Jump to content

End of charge battery drain: workarounds?


Recommended Posts

Hi!

I start to get annoyed by the battery drain following the end of charge and cell balancing of the KS-16X, which happens even on the freshly released 2.03 firmware.
In a video from @Duf, it seems like the KS-S18 is also affected.

First, I am blown away how such an issue can pass QC for months.
Charging a battery pack to its full capacity accelerates battery degradation.
But if you don't take action once the wheel finished charging it will drain back down, wasting cell health and range.

As a developer I am wondering what suitable workaround could be implemented to fix this.

When looking at this graph, despite the noisy data points it looks feasible to identify when the wheel completely finished charging and is starting to drain.
I'm hoping that it would be possible send the "turn off" command over bluetooth before drain occurs, identified by either:

  • amperage not going into negative values anymore
  • voltage going back down

What do you think?

Ping @Seba on implementing this in EUC World.

Ping @Jack King Song on fixing the issue in firmware.

drain-after-charge.png

2020-10-15 143718.csv

Link to post
Share on other sites

In your recording, your pack saturated at 84.5V with the charger connected.
Then, about 2 hours after disconnecting the charger, the system voltage fell to 84.4V.

That is normal for lithium cells...

LG%2018650%20MJ1%203500mAh%20(Green)-Cha

Edited by RagingGrandpa
Ah, you understand this well... leaving it here for reference, carry on
Link to post
Share on other sites

Hi @RagingGrandpa!

Yes the 16X use the MJ1 cells and like every other they should be charged up to 4.2V constant voltage, then the voltage would dip a bit.
To keep the message shorter I didn't specify that my wheel's voltmeter over-reports voltage up to 84.6V (calibration issue) despite the charger only provides 84V, with good accuracy.

I confirm that the issue I'm describing is draining, not the cells settling down to a lower voltage once the charger is disconnected.

In this case, the wheel's motherboard remains powered on, with at least the BLE and Bluetooth audio chips powered on

It drains constant 20W from the pack indefinitely.
It is enough to be at 90 something % after a few hours, 80 something % after a few days.

Link to post
Share on other sites

I notice my 16x will start beeping after a few hours from charging to full. I usually unplug and cycle the power so its really off. 

Link to post
Share on other sites
1 hour ago, Waulnut said:

I notice my 16x will start beeping after a few hours from charging to full. I usually unplug and cycle the power so its really off. 

I installed a hardware switch for the beeper, which I toggle off when coming back home for the neighbours' sake.

So if the charge is completed in the middle of the night, it starts beeping then? 😬

Link to post
Share on other sites
3 hours ago, supercurio said:

It drains constant 20W from the pack indefinitely.

It is enough to be at 90 something % after a few hours, 80 something % after a few days.

You're kidding right?! And I thought the 1% per day on the Z10 was bad!!

Link to post
Share on other sites
1 minute ago, Planemo said:

You're kidding right?! And I thought the 1% per day on the Z10 was bad!!

It will drain battery as long as you don't turn it off from the kind-of-on state activated when you connect the 16X to a charger.

There's no battery drain once off.

Link to post
Share on other sites
28 minutes ago, supercurio said:

I installed a hardware switch for the beeper, which I toggle off when coming back home for the neighbours' sake.

So if the charge is completed in the middle of the night, it starts beeping then? 😬

It is possible it will be on and it trickle charges to make up the small power used with the controller on. I am not sure why it starts beeping. I always thought it alerts that it is going to shutdown because it is idle too long, but it might not be the case. It seems to happen if I leave it longer on the charger than when I pull the plug a little after it turns green on the charger. 

My 18XL on the other hand, sometimes the brake light stays on after charging to full. I then cycle it to turn it "off". But this is noticeable because if i roll the wheel forward and back, the brake light follows the direction like it is on. Weird. Other than that, wheel is fine and it only happens occasionally. 

Link to post
Share on other sites
29 minutes ago, Waulnut said:

It is possible it will be on and it trickle charges to make up the small power used with the controller on. I am not sure why it starts beeping. I always thought it alerts that it is going to shutdown because it is idle too long, but it might not be the case. It seems to happen if I leave it longer on the charger than when I pull the plug a little after it turns green on the charger. 

That would be a different behaviour to the one showed in the graph and wheel log attached in the first post.

The reason why I call it battery drain is that the charge is stopped and the main board keeps running, draining 20W or more while idle.
Attached: a crop of the log viewed in wheellogviewer.net

I stopped recording the log and turned off the wheel manually since I didn't want to drain the battery unnecessarily, but a few previous times it ended up at 98%, 97% or 94%.

