Background:
I own an old house (about 270 years old) where my company and my apartment were located. As the office kept expanding with new employees being hired, I moved into a separate apartment due to space constraints.
However, this was not meant to be permanent. I wanted to have something of my own again, but this time “small but nice.”
After considering various options from tiny houses to mini houses, the decision was made to build a small semi-detached house and use one unit myself (about 65m² (700 sq ft) plus terrace and garden).
My brother is building the other semi-detached house, and both units will then be rented out.
High-quality, ecological, and sustainable construction was important to us from the start. A healthy indoor climate, photovoltaic system, heat pump, and controlled residential ventilation were therefore fixed requirements.
At first, I considered some home automation (implemented, for example, with Shelly devices installed later to control certain functions remotely).
But since our company is involved not only in general IT activities but also in software development, programming, and consulting on IT infrastructure and digitalization for larger organizations such as hospitals, and since my brother is the lead developer of a software that has been on the market for over 20 years and is the market leader in Germany in this field…
Couldn’t we do it ourselves?
Yes, we could!
“Coincidentally,” a bachelor thesis on an open (non-proprietary) smart home system was submitted to my brother (who teaches computer science at FOM among other things).
Based on this and other work that proved certain fundamentals, plus freely available information showing many other possibilities of a smart home, we formulated our requirements.
We wanted a “Smart Home” (as we understood it at the time).
We wanted to at least control lighting and blinds, have a front door with motorized lock, possibly video intercom systems, presence monitoring, etc. The wishes naturally grew as we learned what was possible. So, voice control via, for example, Alexa should also be possible (for those who find it useful).
Our key criteria were:
• Open system (no proprietary/cloud dependencies, etc.)
• All active components centralized in the distribution board if possible
• No “bus switches,” but “normal” Gira, Jung, etc. push buttons
• Affordable!
• The base programming can be quite complex (done by experts), but afterwards the system should be adjustable by “non-technical users” via a GUI (This alone could lead to lengthy discussions)
• We will do the wiring and programming ourselves (my brother and one of our partners are software developers, and I am an IT specialist)
• The wiring will be done with plenty of spare capacity and smart home compatibility.
-> Lighting circuits (lamps) all go directly to the distribution board
-> Outlets are wired room-wise to the distribution board and use 5-wire/three-phase cables so nearly any outlet can be switched later with small wiring adjustments.
-> All push buttons are connected with (plenty of) LAN cable directly to the distribution board
(Discussion about CAT7a/CAT8.x LAN cabling or 10DA took place: Result: Almost everything will be done as CAT7A duplex because, overall, it is easier to handle and cheaper for us.)
The result will soon be installed as a prototype in my unit (half a semi-detached house, about 65m² (700 sq ft)). The photo of the small sub-distribution board is only a small part of the test system currently being assembled! The Eltakos were installed only for comparison (space requirements vs. the Controllino) and will probably not be used in the finished system.
Almost anything “speaking” a smart home language can be integrated into the system via the appropriate binding.
The core hardware (the “brain” of the system) will cost about 1000 to 1500 euros (approximately $1100 to $1650). It mainly consists of the SmartHome Manager (Raspberry Pi) and Controllino Mega pure (Arduino).
The programming of OpenHub, etc. (open source) is done by our software developers/programmers.
Temperature, distance meters, motion detectors, humidity, CO2 sensors, and similar devices can easily be integrated. Small touchscreens (around 20€ (about $22)) as “cornerstone components” usually cost only a few euros.
I will be happy to report on further steps and on the prototype development once progress continues.
Notes:
The advantages of KNX are often mentioned, and some of these certainly apply. In our smart home, if the SmartHome Manager (which is also industrial hardware) fails, almost nothing works. With KNX, all unaffected subsystems continue to run. However, the example described by @Mycraft: “… It’s enough to just connect another push button to the bus cable and configure it…” is hardly feasible in practice. The user cannot do it themselves, the proverbial “unicorn” (ideal technician) is not there, and if the “system integrator” does come eventually, every small change or adjustment is extremely expensive. (These are the experiences of two people with KNX in our circle of acquaintances.)
Please do not fill this thread with pros and cons of the various systems. It only concerns “our” system based on the hardware and software components mentioned at the beginning – thanks!
I own an old house (about 270 years old) where my company and my apartment were located. As the office kept expanding with new employees being hired, I moved into a separate apartment due to space constraints.
However, this was not meant to be permanent. I wanted to have something of my own again, but this time “small but nice.”
After considering various options from tiny houses to mini houses, the decision was made to build a small semi-detached house and use one unit myself (about 65m² (700 sq ft) plus terrace and garden).
My brother is building the other semi-detached house, and both units will then be rented out.
