Crazy Smart Lock Idea - Whole Building Automation Using HA/ZWave

Curious if there’s anyone out there as crazy as me and a couple of my nerd friends at our church… We are attempting to engineer a cost effective whole building automation solution able to be managed in HA with Z-Wave 800 LR devices w/ repeaters (for their reliability) as the primary backbone. (We’ll still deploy Zigbee 3.0, Wifi, hardline POE and anything else necessary to achieve a bespoke solution to a problem though if we have to, our building is just already a 2.4Ghz nightmare from the local Radio Station up the street, so we’re hoping to rely on that band sparingly.)

With a staff of 40+, and volunteer network of about 600+, we easily have over 100 doors on our single building campus with varying levels of access and security needed, and a desire to track access events on as granular a level possible.

Before we jump headlong into the abyss, I just thought I’d ask as general a question as I can reign myself in to ask…

… Is HA (with a properly powerful local server to meet processing demand) robust enough to manage that many Z-Wave enabled smart door locks, with unique access codes (as many types as possible, facial, finger, hand, pin, fob, digital, etc) for that many users… reliably?… or at all?

I’m just curious if anyone out there is as crazy as we are, or if anyone thinks it’s just crazy enough to be possible, based on the fact Home Assistant IS in fact powerful enough to manage that kind of unique level of traffic and variables reliably!

A close second to this question is of course what door lock hardware is available today that is robust enough to handle that level of demand either? Upon initial research, most advanced locks, even Lockly Pro’s suedo-enterprise solution with management software/subscription, seem to cap out at 100 unique fingerprints, codes, fobs, etc or less.

Any insight is immensely helpful, and I so appreciate anyone who took the time to read this in advance. :facepunch:

So Home Assistant is nice project and I think if Z-Wave can handle your project size HA can too, but it is a tinker project and not really for commercial use (yet).
There is no real easy way to manage this amount of different users with different permissions. We dont even have a setting to conveniently set permissions other than admin and no admin although this is on the roadmap. We have no account level user groups.
Most of your access methods might work with some extra modifications but not natively. You can try to make it with a phone and dashboard visibility settings but to manage that is hard.
Then there is security and failsafe points. If you do know what you are doing then thats good and HA is sure good for making your own safe solution through programming, but if not do you really want to be responsible for that amount of money spend into the project when it doesn’t work or breaks and the value of the property?
Company that are specialised on that have insurance and experts working full time so that it works.
But hey I love the enthusiasm and if you go through I am happy to hear about the project. Mind that I am not a core dev and only a tinkerer myself, so dont add too much value to my thoughts. You can look through this forum, reddit or discord where people asked about similiar projects like hotels and got this response. But there are also people who did it and it works and they shared it. Maybe you find a wiser advice there or another one will comment.

At our university we have Simons Voss of you look for an enterprise solution.

The FCC specifies what frequencies can be used for what. There are no radio stations that use 2.4ghz so your interference is coming from something else.

  1. Devices included in Z-Wave Long Range mode cannot use repeaters or be used as repeaters. Essentially you either use a mesh and repeaters to achieve the range you need or you use Z-Wave Long Range.
  2. Battery powered devices can not be used as repeaters. This means that even though you might buy 100s of locks you will need to spend even more money on repeaters.
  3. NOT ALL LOCKS SUPPORT Z-Wave Long Range so you have to be careful when buying your devices. Here is a list of all the Z-Wave Certified Locks that support Z- Wave Long Range. https://products.z-wavealliance.org/Search/DoAdvancedSearch?productName=&productIdentifier=&productDescription=&category=6&brand=-1&regionId=2&supportsLongRange=on&order=. The only locks that I can find for purchase right now are the Philips, The Alfred, and The DEN. All of these locks are arround the $250 price point.
  4. Z-Wave is a wireless protocol. Just like with any other wireless protocol the signal decreases with obstacles, range, and etc. Depending on how big the property is you might have to setup multiple z-wave networks. This is extremely easy to do. Plus you can manage them from a single home assistant instance.

