Seems that search has been broken for quite a while. I've given up on it. (Maybe it was actually off and on, but I just had bad luck?) I just use google or duckduckgo or whatever, and then tell it to search just this site. Works great.
https://www.google.com/search?q=site%3Amysensors.org+sensors
If you look at those two sketches, in the setup() function, they both claim to set it up to use the internal pullup on pin 3. However, they do it differently. That's the first place I would look. Try replacing the non-working one with the same code from the working one.
@monics To know whether the controller was online and working, you could do something simple like request a variable from it, or just check on/request the time every so often. If it responds then things are good, and if it times out then you've got a problem and you can act on it.
@GaryStofer Okay, then sounds like you're doing what I would.
Are you able to turn on debug and connect serial to it to troubleshoot that way? Without getting more info on it, I'm out of ideas at the moment.
@GaryStofer Every time it starts up? That really removes the usefulness of using EEPROM in the first place.
Maybe you're having problems getting a response to the assignment of a child ID, and until it gets that it doesn't work???
@GaryStofer Did you erase the EEPROM? Maybe they stored the child ID number, and since it's a brand new network, they don't connect to it until some kind of timeout when they give up on the old ID? These are just guesses, and not based on any kind of deep familiarity with the code.
I just know that the ID would be stored for quicker connections on the next times it connects to the system, so that would make sense with what you describe.
I don't have a good answer, but since it's still a multiple of 5, I would assume that it's simply missing the other packets, and your sensor really is waking up and attempting? Maybe set it to require a response from the gateway before going back to sleep?
Just a guess.
Well, search seems to be broken again on the site. Hopefully a dev will see this come across their screen and know how to fix it.
I searched the site for "multiple definition" and 'premain' but neither one got ANY hits, even this topic itself!
So if you tried to search the site to see if anyone else had had this problem before, it wouldn't have much chance of success. Anyway, all that is to say that we've run into this before. I'll link below here the thread where I managed to get around it. I'm not sure if there were other threads, but this one should give you some info on how to fix it for yourself. I could find this by going into the history of posts I had made, since search was failing me.... 
https://forum.mysensors.org/topic/10193/stm32f103c8-problem-at-compilation
@igo But in your code you have the wrong calculation. You have (R1+R2)/R2. Whether the names are right or wrong, that calculation is wrong.
But yes, I agree that it is not the root problem, that's why I also talked about trying to find it by checking pin definitions and putting voltage right on the pin.
But what @JeeLet said looks like it's worth checking. Maybe you need to try a loop of a bunch more reads to see if it improves after a bit.
Your voltage divider math is wrong. It should be of the form
R1/(R1+R2)
However, why even bother with a voltage divider? The analog inputs on the atmega328 can handle the full battery voltage from either of your scenarios.
Although, maybe you have the pin wrong? Or your setting of which board you are using is linking in a different connection? I don't know if the PDIP package would cause it to be a different pin. (I find that hard to believe, but I don't know what else to say.) Cause the reading should obviously change if you're really changing the voltage at the pin that you're reading... Maybe if you can feed direct voltage to the pin temporarily and see if it does anything?
It looks like that's the proper way to use the library - the couple of little snippets you added looked like they matched the library on github. But I would use the definition provided in the header:
BME280_REGISTER_STATUS = 0XF3,
Just to make your code easier to read.
Maybe you need to provide the rest of your code to let us see that?
How are you 'combining' the code for the gateway and the BME280 library? It sounds like the NRF52832 is running mysensors, so it should be running the BME280 code, but the ESP32 is running the gateway code. So I don't see how they could be combined? Really, we would need to see code for both boards, I think. I can't think of other things right now to help.
Sure, a resistor could, but then it's always burning power. There are current-limiting circuits that use semiconductors, but I'm not sure how efficient they are.
I'm not sure how pros would do it. There must be good ways to go about it without losing much/any power. Again it comes down to how much power it will burn vs. the practically $0 cost of a resistor vs. other choices.
Probably the only problem with a too-big capacitor is how quickly it might want to charge when everything is connected. You might need to build in some kind of rate limiting for that.
I think capacitors with a blocking diode would be the right call. How long does it need to supply power to the Arduino for? And are you running anything else off of this, or only the Arduino?
Basically, knowing how much power draw (current) that you need, and how long you need it for, and then also how much of a voltage droop is acceptable will tell you how big the capacitors need to be. I'm going to use a bunch of round numbers for easy math, but hopefully you can see how to adapt for your project.
Let's say you're at 5V, and you're drawing 100mA (a lot for just an Arduino, but not so much if you're also running LEDs and a radio, or a servo, or whatever).
If the voltage droop from the batteries due to the motors spinning up lasts for 1 second, then we just need to calculate how much charge that is, and how big the capacitor needs to be to keep it within a certain range.
The total charge, in Coulombs, is 100mA x 1 second = 100mC.
The voltage on a capacitor is: V = charge (in Coulombs) / capacitance (in Farads)
So say you want the voltage drop on the capacitor to be less than 0.5V. Then that's 10% of the initial voltage, so 10x on the capacitor (1/10%) gives you the ballpark. Which means 1F capacitor should be the ballpark you need.
Of course there's simplifications here. (And it's been a while since I did this formally - hopefully I didn't make any big mistakes.) Also real capacitors will not act the same as an ideal one, and tolerances on capacitors are HUGE. Of course the real voltage drop is an exponential, and the current draw will also have part that changes with voltage, but not the same as a resistor, so it would take a bunch of math to calculate exactly.
However, if you can get a ballpark answer and then double it or so, for tolerances, you'll probably be good enough. Heck, for a hobby level project maybe simply going 10x for a few pennies more would be a good idea.
@mfalkvidd Oh, that kind of issue didn't even occur to me. I was thinking that it was still showing up as the same mysensors network, but that now what used to be sensor 10 was now 3 or whatever.
Yeah, my suggestion wouldn't help if Home Assistant sees this as a totally new connection/profile/whatever.
And I use Home Assistant, and I've changed gateways, but I'm always tinkering, so I change things often, and it was no big deal for what I do to go from one to the other. I don't remember having any issues with this. (Could be I still did, but it was a minor enough problem, compared to other big issues, that I simply forgot about it.)
But I don't have an easy way to test going from one to the other right now. I don't have a spare gateway to test on. 
Try using the Neopixel library.
https://learn.adafruit.com/adafruit-neopixel-uberguide/arduino-library-use
https://sbaronda.com/2020/09/20/sk6812-with-arduino-using-neopixel-library/
Those should get you started. Basically, instead of having your code directly turn a pin on or off, you tell it to set a color in the library and then run the show() command and the library will do all the pin toggling needed to actually control the LEDs.
I don't know if there's a way to do it from the Home Assistant side, but it's always possible to hard-code the ID of each individual 'sensor' when you're programming it. If you want them to never change, this is a way, and maybe the only way to do it. (I don't really know, but it's the only way that I know of.)
According to this page: https://www.mysensors.org/download/sensor_api_20
You set MY_NODE_ID to the ID you want and you should be good.
#define MY_NODE_ID 12 // or whatever number you want...
@qqlapraline Those are some quick times for assembly! I usually am doing a board with parts they don't stock, so I have only used them for assembly a few times, and not recently, so it's good to hear.
@NeverDie I've used JLCPCB a bunch of times now, and I have yet to find an error that affected the board electrically. I've found small things like sloppy printing or some rough edges of the copper, but always small, and like I said, it has yet to cause an issue for me. For prototyping they're really hard to beat.
Yes on the flyback diodes.
However, the upper transistors look like they're NPN, not PNP. The lower transistor would be PNP.
Otherwise, it looks like it would work, as long as all of the transistors are sized correctly for the current on the relay coils.
But I would be really hesitant to try to control that kind of power with an arduino. What happens if your code has an error and locks up? What if you have the magnetron running but not one of the heaters? I don't have answers, as I don't understand the working of them that well, but code errors running a high voltage radiator device like that make me a bit nervous.
I like being able to turn on the microwave and then ignore it and do other things while it's running, but it would take me a lot of watching it before I would feel comfortable with something like this.
On the other side, you're only controlling 3 relays, and it's not a nuclear reactor or anything, so I'm likely being overly cautious. But I am an engineer, and worrying about "what-if" is kind of my job, so I lean that way anyway...
