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

Recommended Posts

wake up time set to 60 seconds, battery since yesterday is always at 100%, I am gearing up to sniff zwave, ordered UZB3 stick on amazon, downloaded Z-Wave Programmer and Z-Wave Zniffer. As soon as it arrives I will try it.

 

Link to comment
Share on other sites

1 hour ago, paolo said:

wake up time set to 60 seconds, battery since yesterday is always at 100%

Interesting.

 

1 hour ago, paolo said:

I am gearing up to sniff zwave, ordered UZB3 stick on amazon

Excellent! It usually takes me a lot more effort to convince people to go down this path... maybe you already have seen some other posts and were already considering this to diagnose other issues? Just curious...

 

To be on the safe side, the UZB3 comes in three different "regions", please make sure it has EU in its part number (assuming your location "Italy" is correct) because the hardware differs per region

 

1 hour ago, paolo said:

downloaded Z-Wave Programmer and Z-Wave Zniffer

Okay! Again, just to be on the safe side, you'll have to flash the UZB3 *with the exact same firmware name* as described by tinman (see link below), to turn it into a Zniffer. If you select the wrong one, the UZB3 may no longer be programmable "via USB" after that accident... This will require soldering skills to un-brick the device.

 

If you want to run PC Controller, on that same stick, you will have to change the firmware again, so if you plan to use both it sometimes makes sense to order 2xUZB3 or one UZB3 and a newer one (at the moment that is USB7) from some companies to get "free shipping".

 

If you know how Wireshark works then you'll be able to understand the basics of Zniffer. I wouldn't bother about it yet but to get into greater detail about concepts like "encapsulation" and "command class versions" you can post here or contact me when you have some real world data.

 

Edited by petergebruers
Link to comment
Share on other sites

3 minutes ago, petergebruers said:

Excellent! It usually takes me a lot more effort to convince people to go down this path... maybe you already had seen some other posts and were already considering this to diagnose other issues? Just curious...

I'm a geek lol :-D

5 minutes ago, petergebruers said:

If you want to run PC Controller, you need to change firmware again, so if you plan to use both it sometimes makes sense to order 2xUZB3 or one UZB3 and a newer one from some companies to get "free shipping"

 i'm interested in zniffer only thanks

6 minuti fa, petergebruers ha detto:

Se sai come funziona Wireshark, sarai in grado di comprendere le basi di Zniffer. Non me ne preoccuperei ancora, ma per approfondire concetti come "incapsulamento" e "versioni della classe di comando" puoi postare qui o contattarmi quando hai dei dati del mondo reale

thank you for your availability, if I am in trouble I will contact you, thank you ( know I will definitely contact you so be prepared hahaha)

Link to comment
Share on other sites

I promised to talk about what I consider to be confusing things about the reporting of the old FGK-101 fw 2.1. Some would call it "bugs", some might call them "features".

 

When you remove the battery, and put it back, the device blinks 3 times. If you are like me, you expect this to signal "I wake up" then report state and battery level. But right after inserting the battery this is what Zniffer captures:

 

You are not blind, and I am not mistaken... Nothing there; no data at all.

 

This means, if you got an empty battery and put a fresh one in... You mistakenly assume the new one is empty as well. Or, if you are trying to see if HC2 always reports 100%, you replace the good cell with a know, more or less depleted cell and ... no change. If your wake up time is the default 4000s the battery will magically jump to the correct value after slightly more than 1 hour. Again, if you are like me, but have no Zniffer, you might think "how is it possible, this battery level changed that much in only 1 hour?". The answer, in hindsight, is clearly "it didn't jump, it is a reporting issue'.

 

Sometimes people post "do not worry about battery level, after a battery replacement, check again tomorrow. It will fix itself.". For this device, there is certainly some truth to that...

 

56 minutes ago, paolo said:

I'm a geek lol :-D

Ah, you plead guilty. I have no further questions ;)

Edited by petergebruers
Link to comment
Share on other sites

Next confusing issue: triple-to wake up might (or should I say "may") send a report with the previous value.

