It is nearly impossible not to get a fake these days. I have about 10 of those bought at Amazon with a 3m cable lying around, and they are all fake. They work though, as most fakes do. Temperature measurements might be a bit off, but can usually be calibrated using “known” temperatures like boiling and freezing water.
One thing that often does not work for fakes is “parasite power mode”, so using only signal and ground to power the fakes. Using all 3 wires, so connecting Vcc usually fixes this.
I’m aware of that. So for paying the same amount as I paid for 10 waterproof podded ones in a steel casing with 3m of wire I would get a single original one - unpodded without wire. Adding podding goo, wire, steel casings and a lot of my time, it just did not add up. At least not compared to connecting Vcc as a solution and skipping parasite power mode.
Fun fact: I sourced a DS18B20 for a different project 2 years ago from Conrad - a single one without the requirement of being waterproof and a Conrad store in town (which they closed unfortunately). This is how I got introduced into DS18B20 being faked. I guess they will have their sourcing in order by now though.
But yes, you are right: It is possible to get original ones.
If people are unwilling to pay the equivalent of a couple of sandwiches for a quality temperature sensor they can rely on for a long time and prefer spending cents on some unreliable fake crap while wasting your time with all the nonsense you’ll have put up and live with potentially dangerous failure modes (a common one being self overheating), then that is up to everyones own choice.
But saying that it’s impossible to find non fakes is simply wrong. It’s actually very easy.
Since I didn’t pay top dollar, I can assume mine are “fake.”
OK, I’m not running a nuclear power plant. I’m checking a few temperatures in my hot tub, and (more importantly) around my boiler. If one of these fails, it’ll be bit of a nuisance. That’s all.
Now, catching fire could be a bad thing. Of course, these are potted in a steel enclosure, so the risk has got to be lower. Every device in my house is chock full of cheap Chinese components. I’m quite sure many of them could, in theory, have the same failure mode. Yet somehow houses aren’t burning down all up and down my street. I’m guessing the odds are pretty low there, too.
I mean I get this. None of us here uses DIY electronics or HA to run a nuclear power plant (hopefully ).
Still, there are some other factors beyond being a simple nuisance. The self heating thing seems to be a very common failure mode for fake components, not only the 18b20. I don’t know why exactly, but it’s a recurring phenomenon. I was looking for a max232 awhile ago to connect to an NMEA 0183 device on my boat. Among reviews on Amazon for common breakout boards (clearly with fake 232’s) were things like ‘heated up so much I couldn’t touch it’ or ‘started smoking after a few days’. I ended up buying an original from Farnell. I also had my own hands-on experience with a (fake) SIM800 module which almost caught fire, right in my attic… All that doesn’t seem to be that rare at all. It’s not a problem with Chinese components per-se. Not all components manufactured in China are fake and low quality. But almost all you can get on Ali/AMZN/Ebay are, because there’s no quality control and the aforementioned sites are pretty much the dumping ground for low quality crap electronics that didn’t make the QA requirements elsewhere.
And there’s also another angle to it. This might be professional bias, I don’t know. But counterfeit electronic components are a massive problem for the industry. They stifle innovation, make legitimate manufacturers drop certain older but tried components because they are not profitable anymore due to all the clones, and can be downright dangerous if they end up in your car, in some medical device, or who knows, in the control systems of your local power plant. I just don’t want to support this. Unknowingly buying fake stuff is one thing, but knowingly buying it and even actively seeking out these components because they cost a few bucks less than the real stuff… That’s a mindset I have a hard time wrapping my head around.
I mean, let’s be honest here. This is just peanuts. We’re not saving thousands of USD/EUR/whatever here. Smart home tech is a First World ™ hobby. We don’t need this for our survival. We indulge in this (passionate !) hobby because we can spare the cash for it. Being in this position, the least we can do is try to not make the counterfeit problem even worse than what it already is.
Most DS18B20 chips are fake. Here is an Arduino sketch to test if yours is fake. I stopped testing mine because they are all fake, except for one that I salvaged from an HVAC thermostat.
The biggest risk from using a fake is that it won’t be calibrated and the linearity is questionable. But they are good enough for most of our uses.
If you are using the device in a critical operation, such as watching fermentation temperatures in a bioreactor, then buy the $15 chip. If you just want to know if the hot tub is ready for guests, then the fakes are just fine.
I have five of these types of sensors connected to two different smart implants. Four sensors on one of them and a single on the second. For some reason the second smart implant is just not taking more sensors. I do see that the firmware version is 5.1 on that second smart implant, whereas it is 5.2 on the one that is allowing multiple sensors. Perhaps check which firmware you have, as apparently they did improve the ability to handle multiple sensors. Unfortunately though, if that is the cause, then it is difficult to update the firmware on these because you need a fibaro hub. They don’t distribute the firmware separately.
Never had a device of this brand but essentially they are selling smart home stuff for prices you should get quality.