@NeverDie Yes, though sometimes you have to dig a bit on their site to find them. Here's the one for use with the 6 pin tag-connect: https://www.tag-connect.com/product/tc2030-retaining-clip-board-3-pack
They don't automatically come with the cables, which could be a pain if you didn't know to add them to the order. Though this way I guess if they wear out it's easy to just buy the one without the other...
@NeverDie My personal preference is the Tag-Connect. It costs a bit upfront to get into it, but then there's zero ongoing BOM cost per board, and the connectors are very sturdy, so I don't get bad connections, even while doing extended troubleshooting.
Of course, if I hadn't needed them for a customer project, I might never have sprung for the upfront cost, as the cables are somewhat spendy for the what they are. But they are very well made, and have been great for me. The board footprint is pretty tiny, so that has yet to have ever been an issue in my projects. The footprint is also keyed, so you can't put the cable on backwards. Always a good thing. Also, since the connection points are not through-holes, you still have the back and interior of the board for routing traces. Of course the alignment pins are still through-hole. I prefer the type without the locking pins, and then there's a little board that you slide on the alignment pins from the back to hold everything together.
If I were to go with a standard, I wouldn't be excited about his ESP Flash one, simply because I don't see a need for it. Yet another 'standard' just fragments things worse. I would use the ESP-Prog one if needed, and use the .05" pins if I needed tiny.
@Larson Yeah, I think of "No Clean" as really being "easier to clean." There's still some residue with those that I have tried, but they really are easier to clean up after than the regular flux.
So I still prefer them, but I don't like to leave them uncleaned. Maybe the residue isn't corrosive like regular flux? At any rate I don't trust it either.
@NeverDie You've already tried the tricks I usually resort to. Maybe try to get a hold of an administrator over there? It sure seems like that site should accept all KiCad files, since that's kind of the whole point.
@Larson While I definitely agree about the importance of using plenty of flux - it's almost magic - another very important tool is to have plenty of desoldering braid around. It's super cheap, and along with plenty of flux it makes it easy to clean up excess solder on the board or bridging between two pins, or whatever.
That combination will let you solder all kinds of things that otherwise look very daunting.
@NeverDie They claim that it is already available in a QFN16 package. https://www.iis.fraunhofer.de/en/ff/sse/ic-design/rf-ic/wakeup.html
But they say you have to go through "EBV Chips" to get it, and everything I've seen makes it look like that's for large companies only. I can get to this page at Avnet, https://www.avnet.com/wps/portal/ebv/solutions/ebvchips/ebvchips-overview/ but I can't seem to get past it to any chance to order or see a detailed datasheet or anything like that. A search just on regular Avnet for the part turns up empty with the things I can think of.
This is the best page I can come up with: https://www.avnet.com/wps/portal/ebv/solutions/ebvchips/rficient/ which is cool and all, but still nothing orderable for normal people.
@NeverDie Yeah, but when I priced out domestic places, they all wanted enough that the client just decided to go a different way. This was only going to be for a prototype, but even so, I got prices ranging from I think $10-50 per pin holder, and it was less than 2 grams of plastic each. I just checked again, and JLCPCB would do them for $1 each, and then $20 for quick shipping, and as low as $4 for the slow boat that takes a month. None of those options were very appealing at the time.
Digikey wants $13.66 each for this when I pick qty 10, so still not worth it for this project. Shipping with them is free, but I have no idea how quick they would be.
I also checked PCBWay, just for fun, and they quoted 1.24 at qty 10, for $12.45 total. Interestingly, the quote for 1 was $12.45 also. They wouldn't show an estimate for shipping without clicking the 'Submit' button and having a human engineer review the file for printability. Since I'm not actually going to do this now, I don't want to waste their time, so I don't have a total price from them for comparison.
As far as OP, this is a very old thread from 3.5 years ago, so I doubt that they still need help.
@NeverDie Oh, sorry, I wasn't trying to get proof or anything.
I just was asking for your personal experience. Not doubting, just looking at them now and wondering how it had been for you. Do you mind saying which one you have that you are so happy with? I've been looking at different 'best' lists, and getting some conflicting info on them.
I'd just like a little more input before buying one, but with seeing all the improvements, and how many devices we have that could use it, now I'm wanting to do it sooner rather than later.
Thanks!
@NeverDie I think that usually you can 'rescue' the symbols and part footprints from the files directly as well. I've had it successfully work usually, but a couple times when opening an old file from a previous KiCad version I've had some errors.
Knock on wood, but so far nothing has been unrecoverable. Of course having a dedicated symbol and footprint library would make it even more foolproof, but it might not be necessary.
@NeverDie I'll have to do that next time I upgrade the wifi point. Now based on our chat I've started to look into them, and this might push me into doing an upgrade fairly soon. I see that quite a few of our end devices actually already support it, so I would be well served to swap the router over, too. I guess I had really lost track of the state of the art and thought most of our devices were in sync...
What do you mean by better performance? Is that better bandwidth, or longer range, or something else?
Thanks for the feedback on it! I really appreciate a first hand account!
@NeverDie I would sure hope it would be faster, otherwise no point in doing it. You can change the circuit so that the ESP8266 doesn't even have power unless it gets an external trigger. From the video this needs some kind of external trigger either way.
I would have to test it to be sure, but it sounds like it's booting fresh each time, so I don't see how it could be faster.
@NeverDie I tried to combine shipping and try out their 3d service a few weeks ago, and they said something like 'incompatible' when trying to combine the shipping. They would only ship the 3d print separately from the boards. And I was only trying to print some little pin holders, so it wasn't like it was something large and unwieldy, or even really delicate. Just a basic small part.
So anyway, don't count on being able to combine shipping. I ended up not having them do it, so I don't know how their 3d print quality is. I might still end up trying it just to see, but I have my own printer, so it's not a rush.
@tutur19 I'm sorry, I don't think I have enough details to do much troubleshooting. Though have you looked at the log parser? Here's a link for this specific log that you showed:
https://www.mysensors.org/build/parser?log=16 MCO%3ABGN%3AINIT NODE%2CCP%3DRNNNA---%2CFQ%3D8%2CREL%3D255%2CVER%3D2.3.2
28 TSM%3AINIT
28 TSF%3AWUR%3AMS%3D0
36 TSM%3AINIT%3ATSP OK
38 TSM%3AFPAR
40 %3FTSF%3AMSG%3ASEND%2C255-255-255-255%2Cs%3D255%2Cc%3D3%2Ct%3D7%2Cpt%3D0%2Cl%3D0%2Csg%3D0%2Cft%3D0%2Cst%3DOK%3A
2050 !TSM%3AFPAR%3ANO REPLY
2052 TSM%3AFPAR
2054 %3FTSF%3AMSG%3ASEND%2C255-255-255-255%2Cs%3D255%2Cc%3D3%2Ct%3D7%2Cpt%3D0%2Cl%3D0%2Csg%3D0%2Cft%3D0%2Cst%3DOK%3A
4063 !TSM%3AFPAR%3ANO REPLY
4065 TSM%3AFPAR
4067 %3FTSF%3AMSG%3ASEND%2C255-255-255-255%2Cs%3D255%2Cc%3D3%2Ct%3D7%2Cpt%3D0%2Cl%3D0%2Csg%3D0%2Cft%3D0%2Cst%3DOK%3A
6076 !TSM%3AFPAR%3ANO REPLY
6078 TSM%3AFPAR
6080 %3FTSF%3AMSG%3ASEND%2C255-255-255-255%2Cs%3D255%2Cc%3D3%2Ct%3D7%2Cpt%3D0%2Cl%3D0%2Csg%3D0%2Cft%3D0%2Cst%3DOK%3A
8089 !TSM%3AFPAR%3AFAIL
8091 TSM%3AFAIL%3ACNT%3D1
8093 TSM%3AFAIL%3ADIS
8095 TSF%3ATDI%3ATSL
Sorry it's so long and ugly, but that way you don't even have to cut and paste into it. It handles most log outputs really well and has enough detail to usually help find the problem.
Unfortunately, I'm not really familiar with the W5100 gateway. I've only ever used a couple serial gateways and the MQTT gateways. So I'm not sure how different it is. However, the log shows that the gateway isn't responding to the find parent request. That could be because there's something off with the radio, or there's some programming error in the gateway, or in your sensor code. You might need to provide your code to look at. I'm not sure if a parent request goes up past the gateway to the host controller or not. If the controller needs to respond to it then it could also be a connection issue between the gateway and whatever controller you're using. What do you have set up as the host controller?
