Posts made by wallyllama

How about a sealed tube with 2 copper strips down the inside, then measure capacitance?

Kind of like this:
https://www.instructables.com/id/Capacitive-Fluid-Level-Sensor/

Hmmm. Looking up scintillometers, I see they are advertised as "large aperture", so you may need an array of photo[diodes,cells], which adds a lot of complexity, matching, summing etc,but maybe you could get away with a condenser lens.

Wikipedia has some clues. A scintillometer which measures fluctuations in refractive index of air, just add math.

Satellite imagery +math

Lysimeter - measure the weight of a column of soil, im guessing dry soil is lighter.

All of thoose sound like lots of money.

A scintillometer seems doable as diy. For an area like a tree, I'd start with a laser diode ( i dont know which frequency, but as a complete guess id try infrared, possibly red, and stay away from green and blue), the other end would be either a phototransistor or a photocell. A low noise (probably high gain) amplifier and a quality anolog to digital converter (maybe a 24 bit sound card, probably a pi zero attached not an arduine or esp)

This is all very speculative(pun intended) on my part, it could be completely wrong, i am guessing based on the few words in wikipedia.

Wikipedia

I know WAF is not a joking matter, but I immediately heard Greg Kihn sing, " my WAf's in jeopardy, baby".

Mycroft has some of that. www.mycroft.ai

@naty6458 no need to be sorry. I keep hoping there is a better answer than what I found, and each time it is asked I learn something, so there is at least one person that is happy you asked.

My plan is optocouplers, maybe a rechargeable coin cell to help power them. I want to connect to the pins for the rf receiver. I should be able to send the same code the remote uses to turn it on and off. Since I wont be sending codes often the coin cell should be fully charged most of the time, so the low capacity (~23mah) shouldn't cause a problem, unless someone flicks the lights on and off repeatedly.

I'll power the arduino/rs485 module separately.

This is the $64,000 question. I've seen discussions about it, but no solution. You could build a separate power supply that floats on the ac like the built in one, they work by leaking a small amount of current and can cause flickering, especially with LED lights. Two supplies double the chance of flickering.

If your arduino can spend most of its time sleeping, you might be able to charge a battery or supercap to run it.

There is a lot out there on lowering power consumption on arduinos, remove LEDs, lower clock speed, etc. Maybe a bare bones chip, with just the connections you need.

Or

Use optocouplers to connect the arduino to the livolo, and use a normal power supply for the arduino.

If any of my old x10 gear looked that good I would just make new pcbs and relabel the buttons.

https://www.maximintegrated.com/en/app-notes/index.mvp/id/148

Here is maxim's guide on wiring for 1-wire. There are a few things to note, one is the "weight" of each sensor's cable. Read that section closely. Another is hubs. Hubs are basically a 1 wire switch that connects each segement of the network 1 at a time, so you can essentially walk though each ray of your star. The third is the suggestion above to use multiple pins on the arduino and just make several simpler networks.

Is it any 5th temp sensor? In other words could it be a bad temp sensor? Also 1-wire doesnt like 'star' wiring, are all the sensors wired with short connections to the main cable? More like ,, and less |_| and for sure not |/

I may misunderstand how they work, but I've had the idea that (super)caps are better in short term high current situations. Like regenerative brakes for robot lawnmowers, not slow drain low power devices. It sounds to me like the life boat is the lifepo4 batteries. I'm all for experiments, so please dont stop, but unless the environmental conditions are intolerable for the batteries, I'd say you found your answer.

So these radar motion sensors are actually quite impressive, the ones with just analog output only have 1 transistor, the rest of the circuit is formed by the pcb traces. Weird, but it works. They can be hacked (sawtooth wave for power) to give a distance reading instead of doppler. The signal processing is too much for an arduino, but it is doable.

The other ones actually use the same chip as the most common pir devices.

I'm going to go a bit conceptual here, riffing on some ideas that I believe I got from @dbemowsk. A multi-pronged approach may be useful here, maybe bt for the car and bicycle, amount of time since any PIR device triggered, time of day, etc. Then using the "smarts" of a smarthome a reasonable estimate of home/away can be gained.

I dont think an A.I. like the googazon has is required, but something smarter than a single node might be apropriate.

Would somesort of passive rfid tag attached to your keychain work?

Edit - a quick google search for keyfinder came up with a variety of bluetooth/gps tags, some in bulk for under $2 USD. Probably come with an app, but that could likely be subverted into something mysensors compatible.

