This Home Automation Still Works

[jimkernsjr]

I entirely rewritten all my code from those ESP8266 WiFi functions, errorblink and wifiOTAcheck in a version 2. It lessens the main code and makes the wifi and OTA handling trivial.

I don't want to take the time to write it up, because sadly this forum looks dead now.
If anyone has a real interest, let me know, and I'll put it on github or something.

I also completed my windowshade project, v2! Reproducing it now is trivial. 3d print the parts, and you can order a set of circuit boards I made up off JLCPCB that is a Wemos D1 mini stepper hat with hall sensor for shade up/setting position. If a real interest, let me know.

In light of the inactivity,I'm starting to use Tazmota and Espurna alot now,and I think I'll look into Mysensors again.

I wanted to rewrite this original system with an ESP32+RFM95 and get rid of that outdated RFM69+Arduino Ethernet Combination.

But I'll look into existing systems now since there doesnt seem to anymore activity...  If anyone reads this anymore, any suggestions for some existing project?

miss you all...lol

[papa]

I've been wondering about what you've been up to jimkernsjr. Yes, forum activity has slowed. Though Joshua & I posted some things recently. Also lewishollow PMed me recently about about a User Interface for customizing end nodes that he is developing.

It would be good if you posted version 2 of the ESP82xx code somewhere & put a link here on the forum. (After all, the code probably has a relationship with what computourist offered here originally on the forum.)

I've also been disappointed when people stop by the forum to read, get ideas & help, then don't give back to the forum later with their own ideas & projects. From experience, I know that it takes work & putting yourself out there. But that's the only way things move forward.

What do the rest of you think. Is jimkernsjr right, that this forum is dead & we should just move on?

I'm always glad to welcome new projects & documentation & questions here.  If you have something, please post.

[papa]

Earlier in another thread, jimkernsjr posted: "I've entirely rewritten all my [ESP8266] code from those functions, errorblink and wifiOTAcheck in a version 2. It lessens the main code and makes the wifi and OTA handling trivial.

I don't want to take the time to write it up, because sadly this forum looks dead now. If anyone has a real interest, let me know, and I'll put it on github or something.
I also completed my windowshade project, v2! Reproducing it now is trivial. 3d print the parts, and you can order a set of circuit boards I made up off JLCPCB that is a Wemos D1 mini stepper hat with hall sensor for shade up/setting position. If a real interest, let me know.

In light of the inactivity, I'm starting to use Tazmota and Espurna alot now,and I think I'll look into Mysensors again.

I wanted to rewrite this original system with an ESP32+RFM95 and get rid of that outdated RFM69+Arduino Ethernet Combination.
But I'll look into existing systems now since there doesnt seem to anymore activity... If anyone reads this anymore, any suggestions for some existing project?

miss you all...lol "
=====================================

papa's response: I've been wondering about what you've been up to jimkernsjr. Yes, forum activity has slowed. Though Joshua & I posted some things recently. Also lewishollow PMed me recently about about a User Interface for customizing end nodes that he is developing.

It would be good if you posted version 2 of the ESP82xx code somewhere & put a link here on the forum. (After all, the code probably has a relationship with what computourist offered here originally on the forum.)

I've also been disappointed when people stop by the forum to read, get ideas & help, then don't give back to the forum later with their own ideas & projects, even a quick post that what they read was helpful. From experience, I know that this takes work & putting yourself out there. But that's the only way things move forward.

What do the rest of you think? Is jimkernsjr right, that this forum is dead & we should just move on?

[papa]

I still wonder what others think about "Is this forum dead or inactive?" as jimkernsjr stated "or dying?"

As a whole, the forum has slowed, but as I said, Joshua & I posted things recently about RFM69 nodes. lewishollow made a post & said he is working on a future offering.  Seems that June, July, & August have usually been slower maybe because in the northern hemisphere people are more active outside then.

jimkernsjr has shown more interest & activity with the ESP82xx wifi nodes. After I documented using a number of Sonoff ESP82xx wifi products, there has been little or no additions to the wifi nodes.  On Oct. 3, 2018, Joshua added an ESP82xx project.

