Jump to content
Knifa

Ninebot One S2 - Only one side charging?

Recommended Posts

I have a variant of the above problem.  The attached photos show the state of the two batteries after charging FULLY (they both say "100% charged).  The Voltage on both is relatively equal, but the Current and Wattage show the discrepancy.  The red arc on the third photo grows and shrinks, depending on the state of discharge, but never gets smaller than shown here.  The unit shows all Green lights when powered on, until it runs out of juice.  Then one side shows blinking red.  The time allowed to ride before the unit starts beeping and becomes unstable is pretty short--maybe 1.5 to 2 hours.  The unit is less than a year old.  

Screenshot_20171120-085905[1].jpg

Screenshot_20171120-085856[1].jpg

Screenshot_20171120-085913[1].jpg

Share this post


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

I have a variant of the above problem.  The attached photos show the state of the two batteries after charging FULLY (they both say "100% charged).  The Voltage on both is relatively equal, but the Current and Wattage show the discrepancy. 

Maybe battery 2 is disconnected (bms shut off for whatever reason) showing no current flowing?

Maybe also a connector problem?

Quote

 

Screenshot_20171120-085905[1].jpg

 

Battery 1 has 6387 mAh of a total capacity of 2850 mAh left?

Edited by Chriull

Share this post


Link to post
Share on other sites

It sounds like you could have a dead/dying cell like me? I think remaining power is the same as how much mAh it put in during charging (e.g., my battery with the cell that's slightly larger capacity reads a little more than the other now).

I've had similar experiences with LiPo packs for drones with dead cells. It'll just continuously pump in power for ages trying to balance but it won't go anywhere or take forever.

I think only one battery (specifically battery 1) showing a current pull/wattage while stationary is normal. Mine does the same. See if you can get a look at it while riding it?

Edited by Knifa

Share this post


Link to post
Share on other sites

>See if you can get a look at it while riding it?

Good idea; will try it.

Share this post


Link to post
Share on other sites

>Battery 1 has 6387 mAh of a total capacity of 2850 mAh left?

Yeah--didn't even notice that!

Share this post


Link to post
Share on other sites
On 2017. 11. 13. at 5:32 AM, Knifa said:

With this, we can get a look at the back of the BMS board. The chip on the back here is the actual BMS IC. It's a TI BQ7694003 (datasheet) which is a series of battery monitor ICs providing current monitoring, short circuit protection, low voltage protection, yada yada. The chip on the front of the BMS board is a STM 8L151K6T6 (datasheet) which is a basic 8-bit MCU, presumably doing comms with the main board and handling the BMS IC. There are also two MOSFETs (following the design from the BMS IC datasheet, I guess) one of which is 100V 40A (MDD1902) and the other was larger but I couldn't read the numbers off.

 

The large size IC's (STH15810) STripFET POWER MOSFET.  100V, 110A. It feature  is Gate(1), Source(2,3) Drain(TAB).

MOSFET-STH15810_MDD1902.jpg.707eb0fbef1b501fb96a7395308e56df.jpg

15810FET.jpg.d3f9ae2e70199fba4700662dd93c5945.jpg

 

Have a nice repare and wheeling life..

 

datasheet_15810FET_ST.pdf

Share this post


Link to post
Share on other sites

I received that same error yesterday on battery 1. My concern is that I'm not quite the electrical-hands-on kind of guy to do the kind of troubleshooting you were getting into. At the least, I'd like to remove the good battery and try charging the suspect battery on its own. Which side is battery 1? Left foot side or right foot side? 

IMG_3167.PNG

IMG_3169.PNG

IMG_3170.PNG

Share this post


Link to post
Share on other sites
On 11/12/2017 at 9:32 PM, Knifa said:

So today we replaced the cell and, again, it seems to work! I also got some more information on the BMS during this so see below for details.

We measured the suspect cell (1, bottom left). It read 3.4V, where the other cells were reading 3.8V. Definitely a damaged/failing as mentioned above --- so it was time for it to come out. :)

First step was to liberate the cells and boards from the casing. Thankfully, nothing is glued down and everything is screwed in place with nice holders, etc. There are a good few screws, all covered in silastic, however if you squeeze hard enough with a screw driver (philips), they will twist out. We also had to tear away the silastic around the XT60 and BMS comms cables. Once that's done, the whole thing just kinda lifts out.

remove_screws.thumb.jpg.a14f672a0f75f463b8914d5b16a9b370.jpgfree_standing.thumb.jpg.06ff1a15b045e815d1f2bb801bd2f12e.jpg

With this, we can get a look at the back of the BMS board. The chip on the back here is the actual BMS IC. It's a TI BQ7694003 (datasheet) which is a series of battery monitor ICs providing current monitoring, short circuit protection, low voltage protection, yada yada. The chip on the front of the BMS board is a STM 8L151K6T6 (datasheet) which is a basic 8-bit MCU, presumably doing comms with the main board and handling the BMS IC. There are also two MOSFETs (following the design from the BMS IC datasheet, I guess) one of which is 100V 40A (MDD1902) and the other was larger but I couldn't read the numbers off.

