MS6 including as ID 0 or not including correctly

[AutoFrank 372]

hi,

I bought a bunch of MS6 devices

Some included okay and some are proving a challenge 

I've noticed that some half include (ie I might get temp and motion but no humidity)

I also noticed that some are including as ID 0

 

I tried different power supplies, different cables, close to the HC2 and not so close

I tried the factory reset (hold include button down for 20 seconds) 

 

I just trying to figure out if I missed something or just got a bad batch

 

Thanks

 

Edited January 11, 2017 by AutoFrank

[chaicka 195]

22 minutes ago, AutoFrank said:

I've going over and back with an eng in aeotec - impressive since I just logged the ticket 2 days ago

 

I decided to put the summary in one place.....

The sensors should automatically start to appear as the sensor reports start coming in. If they aren't showing initially when you first pair them, try setting these parameter settings and have them report faster to see if the sensors appear after a report is sent:

Parameter 111 [4 byte dec] = 30 //report all sensors every 30 seconds (apply changes to the master) - This will work if they are USB powered.

V1.07 and V1.7 are the same. Just displayed differently and is the latest rev

Beta 1.08 available (thanks  @petergebruers ) and resolves the following two bugs (amongst others I assume)

Parameter 111-113 report interval bug

- If sensors are reporting around 4 minutes, it is possible that the sensor may send the signals 4 times in a row causing the battery to be used rapidly (users who have done this have seen battery drain within 2 weeks). (bug)

USB Power - Can no longer report sensors at a rate of 1 second, the minimum report is now 240 seconds. (with 1.08 it allows reporting every second)

How to upgrade to beta 1.08 - zwave usb adapter such as a z stick, uzb1 stick, or smartstick. Typically any zwave stick that conforms to zwave standards and a windows pc platform.

Difference between A & C in the serial number -  For model numbers, A stands for US frequency while C stands for EU frequency. B would be for AU.

 

 @petergebruers you were correct

 

but this is worrying ...

I gone back to see how they are working with my system at all.... - I think all of them are A serial numbers 

I haven't integrated them extensively yet but the motion sensor in the kitchen is turning on the light when we walk in 

 

I would have thought that they wouldn't be able to function with my HC2 at all

 

the freq on the label would seem to suggest that it a euro zwave device but the serial number doesn't add up then...

 

Please login or register to see this attachment.

 

Don't worry too much about the "A" or "C"...

 

One more thing to share. I always practice patience with these MS6. They are really quite a 'b i t c h' based on the amount of time I spent testing them and running them in production environment and then finally get down to the fine tuning stage to squeeze most runtime out of the batts.

 

I always have the HC2 nice and sweet ready for inclusion with minimal interference, include it and waits patiently for inclusion to complete which can take a long long long time. At most wake it up if the wait is >2 mins. If I ain't satisfied with the inclusion result (some signs of abnormality), I will wait for several minutes (as peter mentioned, the device may be busy in the background scanning the z-wave network for neighbours), perform proper exclude then wait again... Long story short, the three MS6 I had my hands on, are the devices which I spent the most time on as compared to >40 other devices I have in my setup.

[chaicka 195]

1 minute ago, AutoFrank said:

 

Trying to learn a bit more...

 

Am I correct in assuming that the the sensor will trigger immediately (once motion, lux, temp thresholds are exceeded)  and the wake up interval has nothing to do with that ?,

and that the wake up interval designates how frequently the sensor to "check-in" with the the zwave network/controller, confirm it's still part of the network if it hasn't been triggered in the last wake-up interval...

 

 

 

 

 

USB-powered or Batt-operated? Each has different behaviours. What you mentioned is not correct for batt-operated.

[AutoFrank 372]

Just now, chaicka said:

 

USB-powered or Batt-operated? Each has different behaviours. What you mentioned is not correct for batt-operated.

 

Thanks @chaicka

Mine are all USB powered but would you mind explain how it works in both scenarios ?

[chaicka 195]

The parameter template in HC is quite self-explanation if u really read every parameters. The gist is, batt-operated, it is dependent on wake-up interval so on that basis, tune the parameters accordingly. Tune wrongly = device misbehaves and not deliver what one expects when in fact, there is nothing wrong with the device. I spent >100 hours on these MS6 to really comprehend how it behaves, what to tune, etc. That's how I can understand why peter is saying it takes lots of time to test as he is testing it on an even deeper level than I did.

