Group SD0716
From ECE Department Wiki
Contents |
[edit] Energy Harvesting
Group Members: James Pachan, Manish Patel, Jeremy Voll
Advisors: Dr. Nelson, Dr. Kavasseri
[edit] Project Description
Our aim is to determine whether enough energy can be harvested from an overhead distribution line to power up a transmitter or other signaling devices like a LED array.
[edit] Requirements Capture
• The apparatus must detect the energization of the High Voltage line by detecting the presence of the E-field which is induced around the wire’s space when it is active.
• The apparatus must work even when the line loading is very low.
• The apparatus must work irrespective of the distribution model being used, nature of the conductors and the transmission pole’s dimensions and locations.
• The apparatus must withstand the diverse characteristic of the weather conditions. It must be able to work in highly windy, rainy and snowy conditions. These climatic conditions must not affect the readings from the sensor.
• The apparatus must be able to send all the information sensed by the sensor to the data receiving location at high accuracy and very efficiently. The distance of the data transmission must be large enough to transmit to ground level.
• The apparatus would be consisting of various components like, (I) Sensor (a metal plate/wire) (II) Mosfet (Switch which would detect as to when the line is active) (III) Diodes (For Mosfet protection) (IV) Transmitters/Receivers (For sending/receiving data) (V) Misc. components like LEDs, resistors, potentiometers etc.
These components must be very efficient and consume very small amount of power, so that our device is good enough to work with the minimum power available to us.
• The apparatus may be powered by a power supply most likely a solar panel. The solar panel must fulfill all the energy needs of the sensing apparatus and transmitter while remaining highly efficient.
• The device for the moment may work on an external power source but our ultimate aim is to find out a way by which we can harvest enough amount of energy to drive the sensing apparatus.
[edit] Options Considered
[edit] Research
Our first step was to come up with a method to sense whether the line is energized or not and how we could harvest energy from it. We researched the working of many patented devices for this, and we had two main ideas in mind: one was to detect the presence of E-field and other was to detect the Electro-Magnetic field. Detecting these fields and using them to generate power was the main focus. Since the Electro-Magnetic field depends on the current i.e. the line loading, this varies a lot and is almost negligible during nights. Therefore we decided to sense the E-field which was a very good and reliable sensing agent.
Now, we knew that we had to build a device that detects the E-field and also stores energy form the field. But we could not just build up a circuitry, carry it out and experiment with it around transmission lines, since it is very hazardous to do so. Therefore, we decided to work out some calculations that would give us some close figures which would tell us how much we could actually harvest.
If a Wire shaped Sensor is used,
Capacitance Between 2 Wires
C = { (pi)*ε)/(cosh^(-1)*(h/b)) }
Where, h = distance from the Conductor
and b = Radius of the wire
If a Plate shaped Sensor is used,
Capacitance Between Wire and Plate
C = { 2(pi)*ε) / cosh^(-1)*(h/b) }
MNE(Cap) = { (RL * CGR) / (RS + RL) } * { (RNE * RFE) / (RNE + RFE) }
MFE(Cap) = MNE(Cap)
The Following are the calculations for the Impedance, but we might not use them, since we are interested in the voltage part i.e. capacitive part only.
MNE(Ind) = { LGR / (RS + RL) }* { RNE / (RNE + RFE) }
MFE(Ind) = { LGR / (RS + RL) }* { -RFE / (RNE + RFE) }
MNE(CI) = { RO / (RS + RL) }* { RNE / (RNE + RFE) }
MFE(CI) = { RO / (RS + RL) }* { -RFE / (RNE + RFE) }
Here, NE = near end, FE = far end.
Also,
VNE / VS = jw[ MNE(Ind) + MNE(Cap) ] + MNE(CI)
VFE / VS = jw[ MFE(Ind) + MFE(Cap) ] + MFE(CI)
And,
pi = 3.14159
ε = Permittivity of Air = 8.85E-12
h = Dist from Conductor (km) = 0.001 (1m)
b = Radius of Wire (km) = 0.000006 (6mm)
w = 2*pi*f = 376.9911184
Other values,
RNE = 1.00E+06 RFE = 1.00E+25
RL = 400 + 1000i RS = 400 + 1000i RO = 0
[edit] Rationale
The result from the calculation was very low and below what was required to run the signaling devices.