Why such little activity with wifi nodes even though they are easier (no need us to build a gateway device first)? One main reason, I believe: To my knowledge (willing to be proved wrong) most wifi products focus on controlling AC appliances, which is great. However, I don't believe we've seen them do much with sensors except for one Sonoff product that can include temperature / humidity sensors. Having more documentation to use sensors seems to bring us back to using RFM69 nodes even with the extra hassle of getting them communicating with an RFM69 Gateway.

Again, what do others think?

[flyyboyy]

Wow, I hope not! Unfortunately, I had a hard summer and could not devote any attention to the site/forum nor automation in general. Now that weather is finally getting nice, gonna try to focus a little bit more on last spring's projects.

Need to go back through and reread much of what I had covered before. Or, do I save myself the effort -- ie, is there new tech that has replaced my gateway and RFM69 nodes?

I still have so much to do; proceed forward where I left off or start over in a new direction?

[papa]

Thanks for checking in, flyyboyy: "I still have so much to do; proceed forward where I left off or start over in a new direction?"

Good question. I can say that the previous RFM69 projects, while challenging, still work & I believe are still worthwhile to do.

But it's your time & expense & you & others will need to decide where you invest.

Though I have other commitments besides this forum (one in a few days will tie me up for a couple weeks), I'll do my best to provide & encourage support here.

[flyyboyy]

Glad to know that my Gateway built last spring is still current and usable. I've forgotten so much --  thus trying to wade through the maze of posts, updates, links, etc to find where I left off and where to go next.

No doubt your time is valuable and I understand that keeping a forum going (and current and relevant) just has to be a huge drain on that time. I hope, in the end, that others find it useful as I have, and that it makes a bit of sense to keep at it.

I had made previous attempts using some other well laid-out web sites that seemed to parallel DIYHA, but just could never get anything to work (they seemed to be less focused on / interested in the RFMs). Was so stoked when the GW was actually able to talk to my Pi/Openhabian/MQTT.

Thanks again -- flyyboyy

[papa]

You are welcome, flyyboyy. It's good to get some encouragement & feedback.

[jimkernsjr]

For what it's worth, I dont think it's dead, but we need to breath new life into it.

ESP:
I basically abandoned the ESP code to move to Tasmota, there are so many people that contribute to it and I'm just one person, there is no way I can keep up. It already has a user interface for customizing it, no need to hard code any credentials, and support for so many sensors and power monitoring already built in. You can customize the MQTT topics to appear like these, but I find the node numbers too limiting. In fact, I had to expand mine to 3 digits because I like to categorize by rooms. 20-29, kitchen, 30-39 living room, etc.

RFM69:
As far as developing for RFM69 and that system, I consider that dead. For me, it has no advantage over wifi except battery power. My Ubiquity nanostation can get coverage anywhere my RFM gateway can. This gateway code is stuffed into the full Arduino, with little to no room for expansion.

RFM95:
I started to swap it all over to RFM95. Andreas Speiss on Youtube did a mailbox node, so I was going to combine that with this existing RFM69 code to get a new gateway. For hardware, I developed a Wemos D1 mini shield with the RFM69 on it and placed an order to JLCPCB, and same with my windowshade node with an Wemos. You can build 3 gateways for the cost of the old Arduino/Ethernet combo even if buying clones - and it's tiny! I was considering stopping and moving it to ESP32, but I dont need the features and there is no good shield system for it that I like anyway. I think the Wemos D1 mini's can make great end nodes too with the wifi turned off.

Is anyone interested in this besides me, or should I let it die? I had to back burner the project a couple times already. Honestly too, MySensors already supports RFM95, so I'm even questioning if it's good use of my time right now to do it. I haven't studied enough yet to know what the advantage to this codebase is over MySensors. Anyone else use them both before and see an advantage of this over MySensors that would give me encouragement to prioritize it?