High-quality, ecological, and sustainable construction was important to us from the start. A healthy indoor climate, photovoltaic system, heat pump, and controlled residential ventilation were therefore fixed requirements.
At first, I considered some home automation (implemented, for example, with Shelly devices installed later to control certain functions remotely).
But since our company is involved not only in general IT activities but also in software development, programming, and consulting on IT infrastructure and digitalization for larger organizations such as hospitals, and since my brother is the lead developer of a software that has been on the market for over 20 years and is the market leader in Germany in this field…
Couldn’t we do it ourselves?
Yes, we could!
“Coincidentally,” a bachelor thesis on an open (non-proprietary) smart home system was submitted to my brother (who teaches computer science at FOM among other things).
Based on this and other work that proved certain fundamentals, plus freely available information showing many other possibilities of a smart home, we formulated our requirements.
We wanted a “Smart Home” (as we understood it at the time).
We wanted to at least control lighting and blinds, have a front door with motorized lock, possibly video intercom systems, presence monitoring, etc. The wishes naturally grew as we learned what was possible. So, voice control via, for example, Alexa should also be possible (for those who find it useful).
Our key criteria were:
• Open system (no proprietary/cloud dependencies, etc.)
• All active components centralized in the distribution board if possible
• No “bus switches,” but “normal” Gira, Jung, etc. push buttons
• Affordable!
• The base programming can be quite complex (done by experts), but afterwards the system should be adjustable by “non-technical users” via a GUI (This alone could lead to lengthy discussions)
• We will do the wiring and programming ourselves (my brother and one of our partners are software developers, and I am an IT specialist)
• The wiring will be done with plenty of spare capacity and smart home compatibility.
-> Lighting circuits (lamps) all go directly to the distribution board
-> Outlets are wired room-wise to the distribution board and use 5-wire/three-phase cables so nearly any outlet can be switched later with small wiring adjustments.
-> All push buttons are connected with (plenty of) LAN cable directly to the distribution board
(Discussion about CAT7a/CAT8.x LAN cabling or 10DA took place: Result: Almost everything will be done as CAT7A duplex because, overall, it is easier to handle and cheaper for us.)
The result will soon be installed as a prototype in my unit (half a semi-detached house, about 65m² (700 sq ft)). The photo of the small sub-distribution board is only a small part of the test system currently being assembled! The Eltakos were installed only for comparison (space requirements vs. the Controllino) and will probably not be used in the finished system.
Almost anything “speaking” a smart home language can be integrated into the system via the appropriate binding.
The core hardware (the “brain” of the system) will cost about 1000 to 1500 euros (approximately $1100 to $1650). It mainly consists of the SmartHome Manager (Raspberry Pi) and Controllino Mega pure (Arduino).
The programming of OpenHub, etc. (open source) is done by our software developers/programmers.
Temperature, distance meters, motion detectors, humidity, CO2 sensors, and similar devices can easily be integrated. Small touchscreens (around 20€ (about $22)) as “cornerstone components” usually cost only a few euros.
I will be happy to report on further steps and on the prototype development once progress continues.
Notes:
The advantages of KNX are often mentioned, and some of these certainly apply. In our smart home, if the SmartHome Manager (which is also industrial hardware) fails, almost nothing works. With KNX, all unaffected subsystems continue to run. However, the example described by @Mycraft: “… It’s enough to just connect another push button to the bus cable and configure it…” is hardly feasible in practice. The user cannot do it themselves, the proverbial “unicorn” (ideal technician) is not there, and if the “system integrator” does come eventually, every small change or adjustment is extremely expensive. (These are the experiences of two people with KNX in our circle of acquaintances.)
Please do not fill this thread with pros and cons of the various systems. It only concerns “our” system based on the hardware and software components mentioned at the beginning – thanks!
K
knalltüte3 Mar 2021 22:06Brief update:
Purchased so far: 15 Controllino Max units + 30 relays rated 6A + 5 Rademacher multiple control relays 2751 (each for 2 external venetian blinds). 1 Occy Mini Design PM. Hundreds of meters of control wire/stranded wire. Thousands of crimp sleeves. Three different wire strippers until I was satisfied with the result on CAT7A cables. Matching crimping tool included. Proper woodworking tools for making the required holes in the ceilings. About 2m² (21.5 sq ft) of workshop space occupied by storage of components and accessories. Two desks fully occupied with one prototype each.
We then decided to install at least two Controllino Max units per housing unit (instead of one Mega) to avoid being completely in the dark if one fails. So also at least two power supplies. Additional standard DIN rail / 24V power supplies will be kept in stock as spares. Higher loads are generally not switched directly; if necessary (e.g., outlets with “high” consumption), they will be controlled through separate relays. External venetian blinds will likely be controlled via special multi-relays or several individual relays, but this is still under testing.