Is the level shifter closer to the 5v arduino? Or the 3v? Maybe losses in the cable are dropping the voltage below 3v. So maybe level shift after the cable run.

https://github.com/dukelec/cdbus_doc
This one has some is interesting, not sure if it does multi master or not. There are a few projects that seem to do multimaster/collision avoidance, the trick it to find one with low enough overhead and automatically retransmits if there is a failure, and for this project isolates this enough that my sensors doesnt need restructuring.

Good wiring is crucial so you can use higher speeds. The common ground wire in addition to the differential signals seems almost required.

I think you have found most everything that I have. Something akin to csmacd is what I was thinking. Listen, transmit, back off for random time if a collision still happens. Of course full duplex rs485 is another option, hard to find off the shelf modules.

I think this is the library I had found before.
https://github.com/MichaelJonker/HardwareSerialRS485/wiki

@nofox absolutely that is one of the problems. Polling as you describe could be a fix, another is having each node listen before transmitting, there is a (not mysensors) library that does that, CAN bus( a mutli master 485 comaptible bus) is another option. I think that running a common ground to all the rs485 adapters is also wise. This is all in theory as I havent wired up my home yet....

@schlog you might be able to use a modified swr circuit, an arduino could generate a weak forward signal, and then use an analog input to measure the reflected signal ( sort of a TDR) instead of a meter, the impedance match will possibly be bad(maybe even on purpose to enhance the effect), but each type of antenna should be different enough that you can tell both if it is connected, and which antenna it is.

Another thought, if this is permanantly connected, then you should only need to know the positions of the switches for the transmitters and antennas.

Btw My call is:
KC0QWL

@schlog it sounds to me like you want a basic antenna analyzer. Commercial ones arent cheap, and probably do more than you need. I did find a diy analyser for HF by K6BEZ, maybe it will inspire you.

http://www.hamstack.com/project_antenna_analyzer.html

A time domain reflectometer might work, they are used to measure the length of cables, and dont require a return wire.

it may work to measure the charge time of a capacitor, but I think the far end may need to be grounded.

If you are just trying to see if the antenna is attached, then no antenna would be infinite or open, and an antenna would be somewhere around 50 ohms, wouldnt it?

@constantine-poltyrev i dont have code yet. I havent really started on software yet. My thinking is to start by using the adafruit code as a base, set up the sensor in a likely spot and modify the code to stream data to a pc or r-pi, maybe with a web cam next to it for comparison. Id look to see if there are patterns visually that are helpful, but my real hope is that since the pixels are temperatures, the ones with people are at people temperature, so around 37°c.

Maybe the first filter would call out pixels in the 35-40 degree range, and send 0 for the rest. If that gives a spot that has few false readings it may be good enough for "there is at least one human present" or not readings. A glorified pir.

Likely though a person will be more than 1 pixel, or worse part of more than one pixel, how to know if two cooler pixels represent 1/2 a person each, then there is the problem of the sun heating something to human temperature. My gut feeling is these are simple enough that existing solutions can be adapted. I dont know what those are yet, maybe bayesian filters, maybe some sort of computer vision.

A hardware product called openmv may provide some inspiration. It uses something called micropython and opencv. With 64 pixels we have a simplifies case, but I think an arduino is not going to be enough, a raspberry pi or maybe a nanopi neo should be sufficient.

I think that is the current contents of my brain on this, experiments are the next step. Starting with streaming some data and looking at it.

@ŕähûł-śïñģh if you are successful you'll be rich. There are simple detectors like pir, and complex like cameras with computer vision software. Im working with amg8832 chips with 8x8 grid of temp sensors, essentially a low res camera. I looked at radar "pir" modules and thought modulating the voltage would give data that can be processed, again complex, combining pir and ir door sensors has been suggested. None of this is my original idea, im just trying to piece something together.

Side note. I did a quick search, there is a cec library for arduino!
https://github.com/stefslon/cec-arduino

TL;DR. To me the history of computing is the story of centralizng vs. decentralizing. As a "peripheral" becomes more powerful it becomes a center, that is then decentralized from....

Centralized

Pros
cheap at the edge, center can more easily share resources (economy of scale), one upgrade can improve the whole system.

Cons
edge is less flexible, sharing resources can lead to contention, center is more complex and possibly harder to troubleshoot. Probably more wires between edge and center.