The time constraint and the requirement of a working prototype by the end of this semester made us divide our work on two parallel paths. We decided to be prepared for the worst case scenario, and at least develop a sensing circuit that senses the energization of the transmission line using a power source. We could then implement energy harvesting ideas as we develop them.
1 Given a power source like solar panel or battery, how could we achieve our requirements. And also make the device 100% efficient and reliable just like the original device
2 Try researching as to how we can harvest enough from the transmission line, so that we can just replace the power source of the external powered device with this
In the past few days we came up with a circuit that sensed the E-field. It was based on a design made by Wilf Righter. It used some basic components like Mosfet, Resistors and Diodes.
[edit] Components
MOSFET – Used as a switch, turns on as soon as it is powered, turns off when E-field is sensed by the sensor connected to the gate
IN4004 Diodes – Used to provide the ESD protection to the MOSFET
1Meg Trimmer – It just sets the MOSFET in the Active/Saturation region
Antenna – Senses the E-field present in the surroundings
For the output part we have been using a LED, this would be replaced by a a transmitting circuit.
We finally built up the circuit and carried some tests on it. The circuit was able to sense static electricity induced on a glass rod (by rubbing it), however we did need to find out if it could sense an electric cable.
Working around with the sensor’s dimension, shape, orientation and the material we were able to sense the E-field around a cable carrying 110v AC.
For the moment we have selected the use batteries as the power source, however this would be replaced by the harvested energy. But first we have to make the circuitry low powered and even more efficient in sensing the field and also start working on the transmitting circuit.
DP1203 – C868 Transceivers – 868MHz RF Receiver and Transmitter
XBEE series 2.5 Transceivers – 868MHz RF Receiver and Transmitter
[edit] Field Testing
So by the end of our first semester we had a sensing apparatus which detected the E-field around a standard extension cable illuminating an LED. We also had a working field testing circuit to hand off to Moorhead Public Service to test near live power lines.
Moorhead Public Service field tested our original sensing circuit on live power lines. We originally placed an LED rather than the transceiver we had planned on, just for ease of use while the lineman had the circuit. We had planned on seeing the LED ‘ON’ when it was within 36” of both 12.5 and 7.2 kV lines given results from the 110V line in the lab. The following chart summarizes the results:
Upon viewing these results we noticed that the 30 foot distance they got from the 115kV line was way beyond what we expected to see. After consulting them and reviewing the location, we came up with the explanation that there was existing underground circuitry in the area which could have thrown our results off. Another idea we came up with when we returned to campus and found the MOSFET to be blown and that led us to believe the sensor could have been too sensitive for the field in the area around the 115kV.
[edit] Design
Our design started out as trying to build a passive electric field sensor. We found from the calculation shown above that we are not able to get enough of a voltage on a plate to be able to power anything.
We switched from passive to having an active field sensor. Throughout the 2 semesters we have been adding devices onto our project. Now we have a passive electric field sensor that is able to transmit data about the line being 'online' or 'offline'. This data is going to transmit from pole to every 3rd pole to let the next transceiver know if the line is on or off. A user can then link up to the transceivers via a handheld device and that will also be able to tell you if the line is on or off.
[edit] Pictures
[edit] Passive Ideas
""The diode takes care of the sinusoidal wave input. The storage capacitor will charge based on the input signal. The resistor combo will set the threshold. The smaller cap will give you the time needed for the transceiver to spit out info. Larger cap longer time.""
Jeremy Purcell, Digi-Key
[edit] Microcontroller Code
[edit] X-BEE RF-Modules
Key Features:
+ Long range data integrity, 100-300ft
+ Low Power, 45mA(TX), 50mA(RX), 10uA(Sleep mode)
+ Advanced Networking & Security, Unicast & Broadcast communications, Point-to-Point, Point-to-multipoint and peer-to-peer topologies supported
+ ADC and I/O line support
+ Easy Configuration
+ Antenna options
+ Worldwide Acceptance, FCC Approval (USA)
+ Free & Unlimited Support
Typical Configuration
Router: transceiver that is attached to the sensing device
FIRMWARE: ZNET 2.5 Router/End Device API version 1241
D1=2 this is the ADC input setting. By setting IR=1388, we have set a sample rate of 5 seconds. The module will sample the D1 pin every 5 seconds and create and send an API frame. The DH and DL commands determine the destination address of the IO samples. Here the DH and DL are set to broadcast.
