Jump to content

Welcome to Smart Home Forum by FIBARO

Dear Guest,

 

as you can notice parts of Smart Home Forum by FIBARO is not available for you. You have to register in order to view all content and post in our community. Don't worry! Registration is a simple free process that requires minimal information for you to sign up. Become a part of of Smart Home Forum by FIBARO by creating an account.

 

As a member you can:

  •     Start new topics and reply to others
  •     Follow topics and users to get email updates
  •     Get your own profile page and make new friends
  •     Send personal messages
  •     ... and learn a lot about our system!

 

Regards,

Smart Home Forum by FIBARO Team


Battery levels always 100%


cag014
 Share

Recommended Posts

Posted (edited)

This update doesn't solve the issue of NO BATTERY LEVEL reports.

All battery devices show 100% level, even when they are almost dead.

Fibaro team please roll back this feature as was before 4.620 update

This is not a nice to have option, the devices are dead and the user have no notification on that and everything works badly with dead devices.

My alarm didn't work for a few days because outside motion sensors and the siren have dead w/o any notification from the gateway.... you have added HTTPS for security while basic security (alarm) doesn't supported properly.

 

 

 

Edited by cag014
Link to comment
Share on other sites

Posted (edited)

I'm still waiting for support about this issue, after some comunications with Maciej Kusik (tech support) the curtain has fallen.

Is not professional behavior.

 

 

 
Edited by paolo
Link to comment
Share on other sites

in the last message from the technician I was asked to measure the voltage of the batteries ..... but shouldn't the device measure it?

Link to comment
Share on other sites

  • Topic Author
  • Posted (edited)
    1 hour ago, paolo said:

    in the last message from the technician I was asked to measure the voltage of the batteries ..... but shouldn't the device measure it?

    That's absurd, I have tens of battery operated devices  and no reports from anyone.... should I measure all of them?

    I have monitored the system and no battery reports at all for months...

     

    Have transferred one  sensor to HC3 and have received battery reports.!!?

     

    Edited by cag014
    Link to comment
    Share on other sites

    14 minutes ago, cag014 said:

    Have transferred one  sensor to HC3 and have received battery reports.!!?

    They want me to buy HC3?

    If I was sure everything was working as it should I might even do it.🤔

    Link to comment
    Share on other sites

    Posted (edited)
    2 hours ago, paolo said:

    I was asked to measure the voltage of the batteries ..... but shouldn't the device measure it?

     

    It is complicated. Voltage is not then only factor. There's also the internal resistance of the battery. So it depends on how much load a device poses and that depends on when the battery is measured. For example, load is much higher when device is transmitting than when it is receiving. And that load is higher than when only the CPU is running.

     

    Measuring battery voltage with a multimeter is essentially measuring without load and doesn't take into account its resistance. If it measures low voltage, then it is empty. But when voltage measures high, you do not have a 100% indication if it can handle the load.

     

    IMHO especially the cheap ER14250 no brand lithium thionyl cells can have good voltage and capacity, but too high resistance.

     

    That type of battery is used in the Fibaro D/W sensors which use series 300 and 500 Z-Wave chip, which are comparable. The newer series 700 and 800 chips are less power hungry. I do not have real world numbers on those newer chips. It also depends on the installed memory chip.

     

    Anyone got a multimeter and a 68 ohm resistor? Load the battery with the resistor. If voltage > 3.0 it is OK. If it is between 3.0 and 2.8  it is marginal. Below 2.8 it is NOK. The Z-Wave chip uses about 40 - 45 mA to transmit data and it does not work below 2.8 V (a bit of a simplification)

     

    If you do not have 68 ohm, 100 ohm gives a good indication as well. It gives a better indication of battery quality then no resistor, though the no-resistor measurement is good to know as well.

     

    The internal resistance is a reflection of quality of the chemistry and electrolyte state amongst other things. In other words, it can indicate that a battery has "dried out"

    Edited by petergebruers
    Link to comment
    Share on other sites

     

    I got it

    but the question is : after I measured the voltage of the batteries and found out that some ( definitely) are low, what does it change for me? if the battery charge reading was working before the hc2 update and now it doesn't go, it is a problem of the update.

    Please login or register to see this image.

    /monthly_2022_05/IMG_5823.jpg.d5daeb0512720ebf570e4e478c9e950b.jpg" />

    Link to comment
    Share on other sites

    9 minutes ago, paolo said:

    I got it

    Super!