Decentralized

Pros
nodes are more flexible and easily adapted and customized to the location, smart nodes can respond faster as it doesnt have to travel all the way to the center, wait for attention, then go back (like reflexes, they only have to travel to the spune, not to the brain). A single failure is less likely to take down the whole system. Can probably use radio in place of wires.

Cons
probably more expensive overall, more points of failure, network to center is probably more complicated, required better wiring.

My personal preference is hierarchical. I like nodes that can operate independantly if disconnected, hubs that aggregate and do some processing (like a spine), and a center that has the big view and uses all the inputs to make big decisions. Call me biased if you will, but it is just the way I am made. :c

Use adequately sized wire to make sure you dont have too much voltage drop. There are calculators all over the 'net.

If you have some particularly long runs, consider dividing the nodes between two ( or more) centralized power supplies. It is a compromise, but may be a good one.

@sam9s what you are describing, while a nice way to control things, has the same basic flaw as a PIR device. You have to tell it you are there. The PIR(alexa) knows what room it is in, but you have to signal it some how. Alexa is signaled by a voice command, PIR by motion, but if you are quietly reading a book both of them forget you are there. Alternatively you can have them assume you are there for a set amount of time after the signal, or until they get an off signal.

The trick is to get them to detect you without actively addressing them. If Alexa can detect breathing, or heat or CO2, etc, then it would solve the problem.

you can combine alexa with door sensors. If alexa is triggered and no one has left the room then someone is still here. That is the idea that @dbemowsk pointed out earlier in the thread.

The key to accuracy is repeatability. If you get the same reading from the same sensor for the same conditions over and over, you can calibrate and have great success.

+- .5°c means 1° of range, is that relative to the actual reading? Or the variability of the readings time ove time? The first one is correctable, the second sucks.

Positioning would be critical, but these sensors always make me think of a sort of automated health diagnostic system. It checks your temperature and reports if you have a fever.

I would agree the graph shows resistance vs. grams. It looks your 5kg mass should read about .5 ohms. So if you want force (or wieght) multiply 5 x 9.81. The nice thing about this is f=ma is a linear function so you can make a similar graph for resisitance vs. force.

The force value appears to be logarithmic compared to the resistance read this is where the heavy math is, reporting force or mass is a matter of constants. Hopefully the data sheet gives the formula.

The surface will matter. Nearby objects will matter if you make it sensitive enough to go through 30mm of material. It may also limit how close together the pads can be. I think the important bits are the values of the resistors and the dielectric of the material.

A good general read on this:
https://www.allaboutcircuits.com/technical-articles/introduction-to-capacitive-touch-sensing/

Open energy monitor is both a commercial product and open source hardware/software, they have pages on using ct clamps.

https://learn.openenergymonitor.org/electricity-monitoring/ct-sensors/introduction

There is also a commercial product that uses 2 ct clamps and technology derived from speech recognition to analyze the powrer signal to detect what devices are running and even when appliances are gettimg ready to fail. I'd probably own one of these, but I am allergic to clouds.

The alternative answer is that you chose a hobby that will take up all your spare time and burn all your extra money.

I've been thinking of them as low resolution cameras that can see in the dark, but there may be more clever ways to think of it that I havent come across, but any computer vision algorithm would work. Motion detection for sure.

I have a fairly large living room with a high ceiling, if I mount one in the center, i should cover most of the room, I estimate a person would be about 1 pixel at the floor. The coverage is a pyramid so at the edges height is zero. Corner mounting like the video shows would probably fix that.

I think these work work better than my idea for a giant capacitive touch screen.

I have recieved 2 amg8832 chips for experimentation after some delays because of import rules. The labeling on the bagggie says they need to be mounted within 108 hours of opening the bag. I would recommend getting a breakout board and not raw chips.

I'll post in the appropriate category any additional progress.

@maggie Im assuming the cubes can be in any arrangement as long as faces touch flat. With 9 cubes you have 56 unique sides, if you had a magnet on each face so they make good contact, each cube would only need to know which face is connected to each of its 6 faces, maybe generate an interrupt when two magnets touch, the faces could exchange numbers.... lots to fill in to make this work, but it may be a start. Im not sure where mysensors code helps either.

Is the fact you didnt get it working an OOPic failure? (Not really a comment on what was and probably still is a difficult project, I just cant resist puns)

I dont know. I guess Im wondering what you want from this. Do you care about pets vs humans? Adults vs children? General person count only? If the goal is to not shut the lights off on people then the last one is good enough.