 

Previously I've established that putting a battery in, does not make the device measure and report the battery.

 

But you are clever and after putting un a new battery you 3-click and the blue LED flashes. Unfortunately, the device reports the previous value, which is wrong if you've replaced the battery. I would say this leads to the same confusion as the previous issue, and the same mitigation: wait at least 1 wake up cycle before you draw conclusions from what you see in your web browser.

 

To illustrate my point:

 

  • I set the power supply to 3.2 V which gives 94 % on my device and is expected (see few post back for measurements) after 1 minute. My wake up interval is one minute
  • Power down the device. Note: the power down of a supply is very fast, Zniffer confirms that the device cannot send anything during the transition. So obviously HC still reports 94 %.
  • Set power supply to 2.8 V to simulate battery change (to a worse battery in this case, but imho that is irrelevant). As previously established, the device does not send anything.
  • 3-click the wake up button while pressing the tamper switch at 5-10 s intervals. Observe in Zniffer the device keeps reporting 94. It should be lower.
  • Again, when the 1 minute wake up interval passes, the measurement "jumps" to the correct value, 66

 

Printing the reported value 

Please login or register to see this code.

 

I do not know if the measurement of the battery is tied to the wake up interval but I can conclude: it is not tied to manual wake up during this test. Note: I have not tried different wake up intervals and different manual wake up rhythms.

Link to comment
Share on other sites

Battery always at 100% ...

i noticed anotherr strange things , "

Please login or register to see this link.

The device performs a soft reconfiguration of the Z-Wave network. The parameters and settings (name, assign it to the room, associations) are preserved. ID also remain unchanged, making scenes with reconfigurable device will not require editing." .....I don't think so, after the soft reconfiguration I lost the id the name the room the device was assigned to , and in the case of switch also the toggle switch-temporary switch selection, but how is that possible?? is it my HC2 that went crazy?

 
 
Link to comment
Share on other sites

Nope, my HC2 acted the same after soft reconfiguration door sensor lost ID

//Sjakie

  • Like 1
Link to comment
Share on other sites

  • Topic Author
  • On 5/19/2022 at 1:17 PM, petergebruers said:

    Next confusing issue: triple-to wake up might (or should I say "may") send a report with the previous value.

     

    Previously I've established that putting a battery in, does not make the device measure and report the battery.

     

    But you are clever and after putting un a new battery you 3-click and the blue LED flashes. Unfortunately, the device reports the previous value, which is wrong if you've replaced the battery. I would say this leads to the same confusion as the previous issue, and the same mitigation: wait at least 1 wake up cycle before you draw conclusions from what you see in your web browser.

     

    To illustrate my point:

     

    • I set the power supply to 3.2 V which gives 94 % on my device and is expected (see few post back for measurements) after 1 minute. My wake up interval is one minute
    • Power down the device. Note: the power down of a supply is very fast, Zniffer confirms that the device cannot send anything during the transition. So obviously HC still reports 94 %.
    • Set power supply to 2.8 V to simulate battery change (to a worse battery in this case, but imho that is irrelevant). As previously established, the device does not send anything.
    • 3-click the wake up button while pressing the tamper switch at 5-10 s intervals. Observe in Zniffer the device keeps reporting 94. It should be lower.
    • Again, when the 1 minute wake up interval passes, the measurement "jumps" to the correct value, 66

     

    Printing the reported value 

    Please login or register to see this code.

     

    I do not know if the measurement of the battery is tied to the wake up interval but I can conclude: it is not tied to manual wake up during this test. Note: I have not tried different wake up intervals and different manual wake up rhythms.

    That's strange, I have migrated few sensors to HC3 and have received correct battery values... between 30~80%... 

    Now have returned same sensors to HC2 and the battery level again 100% on all of them!!

    As far as I know when you add device it does reports initial battery value, so same sensor that shows 30% on HC3, now shows 100% HC2.

    ?Could be that HC2 is charging the batteries (even not researchable) ??? ?