Though you say if you buy an 'official' sensor it works. So does that mean that everything in your system stays exactly the same except for swapping the 'official' sensor for one that you're building yourself? If so, maybe we would need to see your code for the good sensor side by side with the code that you are trying for your own sensor that isn't working.
Sorry that I can't be more help, but without seeing the code, it's hard to do more troubleshooting from here.
@NeverDie Yeah, that's a whole lot of trade-offs to consider. I hadn't heard of the C6 variant. I believe the highest number I know of is only C3. But I would have to look up specs of any variant anyway, since I don't know by heart any of the meanings.
I've heard rumblings of wifi-6 and its use cases for battery powered devices, but I haven't had anything to do with it, either on the engineering side or just as a consumer. Wouldn't you also need a compliant AP for it? That's the only way that makes sense to me, but I only know the faintest bit about it yet. I'm glad that more low-power options are being worked on and will be available, but until they're something I can get my hands on and have spent time with, I just don't always have enough time to read up on everything coming.
Heck, I don't have time to finish half the projects I have going at the moment!
@NeverDie Seems that JLCPCB has had great pricing for at least the past 3-4 years or so. It's been very consistently low every time I've gone looking. I think that's about as long as I've known about them.
@NeverDie I don't know. While I've added them to a couple of projects, I wasn't the one that ended up doing the programming, so I'm not as familiar with them.
Though I thought that the options were either 2 or 3 cores. They usually only talk about the high speed ones, but there's also a low-power co-processor. This is not the one that the application usually runs on, but I think it can be used for doing some simple things while the main parts of the chip are asleep. This should open up some nice possibilities.
https://docs.espressif.com/projects/esp-idf/en/latest/esp32/api-reference/system/ulp.html
As far as all your other questions, I just don't know. I think the idea of the multiple cores is that one can be dedicated to either wifi or bluetooth, allowing the other to run the application. But even the single core is able to do those, so maybe a large percentage of the single core is used up by radio housekeeping stuff, which would limit the useful application size/complexity?
Once again, this is mostly from what I've picked up on the side, not directly programming them myself, so don't take any of this as gospel.
@dbemowsk I'm sorry, I really don't know. I've never tried domoticz, not to mention any of its plugins.
I do have Alexa working with my system, but I don't like it, just because of sending recorded voice out of the house all the time. So I plan to replace it once I get time to get Rhasspy going, assuming that I can get it going well enough.
Alexa works very well with home assistant. Though we use a service that costs $5/month to be the cloud piece to interact with the Alexa service so that I don't have to get something standalone working here for that. Though I think that service is really only for integrating to home assistant, so it probably wouldn't do you any good. It's saved me a bunch of time, and as I want to use my time to replace the whole thing with Rhasspy, it's been worth it to me.
But part of the reason I mention all of that is that with a quick search, it looks like controlicz is a similar service. Although they do say that you need to add the skill to your Alexa account. That would be annoying to me - doesn't that mean you need to say something like "Alexa, ask controlicz to turn on the kitchen light" or whatever?? With Nabu Casa, it also just uses the built-in smart home skill, so you don't have to install anything extra, leaving those extra words out each time. After the 10th time of having to do it I bet I'd really be gritting my teeth!
@Trand Well, sorry, but nothing else comes to mind. If you have a scope or logic analyzer then you need to watch the actual signals on the wires and see what is happening.
If you don't, then you should pick up a cheap logic analyzer. You can get one for $10 and it's invaluable for tracking down things like this. I don't know where you are based, but I see them on Aliexpress all the time.
@Nigel31 I also want to play with voice control, but don't want to connect an outside service to MySensors. What I'm hoping/planning to work on when I get a chance is Rhasspy. It's a local voice control which sounds like it might be good enough to be useful. Though I don't know if it will connect very easily to domoticz, since I've never used that. I use home assistant, and it seems to plug in to that very easily.
@NeverDie I've only done my own panelization maybe 2 or 3 times ever. Usually just leave it to the board house, but 10 or so years ago, before the board houses were quite this cheap and I was on more of a budget because of saving for a down payment, I did try it out for some personal projects as an experiment.
But I didn't even try to do v-grooves. Not sure if whichever one I was using would even allow it. I just left extra room between the active parts of the board on the PCB and literally took tin snips and cut them apart with that. With 1 cm between boards it's easy to not hit anything.
Obviously doing it like that wouldn't weaken their board, but if you need precise board shape/size then this wouldn't do it. Ever since then the prices have gotten so low that I don't bother trying. I would be interested knowing what about JLCPCB's process makes them want to limit any order to the 30 pieces. That's always seemed a bit weird to me. I would think they would welcome the extra volume/money.
But weirdly enough, I don't even know where to go for larger orders. I mean, I can do a search like anyone else, but I don't have direct experience. By the time it gets to that qty, then the assembly house deals directly with whoever they work with to get the boards made. At that point it's not me dealing with them directly, except for passing requirements to the client, who then passes them to the assembly house, etc.
I've also entered some of their limitations into KiCad, but I'm not sure if I did a complete job of it. I try to be a bit more conservative than being right at the manufacturable edge. But yeah, get the rules to catch as much as possible. I think my problems came in when I did something that such a minor change I didn't even really think about running the rules again, and of course something like that is usually harmless, but every once in a while it does something unexpected, so obviously the rules need to be checked after every change, and every time before exporting gerbers for manufacturing. And I try to do that, but still these days I'm sure that are times that I forget.
@NeverDie Okay, sounds like you know more than I do about it, I guess. I was sure that I had gotten responses/work done on a weekend, but maybe my memory is just faulty.
But yes, their service is very impressive. And for a 2 layer board, I can get it delivered here on the west coast USA in usually 6 calendar days from ordering with the quickest shipping. That's shockingly quick, and faster than most local places can even just make the board. (Well faster than any that I know of. I've pretty much switched to only using JLCPCB. Hardly even bother quoting other places these days.)
I believe if you really needed a bunch, they would be a decent price even with the penalties. Every time I've had to pay an extra charge for something, it was a reasonable amount. Though I've never tried to order 100 of anything through them, so I could be wrong.
@NeverDie I've never done a LoRa project, but I remember reading somewhere that some packets can take multiple seconds! So yeah, that would take a very large capacitor. I'm not sure what I would do in that scenario. I'd have to give it some thought, but I also might go with a dedicated power supply that could just handle the full current. Of course it would also depend on the specific trade-offs that were best for that project. Interesting problem.
Are you sure about JLCPCB not working on weekends? I 95% sure that I've submitted designs on Saturdays before and gotten a reply that they were accepted later that same day. I've also once or twice gotten things rejected when I made a silly mistake that they caught. I'm sure that it's a person reviewing things, and they have caught a couple errors. (Not in logic, obviously, but I had done a quick change one time, and a trace on another part of the board also got moved somehow, and it crossed over another. That's the only one I remember what the problem was.)
However, you can also always add boards to an order that is in process and they'll ship together for one price. Though if you're adding enough boards then there will be a bit of a shipping differential to pay for.
Those sound like some fun tests, and I look forward to hearing the results!
@NeverDie Oh, whoops, looks like you're probably talking about a different module? I saw the transmit number, and had been recently talking about the ESP32, which has a similar TX power number, and looks like I misread your post.
The overall point remains, but it depends on the TX time vs. the rest of the time for the overall average power consumption and the size of capacitor needed.
@NeverDie But that's only the instantaneous current needed for transmitting data over the radio. The receive current is more like what it will be doing the majority of the time. Put a beefy capacitor or two on the power line and they should handle those short little spikes.
I mean, you're not going to have a super-complicated web page on there or something. The packets being sent out will be what 1kB or something max, even serving a simple web page? According to this page: https://docs.espressif.com/projects/esp-idf/en/latest/esp32/api-guides/wifi.html#esp32-wi-fi-throughput you might get close to 80MBit/s speed. With some overhead, maybe the packet needs 10kBit, so it would only need to be on for 125uS. Even if the packets were far larger, the amount of time the radio needs to transmit is still a small fraction of the overall time. So just choose your capacitor(s) accordingly, and you'll be fine.
@NeverDie Yeah, I think they are the next gen device. I believe that this batch that came out in the past ~6 months was the first production run of the '2' series devices. At least it was the first time that I had seen them available, though it wasn't something I had been actively searching for.
