I've built a water tank monitoring system that logs water usage at a tank 1500' from our house in hilly terrain. The most direct path has lots of trees and is far from line-of-site. Currently, I take a laptop on an ATV to download the data but I'd like to access the data from the house. There is no possibility of pulling a wire to the site but there is (minimal) power at the tanks. (There is power at the intermediate pump house although it is off the most direct path.)
Can anyone shed any light on whether a high power wireless systems such as from EnGenius would work through trees and not line-of-site?
I've also though of using wireless to the pump house then a power line system from the pump house to the tanks. Anyone familiar with the Amped Wireless AV500 power line network adapters?
I've thought of using a MiFi cellular modem but they cost a minimum of about $40/mo. ($40/mo will buy a LOT of miles on an ATV.)
Sounds interesting but not available yet. My depth monitor code requires a mega since it uses multiple sensors (2 ways of measuring water usage, temperature sensor, and pump on/off timing). I've also built in a mini-web server that allows me to configure and debug it while it is running. (I was a computer scientist for 30 years so tend to go overboard on the software side.)
I might be able to use this (not-available-yet system) as relays but at least one would probably have to be solar powered.
Buy it from ebay for about $4 each. You have to wire up your own antenna, so you can look into how to make the signal more directional. It communicates with Arduinos visa SPI.
What are you using for the water level monitoring? An Arduino?
Buy it from ebay for about $4 each. You have to wire up your own antenna, so you can look into how to make the signal more directional. It communicates with Arduinos visa SPI.
What are you using for the water level monitoring? An Arduino?
Tried a free slate pressure sensor but got a lot of drift in the readings that I couldn't get rid of. Currently using a 32" etape but it really isn't long enough. ( Chris Milone, etape's creator, has been great in helping out with some issues with the environment.) I am about to hook up a high res maxbotic ultrasonic sensor. It is built for this purpose and is very accurate (1 mm res). (There are 2 3000 gallon tanks joined at the bottom. They are 7' high so 1 mm resolution should be fine.
I originally ran on an arduino Ethernet but ran out of code space even after using every memory cutting technique I could find and inventing some of my own. I now run on a mega.
Powerline ethernet is one option, but you may have issues with the range and noise generated by the pump. Ubiquiti wireless radios could probably make it. Put a 2.4ghz Nanostation Loco M2 on each side. Theres no way to judge a NLOS link without testing it.
I've been using Omega PX309 pressure sensors to monitor water tank levels for many years (we are on a rainwater system for home and garden) and although they are expensive, they are absolutely rock-solid. I can easily resolve a 1-gallon difference in a 3000 gallon tank over a long period of time, and detect the tiniest of leaks.
The only difficulty I've ever had is with local air pressure differences. The 1500 gallon pressure tank for the house is located about 1000 feet away, 120 feet up on a hill and I monitor its level by measuring the pressure at the house. When the wind is blowing, local air pressure differences sometimes lead to false readings, but for the most part they can be ignored.
You coukd try using nRF24L01+ 2.4Ghz transceivers.
For the distance and conditions you'll need the high power amd pre-amped ones with external antennas. Use some commonly available wifi directional antennas; waveguide, yagi or dish, and you might be able to punch through the signal. Also run them at 256kbs for maximum sensitivity and range.
You could probably get some cheap yagi antennas for less than £30 and the radios wojkd be about £15 the pair. Potentially get it going for around £70 with some pigtails and gender adapters factored in.
If you can get hold of some old satellite dishes then you could even try making your own biquad element to mount and do some diy antennas of about 20dbi gain.
There is plenty of info about on extending 2.4Ghz for normal wifi, which would be applicable.
I thought 900MHz or 400MHz would travel farther than the 2.4GHz that the nRF24L01 runs on. This application isn't transferring a lot of data, so the longer wavelength wireless of the RFM69HW might be better than the nRF24L01. With the RFM69HW, you can select different frequencies (415MHz or 915MHz) during configuration, so you can try out the different frequencies to see which might work better.
The RFM69HW is only $4 per transceiver. Just a bit more expensive than the nRF24L01. It's newer than the nRF24L01, so there's not as many examples of people using it. But it seems like a very capable wireless comm module.
jremington:
What is a "free slate" pressure sensor?
I've been using Omega PX309 pressure sensors to monitor water tank levels for many years (we are on a rainwater system for home and garden) and although they are expensive, they are absolutely rock-solid. I can easily resolve a 1-gallon difference in a 3000 gallon tank over a long period of time, and detect the tiniest of leaks.
