Erratic behavior

[kenflorian]

I have a single HSM100, installed about 6 weeks ago.
It is one of 6 z-wave devices in my system.
It is set with factory defaults for timeout, sensitivity etc.
All 6 z-wave devices are within about 20' of each other
The HSM is in a location which should not be generating any significant interference....an old Hawkeye worked at that location.

Behavior -
It triggers events on motion...most of the time.
In Z-seer, in a 20 packet test, the HSM drops between 5 and 20 of the packets....no pattern to when/how it drops.

The log says "unable to send NULL command to z-wave device 10".

I'm thinking a flawed physical device?

Input?

Thanks,

Ken

[Rupp]

What Z-Wave interface are you using? If you move it close to the controller does it work more consistently?

[kenflorian]

z-troller.

No. Placing the hsm100 within 1' does not change the behavior. I ran another test in z-seer and got 8 good, 12 bad on the hsm100...all other devices were 100%.

[Rupp]

Have you used the Z-Seer software to look at your networked to make sure the network is not having a problem communicating to some of you other devices and causing this one to time out?

[kenflorian]

screen shot of z-wave network

Here is a screenshot of the network's most recent test run.

[nanckekr]

kenflorian: I have seen the same behavior that seems close to what you are seeing. You will get the log entry "unable to send NULL command to z-wave device 10" if you configure HS to pull the HSM100. Reason being, that HSM100 is a battery operated device, and to converse power, it can initiate communication to other systems, but other systems (like HS) can not force a communication to HSM100, as that would require HSM100 to listen to communication all the time and that cost battery.

You can also get the log error above from the recurring sensor readings that HSM100 is sending out. My tests showed 4 out of 40 readings went wrong.
You will also get the log error if you are running z-seer and try to optimize or test the network, as HSM100 is sleeping. To avoid that, click the button on HSM100 before running z-seer and you have 10 minutes of testing time, during which HSM100 will listen to your commands.

I wrote a detailed review of HS, Vizia & HSM100 on this thread. There are many little details about the HSM100 that could frustrate a new user.
http://forums.homeseer.com/showthrea...777#post789777

It looks like your problems are more severe though, as you report 12 out of 20 packages failed. I am not sure precisely what that could be...

[rkaplan]

Unfortunately i agree with your review... this product is not ready for prime time.

I am a new Homeseer user but very proficient with computer technology. So far the system works great with a bunch of other equipment but the battery-operation limitations of the HSM100 make it almost unusable. Do you think these are issues which HomeSeer can fix in software or should I just return the hardware? What alternative motion and temperature sensors work more seamlessly?

[nanckekr]

rkaplan,

If you are skilled in computers and equipped with a good potion of tenacity I would use the learnings I posted in the review above to get a view of the limitations in the hardware, then redesign your system knowing this and test it out.

For me, the hardest part was to get over the hurdle of getting to know the limitations - following that of course getting over the feeling of dissatisfaction that you could not do what you wanted. When experimenting with new technology, you have to be willing to go through a certain amount of "wasted" hours before you finally get something nice going.

I am redesigning my setup now with respect to HSM100 and will test it out over the next couple of days. My thinking is the following
- Timeout: 1 seconds (0 seconds make HSM100 stop communicating to HS)
- Wakeup interval: 6 seconds (get as frequent reports as possible)
- HSM100 is not configured to control any lights directly, but only to associate with HS
- HS is configured to listen to events from HSM100. I only listen to the "motion" event, not the "no motion" event. Then when motion appears I turn the light on, and keep it on for 15 minutes. If another "motion" event appears before the 15 minutes expire, I will restart the HS timer.

I am not 100% sure how to do this in HS yet, but will play with it.

Remember that HSM100 ignores your movement 10 minutes after you changed settings through HS.

If everything fails I will return the HSM100, but I am far from that point still, as there is a lot of testing to do.

BR,
Anders

[rkaplan]

OK I stand corrected.

After looking at the HS2 logs in more detail, I realize that the device listing page only shows the last time a value changed and not the last time data was received. Thus I now realize it is collecting data much more often than I had thought. It also does seem very reliable in waking up whenever I enter the room to turn the light on. So it does seem to do its basic job - some of the setup quirks took some getting used to at first.

[Rupp]

The reviews for these devices have been overwhelmingly positive so far. They are a bit tricky to get setup but once setup they seem to work very good.