Author Archives: Andi

Integrated Motors Simplify Motion Control

With machines getting more robust, smaller, less expensive and more reliable, engineers are facing the challenges of designing newer types of motion control. One way of addressing such motion control challenges, without being an expert in mechatronics is to use integrated motion control systems. Typically, these solutions combine the motor, the drive and the system components within a single unit. The system components include the intelligence or motion controller and input outputs all onboard. The use of an integrated solution allows the designer to focus more on the development of the machine and less on solving compatibility issues between various system components. The integrated motion system usually has all the components within a complete unit and sized for proper use. The decision to use an integrated motion system or an integrated motor usually depends on several factors. Major among them are requirements based on machine size, cost, reliability, modularity and distributed control.

With integrated motors, engineers can reduce the amount of space a machine needs. This is mainly the result of consolidation of components resulting in elimination of cabling. For example, an integrated motor may replace a drive and motor housed in separate enclosures, eliminating one of the enclosures. The panel space required reduces significantly for an integrated motor, while for a multi-axis system the real estate reduction can be substantial. However, an existing machine design must contain adequate space to house the integrated motor as this type of motor is larger than conventional motors.

Using integrated motors results in definite cost savings in contrast to using conventional components. One of the major saving in expenses comes from the absence of cabling that is no longer required with integrated motors. For example, the conventional drive may be located in a centralized cabinet with the motor a distance away on a long conveying machine. This arrangement needs considerable power cabling and feedback wiring between the motor and the drive. With the integrated motor, the drive being directly on the motor, much of the cabling is eliminated contributing to cost reduction.

With improvements in motor technology, the concern with reliability in integrated motors is outdated. The major point of concern earlier was heat buildup and dissipation. With reduced components making up the system, the reliability of integrated motors has improved because of the lower number of wire connections used. Better construction technology has improved the efficiency, decreasing the heat generated and the need for dissipation.

Industrial automation today requires modular machines. That essentially means smaller machines focusing on singular tasks combined to form a bigger system responsible for multiple functions. The smaller machines may operate independent of each other. This arrangement is beneficial because it allows engineers to change on modular section and transform the system into another customized machine. The modular concept is beneficial in shipping individual modules to the factory floor as the motor and drive of the integrated motor is placed directly in the machine.

As more and more industrial control is through PLC or Programmable Logic Controls, motor operations and synchronization through digital data signals is the norm. Since each integrated motor has its own controller, a distributed control system provides faster response and greater accuracies.

Connector Use Lowers Wiring Costs

Contrary to popular belief, hardwiring does not always minimize wire installation expenses. Hardwiring is a popular concept for those who regularly design and build industrial machines. People perceive it as one of the most common ways of saving installation costs when bringing power and signal to the machine. However, when the full range of wiring costs are factored in, these cost savings really seem just as a mirage does.

Installation costs typically involve time and materials, including the cost of the wire, cables, accessories and labor. However, if you look closely, there are less obvious hidden installation costs as well. These have individual considerations for labor and time-to-market.

For example, consider machines that need to be disassembled for shipping and then reassembled before startup. That means parts in the machine will have to be hardwired twice – once while testing and then again after shipping. Additionally, errors while wiring in the field are quite common, mostly when local electricians unfamiliar with the machine are handling the wiring. If you are lucky, such errors may only cause a delay in commissioning the machine. However, there can be worst-case scenarios, and faulty wiring may even damage the machine leading to expensive repairs. Along with such cost of errors, hardwired systems can be complex and expensive to test, so the cost of testing goes up as well.

As a rule of thumb, you can expect the hidden costs to go up exponentially with the number of connection points the machine has. Fortunately, use of connectors can help avoid all these hidden costs. Of course, connector components do add an upfront investment, but this money will be recouped and then some as connectors enable lower-cost machines, the machines can ship faster, they can be commissioned more quickly and offer ongoing savings.

Using connectors, engineers can build modular machines faster and with lower expenses. This approach to machine design allows engineers to pre-build common subsystems and components, and test and stock them for installation. Reusable modules lead to many machines being designed with common control panels, junction boxes, motor assemblies and populated cable tracks.

Connectors are a real advantage for shipping new and large machines, especially if these machines have to undergo some level of disassembly also. Disassembly usually involves unplugging cables from the bulkhead connectors of the panel, while connections and routing internal to the panel remain undisturbed. The process holds true for sensors and data cables, motor assemblies and junction boxes also.