I know a couple of people who actually own various types of this (not cheap) devices and most of the owners are not to happy about them actually. More than one hardware failure and limited firmware updates (they rather want you sell you a new device then just ship you a new software) are things which annoys some owners.
I stick to my dirt cheap esp(home) quality builds because I know what I get
Yes, they have a product page. Lots of pretty pictures, but no technical details at all. And no prices. I guess that “if you have to ask, you can’t afford it”.
They have very extensive documentation for all their devices, down to the byte data formats of their configuration and command classes, just in case you want to implement your own driver… Their devices are extremely configurable, have great build quality, 3 year+ battery life and they’ve been super reliable for me. I’m very happy with them. Their firmware update policy is indeed weird, but I don’t care. Their devices work great for me, no need to update.
Not too hard to search for them on amazon, you know…
They have around 4 times the prices for a wifi (esp) based equivalent when it’s about the type of relay you put behind your wall switch (“shelly style”). It was exactly this type of device which died more than ones at a friends house actually an he meanwhile migrated away from fibaro. On the other hand I run devices with similar form and function which are esp/wifi based and didn’t have even one failure in 5 years. Obviously this is very subjective and I have only around 100 esp based devices which actually less than half are “commercial” products.
That’s well said. Other people/customer do care and there is actually more than one which are not “happy”. Buying “premium” but ether being a beta tester or don’t get updates at all isn’t what I would expect actually if I would be willing to pay top dollar
There are always going to be people who are unhappy with something. Any other news ? I am very happy with my Fibaro devices. Their FGMS001 motion sensor is awesome. I don’t care about the firmware as long as it works. Just like you don’t care about that mysterious binary blob on all your ESPs that you cannot replace, you don’t know what’s in it, that is entirely undocumented and Expressif is suspiciously silent about - yet all your ESP communications run through it. If you were really serious about all that, you would use the Pico W, at least their wifi stack is open source and you can replace it. Anyway, this is really offtopic, so I’m going to leave this here. You should just try to understand that some people are absolutely happy to pay what you call ‘top dollar’ (jesus, it’s like 40 bucks…) for a great looking and reliable off the shelf sensor.
Every one of my ESP devices have ESPHome flashed to it. ESPHome is open source and you can download the source code if you really want to know what’s inside the “mysterious binary blob”.
Appliances like the Fibaro system are made for users who use devices but don’t want to understand how they work. Which is fine. But I suspect that the vast majority of Home Assistant users are more curious about how their devices work, not just that they work. Since I had no clue what a Fibaro is, I am likely such a tinkerer.
So, when you ask hardware questions, you are more likely going to get replies from tinkerers who will reply with DIY hardware solutions that work for them.
That’s not what this is about. ESPHome (or any other user uploadable firmware) replaces only a part of the code running on your ESP. There’s a second part that you cannot replace or access. It manages the wifi part of the device, the wifi stack, the radio packet control, etc. You have no access to this code.
There are some legal reasons for this. In order for manufacturers to get FCC approval, some of the lowest level code (the one that manages the radio) is required to be non-user modifiable. If users modify it, the device loses compliance. But that does not mean that this code has to be closed source or undocumented. And it also doesn’t mean that the wifi stack (the code that manages the TCP/IP layer) can’t be replaced. Yet all that is closed in ESPs and Espressif hasn’t documented anything around this. Who knows what is in that stack. Maybe a backdoor. Since the code is not open, noone but Espressif knows.
Or users who know exactly how they work and could build their own if they wanted to, because creating embedded HW and SW is an important part of their daytime job (someone like, you know, me for example), but who just don’t want to spend the little free time they have on that and are perfectly happy to use a well build off the shelf device. There is absolutely nothing wrong with using off the shelf devices, even if it’s only about aesthetics and WAF.
Had you read any of the links I posted, you would know
Until you posted the links, I had no clue what Fibaro was. That was my point.
No. The build from IDE contains the Espressif Arduino Core for ESP8266
This is from the Expressif site: "Many efforts around the ESP8266 Arduino Core make it worth the effort to use it over the stock SDK, besides simply being as open source as possible. Arduino Core includes C++ classes for a web server (SSL/TLS available), a filesystem implementation, and both a DNS and mDNS server. Client tools for communicating over TCP/IP, including an implementation for making HTTP requests, are also provided."
Since the Arduino code is open source, you do have access to this code.
No. This is just a higher level tool collection and connection layers eventually linking to the ESP NONOS SDK which contains the low level binary components. Even the lwIP-2 linklayer used by the Arduino Core is just a connection layer to the NONOS SDK. Their RTOS SDK provides some more tools with source, but again it ends with the binary low level drivers. See their 8266 SDK docs, chapter 1.4:
ESP8266 FW (Firmware) has been provided in binary format files (.BIN) that can be
downloaded directly to the HDK.
That binary FW contains the actual low level code.
).