Should be about an identical AVR instruction set, but some of the low level stuff like registers might have changed, so it might take a bit of work to convert projects to it. I haven't done it yet to know.
I have a Nortek GoControl USBZB-1 zwave/zigbee hub. I've been very happy with it. I've mostly used cheap devices with it, and only maybe 2 more mainstream things, like a Philips light. Other than lights, I think I only have a bunch of these plugs around: https://smile.amazon.com/gp/product/B08L3K5KPB
Anything branded Zigbee is supposed to be certified to work with any Zigbee network, but I have no idea what the big names do on top of the protocol. I've also never used a name-brand bridge device, so I've never really looked into what it might need. Home Assistant knows that it's a Philips light connected to it, but that's because it asks the devices for their info, and things like brand and serial id get returned to it. Probably other stuff - it's been a while since I looked. Anyway, even though it knows all that about the light, it's just using the basic Zigbee driver for it through the USBZB-1 gateway, and it has full control and it all works fine. I'm pretty sure it's not sending out custom Philips commands to the light, but just standard stuff.
Hope that helped a little!
@NeverDie As far as when the shortage ends, ??? I've seen estimates different times that it's "right around the corner" but nothing that was very convincing to me. Intel just this week or last came out and said that they think it's going to be better soon. NVidia chips/cards are getting to be more available. And random different chips have re-appeared, so maybe they're right this time? I wouldn't bet on it, though. There are still tons of parts that are hard or impossible to get in any quantity. I'm no insider, though. I shop at Digikey, Mouser, etc. same as you. Just maybe more often and a bit higher qty sometimes.
I actually have some of the Attiny3226 here to play with. When they came into stock, I bought a small bag of them. I haven't yet had a chance to do anything with them, though. I do see power consumption tables. pg 477-478 of the datasheet. https://www.mouser.com/datasheet/2/268/ATtiny3224_3226_3227_Data_Sheet_DS40002345B-2887812.pdf Numbers seem pretty reasonable, for a claimed 'low power' family of chips.
@Trand Do you mean that it stops at "Unable to begin"?? I don't see "Enable" in there.
If it's "Unable" then I would suspect that you don't have it wired up correctly. Are you sure about the RX_PIN and TX_PIN? If it were me, I would double check them again. And then swap them in the code after double checking them, just in case.
If there's absolutely no doubt that you have the correct wires connected, do you also have a ground wire connected between the Arduino and the DFplayer? Finally, are you sure that 9600 is a good speed to communicate with the DFplayer? (I have no experience with this module, so these are just general troubleshooting ideas.)
It says that it has a 16MHz oscillator, so it's most likely running at 16MHz either way. Definitely not in spec, but with a cheap clone like this, not surprising. Most people who buy it wouldn't be able to track down a weird bug that happens once in a while back to it running out of spec. Likely it will work okay on a bench, but I wouldn't put anything critical on it.
I wonder if it's a real Atmega328 on it, or some clone? No telling these days what you're gonna get.
@NeverDie Haven't you heard of the chip shortage? It's changed everything, and prices for those things that are even available have generally gone up. (Though somehow magically, the ESP32 seems to be pretty much untouched. You can still buy them, and prices have been relatively stable. I don't know how they've done it.) AVR, PIC, STM chips have all been hard to get specific models, and at times any model that has the features. I expect other brands were similar, but those are the ones that I've looked for.
I hope they someday go back, but for the foreseeable future, expect it to be like this.
Pro minis for close to $1 was always a shock, since even at 1k qty, just the chip itself costs more than that. So to get to that price for the whole assembled board, there must be some kind of gray-market thing going on. If I were to try to make a profit on 1k at a time, assembled and everything, I'm pretty sure the final price would have to be in the ballpark of $10 each.
@NeverDie I think you've pretty much covered it. Yes, sometimes there's no good hobby-level alternative to Amazon/Aliexpress if you want it cheap.
Though someplace like SparkFun or Adafruit can be good for hobbyists, though you'll pay for it.
Along with Mouser and Digikey, sometimes I have good luck with Arrow or Newark. But really, it's nice to do a search with octopart.com or findchips.com if you have a pretty good idea of what you want. They will look at lots of vendors, even some that might be shading towards gray market. Can really save some time, and sometimes find things that I never would have otherwise. Saved me a few months ago on a customer project that would have otherwise had to be completely redesigned.
@NeverDie The characteristic impedance of the coax is what matters. And that is all determined by the construction, which includes the thickness, but also other things, like what dielectric, and probably even what metal is used. (I don't know all the variables that they have to play with.)
So it could be that both of those options you posted are fine. But buying stuff like this on Amazon, especially through third parties is a crap shoot. No telling if it will be good or not, and the reviews aren't usually very helpful, either, on RF stuff. Not saying you shouldn't do it - I will gamble myself, but just go in fully aware that your chances of getting something shoddy are good, and make sure the price you pay is worth the gamble.
@gulsimsur Why are there links to surveyzop in your reply?
@nagelc Multi-transport gateway? So the single chip uses multiple transports? Which ones? I guess do you just have to define the pins to use yourself? (Since that's the only thing his code does for RFM69.) Or maybe I just really missed something...
As far as sleep, it doesn't look like any MySensors code supports it on the STM32. See here: https://github.com/mysensors/MySensors/blob/4afa7a644360c3abcb39d61df7c750c8f0ee19cb/hal/architecture/STM32F1/MyHwSTM32F1.cpp#L95
I've done projects using different STM32 chips in the past (the 'L' variants, usually) that were battery powered and we set up the sleep code. It wasn't too terribly tricky, but I do remember some issue on wake-up that took a bit of troubleshooting. Details are fuzzy, but could have been something to do with clock re-initialization, maybe? But this was with ST's STM32Cube software directly, not through an Arduino core, so I don't know if it's supported. We had to fiddle with registers directly sometimes. (Though the IDE was a lot of help, of course.)
I just went and looked for a core that supported sleep, and this isn't exactly it, but since it's a library, perhaps it would work with any core, with some modification? It does say that it's written for this core specifically. I'm not sure if that core works with MySensors. I'm using the Maple core with STM32. I had an issue with whatever the default was.
At any rate, not something that I have time to go play with right now, but maybe someday. For now, Atmega328 and NRF51 chips are what I'm using for my low-power nodes. I do plan to use some STM32F1 chips (from Blue pills) for plugged-in nodes that are also going to be relays outside, but I figured they might need the extra speed/power.
Hard to tell much from that picture, but it looks like it's just an Arduino and a radio?
An AMS1117, according to this datasheet, should be able to handle 1 amp. That is plenty if I was right about it being just an Arduino and radio. (Also sounds like that's all you've talked about in this thread, so seems like a safe assumption.)
However, the previous suggestions about capacitors at both input and output of the AMS1117 are also good. See what sindrome73 posted. Just substitute the AMS1117 for the LD1117 and you should be good.
@mfalkvidd In looking at the files touched, it does look like that pull request is attempting to support all of STM32. I have no idea if it will work, but at least from the people that have replied to the conversation, it at least looks like it should work.
Though I'm a little concerned by a couple of the include orders. Here: https://github.com/mysensors/MySensors/pull/1422/files
In MySensors.h, at line 71 the STM32 arch is checked for first, but at line 450, the STM32F1 arch is checked first. This would only be a problem if they were both #defined, but I could see that being possible, though unlikely to be common. I think the more specific F1 one should be checked first both places, and then only fall back to generic ST32 if that fails.
I also see that some of the hardware features are disabled, like the CPU temp, voltage, and the random number generation looks vastly simplified/maybe weakened? However, all of those things might have been necessary to keep from having to write individual code for each chip line. If so, that sounds like a decent trade-off for being able to work with so many more chips.
Also looks like the only transport support is for the RFM69, where once again the STM32 arch is checked before the STM32F1.
Overall, I would say that if this pull really supports a bunch of STM32 chips (and passes tests) then it's a good addition, but it looks like it needs work to really be complete.
I'm not really familiar with that board, but it looks like it should be capable.
But I don't see the wiring between the Arduino side and the RF side. Are you sure you set the pins correctly so that the Arduino can communicate with the radio?
Other than that, I think it would have to be hooked to a scope or logic analyzer to do much more troubleshooting.
If you don't already have one, you can get a super cheap analyzer here: https://www.aliexpress.com/item/4000364877295.html
It won't win any prizes for fastest or whatever, but along with the free software sigrok: https://sigrok.org/wiki/Main_Page it can show you what is going on and help to diagnose. You can also capture files and then other people, like me, could help you see what is going on, if you can't figure it out on your own.
