Easy-Access Power Dome

1
Easy-Access Power Dome

• Arjun Khetarpal
• Rushi Kothari
• TA Zuhaib Sheikh

2
Introduction Objectives

• To create a device with multiple touch-sensitive
  regions, which wirelessly communicates with a
  receiver which allocates power to various
  appliances.
• To help individuals with mobility impairments to
  function more independently.

3
Introduction Benefits

• Provides greater independence to individuals with
  limited mobility.
• Creates a safer environment for users, as
  difficult to reach switches can often be a
  hazard, especially for those with limited
  mobility.
• Does not require effort to press or flip, as
  typical switches and buttons often require more
  strength and mobility than an individual might be
  capable of.
• Wireless implementation prevents hazardous and
  messy setup that comes with a wired device.

4
Introduction Initial Design Outline
Power Router
Router-Side Microcontroller
RF Receiver
5
Introduction Final Design Outline
6
Introduction Features

• Four touch-sensing surface regions allow for
  users to apply the slightest touch in order to
  activate up to four desired appliances
  simultaneously.
• LED display on relay-side indicates which
  appliances are receiving electricity. Feedback
  mechanism checks that the appliance is activated,
  and sensor-side LED display indicates that
  appliance is functioning.
• Wireless RF communication between power dome and
  power router allows for a no risk environment, as
  wires can often become tangled and difficult to
  install.
• Portable Power dome is mobile enough to be
  transported, but large enough for easy access
  touch regions.

7
Design Hardware Overview

• 2 x PIC16F877A Microcontroller
• 2 x TRM-433-LT RF Transceiver
• 4 x QT100A Capacitive Touch-Sensor
• 4 x W171DIP-2 Reed Relay
• 2 x F1100E 20MHz Oscillator
• 2 x UA78M33C 3.3V Voltage Regulator

8
Design Reasons for using QT100A

• Small, low-cost package for detecting touch.
• Sensitivity can be adjusted by altering capacitor
  values
• Low-power consumption mode saves battery life.
• Challenge Unfortunately the 10-pin MSOP package
  is no longer manufactured, so the 6-pin WSON chip
  was very difficult to connect.

9
Testing Sensors

• Because we received a 6-pin WSON that was only a
  few millimeters in size, we were unable to solder
  pins to it successfully. Our PCB was finally
  fabricated after the demonstration.
• We were able to successfully implement the sensor
  afterwards.
• More refining is needed, but it shows that our
  sensors are working, and we simply needed to
  connect four of them into our circuit.
• Below is an early capture of response from the
  sensor. We are quickly tapping the electrode, and
  one can see the response on the oscilloscope.

10
Performance QT100A Sensitivity (w/o electrode)
11
Design Reasons for using W171DIP-2

• Challenge To implement a set of relays that can
  connect to our PIC. Mechanical relays require
  large input currents that the PIC cannot supply.
• Solution Reed relays can be directly connected
  to the PIC without using a BJT current amplifier
  circuit.
• The W171DIP-2 only draws 10mA, while the PIC is
  able to supply up to 25mA.
• The relay has a maximum current rating of 10A,
  which is protected by a set of 10A fuses.
• Characteristics Turn-on voltage 2.6V, Turn-off
  voltage 2.1V

12
Design Reasons for using TRM-433-LT

• Initially TXM-433-LR and RXM-433-LR
• Sensor-side transmits data and relay-side
  receives data
• No feedback system implemented
• TRM-433-LT
• Both transmit and receive capabilities, which are
  necessary due to our feedback system
• Power-down mode useful for saving battery
• Challenges
• Only one frequency channel, resulting in
  interference from other projects.
• Switching between transmit and receive is not
  trivial to implement for feedback system.

13
Design Reasons for using PIC16F877A

• Cost-effective solution for our purposes
• USART/RS-232 lines implemented for interfacing
  with RF transceivers
• To be used for taking input from four sensors, RF
  communication, and providing output to four
  relays
• Challenge Implementing a scheme which would turn
  on one of four appliances depending on which
  sensor was touched.

14
Design Sensor-side PIC, transmission, and
reception

• Solution Four input sensors send four distinct
  bytes of data
• Sensor 1 0xD8 0b11011000
• Sensor 2 0xD9 0b11011001
• Sensor 3 0xDA 0b11011010
• Sensor 4 0xDB 0b11011011
• We set values in the PIC code in reverse order
  (e.g. 0b11011000 ? 0x00011011), as the transmit
  line reverses it back to normal when sending.
• Transmitter side is actively high, and receives a
  0 before it starts sending data. After data is
  sent, goes back to 1.
• Receiving module has small data-on-line pulse
  followed by a 0 to indicate that ready to
  receive. Then data is received.