20W+-drain.png

Edited by supercurio
Link to post
Share on other sites
2 hours ago, RagingGrandpa said:

Something's better than nothing, right?

Thanks for sharing @RagingGrandpa

"Optimized Charging Management Mechanism" is the changelog for KS-16X 2.03 firmware as seen in Soft Tuner

This note is part of what prompted me to start this topic as I hoped that the issue was fixed and ended up disappointed.

I'm currently investigating the effect of:

  • Automatic shut-off timer setting values.
  • Using a charger with an self shut-off capability at end of charge (like the adjustable chargers sold by 1RadWerkstatt)

The initial log was capture with shut-off timer disabled (set to 0 in Soft Tuner): the wheel never stops when idle.
I prefer this setting because I can remain confident that the wheel is never gonna stop balancing and fall somewhere.
It might affect the behaviour observed.

The initial log was also captured with a standard charger, that stays on at 84V constant voltage.

Link to post
Share on other sites

I too find it quite whacko that modules stay in a wakeful state when the wheel is off and charged to 100% - the advantage of being able to get data via btooth is way overshadowed by the crazy vampiric drain! This is just poor fware programming by KS - pole the batt state every minute - when 100% initiate full shut down.

Sadly even if they did this I would not likely use it cuz I don't trust fware updates - too afraid to brick the wheel :-(

 

 

Link to post
Share on other sites
12 minutes ago, amelanso said:

I too find it quite whacko that modules stay in a wakeful state when the wheel is off and charged to 100% - the advantage of being able to get data via btooth is way overshadowed by the crazy vampiric drain! This is just poor fware programming by KS - pole the batt state every minute - when 100% initiate full shut down.

Sadly even if they did this I would not likely use it cuz I don't trust fware updates - too afraid to brick the wheel :-(

Thanks for understanding, and confirming it's not just me ;)

I just reproduced the condition.
Despite setting the automatic shut-off to 5 minutes, the wheel seem to remain on, draining the battery for longer than 5 minutes after:

  1. completing the charge fully (end of cell balancing)
  2. disconnecting the charger

I am now trying to see if the wheel will shut-off 5 minutes after disconnecting bluetooth as well - EUC World was used to create a new log.

In case the wheel doesn't shut itself off at end of charge, automatic shut-off set to 5 minutes and no app connected, I will try to shut it off via the standard Kingsong app or SoftTuner.

Link to post
Share on other sites

Bad news

When the KS-16X is in charging-only on state and finished charging fully, sending "turn wheel off" commands doesn't turn it off.
Same as if it is in charging-only on state and finished charging fully, unplugging the charger doesn't turn it off.

However, here are conditions where the wheel shuts off as expected:

  • turn wheel on, plug charger, turn wheel off with the hardware button, wheel reboots in charging-only state, unplug charger: wheel shuts off
  • turn wheel on, plug charger, turn wheel off with SoftTuner, wheel reboots in charging-only state, unplug charger: wheel shuts off
  • wheel is off, plug charger, wheel boots in charging-only state, unplug charger: wheel shuts off

I still have more conditions to test and still have a slim hope to find one that doesn't trigger the issue and could be used to implement a workaround in an app.

Since the behavior becomes different in an unexpected way at end of charge from the charging-only state, I'd guess it is an unintended behavior in Kingsong's firmware programming - a bug essentially.

Link to post
Share on other sites

I noticed the same issue with my KS S18. It drains the voltage down after the charge is completed. EUC world shows a 0.3A constant draw, which I assumed was the BT connection staying active.

The  wheel is on FM 2.06, so I am curious if it will start charging again if left on the charger and if the voltage drops below a certain threshold, which is how I interpret point 5 in the release bulletin. Would have been nice if they stated what the threshold was as well....

I will check what the battery state is when I get home tonight. Yesterday during charging I had the following situation:

  • charged from 13-100% at 4.95A, showing as 83.5V
  • Continued to charge with 3.5A until 84V and slowly dropped off the amperage to 0.5A settling at 84.5V
  • I left it on the charger and checked back on the wheel after 13 hours and the voltage was down to 83.8V and the draw was 0.3A

--> I will edit this post with my findings tonight if the wheel has moved from the 83.8V. I simply unplugged it from the charger, did not power cycle the wheel.