 

I am not able to share usb-pw-operated mode since I barely spent 20-30 hours on that mode during early UAT testing of the "A" unit with fw 1.01 I had gotten hands on. I don't intend to spend >100 hours sitting around observing its behaviour, tuning it, observe again, tuning it again, observe again, ...... since I still have no intention to run them with usb-powered as the current infra I did not build in 5v and/or 12v circuits around the entire place. Perhaps in a year or two I may do so (dependent on a few other factors I am waiting for market to be a little more mature with first).

Edited January 13, 2017 by chaicka

[AutoFrank 372]

3 minutes ago, chaicka said:

 

Don't worry too much about the "A" or "C"...

 

One more thing to share. I always practice patience with these MS6. They are really quite a 'b i t c h' based on the amount of time I spent testing them and running them in production environment and then finally get down to the fine tuning stage to squeeze most runtime out of the batts.

 

I always have the HC2 nice and sweet ready for inclusion with minimal interference, include it and waits patiently for inclusion to complete which can take a long long long time. At most wake it up if the wait is >2 mins. If I ain't satisfied with the inclusion result (some signs of abnormality), I will wait for several minutes (as peter mentioned, the device may be busy in the background scanning the z-wave network for neighbours), perform proper exclude then wait again... Long story short, the three MS6 I had my hands on, are the devices which I spent the most time on as compared to >40 other devices I have in my setup.

 

Yes it would appear that patience is the key with these devices...

 

The other thing I've learned is that I have been looking for the perfect inclusion ( all slaves sensors available and visible in the Web UI) and this is no necessarily what's required or perhaps too high an expectation . As long as I have the master and perhaps one or two slave devices it sounds like I have a good starting point and some tweaking of the parameter, some wait-time sipping the beer (or two) could work wonders..

 

I have three left to tackle this evening

I think I'll start by excluding them and then doing a factor reset on them and them taking the 'beer approach' I'll try and include them one by one

 

One thing I did notice with a USB powered MS6 - If it has started to included and your waiting 2-3 minutes and get impatient and click the action button again, this can result in an error (something like ID 86 not available on the network) and the you have to start again

 

 

 

 

[AutoFrank 372]

2 minutes ago, chaicka said:

The parameter template in HC is quite self-explanation if u really read every parameters. The gist is, batt-operated, it is dependent on wake-up interval so on that basis, tune the parameters accordingly. Tune wrongly = device misbehaves and not deliver what one expects when in fact, there is nothing wrong with the device. I spent >100 hours on these MS6 to really comprehend how it behaves, what to tune, etc. That's how I can understand why peter is saying it takes lots of time to test as he is testing it on an even deeper level than I did.

 

@chaicka

Thanks - I'll dig out the spec and take a read over the weekend.

 

I have read a few zwave online sources but they talk more about the network and not so much on the devices..

Are there other good references that you would recommend if I wanted to deepen my understanding on this subject ?

[chaicka 195]

3 minutes ago, AutoFrank said:

 

Yes it would appear that patience is the key with these devices...

 

The other thing I've learned is that I have been looking for the perfect inclusion ( all slaves sensors available and visible in the Web UI) and this is no necessarily what's required or perhaps too high an expectation . As long as I have the master and perhaps one or two slave devices it sounds like I have a good starting point and some tweaking of the parameter, some wait-time sipping the beer (or two) could work wonders..

 

I have three left to tackle this evening

I think I'll start by excluding them and then doing a factor reset on them and them taking the 'beer approach' I'll try and include them one by one

 

One thing I did notice with a USB powered MS6 - If it has started to included and your waiting 2-3 minutes and get impatient and click the action button again, this can result in an error (something like ID 86 not available on the network) and the you have to start again

 

 

 

 

 

I look for perfect inclusion too. Anything short means that inclusion is not complete and something isn't right. The 2+ years of s h i t early 4.x firmwares trained me well - extreme patience, try again and again and again and forced me to develop a few sets of usual procedures (what I call SOPs now) to apply whenever abnormalities occur.

