I was asked to help lower the building costs of the Building Automation of a ~2000m3 building. I haven’t got the design papers yet, but I’ve been preparing by thinking and reading about this. My hope is to be able to suggest replacing all components with Home Assistant -compatible ones and thus save a lot of money. Traditional, big-name components cost a lot!
What keeps confusing me is that Building Automation Systems seem to be on two separate islands. There’s Home Assistant and the other Open Source systems with pretty compatible ecosystems. Then there’s the conventional / professional systems that are actually used in most buildings. There seems to be a chasm between the two. It’s hard to combine the two into a unified system. Why is that?
It would be easy if I could just start with HA and pick the components, but there’s quite many things that don’t seem to exist, like 0-10v controllers and duct-valve actuators (sorry I don’t know the proper terms). Those would require soldering my own components, which is a bit too DIY -esque.
Starting from the other side, there’s an abundance of choice of components. But they cost lots. Also, they aren’t that much more reliable to justify the 5-50x price. Like 100€ for a H&T sensor. Then if I get all these components there’s no way to get them into Home Assistant. They are supposed to be connected to a 5000€ Building Automation Controller that speaks BACNet or LON. Apparently these are awesome. But then you need a licensed professional with some proprietary software to program it all, after which it’s hard and expensive to make any changes. Of course that’s how they are designed. The system is supposed to work autonomously. I just don’t like that. It gives little room to improve anything.
One thing that seems to be common is tcp/ip communication. The core difference seems to be the communication protocols on top: mqtt / http / coiot vs bacnet / Lon / modbus / 0-10v.
I’m aware that the Home Assistant route would require much more from my part. Lots of care. I’m ready to commit to this. But it looks like I won’t be able to offer a solution.
I would appreciate any help with this. How should I approach this challenge? Do I need to pick one or the other, or is there a good way to combine conventional components and Home Assistant?
I sometimes wonder about the same. I haven’t executed any project but I think the sweet spot would be to treat it as a brown-field deployment, where there’s some stuff and you want to upgrade with more.
Haven’t done a profound market research, but most industrial automation systems are starting to offer ‘gateways’ or translation layers from their own systems into a more open such as MQTT/AMQP. I would consider adding those to the shopping carts, and analyze the data in HA, with simpler, non-critical logic in HA, but still keeping the ‘professional brains’ in the network, specially with critical infrastructure (fire-drills extinguishers, elevators, security door handles, etc…) where time-sensitive is critical to react.
What evidence do you have to support that statement?
The requirements for a commercial structure are usually orders of magnitude more complex than average home automation and require 100% reliability. Paying tenants don’t tolerate unresponsive lighting and one slip/fall incident can cost an owner millions of dollars… In some cases there are regulatory requirements that are much more stringent than residential ones. So don’t just discount commercial quality gear just because you haven’t spent time to investigate the reasons it costs more. If you were in the states and called any regional professional lighting dealer (who knows thier products) they will be able to explain all the differences between thier gear and what you might use in a residential install. Some use different materials some have custom rf specs. Etc. Etc.
So before you try to re-spec ANY of it understand exactly why it was spec’d in the first place.
My father was a commercial lighting dealer for 30 years and I helped many MANY nights in college helping him with ‘takeoffs’ on commercial blueprints. Its part of why I’m so interested in lighting and automation in the first place. You’d be absolutely FLOORED by how many lighting control variants are available specifically because you have to meet a very special building code requirement or client request. (go to your local electric supply houae and ask to see the Cooper lighting commercial controls catalog… Bring a forklift it’s chonky)
Devices are expensive. Controls are expensive. Miniaturization is expensive. Reliability in all of those things at the same time are ridiculously expensive.
Its that three legged stool again. Reliable, cheap, good. Pick two. You don’t get all three at once.
I work in a new building built to the standards required for our government facilities, and stuff breaks all the time. Also electrical equipment built using industrial quality components get failures every once in a while. So just from experience I know you don’t get 100% reliability by paying lots. On the other hand many consumer grade products work perfectly.
Your point about understanding the requirements is good. I take it.