Design Constraint Analysis and Component

Design Constraint Analysis and Component

Selection Rational

Nick McCarroll

ECE 477

Group 11

February 12, 2004

Introduction

This design project proposes to use a global positioning system (GPS) as an aide to personal training. Keeping track of how far and how long the runner and/or biker has traveled, they can easily monitor exactly how much they have worked out, and to see how well they have done. In addition to monitoring the simple racing statistics, this design has a wireless heart monitor as feedback to the user, and possibly a digital thermometer. The user can also choose from a selection of built in preprogrammed distances to runthrough an on-screen menu and input device. All of these devices will have to be as small as possible, not being to bulky as to get in the way of a runner, and use as little power as possible, for longer distance and endurance applications. This can provide the user a very useful tool in training, perhaps for marathon use.

Analysis

With the ease of use of all the data supplied by the GPS, there will not be a huge amount of calculation needed to be performed by the microcontroller itself. However, a considerable part may be to convert all of the units into the common US standard units. It would need to keep all of the information as accurate as possible, and if there is any error, keep it from being accumulated. The rest of the calculations needed to be performed would be the summations of all the history data: distance, time, and average speed. In the middle of a program, a progress monitor will display if the user is ahead or behind of a desired time limit, and calculate the percentage difference. An additional feature includes the heart rate monitor which will calculate the user’s current pulse.

The interfacing between all of the components has been kept to a bare minimum to try to keep the size of the device as small as possible. The GPS module and the display were chosen so that it could be connected directly to the microcontroller via CMOS level signals. The LCD display was chosen because it had a serial interface, so that the amount of pins used up by the microcontroller was kept to be a minimum. By using a CMOS compatible display, the extra chips used for the changing to RS-232 levels were avoided. The GPS unit however does use RS-232 level signals, so a single converter will be needed with 2 channels, one for transmission, and one for receiving signals from it. Keeping the idea of not using extra board space, a rotary pulse generator is going to be used for the user input as a rotary dial for different options with a push button as a selector. The heart rate monitor is going to be a wireless device that outputs a signal around 5kHz. It is powerful enough just to go a couple of feet. The speaker, the only other output device will be made from a PWM output and a digital-to-analog converter as feedback to the user. These last two parts of the circuit, pulse receiver and D/A converter may take up the most space, but will not make the device to big to wear.

Being a portable device, power consumption has a vital role in the design of this product. Every chip chosen has been a low power device and chosen to run at a voltage as low as possible. At the same time, most of the devices chosen should be able to run at 3.0V, except for the display. The LCD display requires a 5.0V regulated power supply, and it will use about 200mW of power. Only the GPS receiver will draw more power than this, due to the active antenna, where it will draw about 280mW nominally. The microcontroller however, being that it can be clocked down, and at a lower supply voltage will only draw about 10mW of power. It is unknown at this time what the power consumption of the receiver circuitry will take up, but with enough batteries, this design will be able to run for the duration of a marathon.

Again, as a device that is to be worn by an athlete, size is of utmost consideration. Weight will not play too much of a factor, most of the parts only weigh a couple of ounces at most. The size of the unit will in the end be determined by the packaging, but a two part device will serve the user the best retaining the most mobility and being as ergonomic as possible. The heart monitor will be a strap that wraps around the users chest, which transmits to the main unit which will be on the upper arm. This part of the device will house the GPS receiver, heart rate receiver, speaker and the microcontroller. The second part will be just the display and user input device which will be attached to the user’s wrist or forearm.

The total cost of the unit, mainly caused from the price of a single GPS receiver would be under $200. The Garmin 15L GPS receiver is $90, the heart rate monitor is $40, LCD display is $13.25, and the microprocessor is $4.50. And in addition, the antenna for the GPS will be about $20. The total cost, not counting the simple receiver circuitry will be about $168. This was a value that was within the expectations of the group. If this unit was in full production, and all components could be bought in bulk, the selling price could easily be between $100 and $150, which is the going rate for many GPS devices.

Component Selection

Of the components in this design, the three main components that needed to be chosen were the microcontroller, GPS receiver, and the LCD display. For the microcontroller, the three viable candidates that the group came up with were the Atmel ATtiny26L and ATmega8L, and the PIC16LF747. All of these micro controllers were low power devices. The PIC stood above the rest with an astounding .2mW power consumption. The Atmels were not far behind where the tiny consumed 2.1mW and the mega controller consumed 10.8mW. These numbers are probably high compared to what will be consumed on our unit, because we will not clock them at the highest speed. The tiny Atmel chip has the advantage of taking up less board space, yet providing all the same functions. However the Atmel mega version provided the best all around features with many more port pins available with all the needed peripherals. Also if any more port pins would be needed, this will allow us some breathing room for worst case scenarios.

The GPS units were very abundant, and there were many from which to choose. The units for consideration were the Garmin GPS 15L, µ-Blox , and the Tyco A1025. Again, these are all low power devices, and the all have the option for active antennas. Size was a big consideration, the µ-Blox unit measured to be 2.8”x 1.6” in area and came with a 2x20 pin header for easy connection. The Garmin GPS 15L-W, along with its great reputation, measures 1.4”x 1.8” and has a header of pins that are more readily accessible for prototyping. And the Tyco A1025 measures 1.2”x 1.2”, and has a surface mount package, which will make the end product fit easier. The Garmin unit only uses about 250mW power, while the µ-Blox and the Tyco unit can consume up to 500mW power, nominally. Knowing Garmin’s reputation with GPS modules, using the least power, and costing relatively less than the µ-Blox, the GPS 15L-W was an easy pick.

For the units LCD output, size and interfacing were two big issues. Three different sizes were considered: 2x16, 2x20 and 4x16. It was agreed that the 2x20 and the 4x16 could display more information to the user, but it was the ease of connection and small size that determined our choice. The 2x16 display by Images SI, Inc was chosen because it can accept true CMOS signals serially, and it was only 3.1”by 1.4”.

References

Data sheets can also be found on the Group 11 website.