vem

---

Recommended: this blog post describes the project in detail.

---

Vehicle Environment Monitor (VEM)

I created this project to remotely monitor temperature in my car when I bring my dog on a trip and have to leave him tempororaly in the car. This gives me a piece of mind knowing that the temperature is good for him to comfortably stay in the car.

NOTE Here cites the laws in different states that prohibits leaving animal in motor vehicle in cirtain situations.

Range Tests

• 915MHz
• Tx Power: 23dBm
• Modem config: Bw125Cr48Sf4096: Bw = 125 kHz, Cr = 4/8, Sf = 4096chips/symbol, CRC on.
• Wire antenna at quarter wavelength

Test1

• Tx: Sammamish Metropolitan parking lot (upper)
• Rx: Sammamish Metropolitan cafe
• Min SNR: -13dB

Test2

• Tx: Sammamish City Hall parking lot
• Rx: Sammamish Library underground parking lot
• Min SNR: -4dB

Test3

• Tx: Factoria Mall parking lot
• Rx: Factoria Mall Nordstrom Rack
• Average SNR: -4.9dB
• Min SNR: -8dB
• Distance: 400ft (120m)

Battery Tests

Both Rx and Tx can be powered by 2xAA battery at 3V, bypassing the voltage regulator. When using non-standard voltage (5V) at Tx, the temperature reading is wrong as 5V reference voltage is hard coded.

Rx Side

0.03mA when idle. 3.5mA when receiving (peak). If Tx interval is 5 sec. Current is 3.5/5 + 0.03 = 0.73mA. For a 500mAh battery, life time could be

500 * 0.75 / 0.73 = 513h

Cannot easily be connected with USB power banks, as the low current causes power banks to cut power automatically.

2xAAA (3v) works fine over JST-PH power socket (tested via BAT pin). The benefit is its low profile.

Li-ion poly batteries with low volumne are also low profile. They are preferale over 2xAAA as being in the recommended voltage range for JST-PH port.

Tx Side

9v Battery

Problem with 9V: voltage slowly goes down making temperature reading gradually goes up.

Measured tx side power consumption from 9v:

• 90mA non-transmission mode
• 215mA transmission mode

Test1: 9v battery (used)

9V battery, almost new, after 70+ minutes, the 5V regulated Vin measured only 3.0V. Battery itself was measured only 4.3V. Not good.

The dropped voltage caused the measured temperature to be misinterpreted and went up from 19C to incorrectly 62C.

Test2: 9v battery (new)

• 9V battery
• Initial: vol=9.46v, 5vin=? temperature reading: 17C
• After 3507s (58min): vol=7.5v, 5vin=5v temperature reading: 20C
• After 5409s (90min): vol=7.3v, 5vin=5.06v temperature reading: 19C
• After 9201s (153min): vol=, 5vin temperature reading: 56C

From the chart it can be seen that temerature is well maintained constant over most of the battery life, but only raised up (due to low voltage) at the end of it. SNR is maintained over entire battery life and cut off when power is not sufficient.

For 9v 300mAH, current expected battery life is

300 / (90 + 200/5) = 2.3hour

(90mA + 200mA/5) is the average current draw. 200mA transmission mode occurs every 5 sec.

This aligns well with the experiment. Now if we put it to sleep during the 5 sec wait, expected battery life is

300 / (200/5) = 7.5hour

Test3: Anker PowerCore+ mini

3350mAh power bank. Running 14.5 hours without any problem.

Test4: Constant from car battery

Test5: All above with IRQ sleep&wake