UHK 80 - Left side battery drains within a day

[kvickyk]

I got my UHK 80 about two weeks ago. It has firmware version 14.2.0 already installed. I charged both sides and started using them wirelessly. I also have all the LEDs and the screen off completely. I noticed that almost every day, the left side completely drains its battery down to zero. The right side still sits around 90% after one day of usage.

Anything I can do to address this issue?

[kareltucek]

Hypotheses:

• hardware failure
• left half eventloop keeps spinning.

I am not sure if it is possible to get to left half logs without special hardware though :-(.

Try to

• connect left half via usb
• open Agent->About
• click the Agent logo 7 times
• scroll the left pane up and open advanced settings
• click zephyr logging
• do you see left half there?

• If yes, then click the checkbox. Do you see lots of warnings of form Warning: following event hasn't been unset:. If yes, what are they?

[kvickyk]

Clicking the logo 7 times does not do anything for me.

[gffranco]

Hey @kvickyk when clicking 7 times it won't change screens or anything, but it should show Advanced Settings just below the Firmware option on the left pane under Device.

[kvickyk]

It only appears if I connect to the right side and not the left side

[kareltucek]

Just a quick update: I am trying to sort out the left half log issue with Robert. I am away next week, so this will probably take some time. Sorry about that!

( @ert78gb referencing just fyi )

[kareltucek]

Hm, I see my left half now in Zephyr logging...

@kvickyk What exactly do you mean by It only appears if I connect to the right side and not the left side?

You are supposed to have your uhk80 connected by two USB cables - one to right and one to left. Bridge cable should not matter. There should be no other UHK connected to the computer at the same time. Agent should be ran without device-selection parameters (the --vid or --pid commandline arguments), but that is probably not your case (as you would know if you did that).

[kvickyk]

Ah! Since the keyboard came with just one cable, I was just using that. I got it logging now. I will let it run for some time before sharing the logs. I do not see any warnings yet. This is what is there so far.

2025-07-09 08:43:49 | UHK 80 left     | Printing trace buffer because: Looks like the firmware freezed. If that is the case, please report bellow trace to the devs:

EV: 512
Tag: v14.2.0
Trace:
[i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1]
[i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1]
[i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1]
[i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1][i1]
2025-07-09 08:44:23 | UHK 80 left     | Battery: ... Voltage corrected because of charging 4184 -] 4116 (88) -] 4108 (87)
Battery:     ... Averaged to 4093 (84)
Battery: Step called with 83 84 -] 83

Battery: corrected voltage is 4093 83

Batteries: 83% 0% (4093 0)
2025-07-09 08:45:23 | UHK 80 left     | Battery: ... Voltage corrected because of charging 4190 -] 4112 (88) -] 4103 (86)
Battery:     ... Averaged to 4095 (85)
Battery: Step called with 83 85 -] 83

Battery: corrected voltage is 4095 83

Batteries: 83% 0% (4095 0)
2025-07-09 08:46:24 | UHK 80 left     | Battery: ... Voltage corrected because of charging 4204 -] 4114 (88) -] 4103 (86)
Battery:     ... Averaged to 4097 (85)
Battery: Step called with 83 85 -] 83

Battery: corrected voltage is 4097 83

Batteries: 83% 0% (4097 0)
2025-07-09 08:47:24 | UHK 80 left     | Battery: ... Voltage corrected because of charging 4200 -] 4128 (90) -] 4119 (89)
Battery:     ... Averaged to 4101 (86)
Battery: Step called with 83 86 -] 83

Battery: corrected voltage is 4101 83

Batteries: 83% 0% (4101 0)
2025-07-09 08:48:25 | UHK 80 left     | Battery: ... Voltage corrected because of charging 4188 -] 4116 (88) -] 4107 (87)
Battery:     ... Averaged to 4104 (86)
Battery: Step called with 83 86 -] 83

Battery: corrected voltage is 4104 83

Batteries: 83% 0% (4104 0)

[kvickyk]

The same logs just keep repeating about the battery, I do not see any warnings

[kareltucek]

Those logs look fine.

I would expect the 4104 mV to be closer to 100% than in the log above, but that relates to the percentage calculation, nothing more.

I will try to implement an additional guard for the event loop as I am not entirely sure that the current guard really catches all cases...

[kareltucek]

We suspect a hardware problem, but before recall, lets do one more check.

Please flash this firmware: uhk-firmware-15.0.0-eventloop_spin_detection_v1.1.tar.gz

Set the oled display on so that you can monitor the info. (Possibly with some timeout.)

Then use it as usual, and check the oled from time to time to see if there is a warning triangle icon. If there is, start Agent and check what it says (there will be a big yellow pane in such case), and let us know.

After a day or so of testing (i.e., after you drain it within a day), let us know.

I noticed that almost every day

Does that almost mean that there are days when it drains significantly less?

[kvickyk]

I will try this out and report back.

What I meant by almost every day was that in the two weeks I had it, every time I used the keyboard in the morning, the left-hand side would not work, and I noticed it was drained.

But at the same time, I use this keyboard at home, and I was heading to work as well, so it was not like the immediate next day. So, I was unsure if it just drained in one day vs two. Also, during this time, the keyboard was disconnected from USB and my laptop was not nearby. Not sure if that played a role here or not.

[kvickyk]

This is what I see

[kvickyk]

@kareltucek Is anything else required here? Can a firmware patch fix the issue?

Currently, I am on the latest build and I have monitoring the battery through the LED screen. After charging both to 100%, today, after 3 days, the left has come down to 38% and right hand side is still at 94%.

[mondalaci]

I think Karel is away for about a week.