Coordinator: transceiver that is attached to the substation computer or handheld
FIRMWARE: ZNET 2.5 COORDINATOR API version 1141
NOTE: only devices running API firmware can send IO data samples out their UART. Devices running AT firmware will discard received IO data samples. This is the reason the coordinator is loaded with API firmware.
NOTE: Analog samples are returned as 10-bit values. The analog reading is scaled such that 0x0000 represents 0V, and 0x3FF = 1.2V. (The analog inputs on the module cannot read more than 1.2V.) Values for the “Output Volts” column are derived by converting the Hex values back to Volts using the following formula. Output Volts= (Hex Output/0x3FF)*1.2V (rounded to 2 decimal points).
THE API FRAME
Below is an example of an API frame sent out the receiving modules UART
7E 00 12 92 00 13 A2 00 40 0A 3D BF 66 BA 01 01 00 00 02 03 FF 4C
Where the UART API data stream can be broken down as:
[edit] Budget
Since our sensing apparatus requires only basic components like mosfet, resistors, diodes and LEDs, which are provided by the university, we did not dip as far into our project’s expected budget as we initially planned on. We also planned on sending out our PCB’s to be done professionally, but we actually had time and focused on building those in house as well, saving more money than we could have expected.
[edit] Lessons Learned
Throughout the project, we had plenty of setbacks and hang-ups that greatly hindered our progress. The first semester was a decent introduction to what we could look forward to. First off, we had designed our e-field sensor but kept blowing the MOSFETs which controlled the switching mechanism.
More research in this, and other aspects of our project could have greatly saved time and money. Many other people could have run into similar problems that we were having, and reading about their solutions could have greatly reduced our efforts.
This closely relates to time management and keeping priorities straight. Juggling between this and classes is not an easy task, especially when you may be working outside of classes as well. Sacrificing time for this or other activities is not always easy or pleasing, but this is what we chose.
Starting the second semester, we decided upon a transceiver to work into the mix so we went ahead and ordered it. We choose this first one based on the fact that we should not need a PIC processor to control every one of the transceivers we used. After more deliberation and looking into the situation, we found that indeed we would need a processor of some sort to work with these particular models. We researched further and eventually found the Xbee RF-modules.
Testing is a great part of a project, and not having an inclusive lab seemed to greatly hinder our achievements. We couldn’t really know what to expect from the field results since our calculations didn’t seem to match too close to field results.
Another major lesson came from the PCB’s and parallel wiring running in close proximity inside the enclosures. Sending wireless data is not that easy, and then needing to display that data through a series of wires can make it nearly impossible if you can’t seem to find the problem. It is just really hard to conduct an experiment when you’re not exactly sure what has changed from the control. For example, during troubleshooting you use the control, power down the working circuit, repower the components and you don’t have a consistent result, having changed nothing. This can be very frustrating and make solutions almost unobtainable.
[edit] Future Works
Our design leaves and endless amount of possible future work for this project. First off, the A/D data sent to the LCD of the handheld could be used to actually determine line voltage using a simple formula. This could use the sensed e-field voltage and distance from line to determine what the voltage on the distribution line would be.
To further improve the senor, it could be powered by solar energy. This would reduce the need to replace batteries in sensors that may be going down every 6 months. Taking it completely the rest of the way to become strictly passive would also be an improvement. Initially that was our idea, but after discouraging calculation results, we chose to spend more time on developing a model we knew we had time to finish.
We could also potentially filter the sensed signal to make sure we are only using E-field from a 60 Hz source, and then rectify the sensed signal to give the transceiver more consistent data to send down line regarding the transmission line.
[edit] Acknowledgements
ECE Department
Moorhead Public Service
Dr. Nelson
Dr. Kavasseri
Joel Aslakson
Jeremy Prucell
[edit] Documents
Final Report: Media:Final_Report_SD0716.pdf
Users Manual: Media:Final_Users_Manual_SD0716.pdf
PowerPoint Presentation: Media:Final_Presentation_SD0716.pdf