A thought experiment. 5 known people and 1 pet in a room. 1 living being leaves the room. IR door sensor in place. What information do we want about the new situation? And what sensors would we need to gather it?....

@gohan Openmv has a single board camera with opencv and micropython, another option.

I think @dbemowsk idea of door sensors fits nicely with mysensors as he has said. Is there a more appropriate forum for the more complex devices that anyone knows of? Im thinking if I come up with a node, I can add it like any other, but there will be a lot of talk that ends up a bit off topic.

On topic, are there controllers that are more amenable to the kind of combining of different nodes to identify people that we are talking about? I've used Mr. House for other things, and domoticz for my one test node, not enough to really have an opinion.

@dbemowsk interestingly, as I thought about these ir cameras, they may require a smarter node (maybe something like a nanopi neo2) to preprocess the data, and then a person tag may be useful.

For example 64 pixels, 2 bytes to encode temperature value, 1hz refresh would mean 1280 bytes/s. Which If I have been reading this right, is pretty high for mysensors. There are some ways to reduce that, but it is unkown if an arduino could keep up.

I've mostly been doing research on sensors, and only built one node and a gateway, so alot of what I have been saying about my sensors is assumption.

Does it have a defined method of extending the data types? Or a board that decides? A Glorious Leader we need to cajole? Maybe "user defined" types?

Im kind of in love with these IR array sensors, and I'm probably not objective about what is best for mysensors as a whole, but I have boxes of opinions I'd like to get rid of, so just ask if you want some

https://forum.mysensors.org/topic/3528/bed-occupancy-sensor

Did you see this project? Seems like it could give a lot of data on sleep.

@dbemowsk said in A smart home vs an automated home:

So looking at the MySensors end of this, would it be too far off to think of adding a new node type, "person". A person node could have customizable properties that would allow you to define different useful bits of data related to that person.

I'd like to hear more about how you would use it. Below is my two pennies worth.

My thinking is that mysensors is a transport for relatively simple data, like state, value, counts etc, things nodes would need to set the environment up, or report back to central command.

A complex object like "person" could have all kinds of attributes and preferences, which would modify values sent to nodes. Example the curtain controller knows to open during the day, close at night, and maybe close for an hour at 10 am in the summer when the sun shines directly in and heats up the house( could also be a light sensor), but if the weather says it is clear, and kent is in the living room and it is night open the curtains up, and that would be an override coming from central. The node controlling the curtain doesnt need to know it is me in the room, it just needs to accept the modifiers.

I say this mostly because as @gohan points out arduinos arent terribly powerful, and telling them too much info may just confuse them.

I liken it to the body. E.g. your finger doesnt have to know if you are walking up as you are pushing a doorbell, it just extends on command and reports that it made contact, moved forward slightly and hit a stop. Your spine may get involved if the finger reports excessive heat, or something gooey on the switch, and pulls the hand back in reflex.

I think @dbemowsk is hinting at something that fits my understanding of "emergent behavior", individual simple things interact and create more complex results. How many, and whom are different questions. Counters in doorways plus a list of whose phones are at home, maybe add in some hostorical data of who likes to sit in which chair. There are probably better combinations, but that is what I got from hisnrecent comments.

@NeverDie data sheet says 7 meters max, there is probably enough margin, at least for typical room sizes in the US. I think the 60° fov will be the bigger issue, getting coverage. Imagine you place the sensor in the center of your ceiling. The room is square, 14ft on a side and 8 ft high (~4X2.4 M), the sensors field of view would exactly cover the floor, but it is shaped like a pyramid with the sensor at the peak, so if you stand flat against a wall, only your feet would be in view.

@gohan's suggestion of bluetooth tags doesnt have that problem, it can be seen anywhere the signal gets to. You can have multiple detectors for coverage and triangulation. If you have a smartwatch or phone you always carry then youndont even need a separate tag. It is relatively cheap and simple, and most of the tech is done already.

(Now here is where I loop around and start spinning in circles) I dont want to have to carry anything, it should be possible to detect my presence by all the signals bouncing off me already, like light, or ir, or wifi, or radar, then the googling happens......

@gohan true for the larger goals, but this sensor is 64 pixels (256 w/interpolation) and we need to track a dot, i think an arduino could gather the data, do a bit of preprocessing, and (the mysensors part) transmit the data to a raspberry pi for "whole house" tracking.