     

    Link to comment
    Share on other sites

    Hi @cag014! Thank you for your time and effort. It is a very annoying problem, isn't it? I am sure we will get to the bottom of this, together.

     

    5 hours ago, cag014 said:

    That's strange, I have migrated few sensors to HC3 and have received correct battery values... between 30~80%... 

    Excellent diagnostic test.

     

    I have to say, to be on the safe side, you quoted a test, by me, of the old FGK-101, version 2.1 and I am quite sure that is one of the more peculiar devices. I chose that one because it causes the most battery issues here and because I am sure it was very popular at the time and plenty of users still have them. It was the first Door/Window sensor that did not look like a brick.

     

    I am reasonably sure that many, if not most devices, manufactured after the FGK-101 (say after around 2014-2015) do indeed wake up and/or report battery after you insert (fresh) batteries.

     

    So, yeah, that points to HC2 but I am not sure if it applies to all HC2 or just specifically your HC2

     

    You see, apart from the anomalies (or features or bugs) that sensor did report values on my HC2 at regular intervals.

     

    The main difference might be, my HC2 got a factory reset and was completely empty and all settings were default.

     

    5 hours ago, cag014 said:

    Now have returned same sensors to HC2 and the battery level again 100% on all of them!!

    Okay, let's find out if I own one of those sensors! I can setup a test.

     

    I don't have a complete list but I am sure I got a FGK-102 aka "D/W 2" and I also have a FGMS-001, but I am sure if it is a V1 or V2

     

    5 hours ago, cag014 said:

    Could be that HC2 is charging the batteries (even not researchable) ??? ?

     

    ?

     

    How could Fibaro Marketing forget about that feature: wireless recharge of Z-Wave devices

     

    Edited by petergebruers
    • Like 1
    Link to comment
    Share on other sites

    @cag014 I have these spare devices, I can thoroughly test them like I tested the FGK-101.

     

    Which one do you want to test?

    • Fib D/W 2
    • Fib Smoke Sensor 1 (non-Z-Wave-Plus)
    • Fib Flood sensor FGFS-101 (non-Z-Wave-Plus)
    • Fib FGMS-001 (probably non-Z-Wave-Plus)
    • Neo (Coolcam) Motion Sensor

    So the D/W2 is the only one using ER14250, the other use CR123

    Link to comment
    Share on other sites

  • Topic Author
  • 18 hours ago, petergebruers said:

    @cag014 I have these spare devices, I can thoroughly test them like I tested the FGK-101.

     

    Which one do you want to test?

    • Fib D/W 2
    • Fib Smoke Sensor 1 (non-Z-Wave-Plus)
    • Fib FGMS-001 (probably non-Z-Wave-Plus)
    • Neo (Coolcam) Motion Sensor

    So the D/W2 is the only one using ER14250, the other use CR123

    All above, have tested on HC3 and are working fine

    Link to comment
    Share on other sites

    3 hours ago, cag014 said:

    All above

    Oh! I better get started then :D

    Link to comment
    Share on other sites

    three days ago my HC2 (rolled back to previous backup 4.62 due to system crash) magically report one of the battery at 30% (note that the day before it was at 100%), today report sensor dead, after a battery change and wakeup of sensor(triple click tamper switch) it report (correctly) 100%.

    I'm stunned.

    I have an inkling that there is a problem with updating the database data but only those who have access to it can verify this, namely the support service which has been silent for over a week.

    Link to comment
    Share on other sites

    Okay... A Quick report on the Neo Coolcam ZW5 Motion sensor. It is going to be quick because during testing I accidentally force 10 V onto the device and it fried the Z-Wave chip, so now I am in a bad mood.

     

    There is a surprising conclusion and you might not like it:

     

    My HC2 faithfully reproduces what the device reports, just like the FGK-101

     

    This means, in case you think: "the values do not make sense" then it is not a HC2 bug, it is the way this device measures and probably averages results over time.

     

    Here are the levels reported by the device (Confirmed by analysing Zniffer report! and checking HC2!)

     

    3.3  100
    3.2  100
    3.1  100
    3.0  100 or another test or  95 or another test 91 or another power on 41 and then lastly, after another power on: 100 !!!!!

