Contrary to appearances, bolts – no matter what their size – change shape under load. They stretch. When the load is removed, they contract. Such behaviour is expected and predictable, depending on the characteristics of the material (and the load).
If we can know elongation based on load, we can also determine load based on elongation. This principle is now being used by a number of suppliers to develop ways of measuring bolt pretension in critical applications where loosening poses major risks.
For example, Danish wind turbine engineering services provider R&D Test Systems has refined an ultrasound-based technique for determining tension in large bolts used for wind turbines and other heavy industry applications called Bolt-Check.
At this scale, bolts are tightened using electro-hydraulic torque wrenches. Measuring pretension torque values indirectly, from the bolting system alone, can be inaccurate. It does not take into account the variance in friction in the thread or between the bolt head or nut bearing surface and flange, or surface finish issues such as tolerance and surface roughness, according to Bolt-Check key account manager Jens Vestergaard Nybo.
Instead, Bolt-Check relies on an ultrasound detector to send a beam through the bolt, and, using the time-of-flight technique, records a kind of echo of the bolt to measure its length. In the load-to-zero method, readings are taken of the bolt when loaded in situ and when unscrewed. Subtracting the difference between them provides what he calls a negative elongation value. Then, once the technician has the value of the material’s spring constant – information available from the bolt supplier or wind turbine OEM – it’s a simple exercise to determine what the clamping force was originally, and whether it was above or below the required tension. Then the bolt, relubricated, can be reinstalled, and the specified tension verified by Bolt-Check.
As the inspection process will be carried out repeatedly, R&D Test Systems simplifies the tracking process by installing traceability tags near each bolt. Technicians scan the tag to bring up the record sheet on a digital device, on to which new readings are automatically uploaded.
To further simplify the job of measurement, some bolt suppliers have started installing the tags directly into the bolt stud. Not only does that save the technician a job, but the manufacturers go a step further. They perform their own time-of-flight length measurement of every such fastener at the end of the production line, and upload that figure into the record sheet linked to the tag. A further benefit of this is that measuring an unloaded bolt is a delicate procedure and shouldn’t be rushed, Nybo adds.
The Bolt-Check system has been offered for several years, and makes use of a Dakota Ultrasonics terminal, specially simplified just for this function. New users need to take a two-day training course to understand the basics of ultrasound, how to use the equipment and how to interpret the readings.
DIRECT SENSING
Two other suppliers are developing a more direct approach: a way for the bolts to tell their story, without having to ask them. Sensors mounted within the body of the bolt sense physical changes physically, and transmit that as a signal whose strength is proportional to the magnitude of the physical stretch.
The advantage of such a technique is that monitoring can be continuous; the devices don’t have to wait for a technician’s probe. That means that, for example, they could send an alert when pretension drops below a certain threshold, allowing time for technicians to intervene before the joint fails. This is predictive maintenance in action.
Another common feature of the two systems profiled here is their in-device data processing. Michele Fabbri, team leader of Sens-In, a start-up subsidiary of Italian fastener manufacturer Poggipolini, says: “We are developing digital bolts, where the signal comes out of the bolt already digitalised. That’s very important because an analogue signal can’t run long distances without being ruined by interference: environmental, magnetic waves and disturbances. You would be bound to be very constrained in installation layout for logging electronics (a ‘box’) to be very close to the bolt. Whereas with a digital signal directly out of the bolt, you can have a cable running to the gateway. That can be many metres, and opens up applications on airplanes mainly, which we are interested in, but also on buildings, structures, tunnels, bridges or power plants.”
The other product, Smartscrew from German firm Bornemann Gewindetechnik records pretension, movement, vibration, torsion and temperature, and sends the information wirelessly to an external receiver via Bluetooth LE. Two ring-shaped sensors in contact with the inner bolt wall measure tension and compression.
To fit the sensor package into the fastener, Bornemann requires a minimum internal diameter of 14mm, so its smallest bolt has 25mm outside diameter. Transmission range is 30m, but there are plans to install a ring antenna to increase range. Also new is a web-browser accessible database, on which data for up to 12 bolts can be viewed, and thresholds set. Sampling rates are up to 400Hz.