     

    9 minutes ago, paolo said:

    but the question is : after I measured the voltage of the batteries and found out that some ( definitely) are low, what does it change for me?

    You can answer the question asked by the Fibaro technician with lots of detail, and hopefully leads to further action by them

     

    I would not only report about those empty, because one might argue "They went from 100% to 0% in less than the wake up time of the sensor so they were never able to TX a message with battery low". This isn't typical behaviour of single use cells, but rechargeable's can have that problem

     

    I think the ones around 50% are the interesting ones, HC2 should not display 100% on those.

     

    Sometimes, I just post some explanation, hoping it helps a few users (this is is, after all, a public forum) based on my experience which hopefully makes users and Fibaro employees think about the issue from different angles.

     

    It can be considered a "mild form of thread hijack" :)

     

    14 minutes ago, paolo said:

    if the battery charge reading was working before the hc2 update and now it doesn't go, it is a problem of the update.

    Sure, firmware update is high on my list too, I think they tuned parts of the zwave engine but it might be in other parts of the firmware :D

     

    But only Zniffer could settle this debate. If Zniffer sees a 50 % battery report and HC2 keeps saying 100 then...

    Link to comment
    Share on other sites

    I performed measurement as follow

    door/windows sensor 1 battery 3,06V   HC2 show 100% (kitchen1)

    door/windows sensor 1 battery 3,178V  HC2 show 100% (kitchen2)

    door/windows sensor 1 battery 2,973V  HC2 show 88% (dinner) always indicates 88 even with new battery (after wakeup of device obviously)

    door/windows sensor 1 battery 3,084V  HC2 show 100% (living)

    door/windows sensor 1 battery 3,0197V HC2 show 100% (entrance)

    door/windows sensor 1 battery 3,129V  HC2 show 100% (bathroom living)

    door/windows sensor 1 battery 3,255V  HC2 show 100% (bedroom1)

    door/windows sensor 1 battery 3,165V  HC2 show 100% (bedroom2)

    door/windows sensor 1 battery 3,079V  HC2 show 100% (bedroom3)

    door/windows sensor 1 battery 3,108V  HC2 show 100% (bathroom night)

    Please login or register to see this code.

    • Thanks 1
    Link to comment
    Share on other sites

    Thank you, @paolo , excellent data.

     

    I would say, based on the loaded voltage test you did, all your batteries are OK, so the FGK-101s should function and report battery data.

     

    I am a little surprised how close the measurements are, but that might be a coincidence like, they are all the same brand, about the same age and all door sensors have about the same average power consumption.

     

    Let's see if I can return the favour. I have a Zniffer, and a spare door sensor and a HC2... I plan to find out what battery percentage the FGK-101 sends when I use a bench supply. Give me a day or so.

    Link to comment
    Share on other sites

    2 minutes ago, petergebruers said:

    I am a little surprised how close the measurements are, but that might be a coincidence like, they are all the same brand

    the batteries were all changed after I discovered that fibaro did not notice the "dead" sensors due to low charge level, otherwise it doesn't work as an alarm.

     

    5 minutes ago, petergebruers said:

    I have a Zniffer, and a spare door sensor and a HC2... I plan to find out what battery percentage the FGK-101 sends when I use a bench supply. Give me a day or so.

    Thank you for your helpfulness!

    I had also thought of installing Zniffer, but I can't find the OSX version assuming it exists.

    ..... and it will sound strange but at home I don't have any windows device, only mac.

    Link to comment
    Share on other sites

    46 minutes ago, paolo said:

    the batteries were all changed

    That explains it.

     

    46 minutes ago, paolo said:

    otherwise it doesn't work as an alarm.

    Ah, yes, makes sense.

     

    47 minutes ago, paolo said:

    but I can't find the OSX version assuming it exists.

    Although it is a "Microsoft dotnet app" which could made to be run on different O/S they never released anything but a Microsoft Windows version and the d4mned thing isn't fully open source, so... I am 99% on MacOS but keep a laptop just to run the non-multi-platform apps.

    Link to comment
    Share on other sites

    ooooooh, this is going to be a long day, to get all the lovely details right :D

     

    I do want to share my observations so far, because this topic is about "HC2 always reports 100%" or not quite the same but close: "HC2 does not report empty batteries".

     

    I cannot confirm that and in fact my test setup shows that my HC2 does report what the FGK-101 says, and the FGK-101 does report battery % (actually, a recalculated voltage level). But there are some gotcha's and I have to retest and make sure about those details, they might explain why some things we do or try do not work as expected.