This is pretty low res and I think a pi could handle it. If not, Intel has a movidius usb stick meant for computer vision/ai acceleration, I believe opencv has been ported to it. So while this is on the edge, some of the blood has dried.

The other plus is houses move slower than cars, unless people are running indoors, a 2 to 3 second refresh rate should be accurate enough.

This is a large project and mysensors would only be a portion of it, so for now I'll try to limit myself to talking about how a node based on this sensor would work and if it fits into mysensors properly or not. There is plenty there to discuss.

@dbemowsk again sorry for hijacking your thread, I'm going to look at the guides for submitting a node to openhardware.io, i dont promise I'll be fast so dont stop working your own ideas.

The amg8833 has an 8x8 grid, and a 60° field of view, so if you have 8' (2.4m) ceiling that will cover a square with 14' (4m) sides at the floor. One pixel will be about 1' 9" ( 44cm) at the floor. That should be plenty of resolution even without interpolation. I suspect interpolation could give an effective grid of 16x16 at least, maybe more.

Careful planning and mounting in a corner or on a wall would have some trade offs, but might allow for covering a larger area with one sensor.

One trade off is identification. Is that heat blob a person or @gohan 's cat? That might be doable, but is it Mom or Dad, or teenager would probably need supplemental information.

Stationary heat sources, lamps, vents etc, could be filtered out, in probably several different ways. I have some large windows that may blur the data, but this isnwhere situational awareness wouldmc9me in. E.g. if (curtains == open && tod == daytime) then apply filter to pixels x through z, maybe time of year etc.

Other obstacles would probably look like cold spots and unless they are large wouldn't affect detection of people. They might dim a bit, so maybe a filter would be needed here.

This is quite doable. I've been thinking about it for a while and seeing usable sensors for effectivley 1/2 price has me a bit excited. I appologize if I have monopolized the podium a bit.

@NeverDie no it was an amg8833 breakout and at adafruit not sparkfun sorry, $39us. Mouser and digikey have just the sensor form$22us in small quantities.

https://www.adafruit.com/product/3538

Once a method of sensing people is selected/found, then mySenors can be used as the transport layer. This leads to the question of the actual "smarts". The various mysensors supported packages seem to track state, allow control, and have scenes, which are good data and tools for the smarts to work on, but dont seem to be smart themselves. Am I overlooking something?

Commercial products use the 'cloud' to gather a lot of data from local devices, and create an AI of sorts that local devices then query for the appropriate response to specifuc conditions. I'm not interested in sending all my data to the cloud, so Im interested in completely local solutions.

Again this doesn't currently exist(that I know of), but many pieces do. Some are just pieces (hadoop for storing data e,g), some are partway there (mycroft ai e.g.), some have large backers (movidius ai accelerator). Some assembly required.

Are there more complete solutions that I may not know of?
What goals do others have?

@NeverDie nice! Sparkfun has a breakout that is 20% cheaper than just the omron sensor. This is is getting closer to my price range, the radar modules are cheap and might be fun, but this would likely yield a working solution sooner.

@gohan do not judge anything by how it bothers cats. In my experience the only thing that bothers them is if you dont feed them or if you throw away their cardboard box.

The rc car would should work for most critters, really anything that makes a sudden motion should.

Side note raccoons are the only animal less bothered by people than cats.

@dbemowsk search "see through walls with wifi". I suspect it could be mounted to the wall and covered with art or something transparent at 10ghz, or with loss of sensitivity in the wall itself.

http://people.csail.mit.edu/fadel/wivi/

Ok this is the one I was looking for

http://witrack.csail.mit.edu/witrack2-paper.pdf

Maybe something like a lytro camera sould be adapted. If the data could be gathered live and processed quickly.....

@wallyllama https://www.researchgate.net/publication/261150852_Doppler_radar_sensor_for_occupancy_monitoring

@dbemowsk there is some promising research using radar, single source, with multiple receivers in a T shape, in theory it is 3dimensional and whole house. Ill see if i can find the paper I read, i think there is similar work that uses of the shelf wifi gear.

Ive also looked into (and not found enough to do anything with) lower frequency rf,but it seems to require running wieenaround the room, and probably a grid on the floor. It also seemed similar to capacitive sensing, but i dont understand the science enough to truly claim that.

I tried to find research on what EM people absorb and emit, but didnt find much publically available. I dont like the idea of carrying a fevice around for tracking like this.