The only difficulty I've ever had is with local air pressure differences. The 1500 gallon pressure tank for the house is located about 1000 feet away, 120 feet up on a hill and I monitor its level by measuring the pressure at the house. When the wind is blowing, local air pressure differences sometimes lead to false readings, but for the most part they can be ignored.
1 gallon in 3000? How do you do that? It exceeds the accuracy of the sensor and the arduino. Perhaps a 1 gal / time period measured over several time periods.
It exceeds the accuracy of the sensor and the arduino. Perhaps a 1 gal / time period measured over several time periods.
l use the analog version of the PX309 (0-5V output) and an external, 12 bit ADC. I take 64 readings of the ADC and average them to produce a 13 bit result, resulting in readings in the range of 0-8191 for a 3000 gallon tank. After subtracting off the zero water level reading, the end result is still better than 1 part in 3000 accuracy, which I have verified by dumping a known quantity of water into the tank. The Omega gauges are high quality measuring instruments!
I've built a water tank monitoring system that logs water usage at a tank 1500' from our house in hilly terrain. The most direct path has lots of trees and is far from line-of-site.
You probably won't find anything wireless that is cheap/easy. Natures warning sign is you don't already see people using this type of wireless setup.
It exceeds the accuracy of the sensor and the arduino. Perhaps a 1 gal / time period measured over several time periods.
l use the analog version of the PX309 (0-5V output) and an external, 12 bit ADC. I take 64 readings of the ADC and average them to produce a 13 bit result, resulting in readings in the range of 0-8191 for a 3000 gallon tank. After subtracting off the zero water level reading, the end result is still better than 1 part in 3000 accuracy, which I have verified by dumping a known quantity of water into the tank. The Omega gauges are high quality measuring instruments!
Cool. Didn't think of using an external ADC. I have to measure the pressure at the tank since we have multiple pressure
reducers between the tank and the house. Can the sensor just be dropped into the tank or does it have to be mounted through the tank wall? It isn't clear from the data sheet what the mounting options are.
Can the sensor just be dropped into the tank or does it have to be mounted through the tank wall?
I didn't see an waterproof sensor and cable option, although I know they exist. I mount my sensor in the exit pipe so as not to compromise the tanks. In that case you have to be careful to measure the pressure when there is no flow, as the flow significantly reduces the pressure.
I think a network of XBee 100 mW 900 MhZ transceivers would work for you. Set up the network of XBee so that each unit sees the unit behind and in front of it. The XBees will set up the necessary networking if you have the correct setup so it is a pretty easy solution for low data rate data collection systems.
Small solar cells and batteries could be used to power the intermediate transceivers and to conserve power the radios could be turned on once an hour to broadcast the collected data and then go silent.
It exceeds the accuracy of the sensor and the arduino. Perhaps a 1 gal / time period measured over several time periods.
l use the analog version of the PX309 (0-5V output) and an external, 12 bit ADC. I take 64 readings of the ADC and average them to produce a 13 bit result, resulting in readings in the range of 0-8191 for a 3000 gallon tank. After subtracting off the zero water level reading, the end result is still better than 1 part in 3000 accuracy, which I have verified by dumping a known quantity of water into the tank. The Omega gauges are high quality measuring instruments!
Which particular sensor model did you use? For my tanks, 5 psi would work but to get that in its most accurate form it costs $750 which isn't worth it for me. I don't see how the less accurate models would work.
I use the general purpose PX-309-05GV5V (5 PSI, 5 V) and PX-309100GV5V (100 PSI, 5 volt) sensors, around US$ 225 each. The absolute pressure accuracy is not important to me because I calibrate the gauge/ADC based on a number of comparisons of the final, averaged ADC values with the actual tank contents.
I find that with averaging, the gauge repeatability and sensitivity is in the vicinity of 1 part in 4000. Omega says the gauge output will drift by a fraction of a percent over a long period of time, but since I'm interested in relative measurements (for example, for relatively short term leak detection), I'm quite satisfied with the system performance.
Finally, as I mentioned in my first post, in the one system where there is a long run of pipe to the pressure measurement station, there are sometimes significant differentials in the local air pressure at the tank and at the gauge. These fluctuations can overwhelm any variations in the gauge output itself, but since they usually coincide with substantial wind velocity, they are easy to ignore.