Since I am planning to build a house, I am naturally also considering smart home solutions. As this topic is completely new to me, I first need to do some research. I would therefore like to clarify some basic questions and facts before proceeding. I will simply write down my thoughts as if they were facts and would ask you to correct me where I am wrong 🙂
I want to leave KNX or BUS systems aside for now and focus only on wireless solutions, so please do not discuss which makes more sense or is "better." One step at a time; this is purely about understanding :p
There are many different wireless protocols. Open ones like Z-Wave, ZigBee; "semi-open" like eNet; and proprietary ones like Somfy io.
Somfy has the advantage that everything comes from one source and it should be plug and play. The ideal solution for beginners—you simply put together what you need, install it, and it works right away. The big disadvantage, of course, is that you can only integrate Somfy or Somfy-approved sensors and actuators and are dependent on Somfy continuing to develop new products and supporting the current protocol.
eNet is a protocol from a few premium manufacturers, and as such, the devices are quite expensive. Although it is an open protocol, it is supported by only a few providers and products. So it has the same disadvantage as Somfy: limited selection and dependency on a small number of providers.
The open protocols, which are basically standards, have the advantage of many providers. However, this is also the biggest disadvantage because it’s easy to lose oversight, you end up assembling the whole system yourself without it all coming from one source, and there are likely to be significant differences in quality.
Is this roughly correct?
I want to leave KNX or BUS systems aside for now and focus only on wireless solutions, so please do not discuss which makes more sense or is "better." One step at a time; this is purely about understanding :p
There are many different wireless protocols. Open ones like Z-Wave, ZigBee; "semi-open" like eNet; and proprietary ones like Somfy io.
Somfy has the advantage that everything comes from one source and it should be plug and play. The ideal solution for beginners—you simply put together what you need, install it, and it works right away. The big disadvantage, of course, is that you can only integrate Somfy or Somfy-approved sensors and actuators and are dependent on Somfy continuing to develop new products and supporting the current protocol.
eNet is a protocol from a few premium manufacturers, and as such, the devices are quite expensive. Although it is an open protocol, it is supported by only a few providers and products. So it has the same disadvantage as Somfy: limited selection and dependency on a small number of providers.
The open protocols, which are basically standards, have the advantage of many providers. However, this is also the biggest disadvantage because it’s easy to lose oversight, you end up assembling the whole system yourself without it all coming from one source, and there are likely to be significant differences in quality.
Is this roughly correct?
HansDampf1311 schrieb:
This is roughly my approach and why I tend to prefer wireless, fully aware that a BUS system is technically better. More important than "systems" and "protocols" seems to me to design neutrally at the level of "network topology". Wireless, in the sense of being consistently and radically completely cordless, is neither feasible nor desirable in my opinion. Therefore, one must always consider how and where to combine wireless and wired components. And although not everything that uses a bus is bad, I generally find a star topology much more appealing. The moment you try to imagine wireless within a topology, most neurotypical people face complete philosophical confusion. When it comes to the number and cross-sections of Siemens air hooks for the wireless conduit, even I gladly use a scratch pad ;-).
But to be clear: I recommend the original poster focus on the basics and essentials for quite a while before attempting to make fine-tuned adjustments at the fifth decimal place.
https://www.instagram.com/11antgmxde/
https://www.linkedin.com/company/bauen-jetzt/
S
Stefan0016 Mar 2021 21:5211ant schrieb:
More important than "systems" and "protocols," to me it seems essential to design at the level of "network topology" in a system-neutral way.Which system or protocol change does not also force a break/node in the topology? Therefore, there are hardly any topologies left that could be designed completely system-neutral.11ant schrieb:
I consider "wireless" in the sense of "consistently and radically cordless" neither feasible nor desirable; therefore, in my opinion, one always has to consider where and how to integrate the wireless and wired components. And even though not everything related to buses is bad: I generally find a star topology much more appealing.Could you please provide a reason for that?Stefan001 schrieb:
Which system or protocol change does not also cause a break or node in the topology? In this regard, there are hardly any topologies left that could be designed to be system-neutral. That is why I also often advise against deciding not to install cables for a possible future system change when currently choosing a bus-based system.
