Micro-Trak 300 - V1.3 APRS Tracker
August  2007

This page describes building a Micro-Trak V1.3 kit and APRS tracker for high altitude balloon use.
A similar page about a Micro-Trak V1.4 kit is described HERE.

Tony required a tracker for future SABLE flights after SABLE-2 was unable to be found last summer and purchased a Lassen iQ GPS Receiver, Embedded Antenna and Byonics Micro-Trak 300 for me to build him a new one with.

 

The Byonics Micro-Trak 300 kit only has a few components and is very easy to build.


As usual, I did some things a little different then in the instructions and made a few  mod's s I felt were important or worthwhile to improve performance.

The first thing I did was set the I.C. socket aside for use with something else.

Sockets are ok for experimenting or when problems are expected that may require replacing a chip, but for projects like this, where reliability is so important, leave nothing to chance and don't use a socket. It's unlikely a TinyTrak chip in something like a Micro-Trak will ever need to be replaced, if one is even just a little bit careful, and if one does, it's simpler and much easier then having to deal with the problems a socket can cause, especially as it gets older.

Below - The board wasn't designed to have an antenna connector installed, but a SMA connector was required for the antenna and one was able to be installed as shown. But this is no longer a problem with newer V1.4 boards that are designed for a SMA connector which now comes with the newer board.

Hot Glue was used on both sides of the board to bond the small connector and provide strength when making connections. Hot glue was also used to embed the red & black leads supplying power to the board to provide strain relief and keep the leads from flexing and breaking at the point where they were soldered.

Micro-Trak Bottom View

( Numbers relate to the changes & modifications described below. )

Micro-Trak Top View

  1. A KF50BDT-TR low dropout voltage regulator was used to replace the 7805 regulator supplied in the kit as I don't like using 7800 series regulators for battery powered projects due to their larger dropout voltage. The minimum input voltage with a 7805 to maintain regulation was 6.4V when idle and 6.6V when transmitting, but with the KF50 it's only 5.1V when idle and 5.3V when transmitting, but even with just a 5.1V input, the output still only fell < 0.1V when transmitting. The KF50 also has better overall spec's and lower quiescent current then a 7805. What this all means is that 1 less 1.5V cell is required, or 4 cells rather then 5, and that 4 cells will last longer with a KF50 then 5 cells will with a 7805 regulator.

  2. A 2.2 ufd tantalum capacitor was added on the regulator input. Instructions showed placing a 10 ufd tantalum on the regulator input, but a revision note indicates it was moved to the output on a board revision so it was placed there. A 2.2 ufd tantalum was then added on the input as 7800 type regulators require an input capacitor for stability if more then a few inches from the power source or if the source impedance isn't sufficiently low, like cheap or weak batteries. One can usually get away without using an input cap, but experience has shown it's always best to use one. 7800 type regulators don't require an output cap, but it's always good to use one anyway to reduce noise and improve transient response.

Some regulators, like the KF50, require an output, rather then input, capacitor for stability so, to avoid confusion and problems, just remember that it never hurts and is always best to use both, an input and an output capacitor, on all 3 terminal regulators.

  1. A DFLS120L-7 diode was added between the KF50 output & input pins for reverse voltage protection to allow the Micro-Trak to be powered with 5V thru the 5 pin connector when programming the Tiny-Trak to save having to make another connection.

  2. A LE33CZ 3.3V regulator with a 0.1 ufd input & 4.7 ufd output capacitor was added, using the 4 solder pads on the bottom side of the board for a 9 pin connector, to supply power to the 3.3V GPS receiver being used via the 5 pin connector (#6). See the Wiring Details PDF for further regulator installation details.

  3. A 3 pin header with a 2 pin jumper was installed for selecting Primary or Secondary configuration. The “Power Enable” feature on pin 13 of the TT3 chip wasn't needed and the solder pad tied to it was needed as a ground so the trace between the 2 points was removed. A 3 pin header, with the lower end of 1 pin cut-off and that pin next to pin 14, was then soldered into the 2 solder pad holes next to pins 12 & 13 and the isolated pad next to pin 13 was soldered to the surrounding ground plain. Pin 12 is the Configuration Switch input so the configuration can now be changed from Primary to Secondary by simply moving the shunt on the header from next to pins 14 & 13 (its storage position) to the position next to pins 13 & 12 (which grounds pin 12).