One thing I like about Z-Wave Locks is the number of events they can generate. Here are some of the USEFUL events the lock can generate

  • It tells you if the door was unlocked or locked using a Z-Wave Command
  • It tells you if the door was unlocked or locked using a Bluetooth Command
  • It tells you if the door was unlocked or locked using a Key
  • It tells you if the door was unlocked using the keypad as well as the user that unlocked the lock
  • It tells you if the door was unlocked using a fingerprint and the user that fingerprint belongs to
  • It tells you if the lock is jammed
  • it generates a tamper alarm when too many invalid codes are entered or someone tries to seperate the external lock mechanism from the internal lock mechanism

All of the locks that I listed earlier support 250 user codes

As I stated earlier the only Z-Wave Long Range locks that are available cost about $250 so for a 100 locks you are looking at 25K and up. I haven’t researched anything else but I feel like you should be able to find a purpouse built commercial solution in that price or for less.

1 Like

I’m with Cornell. One could install UniFis prosumer door controls for those prices.

Remember doing things for yourself is one thing doing things for a business entity or nonprofit - must work, always and liability for not working is a very real thing.

We often shake off the price of pro or prosumer gear as ‘overinflated’ sure. For home use. But there’s a bit in the price for liability and what happens when things go wrong you cannot get away from.

If you’ve ever been in a hospital in the USA you’ve probably seen red outlets those are special hospital grade. Do they cost more than standard? Yep. About double. Are they anything different? Not really just really good quality materials. The biggest one is they’re paying for liability insurance and if that device ever causes an incident. They’re covered. That’s the real difference.

For locks in a nonprofit or commercial space. Same thing applies. Odds are you can put something together that… ‘works’

How ready are you to roll the dice on works right the first time every time and won’t cause an issue one of your patrons ultimately sues you for… (something dumb none of us can even prepare for.)

So commercial space… No not even if you CAN

@JohnFLoki @cornellrwilliams You guys are absolutely amazing. As someone new to this community, your guy’s quick and detailed responses have brought a tear to my eye. My other friends in on this project are gonna make accounts, and we are gonna confer on all these details and try to ask targeted questions based on your responses that try to value your guy’s time.

In the meantime you have given an infinite amount of food for thought.

My lack of knowledge on LR enabled devices versus mesh for ZWave was a glaring wake up call that there is still much to learn. At the same time, that knowledge is extraordinarily helpful, and really freakin cool from a reliability proposition… here I was thinking “yeah it’s a mile, but i’ll still want to mesh” to now “dang, this opens up being able to reach farther places without feeling pressure to build the mesh network immediately”.

One thing we did plan to do is exactly what you mentioned about having multiple ZWave networks throughout the building in each of the server/rack rooms, and awesome to know set up is easy!

Otherwise, many more questions to come and thank you guys so much again for the wealth of knowledge! Thankful for you guys on Thanksgiving :sunglasses:

1 Like

@NathanCu Your point is so well received. This is something my colleagues and I have discussed at length and agree whole-heartedly on no doubt. Access is not something we want to gamble on reliability.

UniFi is 100% an avenue we’ve looked at, and for certain unique latch/bar- style doors, we are going to have to have a solution for similar to what UniFi offers for sure.

I like that Cornell mentioned the cost trade offs in this light as did you. There’s just a lot of extra material and man hours to consider with a UniFi type solution, but we would all agree, wires are the most reliable form of communication and always will be, and less maintenance long term.

I’m not sure what the exact quotes were from Motorola and Siemens sub contractors for doors specifically us when they came out, but safe to say, those price tags are why I’m here and we’re researching other options :joy: (Those companies of course provide HVAC, Lighting, Power Management, all those other types of big picture building automation we will get into eventually as well, we are just starting with the door locks, smart plugs and other tinkery types solutions to minor problems that can be automated as kind of the start of a proof of concept process.

Really heartening to hear there are other people out there who’ve gone down this road with success too, so that’s amazing.

1 Like