15
Testing Sensor-side PIC, transmission, and
reception

• Line 1 (top) is Data Out (TX) pin of sensor-side
  PIC
• Line 2 (bottom) is Data In (RX) pin of relay-side
  PIC
• Sensor 1 0xD8 (1101 1000)

1 1 0 1 1 0 0 0
1 1 0 1 1 0 0 0
Data-on-line pulse
16
Testing Sensor-side PIC, transmission, and
reception

• Sensor 2 0xD9 (1101 1001)
• Sensor 3 0xDA (1101 1010)

1 1 0 1 1 0 0 1
1 1 0 1 1 0 0 1
Data-on-line pulse
1 1 0 1 1 0 1 0
Data-on-line pulse
1 1 0 1 1 0 1 0
17
Testing Sensor-side PIC, transmission, and
reception

• Sensor 4 0xDB (1101 1011)

1 1 0 1 1 0 1 1
Data-on-line pulse
1 1 0 1 1 0 1 1
18
Performance Response Time

• Challenge To run our circuit with low
  power-consumption, while maintaining a 150ms
  response time.
• Our transmitter is brought up from a powered-down
  state before transmission, and we can see that
  the time it takes for the appliance to turn on
  is
• 85ms (worst case sensor response time on
  low-power)
• 10ms (power-up transceiver)
• 8ms (transmit/receive data)
• negligible delay for PIC to
  manipulate data
• 103 milliseconds
• 8ms (transmit/receive feedback)
• 111 milliseconds

19
Design Feedback

• We implemented a feedback method into our device.
  Once data is received at the relay-side PIC, the
  data sequence for the activated appliance is sent
  back to the sensor-side PIC. LEDs change
  accordingly to reflect the actual status of the
  device.
• The transceivers both contain a T/R Select pin
  controlled by the PIC. When low, the transceiver
  is set to receive, and when high, it is set to
  transmit.

20
DesignSensor-Side Feedback Algorithm
0
1
0 1
21
DesignRelay-Side Feedback Algorithm
1 0
22
Testing Sensor-Side Feedback

• Here we can see data being sent from sensor input
  on oscilloscope line I (top).
• Oscilloscope line 2 (bottom) shows the T/R select
  line. Immediately after data is sent, the T/R
  select line goes low to set the transceiver into
  receive mode.
• The T/R select line goes back high after about 1
  second.

23
Testing Sensor-Side Feedback

• Below is the capture of the sensor-side
  transceiver being switched into receive mode (
  the same signal as on the previous) slide.
  However, now the oscilloscope is connected to the
  receive pin of the PIC. We can see feedback data
  being received from the relay-side.

24
Testing Relay-Side Feedback

• Here the relay-side PIC can be seen switching to
  transmit mode after receiving data (not shown).
  During the T/R select high phase, data is sent,
  and the line returns to low (receive).

25
Major Challenge Functionality Issue

• As shown in the demonstration, we were unable to
  get our PIC to decode the incoming data in lab
  conditions, even after much time spent looking
  from various angles at the issue.
• Our receiving PIC code included an algorithm
  which waited until the data-on-line pulse is
  found (kbhit command), and then checked for the
  initiation sequence 110110 present in all of
  our signals. Then it reads the last two bits to
  determine which of the four appliances to supply
  power.

26
Major Challenge Theoretical Cause

• The most probable reason for our issue is that
  noise was fooling our receive code into thinking
  it was the initiation sequence (would explain why
  our relays received so many false signals and the
  output LEDs flickered).
• The length of a 1 or 0 is approximately 0.75
  milliseconds from the PIC/transmitter, but noise
  is many orders of magnitude shorter. Thus, maybe
  the receiving PIC interpreted that incorrectly as
  1 or 0 when we didnt want it to.

27
Performance Power Consumption

• QT100A
• Current draw of 600µA in most applications when
  running in Fast response mode
• Current draw of 6-12µA in Low Power mode
• Sinks up to 2 mA of current
• TRM-433-LT
• Transmit mode draws 10mA
• Power-down mode draws 0mA
• PIC16F877A
• Max current sink/source of 25mA per port
• W171DIP-2
• Draws 10mA of current from PIC.

28
Performance Battery Life

• Because our device runs partially on battery
  power, we want to ensure that the battery life is
  as long as possible.
• Run QT100A in low-power mode
• Power-down transceiver when not used
• Calculations (powered up and transmitting data)
• Capacity of battery 500 mAh
• Battery life Capacity Current Drawn
• Current Drawn PIC(25mA) Transceiver(12mA)
  Regulator(6mA) Sensor(.6mA)
• 500mAh 43.6mA 11.47 hours
• Note that with transceiver powered down, 0mA are
  drawn. And with sensor on low-power mode,
  considerably less than .5mA is drawn.

29
Performance Battery Life
30
Performance Power Distribution

• Relay side receives power from a 120VAC to 5VDC
  adapter.
• Voltage regulated to 3.3VDC
• Receptacles are powered by relays through 10A
  fuses.
• Relays receive input from 120VAC wall outlet.

31
Ethical Considerations

• We have constructed our design with ethical
  considerations in mind. As our design is
  specifically created to improve the quality of
  life of those in need, safety and ethical issues
  are at a forefront of our project.
• We created a battery powered dome in order to
  make our design more convenient for individuals
  with motor impairments.
• Additionally, we made sure that our wireless
  design allows for a safer environment for the
  user.

32
Further Development

• Use a larger touch sensor than the QT100A, which
  was not ideal for development with student
  resources.
• Further develop noise reduction processes to
  ensure data integrity.
• Create longer data sequences to include more
  reliable initiation sequences. Also develop a
  termination sequence and checksum for data
  accuracy.

33
Commercial Viability

• Needs more development in sequence decoding and
  power distribution.
• Develop PCB layout for more compact and cleaner
  circuit.
• Needs functional and aesthetically pleasing
  electrode configuration.
• Use multi-channel transceivers, capable of
  transmitting over less used frequencies.

34
Questions

• ?