  4. A small 5 pin connector was installed using the 5 solder pads on the top side of the board and hot glue on both sides to hold it securely in place. It's ridiculous to use a huge pair of DB-9 connectors, that are as heavy and large as the Micro-Trak itself (when a shell is used for even one of the pair) and a small connector is much more appropriate and lighter for just 3 or 4 small wires between tiny items like a GPS receiver & Micro-Trak.

  5. A jumper was used in place of the 10K resistor in the serial data input line to pin 3 of the TinyTrak chip that protects the pin from RS-232 voltages which can be up to ±25V as only 5V, or less, logic level signals will be used and eliminating the resistor eliminates any chance of having it cause a problem receiving data, especially from 3.3V data sources.

  6. A 4 pin header was installed for the LEDs used to monitor operation of the Micro-Trak. The 3 LEDs and their 1K resistors were soldered to a 4 pin strip and embedded in 5 min. epoxy to create a plug-in assembly.

 
For further information, refer to the
 SABLE Tracker Wiring & Modification Details PDF

The Radiometrix HX1 Transmitter Module

I've been waiting to get my hands on one of these intriguing little modules and the chance to check it out for whatever else it may be useful for.

The Data Sheet says it's for long range data transfer use for up to 10km, but it's SABLE-3 APRS transmissions were received over 600 km away which proves those saying 300mW is not enough power to be useful for APRS -- are wrong.

Value HX1 Data Sheet Spec's Measured Micro-Trak Values

TinyTrak3 Chip section below explains why the Micro-Trak idle current varies by several ma.

Supply Current 140 mA at 5V Idle - 39.2 to 41.5 ma
Transmitting - 134 ma (w/ No LED's)
RF Output Power +24.7 ±1 dBm +23 dBm Output power was the same for two different units w/ Vcc = 5.0V
234.4min / 295.1typ / 371.5max 199.5 mW
Frequency Accuracy ±2.5 kHz

Transmit frequency is temperature dependent.

Modulation BW @ -3dB 0 to 5 kHz

Deviation with 1200 Hz modulation is about  ±2.5 kHz
Deviation with 2200 Hz Modulation is about  ±1.8 kHz

The HX1 module is very temperature dependent. When transmitting both modem tones (using the TinyTrak configuration program) the center frequency was ≈500Hz high at room temperature (≈70°F). (Both tones were used as the center frequency is actually different for each tone as explained later.) The module temperature was then increased by transmitting continuously until the frequency was correct at ≈95°F. 70°-95° is likely the most common temperature range that will be encountered for most 'down-to-earth' uses and 500Hz accuracy isn't all that bad, but when placed in a freezer at ≈0°F (-18°C) the frequency ended up being 1.2 kHz high which isn't good considering this tracker is for high altitude balloon use and may experience even lower temperatures.

On BEAR-3, our first altitude record attempt and with only the tiny amount of heat dissipated by the GPS and Micro-Trak to keep things warm, everything was found covered with ice crystals when the insulated payload box was opened. It's hard to know how cold things got, but we were probably lucky the trackers transmissions were all able to be decoded with how far the frequency was likely off. With more time, it would have been interesting to have plotted the transmit frequency vs. temperature before the flight and to have used a receiver with a deviation meter to know just how cold things got and how high the transmit frequency ended up being.

Accurate deviation measurements were hard to make with the carrier frequency constantly changing as measurements were made, but close enough to determine why things were the way they were. The TXD pin is a DC coupled CMOS compatible input and with the pin biased at 2.5V the carrier frequency was close to 144.390 MHz as expected (depending on temperature, of course). With the pin at 0 and 5V I also expected the frequency to shift ± ≈2.5 kHz, but it actually shifted −3.2 kHz and +2.2 kHz and accounts for why the demodulated tone waveforms are distorted like they are and why the center frequency is ≈100Hz higher for the 2200Hz tone then for the 1200Hz tone.