We (I :eek 🙂 have about 100 data lines per housing unit (there are slight differences) for patch cabling. The number of outgoing lines to the “Smart Home department” is naturally significantly less. In total, each housing unit has 48 TE (module spaces) available in the distribution board for patch cabling and smart home components, plus 24 TE that can be installed hidden with a low installation height. This requires a high packing density for the patch cabling. I have already described how this will be implemented.
The basic programming for “light” on/off is completed, including the appropriate GUI. Programming for “dimming” + “colors” for lighting scenes + “control of external venetian blinds” is in progress.
Since we are all currently extremely busy with work (lucky us!), it may happen that some (optional) components/sensors will only be integrated later during the “actively inhabited phase.”
Purchased so far: 15 Controllino Max units + 30 relays rated 6A + 5 Rademacher multiple control relays 2751 (each for 2 external venetian blinds). 1 Occy Mini Design PM. Hundreds of meters of control wire/stranded wire. Thousands of crimp sleeves. Three different wire strippers until I was satisfied with the result on CAT7A cables. Matching crimping tool included. Proper woodworking tools for making the required holes in the ceilings. About 2m² (21.5 sq ft) of workshop space occupied by storage of components and accessories. Two desks fully occupied with one prototype each.
We then decided to install at least two Controllino Max units per housing unit (instead of one Mega) to avoid being completely in the dark if one fails. So also at least two power supplies. Additional standard DIN rail / 24V power supplies will be kept in stock as spares. Higher loads are generally not switched directly; if necessary (e.g., outlets with “high” consumption), they will be controlled through separate relays. External venetian blinds will likely be controlled via special multi-relays or several individual relays, but this is still under testing.
We (I :eek 🙂 have about 100 data lines per housing unit (there are slight differences) for patch cabling. The number of outgoing lines to the “Smart Home department” is naturally significantly less. In total, each housing unit has 48 TE (module spaces) available in the distribution board for patch cabling and smart home components, plus 24 TE that can be installed hidden with a low installation height. This requires a high packing density for the patch cabling. I have already described how this will be implemented.
The basic programming for “light” on/off is completed, including the appropriate GUI. Programming for “dimming” + “colors” for lighting scenes + “control of external venetian blinds” is in progress.
Since we are all currently extremely busy with work (lucky us!), it may happen that some (optional) components/sensors will only be integrated later during the “actively inhabited phase.”
untergasse434 Mar 2021 19:19I've lost track... is this now a bus topology or a real (complicated) star topology like with Loxone/PLC?
//Edit: Is it possible that sometimes when I click Edit for the first time, a new browser window with ads opens, or am I just accidentally moving the mouse too much lately?
//Edit: Is it possible that sometimes when I click Edit for the first time, a new browser window with ads opens, or am I just accidentally moving the mouse too much lately?
untergasse43 schrieb:
//Edit: Is it possible that sometimes when clicking "Edit" for the first time, a new browser window with ads opens, or am I just accidentally moving the mouse too much lately?No, that’s just the forum.
K
knalltüte5 Mar 2021 13:55I'm also starting to lose track soon. It's a good thing we label everything immediately or promptly :p
The chosen star topology makes a lot of sense for me/us, even if the cabling effort might be higher compared to a typical bus topology.
Our cable trays were once documented in words and pictures in that unfortunately lost thread.
I will, when I have the chance, put together a rough and basic cost overview (infrastructure, hardware). Of course, not everyone can directly apply this since we have different purchasing conditions than a private end customer, and also because we are doing most of the drywall and electrical work ourselves.
The chosen star topology makes a lot of sense for me/us, even if the cabling effort might be higher compared to a typical bus topology.
Our cable trays were once documented in words and pictures in that unfortunately lost thread.
I will, when I have the chance, put together a rough and basic cost overview (infrastructure, hardware). Of course, not everyone can directly apply this since we have different purchasing conditions than a private end customer, and also because we are doing most of the drywall and electrical work ourselves.
K
knalltüte6 Mar 2021 19:07Attached is a photo showing the current status of the wiring in my upper floor. Only a few cables remain to be installed for the motion sensors and ceiling lights. I’m starting to worry a little whether everything will actually fit inside the cabinet. Some LAN cables will be housed separately in a 19" cabinet above the electrical distribution board, probably around 4 or 6 rack units (U) high.
It depends on the UV (distribution board). But from what it looks like, you don’t have much space. If necessary, there are also floor-standing cabinets. If I were you, I would also relocate the network cabling to a small 6U cabinet above. I solved it similarly, except that in my case, the washing machine/dryer is located below the network distribution.
Similar topics