At the destination, the machine requires all disconnected wires to be reconnected once again. A local electrician helped with a set of wiring schematics can simply perform this. Even if the electrician knows very little about the machine and the way it works, there is little chance of them making costly mistakes and adding to startup delays. Most modern connector systems are designed to disallow simple wiring errors. Where large, complex machines are to be installed and commissioned, connectors can reduce the time to a matter of days in place of the several weeks that hardwiring would have taken.

How to Avoid Cable Damage from Oil

Electrical cables are routinely exposed to several kinds of damaging chemicals in the environments they pass through. However, the most damaging of them all is chemical exposure to oil. Many industries and infrastructure settings use oil as a lubricant or as coolants. Such oils react with the polymers used in the cable insulation and jacketing to inflict molecular damage.

If this is ignored, oil can severely damage cables. This ultimately results in failure of the cable, system downtime and replacement expenses. With advanced production facilities such as in automotive assembly, requirements of better performance characteristics in renewable energy and regulatory changes, more people are now aware of oil damage to cables.

Fortunately, better cable manufacturing technology is now allowing cables to resist the effects of lubricating and cooling oils. However, it is necessary to know how oil degrades cables, how oil exposure problems can be diagnosed and how cables can be selected so that they resist oils over the long haul.

Insulation and jackets of cables are typically made of polymer compounds. Although they may have the same family name, not all these polymers show the same physical properties, including oil resistance. For example, some PVC compounds may show better oil resistance, while others have a higher degree of flame resistance. Manufacturers change the PVC formulation according to the properties and applications desired.

For example, addition of certain flame-retardants, stabilizers and filters allow PVC to exhibit enhanced characteristics of this type. However, improving or enhancing one characteristic usually comes at the cost of other performance traits being affected or being completely lost.

That explains why not all wire and cable insulations show equal performance with oil resistance in particular. The chemical, mechanical, environmental and electrical attributes vary depending on the individual compound formulations. To help promote resistance to fatigue and increased flexibility, most insulating compounds have a specific amount of plasticizers added to their individual formulations. When such compounds are exposed to processing oils for coolant or lubrication, the plasticizer diffuses from the compound or the material absorbs the oil.

With the plasticizer diffusing out of the compound, the oil causes insulation hardening, resulting in loss of flexibility and elongation properties. If oil is absorbed, the insulation swells and softens resulting in degradation of tensile properties.

In short, oil causes the insulating compound to lose its primary role virtually as an effective insulator. This creates a hazardous situation not only to the functioning of the industrial machinery to which it is connected, but possibly also to human life. Ultimately, this can result in expensive downtimes, expensive repairs and in the worst cases, replacement of the entire machinery.

Testing can help determine how a cable will react in environments containing industrial oil. UL has standardized these tests and they are commonly known as Oil Res I and Oil Res II tests. In these tests, cable samples are continuously immersed in IRM 902 Oil at elevated temperatures for specified periods. The mechanical properties of the cable samples are observed for physical damage caused by the exposure to oil. The latest UL standard for these tests is AWM Style 21098.

The Human Brain Project: Is the Electronic Brain Coming?

The human brain has always been a thing of extreme curiosity to the students of anatomy. In fact, Einstein’s brain was preserved for future study immediately after his death. Innumerable studies have been done on this part of the human anatomy, yet, we know very little about how its complete range of functions. In the quest to know more about the human brain, resources are being put together for a simulation to study how the brain functions. The HBP or Human Brain Project of the European Union has a primary directive – an artificial brain by 2023. They recently held their annual HBP Summit in Germany, at the University of Heidelberg.

The European Commission Future and Emerging Technologies fund one of its flagship programs, the Human Brain Project. The 10-year-long project has a funding of nearly US$1.3 billion. Initially, HBP aims to simulate the entire human brain functionality on supercomputers, and then replicate the functionality on a special hardware emulator. They expect to be able to reproduce the functions of the brain so accurately as to allow trying out diseases and their cures on the emulator. The long-term objective of the project is to build an artificial brain inexpensive enough to outperform traditional supercomputers of the von Neumann type at a fraction of the cost.