     

    At 3 Volt, I expect around 50% and if you try long enough and hard enough, it will report something between 40 and 50. But many times, when I disconnect and reconnect power, the values seem to be much higher.

     

    I set wake up to 5 minutes, but unlike the FGK-101, waiting a few rounds of wake up does not get the device closer to 50 if it eg started at 91. I got a hunch that (a) the device writes values to EEPROM and (b) It only measures once per 24 hours. I cannot confirm this because I destroyed my test specimen before it clocked 24 hours :(

     

    It died for science but I am still sad because I liked this sensor...

     

    Now, I still have some older devices to test.... Before doing that I will add a voltage clamp or crowbar to my test setup to make sure I cannot go over +/- 4 volt!
     

    Edited by petergebruers
    Link to comment
    Share on other sites

    Next device: Fibaro Flood Sensor, firmware 2.3 (reported as 23.23) - this is the non-Z-Wave Plus version. I think it is one of my oldest devices and I do not have any issues with it. It would accept a 12 V supply but I am not testing that one...

     

    You can probably guess what I am about to say...

     

    My HC2 faithfully reproduces what the Fibaro Flood Sensor reports, just like the FGK-101 and the Neo Motion Sensor

     

    Again, if I see or suspect an "anomaly" on my HC2 then Zniffer proves it is what the device reported

     

    For example, HC2 says:

    Please login or register to see this image.

    /monthly_2022_06/flood_sensor_bat_hc2.png.044868403a32830fc3f4e2decff5560d.png" />

     

    Which matches the report:

    flood_sensor_bat_znif.png.d10a4b60f77106546cb80f5212661fc7.png

     

    But when we look at the power supply, the level should be close to 100%:

    flood_sensor_bat_volt.jpg.d56e993a87b0314b0a5aea8dc5d492ec.jpg

     

    So... How do we get this result? I have to make an educated guess, this device does not measure at each wake up interval, it probably only measures every 24 hours. This makes sense, most batteries would not change that much in 24 hours.

     

    Here, I am using an old, analog power supply which features low noise and I changed to this 30 year old tech to avoid influence from noise that is quite measurable on modern switched mode power supplies. Here is the test setup:

     

    Please login or register to see this attachment.

     

    Yeah those voltmeters are dangling, I need to 3d-print a bezel to mount them properly :D

     

    Anyway, although changing the supply "faster" than a normal battery would deplete is a bit cheating, it is valid to conclude this:

     

    HC2 = Zniffer Battery Report (or previous value if no report was sent)

     

    Triggering a device does not "wake it up" so that is not a good way to check the battery

    Manual or auto wake up does not necessarily trigger battery measurement so still not 100% certain you get a valid measurement

     

    The Flood Sensor is mildly interesting because it uses yet some different rules to measure and report:

    • The device does not send battery level when you insert a battery...
    • I am pretty certain it does measure battery after power on. So if you manually wake it up it will send a reasonable value. It will probably also send this value at the next wake up interval. But...
    • From my observation, even when set to wake up every minute, it still sends battery level from about 16 hours ago, which makes me think it won't take a battery measurement unless 24 hours have passed. I think, for this type of battery, that is really a valid way to do it but it can cause trouble if you're testing

    I am going to leave it at this, testing takes too much of my time and I think so far I have demonstrated that my HC2 does not have a battery reporting issue. Which does not prove that a different HC2 cannot have such issues. There might be something different, like the database, or the mere fact that my HC2 got factory reset ...

     

     

     

    • Like 2
    Link to comment
    Share on other sites

    and just out of curiosity, when you remove battery, push and hold the button(s) and insert battery and release button(s), still no new battery report?

    Link to comment
    Share on other sites

    1 hour ago, tinman said:

    when you remove battery, push and hold the button(s) and insert battery and release button(s), still no new battery report?