There's a bit of a learning curve with sigrok, but it's not too bad. There are protocol decoders in there and you can play around with the settings to see if you can figure what is going on with those nonsense characters.
I can't find it now, but at one point I was also seeing weird things on the serial port. Whenever the library is using the serial port, it will switch the speed. Can be really annoying. In your code, you can do this:
#define MY_BAUD_RATE (115200ul)
and set it to whatever you want. I would suggest having the library use the same speed that you are using in your own code, assuming you are doing anything with the serial port. I also assume you haven't done anything with the number of bits, or the parity bit, right?
edit: I just found this page: https://github.com/nulllaborg/lgt-rf-nano, which seems to show that the pins are all connected just as MySensors expects them to be, with the exception of no Interrupt Request (IRQ) pin. That can be an issue. In case you haven't seen it, the connections are here: https://www.mysensors.org/build/connect_radio
Maybe it's not needed if you're careful?
https://forum.mysensors.org/topic/667/can-i-run-radio-without-irq-on-pin-2-or-have-an-interrupt-on-other-pins-than-2-3/2
But anyway, even if the radio fails completely, you should still get a serial message about that. So whether it's working or not seems like it's probably at least not causing your current issue. Just something to keep in mind later on if you run into problems after this current one is fixed.
@epierre I see some available on Amazon. Where are you? I don't know if USA locations are what would work well for you.
https://smile.amazon.com/12-Pcs-STMicroelectronics-LD1117V33-Voltage-Regulator/dp/B01N09X4E8/
https://smile.amazon.com/Bridgold-LD1117V33-Linear-Voltage-Regulator/dp/B07LG4H4PJ/
https://smile.amazon.com/6-Pcs-STMicroelectronics-LD1117V33-Voltage-Regulator/dp/B01MQF7D9D/
Of course the Mouser, Digikey type places have tons of options for a 3.3v regulator.
@NeverDie Yeah, platform.io is more of a 'wrapper' around the same libraries used in any other place. And of course you can create and use your own. But it's not like they have a ton of code that's created in parallel or something. It's all the same code source. I like that platform.io is based on a more full-featured IDE and also that they hide much less that Arduino does. You can easily get down into the weeds and set the settings you want.
However the MySensors code base has no HAL for the stm32w series. I have no idea how much work it would take to add. The HAL for the stm32f1 series doesn't look like there's much going on: https://github.com/mysensors/MySensors/tree/development/hal/architecture/STM32F1
@NeverDie Probably not every chip, but a huge ton of them. STM32 chips and STM8.
You can also set the framework to something other than Arduino, but of course then MySensors won't work.
In fact, I had to add a build option to make Platform.io use the Maple core, because that was the only one that would work with MySensors. (At least that's what I found at the time.) I don't remember what the default was if you don't specify which core to use, but I was getting some weird errors at compile time from MySensors. I might even have posted about them here somewhere.
This was the option, at the end of my platformio.ini file:
board_build.core = maple
And even though platform.io supports building for STM8 with Arduino, I never tried one for MySensors, so I have no idea if it would even work. I just included the list of them to show how much support platform.io has these days. I'm not sure one of the STM8 chips would even have enough flash space for MySensors, nor whether they support all the same instructions.
Anyway, once you get the settings right, it just works and I'm very happy with it. My sensors are a mix of NRF51822, STM32, and Atmega328 and they all are easily programmable with platform.io. (Though I still don't have much beyond the testing stage, as I keep changing large things like whether my gateway will be serial or mqtt...)
@OldSurferDude It looks like your gateway is set to send to the wrong topic? In the debug log it looks like the gateway is sending to the 'out' topic, while you mentioned that that was output from mysensors. But in your text you said you had it set the other way, so ??
I don't have a system set up with MQTT now. I used it for a while, but got tired of having to remember to update the RPi that was doing the gateway work. I kept letting it get way out of date and that's a bit of a security issue. So I switched to a serial port connection that's powered by the HA server and plugged into a USB port from it.
But have you used the log parser? https://www.mysensors.org/build/parser
It can help a ton with understanding all the messages that you're getting.
But double check your topic settings is the only idea I have right now after seeing those log messages. It looks like the gateway sent to the 'out' topic, rather than just listening to it.
@Larson As far as I know, it takes special probes for magnetic fields to check for EMI. Never done it myself, but I keep meaning to.
But scopes are well worth it to see what's going on. You can make it a fair ways with just fiddling with the dials and some light watching of YouTube. I used scopes a bit here and there before I ever went to school for engineering. (Though there was no YouTube to learn from at the time. The basics are pretty self-explanatory, and the advanced features are still beyond what I really do sometimes.)
@NeverDie Using platform.io, you can still use Arduino libraries with the STM chips. In fact, I've done just that and got a system on my workbench right now with it and MySensors. Seems to work just fine. I'm pretty sure that there's support for doing all this same stuff in the Arduino IDE directly, but I haven't done it that way.
One huge benefit of the platform.io IDE is that you can also specify programmers, on top of all kinds of other command-line scripting that should happen every time. I have it set so that it knows to use my ST-Link clone to program the board, so it does everything automatically every time and is absolutely painless. I again suspect that there's a way to set up the Arduino IDE to do this also, but I don't know how.
I don't have the time to go into details right now, but if you're interested in going this route, let me know and maybe over the weekend or next week I can dig up some more details for you. Though I found everything online, of course, so it's out there. If you have questions or get stuck at some point, let me know and I can also share whatever relevant settings I have that have let it work for me.
I'd say that EMI is just transmitting accidentally. Parts of the circuit are radiating when you didn't mean them to. When that accidental radiation - which is unavoidable at some level - rises above the regulation limit for any frequency band then you have FCC (or other country alternative agency) problems. Of course the accidental emissions can also be just about any frequency, depending on trace length and shape, clock speeds, etc. That again might require antenna knowledge to know what to expect.
I'm an engineer, though not RF. My understanding, from short discussions years ago with RF engineers, was that the ground planes have strong effects on antenna performance.
Depending on the design of the antenna, sometimes you want the ground plane and sometimes you don't.
But that's the extent of what I remember. (And it could be that I've even got errors in that little bit.) But basically to go with whatever the antenna designer recommended, because there are reasons for it. Just because one design calls for it and the other doesn't, it doesn't make either of them wrong.
Without the deep understanding myself, and not having however much testing equipment necessary to check the results, I try to stick as close to what the manufacturer/designer calls for as possible.
@OldSurferDude I don't see a password in your configure settings. Since it is saying that the connection is refused from your MQTT broker, that was the first thing that popped out to me. Do you have access control set up on the broker? I'm assuming that you have the IP address and port set up correctly.
I'm not sure if there would be a different message for other things, but when I had the access control on and messed up the username or password it would tell me connection refused or at least something very similar.
Well, I've narrowed it down.
I got rid of all the testing code that was in that sketch, along with its associated uses of the serial port. (And for some reason at points it used Serial, but others used SerialUSB.) I don't need to have it test the AtSHA204 or the SD card every time it boots - I don't even plan to use those. I guess my plans can change, but at least for now, they just aren't needed.
But getting rid of those took care of the problem! So now I have a working gateway working just like I want it to, and I'm probably not going to do any further work on figuring this out unless someone else has questions and I can help them with it. For my own needs it's now done.
I'm a bit stuck here. I hope that someone else can see what I've missed.
I have a Sensebender Gateway: https://www.mysensors.org/hardware/sensebender-gateway that I purchased from itead: https://itead.cc/product/mysensors-gateway/?acc=eccbc87e4b5ce2fe28308fd9f2a7baf3
I've been battling it for a while now, and have finally narrowed down my problem to one weird combination of settings. First off, it's connected to a NRF24 radio with PA and it is set as a serial gateway and connected directly to the USB port of a HomeAssistant server. Though for the testing I have it connected to a regular computer and watch the serial port. Behavior is the same whether connected to HomeAssistant or my computer.
If I program it to not use any encryption/password then everything works fine, and I'm very happy with it. If I program it to use MY_SECURITY_SIMPLE_PASSWD with the same password as the sensor and have MY_DEBUG enabled, then it also works fine. However, if I then disable the MY_DEBUG, since otherwise it will confuse HomeAssistant, it no longer receives any messages from the sensor. (At least it doesn't report anything.)