[papa]

Jim, good to hear from you.  I also identify with "I'm just one person, there is no way I can keep up [with all the new possibilities in home automation]."

I believe I am like many users: I tend to stay with what is working & what will probably work over a reasonable future. RFM69 Gateway/Nodes & ESP82xx devices have been working well for me & still provide an arena for me to tinker & experiment & expand my home automation.

I see helpful uses & disadvantages to both RFM69 & ESP82xx WiFi devices which I documented here.

I've done some work on developing a bare bones choose_nodes sketch & moving specific node options (sensors, etc) to separate library files.  That may be more readable code, but it also starts to generate many files.  I've paused on that project for a while, but may return to it.  Though my posting the start of the project did not generate much interest.

I am open to new developments, investigating them, trying them & encouraging them here on the forum. However, as I consider investing time & money into new developments, I cannot help but ask "What would I/we gain (beyond curiosity & the satisfaction of trying them successfully)?"

So, Jim, in moving to RFM95, what could we gain in useful function? In remaining with RFM69 & ESP82xx, what are we missing in useful function? Maybe responding to those questions would help you & us respond to the issues you raise.

[papa]

As I wrote above: "I can say that the previous RFM69 projects, while challenging, still work & I believe are still worthwhile to do. ...

Though I have other commitments besides this forum ..., I'll do my best to provide & encourage support here."

Some visitors continue to ask for support. I & others continue to add new material.

[jimkernsjr]

Hi Papa:

The answer to "What would I/we gain (beyond curiosity & the satisfaction of trying them successfully)?"

*The main reason for this upgrade to RFM95 over the RFM69 is the range. In my experience, I get no more coverage for RFM69 than Wifi! I own a Ubiquity AP and it can actually kick the RFM69's butt. You too can get this "Enterprisey" Access point for under $100 if you are open to a used one. I'm a bit snobby about my network equipment especially since I support that professionally. I'd much rather have a professional Ubiquity over a Google """home""" wifi or some other hack. So for me, Why bother with RFM69 especially since ESP is so much easier to set up - unless you have it already. In case of RFM, you have to buy 2 devices, and if not 4 if you don't already have this system for absolutely no gain. RFM95 is another story - it makes it all over my farm, and potentially can reach many miles. You can turn the power down if you don't want it too. So in addition to no better range than wifi, you are limited on space with the little Arduino, The gateway is maxed out, it cost more than ESP, on and on...

To Respond to: "Why such little activity with wifi nodes even though they are easier (no need us to build a gateway device first)? One main reason, I believe: To my knowledge (willing to be proved wrong) most wifi products focus on controlling AC appliances, which is great"

*probably the reason you see no activity on the ESP is because it's so easy. What is there to talk about?

"most wifi products focus on controlling AC appliances, which is great"
Not really:
*Why bother making "sensor nodes" here with ESP anyway? Tasmota has all you need. Tell it what pin the sensor is on in the web GUI and you are done, period. Working on this system here is reinventing the wheel.

For the reasons above, and with the facts I give, I find RFM69 and ESP8266 (with this code here I developed) a waste of time and your resources for *NEW* development. Of course if you have these systems existing already, why change? I have some too, and I will keep them - up until they have their first problem and they will be replaced.

Keep in mind: We are talking on this forum about technology, which rapidly changes. You either change with it, or get left behind. Your not storing your programs on floppy discs are you? For good reading:

connectedworld.com/when-iot-devices-and-solutions-get-left-behind/

[papa]

papa: Thanks, Jim, for taking the time to respond to things I raised above.

Jim: "We are talking on this forum about technology, which rapidly changes. You either change with it, or get left behind."

papa: Truth is that with technology, we are generally behind what is (at least for the time being) the latest stuff. Again like many users, I like using something that is working. I'm not eager to climb every learning curve, but do want to adopt new approaches before old ones become unsupportable.

Jim: "Of course if you have these RFM69 & ESP8266 systems existing already, why change?"

papa: Yes, again, my home automation system works well in my home & yard & the code is so familiar, I can add new features with little effort.