At the end of the first year, all pieces have been assembled. According to the report, all personnel are hired, laboratories engaged throughout the region, and the ICT or Information and Communication Technology set up in place. This arrangement will allow the researchers with their 100+ corporate and academic partners in 20+ countries to collaborate effectively to share data. The projects already running include reconstructing the functioning of the brain at different biological scales along with development of computing systems to mimic the functioning of the brain.

According to the agenda for the ramp-up phase or the first two and a half years, HBP will gather as much strategic data about brain functioning as is known. The project will also develop theoretical frameworks to fit that data. They will also develop the infrastructure necessary for six ICT platforms during the next operational phase to start from 2017.

Supercomputers or high-performance computing will serve all platform builders for the six ICT platforms. These will consist of: the Neurorobotics platform for supporting testing of the brain models and simulations in virtual environments; the Neuromorphic computing platform for mimicking the various functions of the brain; the Medical Informatics platform for cataloging the diseases of the brain; the Brain Simulation platform to assemble the simulation algorithms of different brain components; and the main Neuroinformatics data repository for housing the Brain Atlas.

The first year of the project has some progress highlights. These include: a brain simulation technique for the cerebellum, repurposed from the one originally working successfully for the neocortex; a virtual room for the neurorobotics prototype, where researchers can study virtual bodies with brain models for behavior and cognitive abilities; an HPC or high performance computer successfully retrofitted for interactive-supercomputing – essential for testing brain models; and demonstrations of several new neuromorphic chips and testing them to solve modern computing challenges that only humans can perform today.

Pi Lite: Bright White LED Display with the Raspberry Pi

If you did not know, you can run many LEDs with the tiny, credit card sized single board computer popular as the RBPi or Raspberry Pi. Among the many accessories made for the RBPi using LEDs, Ciseco makes one that is very interesting and useful. This is a display panel using bright white LEDs and aptly named the Pi Lite. You can use the series of white LEDs on the Pi Lite as a scrolling marquee for a Twitter feed, for displaying real-time weather information or stock quotes. You can use it to display static information such as time or functional information such as bar graphs, or other dashboard type applications such as VU meters. On the other hand, you could even play such games as Pong. Pi Lite is strong enough to view in direct sunlight.

Pi Lite is completely self-contained and does not require any soldering. You can get Pi Lite in two colors – white and red. For operation, simply connect Pi Lite to the GPIO pins of the RBPi, and you are set. GitHub has several open-source projects that you can download or you could do your own programming using Python code.

If you are just starting out with the RBPi, Pi Lite is an exciting way to let RBPi do some physical work and generate some fun. The large LED matrix display is easy to plug in and add-on. Since no soldering or any other special skills are needed, anyone can simply start using the Pi Lite for their project.

All the 126 LEDs on the Pi Lite are in the form of a 14×9 matrix, with an ATMega328p processor controlling them. This mixes the highly popular LOL or Lots of LEDs shield of Arduino with the world of RBPi. The Pi Lite communicates with the RBPi via the standard serial communication protocol at 9600bps. That makes it a simple affair to send graphics and text to the LED matrix. With the ATMega processor driving the 126 LEDs, the RBPi processor and its GPIOs remain free for other functions.

The Pi Lite offers several advantages. You can read your emails or tweets from a distance in real time. The firmware being open-source, you can add extra functions as you like. You can achieve multiple functions by sending simple text strings – scroll the text, VU meter, bar graph and or graphics. You can use the well tried, tested and supported LOL shield by Jimmy Rogers. The serial interface makes Pi Lite useful for connecting to any TTL micro radio or PC interface – you can use the popular FTDI cable.

The Pi Lite uses a high quality gold plated PCB. No extra power supply is required, as Pi Lite draws only 49mA maximum at 5VDC, so the RBPi supply can power it. With preloaded software, you can use it out of the box and display variable speed scrolling text, 14 vertical bars as a bar graph, two horizontal bars as VU meter, frame buffer for animation and graphics, or turn on or off individual pixels.

To make a bigger display, you can link up additional Pi Lites with the I2C bus. Each Pi Lite measures 85x55x13.7mm.

Power Supply Ignition and other Switches for the Raspberry Pi

There are several occasions where you may require operating your RBPi or Raspberry Pi powered from a vehicle’s electrical system. To keep your single board computer safe and operational, an accessory is needed to sense when the ignition on the vehicle is engaged and when it is turned off. Accordingly, the accessory will respond by powering the RBPi on or off safely. MausBerry Circuits make such safe power supply ignition switches and other shutdown switches for the RBPi to be used in vehicles.