It makes no difference whether I enable the high power amplifier of the radio or not.
When it is not working, it will only write a little bit to the serial (USB) port on startup, and then nothing else ever. I have let it sit there for hours overnight and never got anything out of it. This is the only thing it will send to the serial port when it has debug disabled and password/encryption set:
0;255;3;0;14;Gateway startup complete.
0;255;0;0;18;2.3.2
So does this tell anyone else what might be going on? When the debug is turned on it will recieve the messages from my sensor just fine, or when security is disabled (or both debug and security are disabled) but it only fails in the actual configuration that I need. (Of course. ) So I know the hardware works, and that the security/signing works. I figure it must be some weird bug/conflict, but that's as far as I've been able to get so far.
I've dug through the source code a bit, but I haven't been able to track down what might be causing this. Of course there are so many files that I'm also not quite sure where to even suspect.
Edited to add: I'm using the example Gateway Serial sketch from https://www.mysensors.org/hardware/sensebender-gateway with the only changes adding the simple signing define and removing the inclusion mode stuff, since I don't have a button on the board and don't plan to use it that way.
Thanks!
@clivec If you look in the MySensors code, wifi is not an option for the radio transport. You'd have to somehow write your own deal for that, and it would likely get really involved.
ESP32 is supported for a gateway, where the wifi is the backhaul transport, effectively, but you would need a separate radio for the MySensors side of things.
To use an ESP32 on its own, without a separate radio, you might look into something like https://esphome.io/ This does not allow for interaction with MySensors directly, though. What is your ultimate plan? An ESP32 is very power hungry when on wifi, so it wouldn't work as a battery-powered device very well. They're pretty cheap, and easy to get a hold of, so if you have them plugged into mains power then you could use them alongside a MySensors network. Maybe with something like Home Assistant controlling both and providing a link from one to the other if needed?
Maybe you haven't been getting much interaction here because it doesn't fit very well with MySensors. I've seen your previous posts, but since I haven't tried to connect any ESP32 devices to the network, I didn't have any input for you. (I prefer my backhaul to be wired, so the ESP32 wifi was a negative, in my book.)
I mean, I have nothing against ESP32 for other things, but for me it has no place in the MySensors network at all.
I haven't used that one, but here: https://www.elecrow.com/water-proof-integrated-ultrasonic-ranging-module-jsnsr04t-p-1151.html it says that 450cm is the max range, so you're aactually getting better than specified, by a good margin.
But with the ESP32 I don't know why it would go down in voltage, though I have a suspicion.
But are you really seeing those kinds of voltages and those aren't typos? I have some of the non-waterproof versions of the sensor, but I've never checked their voltages on the board. It just surprises me that they would have that much boost on-board. That could be quite a safety hazard if someone were to touch the wrong part.
However, what voltage input does the STM8S003F3P6 board run at? I see that it can handle up to 5.5V, where the ESP32 is strictly lower. If the STM8S003F3P6 board is about 5V while the ESP32 is about 3.3V, that's pretty close to the same ratio difference between the output voltages that you're seeing.
If that's the case then you just need to run the sensor at the higher voltage, and deal with level shifting the I/O. Though I saw mention of triggering the board with only 2.5V, so you might only need to worry about putting a voltage divider on the output of the sensor and calling it good.
I'm not really familiar with FHEM, but as far as the MySensors side, you still need the gateway software running. However, this code can run directly on a Raspberry Pi, at least with a radio network for the MySensors transport. I haven't ever used RS485 myself, but I would think it would be similar.
At least the code itself has sections for dealing with running on Linux, which is what is used when running it on the RPi, so I think you're good. https://github.com/mysensors/MySensors/blob/master/hal/transport/RS485/MyTransportRS485.cpp
The gateway code runs as a systemctl service, so it can easily run in parallel with most things, so probably wouldn't conflict with FHEM on the same device.
Oh yeah, and in case you haven't found it, there's this page: https://www.mysensors.org/build/rs485
@KevinT It's sized for whatever size a 60 LED ring is. I bought them off Aliexpress and sized to that. A regular Pi could probably fit in there, though I'm not 100% sure on that. I planned for them to be Zeros. Only thing is, it's been a couple years since I started thinking about it, and I just don't get around to actually doing it much yet. So it might still be a while before I get anything finished.
Though just today I got a marketing email about an ESP32-S3 product that is aimed at machine learning and might very well be able to handle all the needs of a satellite. It would still need software support, so it's not ready to go, but that would be even lower cost and power budget for the satellites.
So I did a bit of searching, and others are also considering it: https://community.rhasspy.org/t/best-esp32-based-hardware-for-satellite/3012
Looks like the chips should have plenty of power for wake word detection, but it would have to be coded. So far if I understood it correctly, they are just streaming everything to the Rhasspy server full-time. Not that it would be a ton of bandwidth, but I'd rather not be broadcasting all audio in every room of my house 24/7. I think it's just the very inelegant design hurts my engineer's brain...
@CrankyCoder as far as the LED project, I'll look into that, thanks! I don't have any plans to go that crazy to need the e1.31 stuff, but I guess support for features you don't use doesn't hurt...
Yeah, it looked like there was just a selection box, and they were both choices. I haven't gone any deeper than that, or installed it on my own hardware yet. Day job and family stuff are keeping me too busy to spend much time on it other than just dreaming/planning.
Some of what you said might have gone over my head with Kubernetes. Makes sense that you would use what you're good at. But by saying you have a 7 node cluster, does that mean 7 physical machines that will actively share the load? If so that's pretty cool, but far more than what I'm needing any time soon. (I think. Unless I get really deep into some big project, but there's nothing on the horizon right now.)
@CrankyCoder Thank you so much for the detailed explanation of your setup! That sounds very much like how I want my final result to be. I still have no idea how I missed Rhasspy as a decent solution when snips got bought out. It was just never on my radar, which in hindsight is ridiculous. I have no explanation for the huge oversight.
However, now I've been looking into it more, thanks to you, and it's looking great! I also like that there's been some collaboration between Rhasspy and the guy behind Project Alice, so it looks like there's no bad blood there. I was looking at the LED control code, and surprisingly to me, I found out that the Raspberry Pi can even control the ws2812 LEDs directly! I thought the timing constraints were too much for it, but I see that people have made clever use of some of its hardware to be able to handle it, even with a non-realtime OS.
So with that, I'll be able to do without a supporting mcu, it looks like. I guess I'm running out of excuses and need to get started soon.
I was comparing the TTS modules, and although the online ones do sound better, as you say, I also want to keep everything local for privacy reasons. I did like the sound of the MaryTTS, but when I was listening to samples online, it sounded like picoTTS was a bit better to my ear.
I haven't figured out my final configuration for sure, but it could end up being a docker instance on an old repurposed desktop that has extra bandwidth or maybe on a dedicated Raspberry Pi 4. I think either one would be enough, because it would be relatively rare for more than one satellite to be interacted with at the same time.
Do you like Kubernetes much better than Docker for any certain reasons? I guess this is mostly curiosity for me - I currently have 5-10 docker instances running on a couple computers and they're mostly hands-off as far as any maintenance, so I would be hesitant to put in the time to start switching at this point.
@CrankyCoder pointed me to another candidate for the speech-to-text part; Rhasspy in a post in another thread.
Funnily enough, in reading through the documentation, it states that it interoperates with Snips, and provides a link. Even though going straight to the Snips website provides nothing but a landing page, using their docs link ends up with all kinds of info. So it's still there, but hidden from a casual search.
@CrankyCoder Rhasspy is a good one to add to my list!
I've seen the name before, but I suspect I thought it was tied to something else? For some reason I've never looked deeper into it. That was a silly oversight on my part. Thanks for the clue!
Would you mind adding your experiences to the Smart Speaker topic that I started? It would be great to hear from another user direct feedback on what that system is like! (And I would like to keep from driving this topic too much off into the weeds. )
@KevinT Yes, that was my question, thanks. I didn't realize that Android had that built in - I thought the voices you hear were generated in the cloud and then the wav or mp3 was sent to the phone. (Like when using GPS to get around or whatever.) Or is it the same voices that you hear while doing that? Is it more basic?
Anyway, thanks for the answer - that's interesting.
Yeah, Alice must be able to run on pretty much any server, but I don't know how much manual effort would be required to alter settings. Maybe they even have an easy-to-use config change that will set up everything?