 

Update: So I checked the wheel when I got home, and the voltage had dropped to 83.2V and still showed a drain of 0.3A. I turned the wheel on and then back off and let it sit for several hours and checked it again, now the voltage was stable at 83.2V --> so it seems that atleast the BT will stay active after a completed charge and the wheel has to be power cycled in order to completely shut it off and stop the draining of the battery. It is small, in my case 0.6V over 10 or so hours, but this equates to lets say 1V per day, so a week would drop it 7V etc.. from now on I will start to power cycle the wheel after a completed charge cycle to prevent this from happening.

Edited by FinRider
Link to post
Share on other sites
  • 2 weeks later...
  • 4 months later...
On 11/2/2020 at 7:20 PM, Jack King Song said:

I‘ve brought this up with my engineers, I'll see how they respond to your results/findings and let you guys know here. 

Really need an update on this.  Its crazy!! got a new KS-S18 in Jan 2021 and I ride to work every day roughly 3 miles downhill.  When I get home on Friday I plug it in an forget it over the weekend.  I started noticing that Monday morning my Battery life is 80-82% before leaving for work.  The weekend drain is nuts!!!! I get enough battery anxiety that i try to avoid running out for lunch on monday or using the thing just so i know I'll be able to get home back up the hill.  It hasn't left me stranded so far but at this point I'm looking at buying another charger to leave at work becuase thats the only option. 

The drain was so bad I thought I had a bad unit.  Like others have said, if I catch the charge at 100% I can unplug the charger and use the hard button to power on and then down at which point the drain stops but that seems like a terrible design flaw.  I paid nearly $2000 for this thing and didn't expect this kind of issue.     

Link to post
Share on other sites

I have never noticed this. I keep my chargers on a timer. I have several outlets on 24 hour dial timers. Through the years I have used them for everything from brass tumblers to water pumps.  I have always just turned the dial to 6 hours and walked away. They come in handy for someone that gets distracted easily.

I even have one on my 3phase split converter that runs the lathe and milling machine ..... Actually I also have a second circuit that turns off the the 3 phase equipment and air compressor when the shop lights are turned off. No more waking up in the middle of the night when the air compressor turns on. 
where was I?? Oh yeah! Timers are awesome. But yeah, Kingsong should fix that. 

Link to post
Share on other sites
1 hour ago, jayd623 said:

When I get home on Friday I plug it in an forget it over the weekend.

Leaving it to charge past 100% wears quite a bit on the battery, and could result in weak cells going down even faster. The battery not charging to 100% (or not keeping the charge) can be a sign of a dead cell(s), which is dangerous.

With some luck it is only vampire drain which with KS seem to be when the EUC controller board doesn't power itself off after charge.

So I would suggest investigating which of these it could be, and if it is the later, make sure it is powered off yourself and hope for a firmware update that fixes it. Anyhow I would recommend lot leaving it to charge much past 100%, whichever method you go for.

Edited by null
Link to post
Share on other sites

The timer trick doesn't work to stop KS vampire drain... when charge is complete the control board stays powered on even if you unplug the charger.

But you might downgrade your firmware... for my 16XS anyway, the release notes for v2.0.2 imply that they will restart the charge cycle if the charger is plugged in and the battery goes below a set level. Doesn't fix the fundamental issue of leaving the control board powered up, but it's something. Still, it's better to power cycle the wheel when it's done and had time to finish balancing the pack.

Over the past few days I've had to mess with charging nearly charged packs and 2.0.2 won't restart the charge cycle until it sees 81.99V—even if you unplug, power cycle the wheel, and plug the charger back in. So I think it'll have to suck off quite a bit before the 'improvement' kicks in.

Edited by Tawpie
add bit about where charging might restart
Link to post
Share on other sites
4 hours ago, null said:

Leaving it to charge past 100% wears quite a bit on the battery, and could result in weak cells going down even faster. The battery not charging to 100% (or not keeping the charge) can be a sign of a dead cell(s), which is dangerous.

With some luck it is only vampire drain which with KS seem to be when the EUC controller board doesn't power itself off after charge.

So I would suggest investigating which of these it could be, and if it is the later, make sure it is powered off yourself and hope for a firmware update that fixes it. Anyhow I would recommend lot leaving it to charge much past 100%, whichever method you go for.

Well, yes and no. The balancing needs to be done every once ina while. However, it appears as though there is a something that drains the battery and the charging cycle wont restart.. my hunch is that it is the Bluetooth that stays active. Cycling power on the wheel after a full charge cycle stops the drain.

ks needs to look into this.

Link to post
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.
Note: Your post will require moderator approval before it will be visible.

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...