Stefan001 schrieb:
Could you please provide a reason for that? Wireless connections – which I like to call “cordless cables” – are more expensive and less reliable than wired cables. They have their justified applications but are often used out of convenience even where they are unnecessary, especially when connecting components that are fixed in place at both ends. In practice, signals rarely travel the entire way through the house wirelessly; instead, they switch between wireless and wired segments.
https://www.instagram.com/11antgmxde/
https://www.linkedin.com/company/bauen-jetzt/
A
AllThumbs7 Mar 2021 14:03T_im_Norden schrieb:
You don’t have to use KNX even if the wiring is installed, but by centralizing the wiring in the distribution box, you have access to everything.
[..]
If, for example, you run the wiring from the light fixtures to the distribution box and also the wiring from the switches there, then you can:
- Use a regular switch
- Use a push button
- Use a sensor
- Use an automation system But doesn’t that make it even more expensive than starting with KNX from the beginning?
With KNX, at least I can wire the push buttons in series. Wiring them in a star topology now requires even more cables.
bauenmk20209 Mar 2021 09:56A general question for those who reject wireless and believe that only "wired" is the way to go (in new construction):
Do you not use a single (smart) device connected wirelessly? For example, something like a robot vacuum cleaner or similar, which only works wirelessly—am I mistaken?
What do you do with cables that aren’t installed? Do you add a bus cable afterward, or does it end up being a wireless device after all?
Do you not use a single (smart) device connected wirelessly? For example, something like a robot vacuum cleaner or similar, which only works wirelessly—am I mistaken?
What do you do with cables that aren’t installed? Do you add a bus cable afterward, or does it end up being a wireless device after all?
Here you need to differentiate. A robotic vacuum cleaner or a lawn mower, for example, are just a few individual devices that are usually unproblematic.
In a smart building, however, you may have hundreds of devices interacting with each other, making smooth communication essential. Unfortunately, beyond certain amounts and distances, wireless transmitters can start interfering with other devices in the system (essentially interfering with themselves), leading to connection drops, collisions, and so on.
Sure, garden lighting can be easily realized wirelessly, but even just the wireless connection between a basement and an upper floor can sometimes be nearly impossible to achieve (with limited resources). Wired connections are simply much easier and more straightforward in such cases.
Additionally, wireless devices are often visually unappealing. For example, in a modern house, I want to have as little visible technology as possible. Surface-mounted switches, sensors, actuators, and similar devices are the complete opposite of that.
Simply put: build a backbone with cables first, and later you can add some wireless elements here and there. If you set everything up wirelessly from the start, you will sooner or later reach the limits of the system—sometimes faster than you would like.
On the other hand, if you only plan a small amount of automation on the level of a model railway setup, you can of course go fully wireless. Unfortunately, it still happens quite often that manufacturers discontinue certain wireless product lines. So careful selection is necessary. With wiring, it’s simply easier.
Please provide one or more examples where cables might not be installed. Preliminary note: outdoors, such as in gardens, this can happen relatively quickly.
In a smart building, however, you may have hundreds of devices interacting with each other, making smooth communication essential. Unfortunately, beyond certain amounts and distances, wireless transmitters can start interfering with other devices in the system (essentially interfering with themselves), leading to connection drops, collisions, and so on.
Sure, garden lighting can be easily realized wirelessly, but even just the wireless connection between a basement and an upper floor can sometimes be nearly impossible to achieve (with limited resources). Wired connections are simply much easier and more straightforward in such cases.
Additionally, wireless devices are often visually unappealing. For example, in a modern house, I want to have as little visible technology as possible. Surface-mounted switches, sensors, actuators, and similar devices are the complete opposite of that.
Simply put: build a backbone with cables first, and later you can add some wireless elements here and there. If you set everything up wirelessly from the start, you will sooner or later reach the limits of the system—sometimes faster than you would like.
On the other hand, if you only plan a small amount of automation on the level of a model railway setup, you can of course go fully wireless. Unfortunately, it still happens quite often that manufacturers discontinue certain wireless product lines. So careful selection is necessary. With wiring, it’s simply easier.
bauenmk2020 schrieb:That almost never happens because the cables are usually already in place. Otherwise, either poor planning occurred (or experience-based recommendations were ignored), or it was planned ahead and cable installation is still possible (which is highly likely).
What do you do with cables that are not already installed? Do you pull in a bus cable later, or does it end up being a wireless device after all?
Please provide one or more examples where cables might not be installed. Preliminary note: outdoors, such as in gardens, this can happen relatively quickly.