[AutoFrank 372]

2 minutes ago, chaicka said:

 

I look for perfect inclusion too. Anything short means that inclusion is not complete and something isn't right. The 2+ years of s h i t early 4.x firmwares trained me well - extreme patience, try again and again and again and forced me to develop a few sets of usual procedures (what I call SOPs now) to apply whenever abnormalities occur.

 

thanks

So maybe I shouldn't lower my expectations after all

I'll try both and see if I can achieve the same result

 

will keep people posted

 

[chaicka 195]

5 minutes ago, AutoFrank said:

 

@chaicka

Thanks - I'll dig out the spec and take a read over the weekend.

 

I have read a few zwave online sources but they talk more about the network and not so much on the devices..

Are there other good references that you would recommend if I wanted to deepen my understanding on this subject ?

 

The Aeotec support portal I shared, there are few articles by Aeotec themselves that explains device behaviours when on batt or when on usb-power. Have a read as they are useless in better understanding these MS6.

[AutoFrank 372]

Just now, chaicka said:

 

The Aeotec support portal I shared, there are few articles by Aeotec themselves that explains device behaviours when on batt or when on usb-power. Have a read as they are useless in better understanding these MS6.

thanks

 

[chaicka 195]

2 minutes ago, AutoFrank said:

 

thanks

So maybe I shouldn't lower my expectations after all

I'll try both and see if I can achieve the same result

 

will keep people posted

 

 

What you mentioned in a post earlier up about getting patience and get errors. Just want you to know I encountered that many many many times. That's how I developed patience.

 

What peter calls 'take a beer' approach is what I documented as 'wait or allow system time to run background invisible activities/tasks', aka take a smoke break to me. Hahaha... You need plenty of these when dealing with Home Center and all the various 3rd party z-wave devices. They can be as frustrating as having the urge to just smash it against the wall or floor at times. At least it is for those of us who take the early adopter route and end up becoming bug reporters to Fibaro and other manufacturers of the device.

[AutoFrank 372]

26 minutes ago, chaicka said:

The parameter template in HC is quite self-explanation if u really read every parameters. The gist is, batt-operated, it is dependent on wake-up interval so on that basis, tune the parameters accordingly. Tune wrongly = device misbehaves and not deliver what one expects when in fact, there is nothing wrong with the device. I spent >100 hours on these MS6 to really comprehend how it behaves, what to tune, etc. That's how I can understand why peter is saying it takes lots of time to test as he is testing it on an even deeper level than I did.

 

I am not able to share usb-pw-operated mode since I barely spent 20-30 hours on that mode during early UAT testing of the "A" unit with fw 1.01 I had gotten hands on. I don't intend to spend >100 hours sitting around observing its behaviour, tuning it, observe again, tuning it again, observe again, ...... since I still have no intention to run them with usb-powered as the current infra I did not build in 5v and/or 12v circuits around the entire place. Perhaps in a year or two I may do so (dependent on a few other factors I am waiting for market to be a little more mature with first).

 

Understood

 

Re " did not build in 5v and/or 12v circuits around the entire place "

I didn't either as I decided that it would be too complicated and DC doesn't travel well 

 

I have 8 MS6 sensors upstairs (installed in recessors in the ceiling ) and have them powered by 4 of these distributed around my attic space

Please login or register to see this attachment.

It took a bit of fiddling as the house is practically airtight and I didn't want to compromise that integrity. The recessors installed neatly under the membrane above the ceiling plasterboard and a little airtight tape to cover hole for the usb cable back into the attic space brought everything back to normal. 

 

Downstairs is proving a little trickier as I've put a lot of effort to hide all the technology in the house so in some rooms they'll sit on shelves until I can figure out how to get them into the ceiling spaces.

 

..the only problem with this is that when my wife catches me in a room staring at the ceiling she is ready to have me committed

 

Edited January 13, 2017 by AutoFrank

[AutoFrank 372]

8 minutes ago, chaicka said:

 

What you mentioned in a post earlier up about getting patience and get errors. Just want you to know I encountered that many many many times. That's how I developed patience.

