Should i have really spent my money on a HC2?

[Produno 15]

Im hoping to outfit my new home which im currently renovating with full home automation. I was sucked into Fibaro due to their glossy advertising which inspired me, so i decided to go with the HC2.

Though after reading these forums a lot lately I see a lot of concerns with Fibaro and their system.

Should I really have gone with the HC2? Is my smart home not going to be so smart and just a massive pita? Im hoping a few here can give me some reassurances.

Also what Fibaro products come recommended and which should I avoid like the plague?

Cheers.

[Dalle1985 16]

The panic issue should be solved with thermostats of version LC-13. I have a zniffer set up and I do not see any panicing on the devices with the LC-13 firmware. My old LC-12 however... That is a different story.

I have a hard time seeing though how you'd be able to have a house where 20+ devices can roame on the same network. If your house is large enough to have 20+ radiators, then it is probably also too big for a single controller. I have 360 m2 and 18 radiators - however, one floor is served by a separate controller (not Fibaro).

[robmac 173]

robmac, in most you got right but please remember that battery device == battery device

for example some of devices need wakeup every 4000+ with them should not be a problem

some other like zwave locks use FLIRS - with them should not be problem too

and here we got a group with device with 1 -1000s wakeup we can say there is two groups in that group

- Horstamnn thermostat send wakeup notification and wait for controller - with this device should not be problem

second group is for example danfoss where he send wakeup notification frame and give some time to answer controller, when controller will not answer danfoss starting panicking.

In theory should be ok but there is many situation when we got that problem

for example when you want do action like turn ON group of devices

- controller is talking with other battery device

- controller is rebuild mesh

- controller is learn mode

There is no good solution for this situation if we give danfoss bigger priority then when you will want turn On light you will saw lag because controller first must answer all danfoss with wakeup notification frame. If we give bigger priority to control frame like turn ON/off then change that some danfoss will get panic is higher. Even Danfoss with his own protocol and controller not support more then 20 devices on it So every installer should know that when client want more thermostat he should group them in diffrent zwave networks

That is all good Andrew and I understand when there is a rebuild or configuration going on some devices could panic if they wake and get no response. If the network and nodes in the network are busy I can understand but this is a special case and not a real issue.

But why should the limit be only 20 in normal operation. There should be no issue other than when we get that strange pause moment that impacts all traffic that I hope the new engine will fix.

If you consider that Danfoss by default configure to either 900 or 1800s wake depending on version. As the valves were all likely installed at different times or had batteries changed or wake set they should all be randomly out of sync. Even if we could set the internal clock and send a schedule I am not sure they would wake in sync.

So we have 20 odd devices configured to default waking time we get at worst 4 times an hour from the old and new ones 2 times per hour. So that is only 80 or 40 wakes an hour for 20 valves. So probability of many of the devices awake in the same minute is low.

We have all observed that the controller can send a message in a fraction of a second so why the issue?

From the manual "The nodeID set in the WAKE_UP_INTERVAL_SET command must be for a permanently listening device which reponds to the commands sent from living connect Z. This means PC’s with USB sticks will only work if the PC is never turned off. If the controller is turned off for extended periods, living connect Z will use the batteries too fast."

I only get a few months life out of my batteries even during the summer months when the set-points are never changed from 11C so no messages to send from HC2.

Does this indicate that the valves regularly are not getting a response fast enough from HC2? That would be strange as 99.99% of time my network is lightening fast and my HC2 is never turned off.

[alex88 71]

Back on topic… my 2 cents worth on “Should i have really spent my money on a HC2”

Yes, if you like tinkering, are tolerant with things going wrong and realise that it’s not a finished product

No, if you want something that just works OOTB

It’s great to hear some people sharing their positive experiences with Fibaro however for the rest of us it has been a painful journey.

To give you my latest example and it’s just an example …. I went to dry my hands this morning and noticed that the towel rail in my bathroom was hot, so I went to the HC2 event panel and noticed that the heating had been turned on in the bathroom from 8 am to 8:30am. This aligned exactly with the heat sensor in my bathroom, which is a Fibaro Universal Module with a DS18B20 connected to it, reporting a temp of 0C at 8am. At 7:30am it reported just under 22C then at 8am it reported 0C and then at 8:30am it jumped back to over 22C.