What I like the most is that the 'satellites' don't need much power, because all they really need to do is wakeword detection and then piping audio back and forth. Just need one with enough grunt to do all the speech to text and back. With our use patterns, that one wouldn't even get worked all that hard for the most part.
Actually, I'm really looking forward to this, because I have a Death Star 3d printed to hold all the electronics, and an addressable 60-LED lightring to go around the outside. So it will also function as a clock and timer/stopwatch with the ring.
And I'm also adding a piezo mic to act as an impact sensor, so you can use them as targets for nerf darts. I got that idea from someone who made an alarm clock that you turn off by shooting. As soon as I saw that I knew I had to add it to my speaker.
Oh, and the depression for the laser beam from the Death Star is where the speaker mounts. I have all that 3d modeled and printed, just need to do all the electronics and programming... I just need to figure out a way to make some simulated lasers and to be able to light them up. That would also work as a bit of a speaker grille to protect it. I haven't spent much time on that part, but I think it would be a nice addition, and I need to see about it...
@KevinT What is it that actually does the conversion from text to speech? Is it on the phone? Did you create that part, or did it come with the phone, or something else?
Yes, I'm aware of the keyword spotting, but I can't think of any applications in my house where that would be enough, except for using it to get the wake word for a more fully-fledged speech to text. I hadn't see the Edge Impulse learning site, though, so thanks for that!
I also hadn't even considered something like the mid-range like in the Wio Terminal link you have. I have to study that some and think about it. That could be enough for what I'm doing... But in any case I want to work through the tutorial and see what I can learn.
But overall I'm still a bit set on the full-fledged idea with Project Alice. Which might actually not be reasonable. I need to think about it. It could just be that I set myself on that a while back and now going with anything else feels like giving up? But in any case, these other options will be great learning experiences, so I'm going to look into them.
Thanks!
@KevinT yes, besides electrical issues and fires, there's not much that will destroy a house faster than water! Very important to catch it quickly!
As far as the speakers, I did start a new thread. https://forum.mysensors.org/topic/11853/smart-speakers
Well, this is not super on-topic for the site, but there's probably a lot of cross-over interest with people here on doing a local smart speaker.
It came up in another thread with @KevinT, but was really not the topic of that thread, so I'm starting this one to have a better place to hold the discussion.
So the current "smart speakers" of yours are really just output devices? Is the MQTT to speech internal to the phone, or does something else do that?
I'd definitely be interested in continuing the discussion. Kind of off topic for this whole site, since its use of MySensors would be minimal. There would be a two-way integration, but it wouldn't be very tightly coupled.
My thought was a raspberry pi, or even a powerful server if needed for the speech recognition and also for responses, and then just raspberry pi zeros for the satellite devices. (They would be doing wake word detection and then simply passing audio back and forth, so not a lot of power needed here.) It would be able to be a pretty cheap overall solution, while keeping full privacy because everything would be done here at the house, not sent out to the cloud.
I'm going to add links here to what I think are relevant projects.
https://snips.ai/ is what got me started on the idea and general architecture, but they got bought out and shut down before I could really get my whole plan off the ground. I was still fighting the learning curve and only had a single device kind-of working like I wanted when the announcement was made. Their website now is just a landing page that points to sonos, so it's useless. Used to be that you could still view the old info there for a while.
Project Alice is an open-source fork of what was snips. Snips had been mostly open-source, with the large exception of their web-configurator tool. Project Alice is not a finished product, but it looks like it's in a useable state, though it also looks like it's mostly the passion project of a single developer. I think this is the way that I'm going to go, once I get enough free time to wrap my mind around it all. https://github.com/project-alice-assistant/ProjectAlice https://community.projectalice.io/
Then there's Mycroft. https://mycroft.ai/ It looks reasonably good for a front-end, but they use Google on the speech recognition. At least they realize the privacy downside of this, and they aggregate everyone's speech snippets and proxy it through their own server, so that minimizes the amount of info that Google could get out of it. They have been supporting Mozilla's Deep Speech project, but I don't think that you can send requests to that over the cloud. Though they also say that if you have hardware with enough power to run the Mozilla service locally that you can set it to do that. https://mycroft-ai.gitbook.io/docs/using-mycroft-ai/customizations/stt-engine#default-engine
While Mycroft will run on a Raspberry Pi, it needs to be a larger, more powerful one. It can't just run on a zero. https://mycroft-ai.gitbook.io/docs/using-mycroft-ai/get-mycroft/linux#system-requirements Although, now that the Zero 2 has come out, it might be enough.
Finally, just to be complete, there was the Jasper project. https://jasperproject.github.io/ I was especially interested in this one when it came out, both because it ran on a Raspberry Pi and my son is named Jasper. I thought that would be cool. It seemed like a great start, but then it died off pretty quickly. I got the feel like it was a student project and the students then were done with school and it didn't go anywhere.
That could be totally false - I didn't follow it closely. But it's been basically abandoned for quite a few years now. I didn't understand the code well enough to just pick it up and run with it, so I never did anything other than read up on it every year or two.
So that's my list for now. If nothing else, it's nice to have this all in one place for future reference. Right now I still think I'll end up going with Project Alice, but with how long it takes me to get to these things it might be a long time before I actually get anything going. If anyone else has suggestions, or things that I've missed, I'd love to hear about them!
Yes, that does look nice, and I'm going to have to think about doing something similar. Water leak can really ruin your day!
However, how can you just casually mention "home made smart speakers" without giving more detail??? That's what I want to hear about!
I was about ready to make a bunch of them using snips.ai and was even starting on 3d printing enclosures when they announced that it was being taken over and was no longer available.
I still plan to do it, but I haven't been able to make the time yet, and I haven't settled on which setup to use instead of snips. Though most likely is Project Alice, which is kind of/sort of an open source fork of snips. (With a lot less polish. We'll see how it goes.)
@kasparsd Yes, this is a very disheartening thing to see.
I would rather not have to maintain a local version with some of the fixes that I see talked about on the forums, but since there's no movement on the code, at least from looking at github, I'm left with doing a bunch of it myself if I want to incorporate the changes.
Especially bad when the changes are in a pull request and just languishing there. It's discouraging when I want to go work on something but I need to deal with the core mysensors code before I can even get to that step.
I mean, I'll deal with it cause I still appreciate mysensors and find it a net benefit, but I can certainly understand why some people either give up on it or maybe never pick it up in the first place, seeing stuff like that.
I don't know how well it would match just what you want, but there's a framework built with MySensors that is supposed to make this kind of thing pretty easy.
Check out: https://www.mysensors.org/download/node-manager
I haven't actually used it yet, but I plan to once I get some of that type of sensor up and running in the next little bit. (So I can't personally vouch for its quality or anything, but it seems like a reasonable way to get started, and for it to be on the site it has to meet some minimal level of working well, I expect.)
@nexus1212 You're using a 3V Arduino. It looks like all the patterns are correct, but it's just not getting things right. A lot of the unprintable characters in ASCII are at the start of the table, so I'm thinking maybe you have a serial connection that is expecting 5V as the 'high' voltage and what it's actually seeing is so borderline that it's not always registering correctly.
So look if there's a setting on whatever you're using to read the serial data to set it to be 3V compatible. (Maybe a physical jumper on the board?) It's so close to getting it that I bet it's something super simple like that and your Arduino is really doing it's job pretty well.
@perIpI I don't know if this will be helpful - I haven't used an ESP8266 on MySensors, nor have I really messed with one hardly at all.
But when I was troubleshooting my NRF51822 with MySensors, I was surprised that there were two different baud rates. I had MY_DEBUG enabled, and during the library initialization and waiting for the controller to hand out an ID it had one speed, while during normal use it had another. (I had other issues, so this took a while to resolve.) I also had Serial.begin(9600) but during startup it ran at 115200. Once it got to my code it then switched to 9600 and ran the rest of the time at that speed.
So all of that is to say that it's possible that if you let it run until you stop seeing responses at the faster 74880 speed, you might then get some results by switching and watching it at 9600. (Of course if you've already tried that then this won't help.)
Alternatively, you could simply switch the speed in your code to match the 74880 speed that you're seeing results at, and maybe you'll then see everything in the same window.
@martin3000 said in DIY zigbee sensor?:
@frits thank you, I tried to flash the ptvo firmware, and after flashing I found out that the firmware does not contain a boot loader. So my zigbee stick is lost...