There are methods that are sensitive enough to detect respiration, used in fire detection that may be of use. Maybe extremely sensitive microphones that can detect heartbeats or footsteps. Some serious signal processing there, which maynlead to DSP devices or maybe some gpu acceleration.

Any of these could be doctorate earning projects, but I keep hoping . Things like intels movidius usb stick may make this stuff more accessible.

Sorry these arent much more than speculatiom, but it what I have found when looking for similar solutions.

Oh 2 more, multi "pixel" ir, omron makes some that are 4x4 or 1x8, several of these, or some kind of scanning to get enough resolutions, or maybe webcams and opencv type computer vision.

I've come across a diiferent rs485 module labeled XY-017. T claims to be "stable" and or "reliable" and have a uart. Has any one tried them? All sellers I've found seem to use the same (probably google) tranlation, and there is no mention of the chip other than "imported". I'm guesing they mean not counterfeit, but are they max485 or other?

They have a trrminal for A B and ground. Ive seen ground recommended for long runs. Any thoughts? With a uart it looks like it handles the di do de re lines for you, which may be a blessing or a curse, certainly would need changes to mysensor code to use them.

My plan is to use rs-485, but my experience is all still theoretical so confirm anything I say here.

Higher speed and longer distance require better wiring, but a house should be well inside the limit even, if you snake a single wire all the way around the house. According to this link you may even be able to get away with some star topology
https://arduino-info.wikispaces.com/RS485Info

It seems the biggest issues on this forum with rs485 are collisions, and correct use of terminating and pull up/down resistors on the line. There is a lot of information and solutions to these out on the net. A separate ground wire is often one of the recomendations, but the inexpensive modules i have only have A and B, so im not sure exactly how the ground is wired in.

My thinking is I will divide my house into zones, with a separate rs-485 net for each, reducing length, number of devices(collisions), and possibly less grounding issues.

Ive heard the max485 chip isnt the best, so possibly buying better modules would be helpful. Maybe rs-422 or CAN might be other options.

Do pir modules trigger when power is first applied? If you are cutting power to it when you sleep, maybe it needs time to settle before you take a reading.

Shouldn't that be 192.168.1.7, not 192,168,1,7?

@Andrei-Călin-Tătar b1 is the buzzer. When you hold a touch pad down for more than 5 seconds, the switch goes into learning mode and the buzzer makes a sound.

@Andrei-Călin-Tătar said in livolo Glass Panel Touch Light Wall Switch + arduino 433Mhz:

Also the 433MHz radio signal is decoded by the touch MCU not by the dimmer MCU.

The touch mcu does this for non dimming temote modules also. I think it uses mostly the same system for the 2way switches also.

Under the mixed section, "double antenna", a passive re-radiator may work. I dont know enough to give details, but it is two antennas connected by a wire. You may have to drill a hole in your wall to pass the wire through.

Short answer about the additional components is yes. The datasheet has a recommended circuit and about 1/2 the parts listed are noted as required.

@parachutesj so I mostly just spewed random thoughts, And looking at them again I came up with some more focused, probably not original, ideas.

Some kind of watchdog/debug output is needed on actuator type nodes(may already exist)

Rtl_sdr dongles are cheap and some signal analysis tools exist. Someone that knows what they are doing (not me, I just have a big mouth) could probably cook up some useful tutorial that would shed light on what looks to me like the biggest issue in mysensors, signal reliability.

In the days of analog tv/radio dead spots and multipath were easy to detect. Digital isnt as susceptable to multipath, but isn't completely immune either. An rf field meter might help. Directional antennas probably will help. If these nrf devices can measure rssi it may be good to hook up and led as a signal quality indicator of some sort. It sounds like you have a strong signal near the problem, could be overloading the receiver, could be IM ( intermodulation, not instant messaging). It could be an rf shield on all but the antenna may help.

My sensors may have some of this already, if so ignore me, Im using rs485.

Note to self, check if "myactuators.org" is available.

Filter caps on the actuators? I think someone mentioned this, they can be a source of noise in the power, or rf hashing.

An rtl_sdr sort of device might be a cheap rf signal analyser.

Assuming the "in" pin is ttl on your relay as well, then yes. You'll want a 5v arduino, or a level shifter. Some ttl will work with 3.3v logic, but it is outside my knowlege what the long term effect is.

@moskovskiy82 the picture looks like a 5v relay. Unless the ground on the power supply is the same as the common on the load side
you'll want to remove the jumper.