From the data sheet information, the HX1 is obviously designed for digital, rather then analog, modulation and with digital modulation the non-linear deviation would make no difference, other then making the center frequency appear ≈600Hz lower then what it is with the TXD pin at 2.5V which would make it almost correct at room temperature.


The 1200 Hz tone waveform from the TinyTrak 4-bit D/A converter.

The 1200 Hz tone waveform after low-pass filtering in the transmitter.

The received waveform is distorted with the upper half more rounded then the lower half from the unequal deviation resulting from each half.

2200Hz tone deviation is less then for the 1200Hz tone due to the transmitters low-pass filter which starts at ≈1.8 kHz.
Confused and wondering what all of this really means? It simply means that the Micro-Trak is a quick, simple and easy way to build a tracker that's good for most applications, even for tracking high altitude balloons in many cases, but to expect problems, especially in cold situations, and that there are better transmitters to use. A Balloon_Sked entry at yahoo groups once questioned why a tracker had failed above 60K ft. (thus proving the GPS was ok for such altitudes), but started working again after returning much lower. I first guessed that the batteries likely froze from too little insulation, but upon learning it was a Micro-Trak it's much more likely that the cold had simply caused the transmitter to drift too far off frequency to allow the transmitted data to be received and decoded.

 

The TinyTrak3 Chip

Some may be able to program a TinyTrak using a RS-232 port directly, but I have yet to find a PC or laptop RS-232 serial port that will accept logic level data and TinyTrak's use of inverted logic level data has only caused problems and confusion for myself and all those I've helped. TinyTrak3 is a big improvement over previous versions with being made to work with either normal, or inverted, logic level GPS data, but it would have been nice to see the same done for programming data to save having to provide extra circuitry (inverters) following a RS-232 or USB interface chip in order to accommodate TinyTrak's inverted data. Using inverted data to save having to use a RS-232 interface chip is fine for personal project, but there are standards for a reason and they should be followed for things being marketed to prevent confusion and problems.

As for why the Micro-Trak idle current is listed as 39.2 to 41.5 ma. in the HX1 transmitter section above, it's because the four TinyTrak D/A output pins are left in 1 of 16 different random states after each transmission which results in 16 different levels of standby current. Not a big deal, but it would have been nice to have them always left in a high state to reduce the standby current a few ma. and make it consistent for easier measurements.

 

Programming Interface

RS-232 serial port interface cables were used in the past for devices with a logic level interface, but I now use USB serial ports as I find them much simpler to use, my lap top has no RS-232 port and USB ports are often able to power what they are used with which saves having to deal with a another power source and cable. And a Sparkfun FT232RL USB to Serial Board solved the problem of dealing with TinyTrak's inverted data as it's software configurable I/O pins allowed me to simply invert the RXD & TXD pin signals to make them compatible with a TinyTrak.

This is not unlike installing facility maintenance software on a business's computer systems but only on a much smaller scale.

The description appearing in the list of connected USB devices on a PC can also be modified and was changed to "FT232R TinyTrak Interface" to indicate that this board was modified for inverted data and TinyTrak use.

A Red Tx Data and Green Rx Data & Power LED was also added and the board was modified to supply 5V, rather then 3.3V, via the 4-pin connector.

 




Note: Sparkfun now supplies a new version of the FT232R
USB board that includes Tx & Rx LEDs.

NEXT SEE:

 This Micro-Trak with a Trimble Lassen iQ GPS as used for SABLE-3
   This Micro-Trak with a Trimble Lassen iQ GPS as used for BEAR-4
  A similar V1.4 Micro-Trak with a Trimble Lassen iQ GPS was used for BEAR-3

The SABLE Tracker Wiring & Modification Details PDF has wiring & further Micro-Trak, GPS & USB Board Details.


  
To BEAR Home Page