If you have hardware that can flash it, then you can still recover. Not having a bootloader is only an issue if you don't have the right hardware. And as @frits seemed to be getting at, there are many choices these days where common hardware like a Raspberry Pi or even Arduino can be programmed to act as that special hardware. But even if not, some of the more common platforms you can buy pretty cheap programmers that will do the trick.
In my experience it's actually fairly rare to truly lock a device to the point where it's not recoverable. At least at this kind of DIY level. (Burning out a chip by putting the wrong power to it is another story... )
@Dizzwold I see lots of different screens out there, but when I search for "Raspberry Original 7" Touchscreen" it's not HDMI. I can't find a Raspberry Pi branded touchscreen that is HDMI in a quick search. Could you share a link?
However, it's most likely that the USB would be needed for the touchscreen part of it. HDMI couldn't carry that info easily, at least. (I'm not very familiar with the standard, and I don't know the limits of what info it can pass back through CEC.)
The USB cable could provide power, but maybe the limit through an individual port is too low and it needs the beefier connection? Maybe it uses both?
Well, it could be that the issue is the voltage regulator on the Arduino board. The higher the input voltage, the more power/heat that needs to be burnt at the voltage regulator. Maybe it's self-limiting because it's getting too hot?
The Arduino website lists this part as the voltage regulator:
https://www.mouser.com/datasheet/2/146/SPX1117-1889129.pdf
On page 6 it claims a thermal resistance to ambient of 46 degrees/W. If you're using 1A and have 12V input then you're dropping 7W across the regulator. This is a temp rise at the junction of 322 degrees over ambient, which it obviously won't do. The datasheet also says that it's internally regulated and will current limit at 155C. Assuming 25C ambient, this only leaves a margin of ~2.8W that can be dropped over the regulator. So you'd have problems with any more than around 400mA. To have a safety margin, I wouldn't plan on more than 300mA or so, or maybe even down more towards 200.
This voltage regulator is a nice, low dropout one, so if you could power it from a supply that gave it 6 or 7V instead of the 12 that would give you a ton more thermal margin and the regulator could then allow a lot more current through.
mfalkvidd gave you a great link, but if you ever need to see any possible choices, you can run
./configure --help
and it will show you. The extra flags section is one thing (of many) that it says.
Actually, this is very timely for me, as I'm trying to troubleshoot my own radio issues, so I'll use those extra two flags that you mentioned and see what they say.
Thank you!
Oh, and Skywatch's tips are also good. Solid power is very important. I would put a few different capacitors as close to the power pin of the radio module as possible. 100n, and then a selection from: 1u, 4.7u, and 10u. I would likely go with 2 of those microfarad capacitors, and maybe all three. I like to err on the side of overkill where power capacitors are involved.
I think your numbers look good. Nice job getting the trace much shorter - that should help a bunch.
Just a warning upfront - I'm an engineer, but not an RF engineer, so I do see some things, but I'm almost guaranteed to miss others.
As far as your highlighted line, it is a trace, but not a via. A via connects the trace on one plane of copper with a trace on another plane.
That trace should really be as short and direct as possible. Wrapping around the back of the module and along other signal lines is not a good idea. I see that the closest line to it is 3.3V power, so that acts basically as ground for small signal, but any power spike is bound to couple into the antenna line as well, at least some. If possible, it would be best to have ground on both sides of the antenna trace on that side of the board, as well as the whole surface on the other side of the board from it. Might also want to guard it by having vias connect the ground planes on either side of it to make kind of a 3d cage around it.
Additionally, you should try for 50 ohm trace impedance on the antenna line. However, there is no one-size-fits-all answer for the width of this line. It depends on the exact board parameters, and even to a small amount on the frequency of the signal. There are trace width impedance calculators that you can use to get this answer. https://resources.pcb.cadence.com/blog/2019-just-how-wide-should-a-pcb-50-ohm-trace-width-be
Almost guaranteed that it's correct to ground those 4 outer pins of the antenna connector. View the datasheet of the specific connector you are using to be 100% sure.
@mutantx Is there something lacking from the circuit diagram on the openhardware page? I'm not the original creator, but it seems like everything is on there.
https://www.openhardware.io/uploads/58f4f976bacb0bcc79a15e93/394/DFPlayer_NRF_ProMini2_4.5V_Steckplatine.png
But back on topic - that device looks pretty cool, if it does half of what it claims. I'm a little tempted, even at my own cost. I won't, but if my insurance company ever offers it, I'll jump at it.
Yes, it's a little 1984. But I also already signed up for an app with my car insurance that monitors my driving. It doesn't have any hardware, just GPS and accelerometer from my phone, but it doesn't cost me anything other than some power and I guess privacy. But it saves me something like 15% and I don't have to do anything. I guess if I ever really care about it watching me go somewhere I can disable it for that amount of time.
A little weird what we're willing to put up with, I guess. Though I also like the fact that with a record of how I drive, if there's ever an accident, it might also be possible to subpoena the data in my defense if really necessary. Or maybe that's just what I tell myself to help justify it?
At least here where I am - West coast of US - they're allowed +/- 10% and still be counted as within spec. So those numbers are fine.
The thing is, they have voltage transformers with multiple contacts inside, and they will step the voltage up or down on the neighborhood feeders as necessary to account for varying voltages on the higher voltage lines. Actually, I think the switching happens going from high to medium voltage, and there aren't really that many of the switching transformers in the system. All of your local transformers for the houses wouldn't have them - that would be far too spendy. They do it on a large scale.
But as the residential power demands change over the day, they use the switchers to keep everything balanced. They could keep the voltage variance smaller, but at the cost of more frequent switching, and every switch of the contacts creates some wear, so more maintenance eventually. Every switch also creates some ripples in the rest of the system, I'm pretty sure. But those are probably pretty minor - it's a very stiff system. Also having higher resolution on the switchers would be more complicated and have more parts/windings inside every one of them. They keep the frequency very tightly regulated, but allow the voltage to float a bit. The frequency variances they have to make up for throughout the day so that over the course of a whole day there's practically no net change at all. (For the sake of clocks that literally count cycles and therefore don't need a crystal.)
Actually, allowing the voltage to vary also has another nice benefit to the system - it allows for a natural damping of disturbances. When the voltage dips a bit because of increased demand, the power used by many resistive appliances will also dip, so that can help the system in recovering as potentially more generation is brought online. I don't think something like AC that has a spinning motor helps much, as the frequency won't vary much.
Sorry if you're aware of most of this. I worked for the power company for a while, though that has been some years ago now.
I wouldn't even bother with an Arduino/microcontroller at all for this. Just some diodes on each circuit and maybe some transistors for powering the large LED strips, or could also use tiny relays.
But I don't see where it has anything nearly complicated enough to have code running on a chip to run anything on it. @itjobhunter if this is not helpful and you need more specific details and/or a quick napkin sketch I can throw something together to show what I'm trying to say.
Well, I don't know what all is in the video, because I'm not going to watch a 27 minute video about it.
However, those modules often fail due to bad power supply. Adding some capacitance to the power rail is supposed to take care of it. I've read about it multiple times, but I haven't used them hardly at all myself.
Oops, I missed the latching part somehow, even though it was stated multiple times. That would take a few more transistors. Though the way I would do it would be to just use an H-bridge motor controller chip/board. They would have everything you would need, as well as most having the free-wheeling diodes built in.
I haven't checked it all out, but I do like that you have a diode on the relay coil to keep the voltage from spiking.
What voltage is VDD? Maybe it says somewhere, but if so I missed it.
But in general, it's better to have a combo of mosfets so that the relatively low voltage out of the Arduino can control the higher voltages safely.
The problem is that an N channel mosfet needs a positive voltage between the gate and the source. You've only got 3.3V available for the gate signal in your diagram, which means that best case you're going to have a threshold voltage difference. So the source voltage is down to say 2.8V. That's not going to do much for your solenoids.
You can rearrange the system and get rid of the relay so that you are only switching the low side of the solenoid and you have diodes for the high sides. (I haven't drawn it out, but I think it should be possible.)
Alternatively, you can have N channel mosfets or NPN bipolar transistors controlling the gate of a P channel mosfet that does the actual high-side switching. Figure 3 of this doc shows the basic circuit for mosfets. If using an NPN in place of the N channel mosfet, it only requires small changes to the circuit. https://www.vishay.com/docs/70611/70611.pdf