The power supply ignition switch attachment from MausBerry features a built-in step-down converter that produces 5V from the 12 or 14V of the vehicle’s power supply. Once connected with wires behind the vehicle’s radio, the attachment provides the RBPi with instructions based on the vehicle’s ignition status. It communicates with the RBPi using two of its GPIO pins.

An added advantage of the ignition switch attachment is it can retain power for about 20 minutes during its power-down cycle. That means the RBPi will remain powered for 20 minutes after the vehicle’s ignition is switched off, so waiting for the RBPi to boot is not required for those making frequent stops. A selector switch on the device will allow you to reboot the RBPi, if required. Even if the RBPi was left in the vehicle and not shut down, there is no cause to worry. The automatic shutdown feature of the device will kick-in to shut the RBPi down after four hours of non-use, thereby preventing drain on the batteries.

MausBerry makes many other similarly useful attachments for an RBPi to be used with vehicles. One of them is the 3A car supply that can sense the car ignition to shut down the RBPi safely when the car is turned off. The unit has two USB ports and communicates with the RBPi using two GPIO wires. The unit is to be wired to the vehicle’s battery and the 12V ignition source. Ground and power wires, both 18AWG and 18-inches long, are included.

If you are looking for an on-off switch for your RBPi, MausBerry has an illuminated LED type switch. Plug this unit into the RBPi power port and it accepts your existing micro-USB power cable. To turn the RBPi on, simply press the button. To switch off, press the button again briefly – the operating system senses the button and safely shuts itself down. After a safe shutdown, the switch will cut off all power to the RBPi. When illuminated, the LED gives off a bright blue light, and holding the button for five seconds performs a hard-reset for the RBPi.

Although aligned to the layout of the RBPi models A and B primarily, the illuminated LED shutdown switch will work directly with all models A, A+. B, B+, RBPi2 of the RBPi series. For the B+ models, the new power port location may make the switch stick a little out of the side.

Another shutdown circuit from MausBerry allows you to use any custom switch for operating the RBPi. The circuit plugs into RBPi power port and accepts the micro-USB power cable. This circuit is useful when installing the RBPi into a case, as the switch can be installed separately.

How Opto-Couplers Help with Intrinsic Safety

Electrical equipment and wiring are used in different environments, including hazardous locations, where there is always a risk of explosion due to any malfunction in the wiring or equipment. To mitigate this risk, electrical and thermal energy generated must be limited to a level below that required to ignite a specific mixture of the hazardous atmosphere. This technique of designing electrical equipment and wiring to be safe under normal or abnormal conditions is called intrinsic safety. Therefore, intrinsically safe wiring and equipment are incapable of releasing adequate thermal or electrical energy under any operating condition to cause a combustible or flammable atmospheric mixture to ignite.

Independent third party agencies such as the UL or Underwriters Laboratories, CSA or Canadian Standards Association, FM or Factory Mutual Research Corporation and the MSHA or Mine Safety and Health Administration, test and certify equipment for intrinsic safety. For use in explosive atmosphere, the agencies test and verify equipment for compliance to IEC international standards. Within the IEC 60079 series, the standard IEC60079-11 specifies the construction and testing of intrinsically safe apparatus intended for use in an explosive environment.

According to IEC60079-11, the basic principle in achieving intrinsic safety is for limiting the energy in the power circuit, preventing unusually high electric arcs, ignition sparks or high temperatures that could create ignition energy required to cause an explosion. For limiting the power or energy, designers should implement a resistor or fuse in series for limiting the current and a Zener diode in parallel for limiting the voltage.

Additionally, IEC60079-11 also requires that conductive parts of intrinsically safe circuits be separated from the conductive parts of non-intrinsically safe circuits. The separating distances have different requirements through insulation structures, and this includes clearance, separation and creepage distances. For example, casting compounds specified includes epoxy resins, while solid insulation specified include silicone and polyester film.

Apart from providing galvanic isolation, opto-couplers have internal clearances, which include DTI or distance through insulation. This is a part of the insulation and safety related specification of opto-couplers. DTI provides galvanic isolation through optical technology, forming a straight-line thickness distance between the LED emitter and the detector within the opto-coupler. The DTI of the opto-coupler meets the separation distance requirements 0 to 2 of the gas zone classification. This depends on the voltage level of protection required.