This is the official (US)National Weather Service definition of wind gust. It looks like they dont report gusts until wind speed is above a certain amount. I like to overcomplicate things, so I would report the average, max in a minute (like Yveaux), and a standard deviation(just to be fancy)

https://graphical.weather.gov/definitions/defineWindGust.html

@MasterCATZ said in help needed with multiple temp sensor data logging project:

@wallyllama thanks that is what I feared, and would have made the job a lot easier
how ever I was thinking about doing it the other way to keep the other probes already made backward compatible as they used pins 1-4

just trying to weigh up the pro's and cons
also out of all those 3 way splitters only 15x actually have all pins functional ...

The big "con" is the risk that someone will come along and try to use it in a standard way.

so trying to think of a way to still use the dodgy ones by having redundancy's in place

I think the way the PSU is wired up at the moment is the correct way
things don't seem to be stressed out this time

I am also raising my polling times as I think I read somewhere it takes the sensors 750ms to perform a function and uses 1.5 mA each?

Sensor Poll Period: 1000 to 30000
Switch Poll Period: 100 to 1000

my way of thinking this should still have all 60 sensors read in 1 min?
60 sec / 1000ms for Switch Poll period?

There is a function that tells all the sensors to start a temperature conversion, you could then read each one individually. So it would be 750ms + (60 * pollingtime) to get them all. I dont know if you trade speed for reliability or not. You may be able to interleave conversion requests and polling.

Convreq1 poll30 convreq2 poll31 8convreq3 poll 32 .... convreq 60 poll29 - loop

Maybe overkill if 1 reading per minute is enough

It looks like the standard for power over ethernet uses a twisted pair as a "single wire". So blue/blue-white is + and brown/brown-white is -

Seems like a reasonable way to do it, even though you arent wiring ethernet.

@mtiutiu I also assumed that I can spell and proofread. Thank you for this info. It is more helpul than a month of reading google translated russian sites about livolo.

I am mostly trying to push data to the livolo, so the power to drive the optocoupler was coming from my ftdi. Inassumed (wrongly?) that with a reisitor the data pin was vbing driven about the same as if a radio module was there.

@MasterCATZ

I found that out, it has to do with how they pull power when the switch is off. I did get some optocouplers, i did get some results, but I havent make it work yet. I did find the difference of regular and 2way is a diode and resistor. (And the buzzer). Oh and like Tesla and edision, the rule is to work with one hand behind my back when AC is live.

@gohan if not you could add.... wait for it.... a capacitor. (Weak, I know, but I couldnt resist)

The project you linked to uses a feature of the atmega to speed up the a/d sample rate, leaving little processing for other work. It also sends a 512 byte payload, whih is nice for graphing, but Im guessing you'd like a count of lightning, strength, and distance. This one would give strength and count. If I were doing this, I'd use 2 arduinos, 1 as a detector, 1 to parse the data and act as the mysensors node. Look through the apu docs. There is probably a reasonable data type to send the processed data to domoticz or what ever you like.

If I were really doing this, I would look for a more robust detector using discrete components. And connect it to an arduino/mysensors node.

This one may work. (Maybe more complicated to build)
http://www.techlib.com/electronics/lightningnew.htm

An old am radio tuned to an unused frequency, could be connected to an arduino analog pin with some resistors and capacitors, and sampled at 9600 samples/sec or less and could be done with 1 arduino.

If you search "1-wire hub", there are premade options, and diy.

There are 1wire "hubs" that would help, they basically turn on or off parts of the network, so each run could be the max distance.

Each network would need only 1 digital pin and resistor, i think you could define multiple pins and do a onewire.begin for each (using different names each of course) i think you would run out of memory or clock cycles before you ran out of pins even with a pro mini.

Using a dedicated 5v line is probably better than a capacitor and parasitic power, especially with long runs and lots of sensors. Unless you can live with really slow polling times. With the complexity you are talking about, I wouldn't be surprised if takes a couple of minutes to poll everything reliably with parasitic power. If it works (i havent tried it) using several pins on the arduino, each controlling a 1wire network might help out considerably. You could service 1 network, while the others are being "charged".

This is a nice simple solution. There is one downside that is worth noting, pushing multiple buttons at the same time will throw it off. If that isnt a concern it will serve well.

Finding things to make into cases is fun. Here is my raspberry pi.

Maximintegrated

Read this, there is a formula for what they call weight, and they talk about stubs. That would be the part from the main bus wires to the sensor.