HOWEVER ... Taking your nudge, I will look again at trying the approaches you referenced in this post above.

Does anyone else have a comment on this discussion?  Are you using the documentation on this forum?  Are you using RFM69 radios & ESP8266 devices?

[papa]

    Comparing MySensors.org & DIY Home Automation (DIY)

                RFM69 Gateway

MySensors has more radio options listed, but for now, I can only try RFM69. I'll be away from my workshop for another month. On hand, I have an RFM69 Gateway & some RFM69 nodes with various sensor options. So for several hours over the previous two days, I tried what MySensors had for RFM69.

First I tried GatewayW5100MQTTClient.ino on my Arduino comptatible RFM69 Gateway build. After activating the RFM69 & LED defines, the sketch used 31066 bytes (96% of available space) & Global variaables 1377 (67%). The latest DIY Gateway takes only 27754 bytes (86%), Global variables 1449 (70%).

MySensors gateway generates much more serial monitor (SM) debugging info, though cryptic. www.mysensors.org/build/parser is helpful for decyphering.

For good or ill, I used my DIY knowledge of what was needed for a working gateway, but I needed to hunt a bit to configure the files. According to SM the install failed at first.  At first, I could not find how to use MySensors to change RFM69 chip select to avoid SPI pin conflict with the W5100 Ethernet shield. I finally found a hint here in a user comment NOT in the sketch.  Using that hint, I did not see SM errors for the gateway.

Along the way, I noticed that for sensor & other libraries, MySensors tended to use supplement & substitutions (including for RFM69) for the official libraries. It could be good that they provide a package of libraries that might be needed. However, at least one of the libraries changed keywords from the official version.  This approach may cause confusion & conflicts.

    Continues ...

[papa]

        Comparing MySensors.org & DIY Home Automation (DIY)
                        RFM69 Sensor Nodes

Next I tried two sensor node sketches on my RFM69 node builds: The DHT sensor & the binary switch. As I wrote above, the DHT sketch expected a modified unofficial library & used its non-standard commands. I used the official DHT library I already had & changed the sketch to use the official commands for getting temperature & humidity. After tinkering a bit, I could see no Serial Monitor (SM) errors, except I was getting empty payloads.

Next I tried the binary switch sketch, which needs no sensor library, just the Bounce2 library which I installed. Using the sketch, I again got no SM errors, but also empty payloads.

The Gateway sketch offered radio options of RF24. RF69, & RF95. The sensor sketches only offered RF24 & RF69 options.

Some of the forum comments made me wonder if only a few people got the RFM69 approach to work & maybe they've moved on from that. So trying MySensors for RFM69 might not have been the best test, but that's what I could do for now.

        See better RFM69 results below.

[papa]

        Comparing MySensors.org & DIY Home Automation (DIY)

OK, now from MySensors, I have the RFM69 Air Humidity DHT node & the binary (door, window) node working.

I had kept OpenHAB running with its embedded broker.  I believe that broker's topic may have conflicted with the MySensors MQTT Gateway topic. So I shut down OpenHAB & started the still existing Mosquitto MQTT broker.

I also missed this important note here: No controller supports dynamic ID assignment through MQTT. All nodes must have MY_NODE_ID defined in the sketch to work with MQTT. If you don't set MY_NODE_ID, nodes will complain with the message "!TSM:ID:FAIL".  So I defined MY_NODE_ID in both node sketches (range 1-254).

Since this node ID assignment is so important, WHY was there not a note & a commented default value included in the node sketches? That's the second key piece missing after no easy place to move the RFM69 slave select pin on the Arduino.

Another annoyance: I wanted imperial units for the temperature. In the sketch, I set the bool metric value to false.  No change. Then I discovered another line: metric = getControllerConfig().isMetric;  When I disabled that line, I got imperial units.  MySensors sketches & libraries may define something in more than one location.

Anyway, now the MySensors DHT node & the binary node are both working & I had a better test of MySensors.

