuits.
An example would be the use of electrical circuits in an electric pitch system that uses low-voltage communication to control higher-voltage and current pitch operations. In a system like this, you have a lot of digital input and output signals that have to be present before the higher-voltage system even thinks about doing any work.
Recently, I had a group of techs go up to find the situation I previously described with the step-down transformers braking loose and shorting themselves out. They had seen this happen in the past, so they changed out the transformers and all of the fuses that had blown and cleaned up the area; all of the faults cleared, allowing the turbine to be returned to service.
When they returned down-tower and ran the turbine up, it went online, as to be expected, but it didn’t last long. The turbine faulted on an axis-specific fault that had been misleading, causing them to troubleshoot an area of the turbine that was really not the source of the issue. Eventually, the techs found the source of the problem in another cabinet.
They found that the safety loop had dropped out to all of the axis control boxes, but the source of the 24 V supply was still intact. At this point, it came down to looking at the entire circuit from beginning to end and cutting the circuit in order to find where it had been broken. They determined that the power was going into the control card that distributed the power to the rest of the three axis controllers but was not coming out. Just change out the card where the voltage stops. A simple fix, right? Well, they did that, and the same thing occurred.
When I decided to go up, the first thing I did was gather the team together to discuss all paths that had already been taken and come up with a path going forward. We knew where the safety chain was stopping, but we didn’t know why. The first thing we did was measure the source of power, then the resistance of the conductors where the power was flowing on both sides of the break. When I went to go measure the current being drawn between the card that was expected to be bad and that had been previously changed out, the hub came alive, and the fault was clearable. Knowing I had just completed the circuit with my meter by giving the voltage another path to take, we decided to test the resistance of the card in question. Sure enough, the card’s resistance readings were extremely high. We replaced the card that was just installed the day prior with a new card, and the fault cleared.
Never assume. I don’t know how many times we see a fault and jump to the conclusion that it is a particular part that has failed because it has done so in the past many times before. Field technicians must go into each situation with an open mind, or we can find ourselves following a rabbit hole that leads further away from the fix. Remember the acronym SLC: source, load, conductor. Identify the source of the power, the load that is drawing the current and the conductor the current is passing through. You will find the source of the problem if you understand your system and what it is intended to do.
There is nothing better than that feeling you get when you fix a turbine that has been down for a while. Using basic electrical theory, cutting the circuit, trusting but verifying, communicating with everyone involved, looking at the most obvious solution, and not being afraid to try the same thing twice all helps to get a turbine back online.
Unfortunately, sometimes you get parts that are bad out of the box, and they can cause some unnecessary headache, but it’s part of the game we play. Always be open to suggestions from your fellow techs, and don’t be afraid to ask for help. A fresh set of eyes on a fault can go a long way. Stay safe out there, and stay positive.
Neal Gyngard is an experienced wind turbine technician and founder of Tower Climbing Grease Monkeys, a user group for wind turbine technicians. He can be reached at firstname.lastname@example.org.