    Ah, excellent question and I did try that on some devices and indeed that might/may work on specific devices. I think I know what you are testing: by clicking the buttons you discharge the capacitors, forcing the device to go completely dead and detect a new battery. Unfortunately for a device like the old Fibaro Door/Window sensors that trick did not work, and also the Neo does seem to have its own "logic". So I stopped testing that. To determine Voltage versus Percentage levels of other devices I did, often, disconnect the supply between tests for about 30 - 60 seconds which seemed enough to let the device self discharge.

     

    I am sure on some devices, probably more recent ones (I mean, at least Z-Wave Plus or even more recent ZW700 stuff), it can make a difference. I mean, if you are really fast, you could probably swap a battery without the device noticing that... Which would make the level wrong for at least one "wake up interval (randomly expiring)" or even longer, like 24 hours (that is a bit of a speculative statement).

     

    I will test one if I find the time, but I think I have a limited selection of candidate devices. A Fibaro D/W version 2 and a Neo ZW700 probably.

     

    As a rule of thumb, to keep it simple, I would advise users "do not worry about battery levels after replacement, start worrying after at least 24 hours have passed"

     

    Then again, an equally important rule of thumb, as @jakub.jezek discussed with me in private, ... The second advice is "buy good quality batteries" in the first place, and if you do not have to tools to check the quality that more or less translates to "buy reputable brand from reputable source".

     

    What do you think? Shoot!

     

    Link to comment
    Share on other sites

    just one thing to add, our devices won't ever report battery level higher than measured before.
    So if you will set the voltage to 2.9, get low battery report, up the voltage to 3.3 it will stay on low battery forever.

    It will measure new battery level during power up (placing the battery in the device).

    It is that way because of characteristic of most batteries, if battery is pretty low it we can measure on it both 2.8V and 3.1V (if there was no high load for some time).
    So, with one inserted battery its level can go only down :)

    • Thanks 1
    Link to comment
    Share on other sites

    The FGK-102 aka Fibaro Door/Windows 2 Z-Wave Plus adds an interesting twist to the reporting...

     

    When I set voltage to 3.0 to simulate a cell that is about 1/2 full then turn the power on:

     

    Please login or register to see this code.

     

    The 100 arrives a few seconds after power on and is confirmed by Zniffer. About 12 seconds later a second report gets send containing 60, which is reasonable.

     

    It is not a fluke, it is repeatable

     

    Please login or register to see this code.

     

    Let's try 2.80 Volt

     

    Please login or register to see this code.

    Yep, "battery full" at power on then 12 seconds later "battery empty"

     

    Let's do one more, to get an idea of how the V translates to %... This is 3.05

     

    Please login or register to see this code.

     

    I did measure the time it takes for the Door Window 2 to detect a battery change: it is less than 2 seconds. I would say it is nearly impossible to swap batteries on this devices, without it noticing the change.

     

    The device does send reasonable values after about 20 seconds.

     

    Here is my test setup with a low noise supply and a micro switch:

     

    Please login or register to see this image.

    /monthly_2022_06/IMG_20220603_084421b.jpeg.fb4877d079af649847a691fb42fbc4b2.jpeg" />

    Link to comment
    Share on other sites

    • 3 weeks later...

    I forgot to report about one of my Favourite Fibaro devices: the Key Fob

     

    Please login or register to see this image.

    /monthly_2022_06/fibarosmart-homefibaro-keyfob-z-wave-plus-telecheiristerio-fgkf-601-a2182-550x550.jpg.51c2c62419e9c9c987b82763f01ff976.jpg" />

     

    The takeaway for this device is this: the key fob sends 100% when you insert a battery, even when that battery is nearly depleted. After 5 minutes the Key Fob sends a real level, based on measured voltage.

     

    As far is I can tell, you cannot speed up this process by entering the LED menu and selecting a "battery status".

     

    I bet this has to with the optimised circuitry, which allows it to take every bit of energy from a CR2450.

     

    I love this device... Excellent battery life, the LED feedback, settings, the 6 buttons, the sealing to protect against lint accumulating inside, and I probably forgot some more positives. As to negative aspects, I cannot think of any.

     

    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.

    ×
    ×
    • Create New...