In the highly competitive world of system design, there are times when you face a challenge that seems impossible to tackle with conventional wisdom. What if you could use your humble automotive solution to tackle a military design challenge? What if a Raspberry Pi could do it too?
P. Chow Reddy, manager-R&D (Power Division), ICOMM Tele Limited, spoke to Abhishek Mutha of EFY on designing solutions deduced from existing solutions in other disciplines
Q: What are the most exciting trends in R&D?
A. There is immense requirement of USB, Internet connectivity and RS232 in power distrib
ution boards (PDBs) and automatic changeover against mains failure (AMF) panels for their access to the media and system connectivity. Apart from this, I have identified the utility of ruggedised, dedicated computing systems for military applications. For all military applications, one single solution would be Raspberry Pi.
Q: What makes Raspberry Pi work for such serious applications as in military and defence?
A. We have been working on Raspberry Pi. It consumes less power, and is virus-free (since it is Linux-operated) and rugged (as it uses solid-state memory in the form of a SD card compared to a hard disk). It is small and minimally configured, and can fit into a small cabinet or on the back of a monitor. Its HDMI is faithful enough in giving a marvellous picture quality. Above all, it is very inexpensive and customers accept it.
Q: What kinds of Raspberry Pi projects could help design engineers?
A. We have raised proposals of the projects based on Raspberry Pi. In parallel, we have also developed a small application that involves data storage memory, display, RS232 and input/output (I/O) port handling. All this is operated on Python and has been found to be operating up to the mark. One thing that an engineer has to keep in mind is that it lacks real-time clock (RTC).
Personally, I am very keen on using Raspberry Pi for military applications. There are numerous other applications and fields where it can be utilised. It has a big future ahead and if its cost reduces further to the declared figure of $25.00, one can think of replacing microcontrollers with Raspberry Pi.
Q: Could you elaborate on incorporating USB, Internet connectivity or even RS232 in PDB and AMF panels by utilising a Raspberry Pi?
A. AMF is used for changeover from mains to a diesel generator (DG) set and vice versa depending on the conditions. In these panels, USB is required to store the logged data like power fluctuations, DG run time and fuel filling updates to estimate the consumption at all load levels. All this is saved into a pen drive.
When it comes to RS232, I personally would require it for communication between other panels, LCD instrumentation displays and annunciation through GSM/GPS. Internet connectivity is used to take the stored data for a view on the performance of the panel, loads and power sources. All this exists as hardware in Raspberry Pi. Additionally, all the logical decisions, and activation of relays, contactors and solenoids are possible through software.
Q: Could you tell us about a recent cost-effective solution developed by your team, deviating from existing approaches?
A. Mil-grade air-conditioners are designed and manufactured in monopoly by very few companies. These are costly and use hermetically sealed compressors that are designed and integrated into the application cabinet to withstand vibration, dust and harsh climatic conditions.
I have come up with a cost-effective solution based on a pre-existing automotive solution. It comprises an existing automotive air-conditioner compressor and an AC induction motor for driving it. The motor can be controlled to operate for a select temperature required in the cabin. These compressors can sustain dust, high temperatures and vibration. Leakage of R-134A gas from the compressor is also almost impossible for applications in shelters. This serves the purpose in shelters as well, as it is a cost-effective solution.
Q: What were the biggest design challenges faced while re-engineering the system for a military application?
A. In military applications, the compressor must withstand the following conditions:
High temperatures. According to MIL-STD 461E, the compressor should withstand up to 55°C. This temperature range is already available in a vehicle’s engine with temperature running up to 100 degrees centigrade.
Dust test. The compressor is already proven to dust of all terrain conditions where the vehicles run around and still perform without any hindrance. This explains why I use it in my solution.
Leakage of R-134A gas. Freon gas is the standard used for refrigeration. Its grade is numbered and legally authorised to use in refrigeration. Earlier, it was R-12A for applications. Now it is mandatory to use R-134A. It is prone to leakage from capillaries, so care is taken in automotive systems to sustain all the vibrations and harsh climatic conditions. This has performed well all these years in our vehicles, which proves it is rugged and can sustain the required conditions in shelter applications.
Variation of RPM. In vehicles, RPM of the engine varies depending on the run of the vehicle and has a governing arrangement to sustain the stress. This is a boon for my application where I am running the compressor on an electric motor. Moreover, I am running the compressor only in standstill conditions. It is carried in a sheltered vehicle in turned off condition.
Q: What kinds of design challenges did you face while addressing the power needs of shelters with this solution?
A. One condition for design and integration of AMF panels for DG sets used at high altitudes is that the DG set should be able to operate, sustain and survive at low temperatures of high altitudes. The main problem here is the availability of oxygen and very cold and dusty climate.
Specific to the usage of the DG set, I design monitoring circuitry dedicated to look after the operating temperature conditions of the engine. Starting the engine when the coolant and lubricating oil are in ‘freezed’ condition will damage the rings, piston and cylinder block. We have also added a monitoring circuitry that melts the liquids before the engine is started.
Q: You must be incorporating microcontrollers to make such critical decisions?
A. Microcontroller is a must in such applications. Specifically talking about this critical decision, it is a simple thing that when starting the engine, first a microcontroller senses the temperature of the engine at its lubricating oil, coolant and fuel tank. The microcontroller cranks the engine only if the liquids are at a temperature where they will be in liquid state. Otherwise, it first turns on the heater plugs at all or required points. Once they have reached the set temperatures, the engine is cranked.
Q: What MCUs do you recommend for these applications?
A. Usually, I select a single microcontroller with minimum configuration or architecture required for the application. Such applications can be catered with an AT89C52. Now I am trying to apply it with Raspberry Pi where I can play big logics, conditions and much smarter solutions.
Q: How is the distribution of power managed? Also, do we see Raspberry Pi anywhere in the power electronics picture?
A. Use of Raspberry Pi in our projects is in our future scope and still in proposal stage. But we have started working on the applications as I have mentioned earlier. For instance, it could be a dedicated ruggedised computer module mountable in a 48cm (19-inch) rack. It goes with Linux and is very convenient to use in connectivity through Ethernet and solidstate data storage in real time with some external hardware. Also, it is a low-cost solution for my applications.
In power distribution and switch gear handling, Raspberry Pi has a big future in our organisation. We are mostly working on Python programming for our applications. Python is much simpler and allows fast development. Our engineers have done some creative work on it and are confident about it. Currently, we are focusing on the applications of power distribution from a PDB to be used in shelter.
Q: Once the concept to product development stage is accomplished, what approach do you recommend for selection of components and modules in the design?
A. Every design engineer has his own hand and way of getting a solution for the requirement. Likewise, I go with my own approach. Component and module consideration starts at the concept stage itself. Depending on their availability, time to procure and the cost, which is the most important thing, I select components at the concept stage itself.
While considering modules, I look into two things. One is the performance in various applications and climatic conditions. Second is the module’s time of existence. Sometimes a module could be a solution in some other application while also being suitable for my concept. Sometimes a module cannot be related to electronics but it still could be suitable in electronic applications.