Typically, isolators with structural DTI less than 20µm cannot achieve the stringent separation distance requirements of intrinsic safety criteria. Special opto-couplers, such as the ACNV series from Avago, have a 13mm creepage/clearance, with insulation material classified as casting compound. This allows the ACNV opto-couplers to achieve up to the 375V level of protection. Similarly, ACNW/HCNW opto-couplers from Avago, with 10mm and 8mm creepage/clearance, can meet up to 60V level of protection.

Such intrinsically safe opto-couplers are routinely used in applications for measurement of level, pressure and temperature in flow meters and transmitters. Meeting safety requirements, these opto-couplers provide the reinforced insulation required between field sensors and micro-controllers on control boards. Typical examples of such applications are in the explosive atmospheres of petrol stations and sewage, where fluid pumps and flow meters are used.

Computers Can Beat Humans in Image Recognition

Every day, computers are getting smarter. So far, it is not clear whether the smartness is moving towards something as depicted in the Terminator movies, but computers are beating humans in chess, poker and Jeopardy. The next hurdle that computers have crossed is image recognition. Microsoft claims to have programmed a computer that can beat humans at recognizing images.

Although the final competition is going to be held on December 17, 2015, already there are claims that computers are better than humans are in visual recognition. The ImageNet Large Scale Visual Recognition Challenge will do judging for the final competition. The first claim about computers beating humans came from Microsoft. They claimed that while humans made 5.1% errors in recognizing images, computers failed only in 4.94% cases. After 5 days of Microsoft announcing their feat, Google announced that they have bettered the Microsoft claim by 0.04%. That means the competition is getting fiercer every day.

Since 2010, more than 50 institutions take part every year in the competition for image recognition. ImageNet runs this competition and they have hundreds of object categories and several millions of example images. So far, humans have scored the most, but this year a computer is expected to take the crown. Typically, contestants use the latest deep learning algorithms. Derived from different types of artificial neural networks, these deep learning algorithms mimic the way the human brain works to a varying degree.

Although no contestant actually offers their exact code, they provide papers that freely describe their algorithm in great detail – similar to the spirit of open source – explaining the advantages of their algorithm and why it is expected to work so well. As Microsoft explains in their paper, they are using deep CNNs or convolutional neural networks that have 30 weight layers. Google have revealed that they are using batch normalization techniques, and these do not allow neurons to saturate during initialization.

Usually, the conventional way of using neural units involves hand designing them and fixing while training. However, Microsoft has deviated from this path and made the neural units smarter. They have done this by making their form more flexible in nature. According to the principal researcher at the Visual Computing Group of Microsoft Research, Asia, each neural unit undergoes a particular form of end-to-end training that imparts the learning. The introduction of smarter units improves the model considerably.

However, the reason for the ability of current neural networks being able to beat human experts lies in the algorithm of Microsoft’s Deep Learning. This algorithm usually initializes and trains on 1.2 million training images and verifies its learning on 50-thousand validation images. For the final application of its learning, Deep Learning uses 100-thousand test images from the main image database. However, Microsoft did not actually follow this standard route.

As training of very deep neural networks is rather difficult, Microsoft used a robust initialization method. As with other contestants, Microsoft did buy Nvidia’s access to their arrays of graphic processing units. However, they also bought and configured their own supercomputer. They simulated parametric rectified linear neural units and that helped them finally to beat the human experts for image classification.

Graspinghand’s SweetBox, ScorPi and Heatsinks for the Raspberry Pi

Those who need a casing for their Raspberry Pi or RBPi are rather spoiled for choice. There are so many types of casings available, and that makes it so difficult to settle on one. Sometimes, you need a casing that does not take up too much space, but is able to protect your RBPi from sundry damage. If you want the smallest case on the market, try the SweetBox from Graspinghand.

Besides being the smallest on the market, SweetBox is injection molded with high-performance nylon, and is compatible with RBPi models B, Rev 1 & 2. It has several features such as it allows the insertion of a Micro-SD card into its adapter and the mounting of the RBPi camera. A rubber cap protects the GPIO pins when not in use, and is easily removable to allow connections.