The ds2450 adc is out of production. The ds2438 battery monitor is ehat you'd have to use.

This person was going to use the adc, but ill bet the batter monitor could be made to work.
Reefcentral

For connections, a 3 port rj11 might work, in from previous tank, out to next, and one to the tank sensor. I think if you keep the line to the tank under 1meter you are ok, but i have never built a network as long as you are proposing.

Other answers. 1 resistor per 1-wire network.

I believe you can still get a/d converter chips, they could probably be used to read an orp or ph sensor. Water level could possibly be read with a capacative sensor, but you are going run into your power problems again.

@MasterCATZ 1-wire hubs are active, sort of like usb. You dont want your one wire network to be a star topology. You want some thing like railroad tracks, long wires(rails) with short connections to the sensors(ties). The guides i mentioned really are worth a read before you plan too much.

@wallyllama awww the spaces on my cheap ascii graphic got crushed.

how do I go about Daisy chaining the probes

1-wire devices (actually 2 or 3 wires) are designed to be linked in a daisy chain, or really just connect in parallel. The wires are data, ground and +5v. The chips can run off parasite power, they charge when you put a "1" on the data line. Parasite power will affect speed, length and reliability of reading. Do a search on "building reliable 1-wire netowrks". I'm far from an expert, and there are a couple of sources that explain it all nicely. One link is on maxim's site, and I dont remember the other, they cover most of the same information.

In your case you'll want to run the +5v wire if at all possible, you may want to run a hub with switching, or it may be easier to run several separate busses, with just a few sensores on each, you could still wire them to 1 arduino, just use 1 digital pin for each. It will make the software more complex, so you may want 1 arduino for wvery 3 or 4 tanks. Technically you can get away with all of them on one bus, ( aka run 3 wires all around from one tank to another.

Cheap ascii graphic ( hopefully it helps)
Arduino _____________________________________________ +5
===================================== data/gnd
||| ||| |||
||| ||| |||
Sensor1 sensor2. Sensor3 ....
Tank1. Tank2. Tank3

@zboblamont the otiginal poster specified hard wired for the sensors, that made me think 1-wire, if it is truly 100 tanks, I would probably put 10 tanks per arduino/nrf, the basic mysensor temperature node mentioned above with a bit of modification to allow multiple sensors would do the trick. The arduino ide has examples that enumerate the sensors and poll them. There are some threads here in the forum where people have done that, and it may be just copying their code would suffice

I'd search "reliable 1-wire networks" there are guides that help with the distance, and how many per network etc. Better to read the original than my summary on that.

Do you have and idea of what temperature sensors you will use? The 1-wire ds18b20 have reasonable accuracy, there are inexpensive versions that are weatherproof, and probably submersible. They can all reside on a single bus, so you would need 10 of them and 1 arduino (plus some wire).

I think the dhtXX and sht sensors would require some waterproofing, but may work.

@dbemowsk looks promising. A good (re)start.

@wallyllama i should have read @Nca78's post before adding mine.

Message for postman or other daytime visitors. A drag racing game in the vein of the old mattell electronic football(not really mysensor). With redefined characters it could display a good amount of rrd or other graph data, vertical a very accurate battery level.

Do they fit close enough to create an 8 line display? All of them together would be 40x40.

These arent great ideas, but maybe they will inspire better ones.

The idea is good, but I have seen posts from at least 3 continents, so getting people online at the same time is difficult. IRC^h^h^hslack has a nice immediate quality, forums are less ephemeral, and easier to search and understand later, and for newcomers.

There is good in both, if you really want a chat like forum, it will take more time and effort. Maybe work with the creator of mysensors (that is @hek, right?) to see if there is a slack plugin for the website, or try some events in slack, etc.

What about irc?

Most important:
Community is helpful and enthusiatic
Open source/hardware
Doesnt use the cloud (unless you choose to)

Nice to have:
Mature enough to be usable
Wide hardware support, with ability add new
Uses inexpensive commodity hardware

@NeverDie you mean a like battery from a hybrid car? Cells form a pack, but monitored(at least) and [possibly] charged individually. I believe that is how they are set up. It might be a way to find a reference on how to do it.

I meant that using an optocoupler was an obvious solution, and I should have thought of it.

I think most of the problems could be overcome, but they will take time and effort, which would take away from other projects unless a volunteer comes forward. Im not ready for something like this yet, so I should propbably keep my mouth shut.

@mtiutiu i usually need help with the obvious. Thank you.