We recently had a discussion with Sri Peruvemba, VP of Strategy at BeBop Sensors (Berkeley, CA), about the technology and applications for their two-dimensional fabric sensor material. I started by asking Sri about the unusual name for the company.
Sri Peruvemba: Music has always been a passion for our CEO, Keith McMillen. So he named his latest company as a tribute to the avant-garde bebop (or “bop”) jazz style that was developed in the 1940s and continues its influence to this day. He founded a series of companies, starting with Zeta Music in 1979, all related to applying electronics to the music industry. BeBop Sensors was spun out of Keith McMillen Instruments (KMI) in 2014. Keith had developed the BopPad, a unique electronic drum pad that uses patented smart fabric sensor technology to provide drummers with a wide range of musical effects depending on where and how hard they hit the surface. He created BeBop Sensors to expand the range of applications for this unique technology.
Tech Briefs: Could you tell me about the sensor fabric.
Peruvemba: It uses a nanomaterial, which is a two-dimensional conductive polymer that can be coated onto pretty much any fabric that is made up of threads. The polymer is a piezoresistive material, so its resistance varies with applied pressure.
Tech Briefs: How do you apply the polymer?
Peruvemba: We can do it in multiple ways, depending on the type of sensor. For most applications we have a roll process that can yield a square of about 2000 meters. Experimental materials can be produced as small as a 12” square.
Tech Briefs: Can you adjust the amount of resistance per sensor layer?
Peruvemba: Yes, by how much more or less we coat the material. The fabric comes out of a fixture that can apply up to hundreds of layers so we can adjust the amount of nanoparticles we apply.
Tech Briefs: Can you explain how the coated fabric works as a sensor.
Peruvemba: When you push against the material, you’re changing the resistance. The resistance is a function of the pressure applied to the fabric — the greater the pressure, the lower the resistance. Typical deviation is 25% to 75%. The resistance range depends on the product design. In a recent design, for example, the deviation was from 10K ohms to 2500 ohms. More typically, however, the resistance is much lower than that. To measure the resistance, we apply a current, which is usually in the range of a few hundred microamps, and measure the change in voltage, which usually varies a few volts in a 3.3V system.
Tech Briefs: What kind of accuracy do you achieve?
Peruvemba: Our long-term accuracy is +/- 3%. We’ve even had a customer who tested our product by putting 32 kg of pressure on our sensor on a 25 – 30 mm area up to 600,000 times. Although the mean variation in accuracy over that period was +/- 5%, the product recovered, and the accuracy returned to +/- 3%.
Tech Briefs: How do you connect your electronics to the fabric?
Peruvemba: We have four methods to inject current and measure voltage from the fabric. One method which is very flexible and environmentally stable is to print conductors on both sides of a polyurethane (PU) film and melt that onto the fabric. We also use polyethylene terephthalate (PET) in some applications. The printed ink connects our sensor to the drive electronics, which is neither flexible nor very thin. It’s miniaturized, but chips are rigid.
Tech Briefs: Could you describe some of the applications for this sensor material.
Peruvemba: One application is hospital beds, to measure where pressure is being applied, and how much. Knowing the location and amplitude of pressure points could help determine if there is a potential for bedsores. Similarly, for wheelchairs — when someone is sitting in a wheelchair for a very long time, the patient can become uncomfortable or even develop “bedsores.” The bed or wheelchair designs can be modified to mitigate the problem. One customer is even using them in infant car seats for automobiles.
Tech Briefs: You’ve explained how your sensor can be used to measure force, but how can it pinpoint location?
Peruvemba: The entire sensor is a grid, so we know which part of the sensor you are touching very, very, accurately.
Tech Briefs: Is it like pixels in a video image?
Peruvemba: Yes, exactly like pixels. Because we use nano material, we can sense down to a very high resolution. We have a multiplexed grid with conductive inks that connect from each set of fibers on our sensor to the connector that leads up to the drive electronics/processor. This accurately delivers the x-y coordinates of every location on the sensor.
Tech Briefs: What are some other applications?
Peruvemba: We’ve used them for football helmets. When you take a hit, it can sense exactly where on your head that the hit occurred. Typically, between the upper surface of the helmet and your head, you can’t tell exactly where the hit happened because of the gap. With our technology, the sensor actually touches the head. It’s not on the outer surface of the helmet. The doctor will be able to know exactly where the hit came and also the extent of its impact. The helmet manufacturer can see that the hits always come at certain locations on the helmet. As a result, you can design better helmets. There is now a company in Europe that’s also putting them in shoulder pads.
One of our other products is a set of sensors installed in a glove, which can be worn, for example, by a factory worker. Imagine you’re building a machine. You need to apply a certain pressure to a panel. If you don’t apply enough pressure, you will have a quality defect. If you apply too much pressure or apply it to the wrong part of the equipment, then the worker might be injured or have long-term problems with his or her hands. Not only can these gloves be used for injury protection, but also to increase productivity and improve quality control.
One of the unique features of our sensor is that it can also measure twist and bend, so it can convey dynamic information by measuring these deformations as a function of time down to milliseconds. We can actually measure four different things: force, bend, twist, and stretch. So that, for example, since in our glove, the sensors are in the fingertips, along the fingers and all over the palm, we can measure complex gestures.
Still another application is to replace the insole of a shoe with our sensor, which gives you a very accurate measurement of your foot. From that data, you can determine the location of pressure points in order to know what kind of shoes to wear. It can also be used for medical treatment, and it provides valuable information for athletes.
The virtual reality industry has recently been adapting our technology for their headsets. If you look at the virtual reality environment, you’ve got very good visuals because of the high resolution, and also good audio when you use a high-quality speaker, but you have no touch feedback. Let’s say you’re shooting a gun in a game. You can see the gun and your hand and you can hear the sound. But by adding haptic actuators to our glove, you can feel the trigger and the bullets leaving the gun. One of the challenges for this application is to match the speed of the sensor with the speed of the visual, which we were able to accomplish.
Tech Briefs: Could you tell me more about your haptic actuators.
Peruvemba: In the VR shooting application, for example, haptic actuators are in the fingertips. The sensors will tell you where the finger is and when you move a finger, it will show on the visuals. When the haptic actuators engage, they can put out 32 simultaneous unique feelings. That will allow you to literally feel the trigger.
There’s also a company that’s putting them in steering wheels for driver assist systems. When I put my car in diver assist, I’m still expected to hold the wheel, otherwise the car should stop. When I cross out of my lane, the steering wheel vibrates. This one will do it much more accurately. It will give very good feedback.
Tech Briefs: How are the different haptic feelings generated?
Peruvemba: We feed the haptic actuators waveforms that are basically audio files. Take for example, a VR environment where someone slams a door. We record an actual door slamming and feed that audio file to the actuator. There is a menu of haptic files that go to the fingertips depending on the user’s journey.
Tech Briefs: Are there any new applications on the horizon?
Peruvemba: We’re working with potential customers on some new ideas. Imagine you have a little pad of material on your sleeve. Since the sensor is a fabric, it can go around your sleeve like a shirt. You can write on that with your fingers. That information can wirelessly communicate with another person who is wearing a similar sleeve. We’ve actually built this as a prototype.
Another application is a sort of Bluetooth headset. It sits in your ear and there is a small area on which you can move your finger around and navigate a screen. We built our sensor and also the software to demonstrate that you can do these kinds of things.
We’re constantly seeking new applications that would be a good fit for our fabric, which can uniquely sense both deviation and location.
This article was written by Ed Brown, Editor of Sensor Technology. For more information, visit here .