Create Project Need help?
Show more (3)

Remote telemetry on LoRa - 433 MHz

Status: Proposal
August 3, 2016
LoRa spread spectrum is the answer

How long shall it continue ?
Ever since the new project has come in front of the stage-1, the problem started – the status of chlorination plant chlorine leakage along with chlorine cylinder temperature was not available at the control room of the new chemical plant which is situated 3.5 kms away from the chlorination plant. Earlier the status used to come vide underground control cables to the nearby old chemical plant but the erection of the new stage & new chemical plant has made all these wires defunct.

The space in between has been taken by the newly laid 800 Kilo Volt DC switch yard besides the existing 400 Kilo Volt switch yard, not to mention the huge 30 mtr wide circulating water channel from which the huge hot water fumes pollute the atmosphere 24*7.

Since this requirement was a statutory audit point, effort was made to bring the data wirelessly using some RF link but the interference from the huge power line switchyards made it difficult to cross to this side. Some vendors claims to make it using UHF set up but they demanded a prohibitory cost for that. Hence the problem remained unattended for quite some time until …

LoRa spread spectrum:
The secrets of spread spectrum was not unknown to US military. Since world war-II they have been deploying spread spectrum for their torpedoes & missiles to hide them from deciphering , only at great cost. But as the chip technology matured, the cost reduced and for the last 3 years few companies like – Dorji, Hope, Semtech has released small hardwares which can produce spread spectrum signals at a very low cost.

Technique of SS: Though this article is not the right place to talk about it but a small intro will help the readers to understand it better. The Spread Spectrum is a signal below the noise level that is why it's not decipherable neither it can be jammed intensionally or unintentionally. It's immune to atmospheric conditions other than LOS. First the signal is spread to a big bandwidth below the noise level with the help of a dynamic hardware key set. Then it propagates below the noise level towards the destination. At destination the same hardware produces the key set to de-spread the signal to reproduce the original signal – something like a helical coil spring is first expanded to a great extent , then it is delivered to the destination and then at destination it is compressed again to make the original healthy looking helical spring !

It cannot be jammed because even though the jamming noise can be introduced but since it does not have the key set it will not be able to spread & mix with it thereby jamming it without any effect. Since the signal is spread below the noise level the power density is also very less. If somebody tries to eavesdrop it, he may only notice some increase in noise level nothing more !

Various type of spread spectrum techniques are - Frequency Hopping spread spectrum, Time hopping spread spectrum , time division spread spectrum , mixed type etc.

This Project: This project will describe how to deploy this technology to solve the above problem and many other possibilities related to remote telemetry. The signal for chlorine leakage is a 0 to 24 volt analog value which was already available at the chlorine plant. It is converted to 0 to 3.3 volt signal by voltage divider using 27K & 170K resistor. These values may change depending upon the actual voltage. Do some trial & error to reach the 3.3 volt level when the signal voltage is 24 volt. Tantalum resistors are preferred for this voltage division.

The other signal is a chlorine cylinder temperature which was defunct for many days. We have deployed a water proof Strainless Steel type DS18B20 which can measure temperature upto -55 degree centigrade very accurately. Since the atmosphere is chlorinated, the corrosion is a big issue for everything metallic here.

[The leaking chlorine gets enveloped in air very fast at 1:1000 by volume ratio. Therefore, in the event of chlorine leakage one has to see the air direction and then run opposite to it. Running towards air direction will make sure the poisonous gas catches you. As the chlorine cylinder gets emptied fast on the event of a leakage ,the cylinder temperature falls very fast . The measurement of chlorine cylinder temperature is also an important parameter for that identification. Chlorine is used to doze in limited amount into the circulating water to kill the micro bacterial organism ]

01. LoRa 433 mhz transceiver module – 02 nos $20.00 link:

02. Arduino clone or ATMEGA328P-PU – 02 nos $3.00

03. DS18B20 ,water proof -01 no $1.20 link:,searchweb201602_3_10037_10049_10017_405_404_407_10033_406_10032_10040,searchweb201603_7&btsid=4c2bf2d0-c91b-490c-8bb6-a1e1e3f16975

04. 433 mhz 6 dBi rubber antenna – 02 nos $4.5 link:

05. SMA female plug – 02 nos $3.0 link:

The LoRa module has the pin up connections of 2mm pitch , which may cause problems while connecting them on standard PCB of 3mm pitch.

A typical Hope RFM98 module:

To cater to the problem look carefully into my prototype. I used 2mm pitch female pins which is then twisted and entangled to connect to the 3mm pitch. Not a very elegant work but it works & robust too.

Installation: The Tx unit is installed on the corner of the chlorine plant roof on a mast of 4 mtr height stick. The small 433 MHz antenna is placed strategically in line of sight of it's brethren unit placed 3.5 km away across the switch yard & the water channel on the new chemical plant unit, roof top. On test run we found every data was arriving in a gap of 2 seconds.

Schematic of Tx: Runs on 3.3 to 3.7 volt Li-Ion battery

Schematic of the Receiver: Runs on 7.4 to 11.1 Volt Li-Ion battery

[Since we could not get a 3.3 volt LCD panel, we end up using two voltages – 3.3 volts & 5 volts ]

Software: HOPE LoRa transceivers that I've used are connected on SPI bus. There are lots of softwares available in the internet to make the life easy while handling with LoRa radios. Out of which I prefer RH_RF95.h package where commands are really very neat and clean. In less than 10 lines one can make these radios up & running using this headers. There are many choices available in the header file to make the transmission like – default & medium range, fast & short range or slow & long range etc. The bandwidth, spread factor and CRC can be adjusted depending upon the options set in this file. I've used all three modes for this short transmission on LOS (line-of-sight) and in each mode it worked. The transmission power can be set from 13 (default) to 23 dBm. The transmitted data is received and then deciphered and placed on the LCD.

Alarm generation:
The deciphered comma separated data is received as byte which is then regenerated and reconstructed to suitable float variables. This float values are then used as alarm generation and other purposeful usages. The alarm / relays are connected to the A1 pin through a small transistor switching – BC547 & 10K resistor. The relay is 5Volt whose one end is connected to the 5V supply. However, we are still in the thought process to multiply the alarm to some more strategic locations. The air direction is taken from win sock placed at the chemical plant and chlorine dozing plant.
My Prototype: The Sender

[GPS receiver shown in the picture is optional ]

The Receiver:

Shortly we are going to deploy LoRa for some more areas where remote telemetry through heavily interfering atmosphere is a challenge like bringing remote cooling tower individual cell temperature data to the control room, remote ash slurry line temperatures etc. We will place the receiver at the height of boiler where from it will be in LOS for all the respective senders.

bye bye,
S. Bera / Vindhyanagar
Read the full post
Show less
It transmits data
It transmits data
Receiver schematic
Transmitter schematic
For precise sonar level I've used a multi point ultrasonic HC-SR04 set. It gives a much better result.

Loading comments...