 

What peter calls 'take a beer' approach is what I documented as 'wait or allow system time to run background invisible activities/tasks', aka take a smoke break to me. Hahaha... You need plenty of these when dealing with Home Center and all the various 3rd party z-wave devices. They can be as frustrating as having the urge to just smash it against the wall or floor at times. At least it is for those of us who take the early adopter route and end up becoming bug reporters to Fibaro and other manufacturers of the device.

 

My system is almost running what I want apart from integrating these new sensors in the setup and tweaking some of my scenes to make me even more lazy

 

A closer look a the devices and helping improve them through feedback sounds interesting but I have a lot to learn before then

[chaicka 195]

5 minutes ago, AutoFrank said:

 

Understood

 

Re " did not build in 5v and/or 12v circuits around the entire place "

I didn't either as I decided that it would be too complicated and DC doesn't travel well 

 

I have 8 MS6 sensors upstairs (installed in recessors in the ceiling ) and have them powered by 4 of these distributed around my attic space

Please login or register to see this attachment.

It took a bit of fiddling as the house is practically airtight and I didn't want to compromise that integrity. The recessors installed neatly under the membrane above the ceiling plasterboard and a little airtight tape to cover hole for the usb cable back into the attic space brought everything back to normal. 

 

 

Hmmm...interesting. I didn't know this exists nowadays. Good thing it is also UK sockets which is what I have here. I learn something from you too. Isn't it a beauty of having forums.

[AutoFrank 372]

3 minutes ago, chaicka said:

 

Hmmm...interesting. I didn't know this exists nowadays. Good thing it is also UK sockets which is what I have here. I learn something from you too. Isn't it a beauty of having forums.

 

That's what it's all about...

.. and I don't intend to take up golf  ⛳️

Edited January 13, 2017 by AutoFrank

[chaicka 195]

Not sure if I mentioned before... this MS6 is a pain to include in secure mode. Inclusion in non-secure mode works better.

[AutoFrank 372]

 

 

Hi

I gave this another few hours last evening and none of my remaining 3 would include successfully.

I gave it each one another go this morning after a stern talking to and still now luck so back in the box they have gone.

 

I'll be returning them monday for replacement

 

thanks everybody for your help

Hopefully the replacements will include without issue

 

-f

 

[petergebruers 1,261]

On 13-1-2017 at 11:57 AM, petergebruers said:

Regarding "MultiSensor Gen 5": toyed with p 3, 4 and 5 and can't get "motion" to work. Tried unofficial firmware 1.3 released last November, can't make that work either. This sensor is now punished and has to sit on my bench for at least one month.  ...

 

Problem solved. Cleaned PCB and fiddled with the sensitivity knob. I think the sensitivity knob was the culprit. How I diagnosed: Tried sensor, FW 1.2 and 1.3 on a RazBerry and got sensor data, but no motion. Opened sensor, noticed the "out" signal of the PIR stayed low. Fiddled with potentiometer, saw red LED flash and output PIR chip OK. Then included on HC2, set p3=4 to get faster updates and p4 to 2 (it defaults to 1) to get "Sensor Binary Report" instead of "Basic Set". Voilà.

[petergebruers 1,261]

Regarding accuracy of "Light" measurement of various sensors... I'd agree that MS6 isn't very accurate but I'd like to warn you that you cannot compare the "luminance" of different sensors. You can compare the output of tools used by photographers: handheld exposure meters, the ones with a half dome, but not the not so standardised Z-Wave sensors. The Z-Wave sensors point their photodiodes at the white semi transparent material. The angle of different models is different, the angular response is different and spectral response is probably different too. The shape of the dome is different (MS6 = almost flat, FGMS = small part of a hemisphere, old MultiSensor =almost half part of a hemisphere) so it captures light sources in a different way. If you wan't some reasonable comparison to begin with, you'd have to use them in a white room with very equal lighting.

 

This Wikipedia article might inspire your next DIY project:

 

Please login or register to see this link.

 

Also, because the nature of our perception, it makes sense to look at the logarithmic of the luminance level. Somehow I got the feeling that talking about "spectral response", an "Integrating Sphere" and "logarithmic" on a Sunday afternoon is more than enough. So I'll shut up now  

 

 

 

 

[robmac 173]

It is interesting that people are worried about an accurate reading for lux in an automation system.

 

What are you all doing that needs more than when a sensor in a room reads a value it is dim or bright and something needs to happen?