A smart HA solution knows that the temp can’t drop by 22C in 30 minutes and wouldn’t turn the heating on. Instead it would give you a warning that the temp module is faulty. Now the question as always with my HC2 is, do I write a custom script to warn me when the temp changes by more than xC and also write a custom heating script to control my heating so that this doesn’t happen again (tinkering) or do I just live with it (tolerance).

It’s not the best example as there are lots of serious issues out there and I for one have a tally of issues running at around 15+.

We all patiently (or not so patiently for some) wait for a new stable release of v4.

[robmac 173]

No, if you want something that just works OOTB

Totally agree I tried to put it all in perspective a while back.

Please login or register to see this link.

While others here have shouted how wonderful other systems are, you go on the vendors forums and you see similar issues or the packages are out and out technical knowledge required. I don't think anything is currently OOTB. Before some Cupertino fan bleats about homekit. It is an API to develop on ... hardly OOTB.

So IMO if you bought it thinking it would be OOTB you should not have spent your money. The only way to get OOTB automation just now is to buy it from someone who does all the work for you.

[A.Socha 133]

robmac, yes but please remember that most client want fast update with new temperature, they don't want wait long for action

Number 20 is danfoss limitation

[robmac 173]

robmac, yes but please remember that most client want fast update with new temperature, they don't want wait long for action

Number 20 is danfoss limitation

sorry still off topic ...

But that should not be impacted by the number of devices. You set wake to 900 on a device then when you send it a new value it updates within 15 mins. Could be < 1 min or could be all 15. How does 20 make a difference.

20 is the limitation for their controller. It is not intrinsic to the valves. I have no idea what the reason for the limitation is in their controller but can not see why it is the same limit on HC2.

[alex88 71]

robmac, i am sure there's a very good technical reason. Maybe it's best to ask Danfoss themselves. If they set the limit at 20 then there is a reason for them to do it.

EDIT: What i'm saying is that I am not sure Andrew knows... correct me if i'm wrong Andrew.

[stevenvd 51]

why is it not stated anywhere that the limitation of Danfoss valves is 20?

[pathduck 18]

I am On version 3.590 and although a beta I would call it stable. I have tried both raspberry pi and Vera lite. Fibaro is the most complete setup in my mind. There hardware has the best design in the industry. I have mostly fibaro devices and that is probably why it is so stable. I have over 80 z-wave devices and around 30 virtual ones.

Have I experienced buggs and lags? Yes off course but so do all z wave smart home owners. But as I said for me this is the best complete situation I have used so far. I agree with the new functions freeze idea and that fixing bugs is more important but there is a balance there as well. If we look at Smartthings they have implemented a lot of new functions and are very quick at implementing new things but their z-vawe engine sucks.

I know version 4 will give addons and that is the most needed function in my mind. Fibaro need to get their API well documented and their system open and allowing addons/plugins. That way they can let developers and the community develop cool addons and functions and the Fibaro team can concentrate on development of the basis on the platform making that stable. But to make this happen developers must be able to access the whole platform. I am eagerly waiting for version 4 and I hope the final version is going to be plug and play because beta 4.017 was not.

For those of you complaining about the system, have you tried using the same amounts of units in a another controller, with the same nice design and functionality of the HC2?

[akatar 208]

For those of you complaining about the system, have you tried using the same amounts of units in a another controller, with the same nice design and functionality of the HC2?

yes, i have

that worked well,

for the moment hc2 is (almost) zwave only, while others can handle all kinds of stuff

and what does it matter how the controller looks like?

it should work and in fact it has to work that good that you forget it's there

[pathduck 18]

For those of you complaining about the system, have you tried using the same amounts of units in a another controller, with the same nice design and functionality of the HC2?

yes, i have

that worked well,

for the moment hc2 is (almost) zwave only, while others can handle all kinds of stuff

and what does it matter how the controller looks like?

it should work and in fact it has to work that good that you forget it's there

I did not mean the physical design of the controller but the GUI. Both my Vera and raspberry pi were painfully slow and with lots of buggs. A lot more than what I have experienced with fibaro. But with less functions though like you pointed out. I use the vera as my z wave machine and my vera as a secondary controller where all devices from HC2 are imported and I am able to run addons that are not supported through HC2 on the vera.