[papa]

        Comparing MySensors.org & DIY Home Automation (DIY)

MySensors offers sketches & wiring for many sensors, some we do not offer here at DIY Home Automation. We here offer some sensors that they do not. They offer one sensor per sketch which may be easier to follow than my choose_nodes approach. I generally document one sensor or output at a time, but I like the sketch to have the ability of combining more than one sensor or output in a node.

I believe that (via RFM69) I made a better test of what MySensors offers.

I know I am biased, but I still feel that our RFM69 approach still works & offers several possibilities for Home Automation.

At the same time, if options like using an RFM95 radio could offer more transmission range or other improvements, I'd be happy to help those be documented on this forum. I would try an RFM95 version of our code. When I return to my workshop, I may try RFM95 at MySensors.

[papa]

I have been mostly busy with things other this forum (except with regularly knocking off spam posts & blocking people who post them)

At the same time, my DIY home automation system that I've documented on this forum is still working.

Perhaps when I get a chance, I'll review the system here.

[papa]

My system to monitor a remote location through openHAB & MQTT is still working. I have a mixture of node devices: some ESP82xx devices (connect to MQTT/openHAB via WiFi) AND some Arduinos plus RFM69 radios (connect to MQTT/openHAB via an RFM69 Gateway).

Three nodes monitor if doors' deadbolts are closed. One node monitors whether the garage door is closed. Another node commands the garage door opener. Several nodes turn lights off & on & openHAB rules toggle them pseudo randomly. Some nodes monitor room temperature & humidity. One node monitors the temperature inside a freezer. One node monitors the water level & pump in my sump cistern. I have a programmable cooking controller for a crock pot that displays states on the openHAB User Interface. In the warm season, three nodes monitor soil temperature & moisture & trigger watering through drip irrigation in garden beds.

I & others have documented many other nodes using several different sensors, an LED screen & a servo motor. I've also documented camera nodes, including one with a PIR motion sensor & OLED screen.

Security:  All these nodes were flashed with our forum's programming. If you choose, you can monitor your home automation only within your own network (no connection to anyone's cloud).  Or if you choose, you can monitor your home automation from any internet connection & be connected to only one cloud (openHAB's).

Though this forum may not include the latest bells & whistles, it provides several detailed, workable examples of home automation.

See this post for a linked list of home automation node possibilities.

[Knowyourbuilding]

1. Choose your first smart device

There are thousands of ways to automate your home—making it easy to go overboard on smart devices you don’t need. We suggest starting slowly with one or two smart devices before gradually adding more.



Top smart home device factors to consider

Cost: You can buy smart devices on almost any budget. Decide where your limits lie.

Setup: Stick to intuitive smart devices that match your skill level—some electrical devices (like smart thermostats and smart outlets) require more technical skills to set up than others.

Lifestyle: Everyone is different, so the smart gadget your neighbor swears by might not work for you. See our section on a balanced smart home for basic lifestyle categories.

Research: Before buying, read customer opinions, buyers guides, and independent reviews. You don't need to do as much research as us (that's our job), but different perspectives always help.


However, There are many companies in home automation and energy management for household and commercial buildings. But I would recommend a company called Know Your Building, they are experts in creating low consumption efficient energy systems.

[papa]

KnowYourBuilding,

So far, your above post seems related to the topics of this forum. From my first glance at your web site, your business focuses more on commercial applications than on the "Home" in our forum title DIY Home Automation.

For now, I'll leave what you post, but I'll watch to see if things get "spammy."

Forum members, what do you think?

[fito]

I appreciate the time you spent finding that information for me.

[papa]

Thanks for checking in, fito, & expressing your appreciation. I hope this forum's contents are helpful to you.

Papa

[teresa]

Web development is the process of creating websites and web-based applications. It typically involves planning, designing, coding, testing, and deploying web pages. It often utilizes frontend technologies such as HTML, CSS, JavaScript, and user interface design, as well as backend technologies such as server-side scripting languages like Ruby on Rails or ASP.NET. Many web development companies also provide hosting, support, maintenance, and other services mlsdev.com/.