     

    So HC3 is factory reset and has been updated to 4.630

    Door/Window sensor first generation has firmware 2.1 which means it is not Z-Wave Plus and based on the old series 300 chip.

     

    The wake up interval has been set to a very low 60 seconds to speed up the test and to confirm a theory: the device measures battery level at (auto wake up interval).

     

    I use a power supply to force a voltage, from high to low, while observing the reports in Zniffer and the JSON attribute on the HC2

     

    If I start at 3.6 V, power on and wait one minute, Zniffer shows the HC2 doing a "battery get" and the device responds with "battery report 0x64". The spec says this value goes from 0 to 100% and hex 64 is 100 in decimal, no surprises there. Spec also mentions, a device can report 0xFF = 255 to mean "battery empty". AFAIK gateways treat 255 like 0 meaning "empty".

     

    Then I do:

     

    print(api.get("/devices/24").properties.batteryLevel)

     

    And it says... 100

     

    No surprises there.

     

    If I drop to 3.3 V, Zniffer says 0x63 and HC2 has 99, makes total sense. Let's do a few more

     

    Please login or register to see this code.

     

    I conclude that the reports make sense, and HC2 faithfully reproduces them...

     

    Now don't quote me on this yet but based on reports from other users and my own observations (not measurements) I have always thought the batteries go from 30% to 0 really fast (days, maybe 2 weeks) and these measurements seem to confirm that.

     

    I am unsure, if this device is able to faithfully report levels below 30 % because I can see the signal getting weaker and some resends happen. That is one of those details to keep in mind.

     

    Tell me what you think...

     

    Link to comment
    Share on other sites

    Ok makes sence.

    When I used HC2 I replaced batteries if reported empthy but used the till zero in my easy to reach door sensor(s).

    In both doors the lifetime of "empthy" batteries is various. Same time usage.

    So for me 30% is not always 30% in practice.

    Same as you no measerment done will try with my HC3 later this year.

    //Sjakie

    Link to comment
    Share on other sites

    22 hours ago, petergebruers said:

    conclude that the reports make sense, and HC2 faithfully reproduces them...

    how is it possible then that my sensors whose batteries are about 3 v always score 100%?

    I'll try setting the wake-up time to one minute and see what happens (other than the battery will drain quickly)

    Obviously I realize that if the wakup time is very long and if the sensor transmits the battery status at wakeup, it could happen that in the meantime a myriad of factors including temperature , could completely discharge the battery. The fact remains that my HC2 cannot always report 100% batteries ( even in the sensor which reports 88 if I put a new battery it always reports 88).

    Needless to say, fibaro technical support, which asked me to measure the voltages of all the batteries, has not yet replied.

    Couldn't technical support access my HC2 and do all the appropriate checks?

    Link to comment
    Share on other sites

    Posted (edited)
    27 minutes ago, paolo said:

    how is it possible then that my sensors whose batteries are about 3 v always score 100%?

    Excellent question.

     

    With the 68 ohm load test they show 3.1 to 3.2 V and the 68 ohm resistor may represent a somewhat higher load than the average FGK and so the result may be pessimistic

     

    Without Zniffer we cannot establish the truth, we need to see what the FGK is reporting (EDIT: as an end-user, you don not have access to Z-Wave log files)

     

    To me, it is very possible that your batteries (see your previous post) measure 3.3 V or slightly above "in operation" and my tests (see my bench supply tests) turn out to report 99 to 100%, so, ...........

     

    Do you have a battery that clearly fails the "68 ohm test" or at least values closer to 2.8 V? I think we might learn something from that.

     

    Be careful not to accidentally reset your FGKs, they are know to easily reset if you follow another order than the one in the manual. I man, I totally understand if you do not want to do a test due to this possibility.

     

    BTW are all your FGK-101 sensors firmware 2.X or do you have some 3.X? The 3.X is a different device, it is the Z-Wave Plus variant. Looks identical to the older model.

     

    27 minutes ago, paolo said:

    I'll try setting the wake-up time to one minute and see what happens (other than the battery will drain quickly)

    Good idea. Can you measure the voltage of one of those sensors, without 68 ohm? No need to take it out from its FGK, measuring in circuit is OK (idle current is microamperes)

     

    27 minutes ago, paolo said:

    Obviously I realize that if the wakup time is very long and if the sensor transmits the battery status at wakeup, it could happen that in the meantime a myriad of factors including temperature , could completely discharge the battery.