Slots on the casing allow easy access to the DSI or Digital Serial Interface for attaching an LCD panel to the RBPi and the CSI or Camera Serial Interface for attaching a camera. Other mounting holes are available on the base, while the entire casing allows simple opening and closing without any screws or tools.

SweetBox is made from high-performance nylon, the EMS Grilamid type typically used for glass frames, electrical equipment and tools. This material makes the casing nearly unbreakable. The material is also lightweight, and the casing is only 35gms with dimensions of 95x65x25mm.

However, one of the most remarkable features of the SweetBox is it allows heatsinks to be mounted, so that your RBPi can operate within the casing, but without getting all heated up. Graspinghand offers three CNC machined heatsinks that you could use with or without SweetBox. The three heatsinks come with ready-to-mount thermal pads. With the heatsinks fitted, your RBPi will run at least 4°C cooler at full power.

Placing the heatsinks requires some dexterity. First, you must peel off the protective film off one side of a thermal pad. Then fix the heat sink very carefully in the center of the uncovered surface – this will stick the thermal pad to the heatsink. If there is excess thermal pad protruding out around the heatsink, use scissors to cut it off. Now peel off the remaining protecting film from the other side of the pad and place the heat sink and pad combination very carefully on top of the IC to be cooled. Use the same procedure for mounting all the three heatsinks, taking care to keep the same orientation of the fins for all the three.

Graspinghand also offers ScorPi, a flexible gooseneck arrangement for holding things such as the camera board on the RBPi. A brass fixture allows the ScorPi to be attached to SweetBox, while the brass fixture on the other end of ScorPi attaches to the camera board. You can flex the ScorPi to position the camera at any angle required, and it will remain in position to allow capturing images without any blurring due to shaking.

Cleaning the ScorPi is also very easy, as you can loosen all parts and clean them with a soft wipe using a mixture of white vinegar and salt.

Adding a Reset Switch to your Raspberry Pi

Normally, shutting down the tiny credit card sized single board computer, the RBPi or Raspberry Pi, involves pulling the plug. That means disconnecting the power cable from the RBPi board. However, that is a risky way of shutting down the SBC, since it may be in the process of transferring data to the SD card, and the power interruptions may cause corruption of the memory card. Another problem with frequent removal and re-insertion of the power cable is the damage this may cause the connector port. Program development on the RBPi may cause it to hang occasionally. Therefore, frequent restarting via power cycling with removal/re-insertion of power cable will be a problem. A simple fix is to add a simple reset function to the RBPi. You can do this in one of three ways. The first is to use a USB reset button. The second is to use a motherboard jumper on the GPIO bus. The third option is useful only for RBPi Models B Rev2 and B+, where you solder pins on the P6 header and connect to a momentary button. The third option is the most complicated, requiring soldering on the RBPi.

Although the first option of a USB reset button is the simplest, it also ties up one of the USB ports on the RBPi. With only one or two USB ports available, depending on the RBPi model, this may not be a very viable option for many. However, in case it works for you, get a USB reset button from any specialist online stores. Those who want all their GPIO pins available or those who are averse to soldering may use the USB reset button connected to the RBPi for scenarios when the device needs to be booted.
If you can salvage a jumper from an old motherboard or an HDD, connect it on two pins on the RBPi GPIO. All RBPi models have GPIO pins – models A & B have 26 pins each, while the models A+ & B+ each come with 40 pins. You need to place the jumper on the GPIO3, pins 5 and 6, counting from the left while holding the board the right way around.

However, you will need a script to detect the jumper. Make the script executable before running – use ‘sudo chmod 755’ for this. You will also need to run the script every time you boot up. For this, add the following line to /etc/crontab –

@reboot root /home/user/scripts/gpio_actions.sh

Whenever you place the jumper on the specified pins of the GPIO, RBPi will sense it and will shut itself down.
The third option involves using the P6 header, which is available only on the latest models of the RBPi – models B Rev 2 & B+. On the Model B Rev 2, you can locate P6 next to the HDMI port. On the model B+, you will find P6 next to the label marked as ‘Raspberry Pi 2014’. Normally, the RBPi does not come with pins soldered on to P6, so you will have to do the soldering.

Once you have soldered the pins, install the jumper with the switch to reset the RBPi. However, use this switch with caution, only when the RBPi is not responding.