The sensor can be powered by a rechargeable AAA capacitor or through induction charging – during which the sensor receives charge during bolt tensioning. At data transmission rates of once per minute or per hour, the battery should last for years, though the software can be set to send an alert when the battery is low.
Farther to the south, Sens-In is developing two types of in-bolt sensors to measure either pretension or vibration (in addition, both also measure temperature). Initial proofs of concept are wired; a wireless version is also in development. Minimum bolt sizes are said to be an M8 bolt, with an extra 10mm added to the bolt head for the electronics. It has taken out a patent on its technology, which includes automated production. Another current design project intends to further reduce the size of the electronics.
Fabbri says: “Adding a sensing point to the aircraft without thinking about intelligent fasteners would not only increase weight but also complexity, because you would have to find a strategy to install this new sensor somewhere. Making another threaded hole somewhere in already-established designs could be a nightmare. The fact that we bring intelligence inside bolts inherently lends itself perfectly to retrofit. Replacing the existing normal fastener with intelligent one is an extraordinary concept; it’s very simple; possibly revolutionary.”
Referring to potential applications, the managing director of the German supplier, Moritz von Soden, says: “We do it for lifting jacks for heavy applications.” He adds: ”The tension gauges measure the force applied to a 50t diesel locomotive. Instead of including an extra gauge on a lifting appliance, why not add it into the screw, and save a component?”
Now being fitted in novel applications, the sensors are providing data never seen before. For example, Fabbri mentions some exciting results from an oil and gas application that bear on pretension practices. Sens-In provided four instrumented bolts for a critical eight-bolt flanged pipework connection. Because the load distribution changes on the whole crown as you tighten one bolt, there are complicated procedures aimed at equal distribution of load on all bolts, to reach the target level of preload intended by design engineers. They involve tightening one, then going to the opposite side of the flange, and then turning at a 90° angle to select the next one to tighten.
Monitoring load data in real time, Sens-In found that the standard operating procedure for a manual bolt tightening sequence was producing pretensions in some cases 40-50% below the target value. Further manual tightening improved results but were unable to reach the target values. (Additional tests are planned).
Fabbri observes: “The normal procedure didn’t produce what was expected. This [technique] is something that I would say could be extended to all similar flanged couplings, for example in the helicopter industry. We are very confident that with this technique our customer will be able to be much more effective in the health of the coupling that they set up.”
COOPERATIVE APPROACH
Sens-In is also working closely with customers to develop the technology, and find the right applications. Like Bornemann, it is also considering a service model, in which it provides not bolts but predictive maintenance information.
Fabbri states. “We are targeting the most demanding applications, where a real added value can be created through the use of these systems. And the more technically demanding and complicated the application gets, the more you need cooperation between the two teams to make sure of success.”
Adds Von Soden: “We get ideas from customers. We also make versions for feedscrews for medical and food applications. There’s one from the kneading machinery of an Italian pasta brand. They worked out that because the feedscrew senses resistance, it can indicate when the dough is at target consistency, so better control the process.”
Both companies are currently in low-volume mode, partly because of the need for manual work in assembly of the bolts, which feature a tight fit between sensor and bolt.
Another reason why is because both companies were only set up three years ago.
Von Soden says that his business had the bad luck to come to market in early 2020, just before Covid. Since then, the company has made less than 100 of the instrumented fasteners, and there are 30-40 in the field for testing.
Sens-In was also set up in 2020. It currently employs four people, and is hiring four more. The project is hugely ambitious because of its technical range, Fabbri states. “It brings together different and quite diverse areas of engineering: mechanics, structural mechanics, sensors, electronics and also software. We are developing a cloud platform, a digital hub to host the large amounts of data the device is able to generate. That’s quite a complex effort.
“That’s why we have to be very careful to develop a well-targeted product, and be careful not to make a wrong choice in how the product is thought about and its features. We will have to be for the first 10 years in my estimate, in specially tailor-made system mode.”