    Plausible! The default is 4000 seconds iirc and I often set them a little higher, to 7200s =2h. There is not a lot of battery life to gain above that, though imho you can still safely double that. I never truly investigated that (for various reasons). On devices with smaller sells like CR2450 it might make sense to set it very high or disable periodic reporting.

     

    27 minutes ago, paolo said:

    The fact remains that my HC2 cannot always report 100% batteries ( even in the sensor which reports 88 if I put a new battery it always reports 88).

    That is interesting but I promise you to post a few details about this FGK-101. Does not explain everything but you might have a few "aha" moments.

     

    27 minutes ago, paolo said:

    Needless to say, fibaro technical support, which asked me to measure the voltages of all the batteries, has not yet replied.

    Thanks. You have done your part of the deal and we can benefit from your feedback.

     

    27 minutes ago, paolo said:

    Couldn't technical support access my HC2 and do all the appropriate checks?

    Yes, they can see what the device reports, when it wakes... So if you set up a few device to wake up 1 to 5 minutes then that should not be too much of a trouble to them imho. It is a simple transmission, with a byte indicating 0-100% and 255 for battery empty.

     

    Obviously, if that number would be 100 or 88 and HC2 GUI shows the same then the device is to blame.

     

    If there is a discrepancy, some filtering, cacheing or database issue might be the problem.

     

    could try something on my "production" network but that is a HC3. My HC2 is a test system, it has kind of ideal situation. I mean, it was factory reset, never had a database upgrade, very few devices... It might not have the issues you have.

    Edited by petergebruers
    Link to comment
    Share on other sites

    magically even the sensor that has always reported 88% now also reports 100% (with the same battery as always, the lowest of all). but is there any way to reset the table of readings ? I don't know how HC2 works , I don't know if it has a database or what else

     

    Link to comment
    Share on other sites

    1 minute ago, paolo said:

    magically even the sensor that has always reported 88% now also reports 100% (with the same battery as always, the lowest of all).

    We were probably posting at the same time so you haven't read my answer, so I'll give the gist. (1) the test you and I do with 68 ohm is probably "pessimistic" - but it is better than no resistor to detect really bad batteries so I still recommend. All your measurements suggest the batteries should indeed report 90 to 100%. (2) I fully agree with what you said yourself, there is some variability due to "a myriad of factors including temperature" and chemistry play a part in that.

     

    5 minutes ago, paolo said:

    I don't know how HC2 works , I don't know if it has a database or what else

    It use a SQLite database, but only tech support can access it. You are looking at that database, because sleeping device cannot be contact so you cannot ask its battery level, you have to wait until a wake up interval passes.

    Link to comment
    Share on other sites

    5 minutes ago, petergebruers said:

    BTW are all your FGK-101 sensors firmware 2.X or do you have some 3.X? The 3.X is a different device, it is the Z-Wave Plus variant. Looks identical to the older model.

    three sensors of the 11 are new model

    this evening when I leave work and come home I will try to set the short wakeup (also because if I do it now remotely but do not wake up the device it may not acquire the changes right away, due precisely to the current wakeup time) and let you know if the battery readings change.
    for now thanks for the clarification.

    Link to comment
    Share on other sites

    3 minutes ago, paolo said:

    three sensors of the 11 are new model

    Thanks. I don't own the FGK-101 fw 3 model, but it is probably closer in operation to its successor, the FGK-102, the complete redesign with built-in temperature sensor. I do have one of those on my bench already ;)

     

    4 minutes ago, paolo said:

    also because if I do it now remotely but do not wake up the device it may not acquire the changes right away, due precisely to the current wakeup time

    Spot on! Let's say your current wake up interval is 6 hours. Let us assume, you happen to see in events, this sensor woke up at 10:00 today. Then the next wake up is at around 16:00 and then it will get the queued command "set wake up to 1 minute". You can set it, and just wait and see what happens, it is not a complete waste of time.

    Link to comment
    Share on other sites

    Join the conversation

    You can post now and register later. If you have an account, sign in now to post with your account.

    Guest
    Reply to this topic...

    ×   Pasted as rich text.   Paste as plain text instead

      Only 75 emoji are allowed.

    ×   Your link has been automatically embedded.   Display as a link instead

    ×   Your previous content has been restored.   Clear editor

    ×   You cannot paste images directly. Upload or insert images from URL.

     Share

    ×
    ×
    • Create New...