Some interesting tidbits from the BMS IC:

  • There's mention of a SHIP/NORMAL mode (low power for shipping vs. normal operation) and a button shown in the example circuit. Maybe this is what the button is for?
  • The IC is capable of balancing using either its own internal drivers or externally (I guess a whole ton of external MOSFETs like a normal charger?). In this case, it looks like Ninebot have opted to use the internal drivers as there isn't really anything else on the board. Internal balancing is limited to 50mA which is gonna be hella slow --- is this what happens during trickle charging? Balancing is controlled by the host (the STM) rather than by the BMS IC

bms_back.thumb.jpg.c28099cf4cf554b18eca749dea5260a7.jpgbms_ic.thumb.jpg.fbdd4bf4bb27b4c73917405fecb82c25.jpgstm.thumb.jpg.bdc8ea36be3da3a184afc76fe5038065.jpgmosfet1.thumb.jpg.c0f696030ac7d18a20ff19e3f7a72af8.jpgmosfet2.thumb.jpg.2913a30d510cba5815a9872e18af29a5.jpg

After that adventure, we ripped the dud cell out. We pried at the welding tabs with a stanley knife until it simply came free. There is enough room to get in there (at this edge cell, at least) however I'm not sure what it would be like trying to do it in the center. The plastic casing around the batteries is weirdly flexible so it makes it pretty nice to work with. At this point, the LED started blinking red (surprise!) which at least confirms it's not pulling all power from one cell. :) Pushing the tab back into the battery made it turn a happy green.

tabs1.thumb.jpg.0b24082fa2ae958cc3b5f3d8716b87ba.jpgtabs2.thumb.jpg.234a1ecc6263abe8ab148e961cdfdb36.jpgangry_red.thumb.jpg.63588c90bdb2f64f7e59f92afc1413cd.jpgcell_gone.thumb.jpg.58b8a1868f13bb381aef8ffbc9eff5ed.jpg

With the cell gone, it was time to put in our new cell. The new cell is a LG HG2 Series 3000mAh from Amazon directly, which fits the bill for this perfectly. As far as I could tell, the HG2 is the step up from the cells that are already in this. The chemistry is the same (INR) and it provides the same/higher constant discharge rates so all seems good. I should have really double checked the capacity at this point, to make sure it was legit and not a fake, but some visual checks (plus coming straight from Amazon) made me feel OK. If I was doing it again, I would double check.

I didn't get any pictures of the soldering process, unfortunately, but the gist is this:

  1. We used a good temperature-controlled soldering station (Hakko 888D) with a very large tip, good solder (MG 63/47), and good flux (MG liquid something...). The iron was set to 450C.
  2. The positive and negative sides of the battery were tinned with a reasonable dollop of solder, after applying a lot of flux. We double checked for heat at this point, and the battery didn't get hot at all.
  3. With lots of flux on the tabs, the battery was placed in the plastic holdery-thing and held firmly in place. The positive end was soldered first (near the BMS board). My friend pushed the tab firmly towards the battery with a stick while I gave it a quick tap with the soldering iron.
  4. Repeat above for the other end. :) The battery had a warmth to it afterwards but not warm enough to count as "hot". The light started flashing green again, so seems like everything went together well.

 After this, we basically did everything in reverse. Everything back into the casing, screws all in, bit of hot glue to make up for the missing silastic. Nothing in particular to report here.

finished.thumb.jpg.e32316b18b11251230ea6655a1046ace.jpg

Once all back together and screwed into the Ninebot (on it's own, other battery disconnected), it all seemed to work OK! No shouting from the app, reporting around 60% capacity. I rode it around for a good bit, then to the store and back, and it seems to be dropping at a fairly reasonable rate. Battery temperature also reports reasonable temps, no huge voltage sag, etc.

So that's it! All seems to be in working order, for now. That said, a couple of things still to test:

  • Run the battery to near empty, and charge it up again. Haven't tried charging it with this new cell yet.
  • Test with the other battery installed. Do they drop in capacity at the same rate?
  • Go through a couple of cycles, crack open the battery again and check cell voltages. Are they all within reasonable limits?

 

 

Hi there,

Thank you really much to Knifa for this topic, a great piece of work ! 

I'd like to repair a battery with a similar problem. Following the method, I just don't know how to remove properly the "welding tabs" which are really tight and I'm personnaly afraid to break them by following the same method with the knife... Does anyone have any advice or clue to make this step easier ?

Thank you in advance.

Share this post


Link to post
Share on other sites
Posted (edited)

Thanks for sharing all your info @Knifa :)

I am in the same operation (supposedly, dud pack, opened it) Just a small technical question: do you remember how you managed to measure the batteries that are not on the edges? The four accessible ones of my pack have different voltages, so I feel I should measure them all.. Thanks and sorry for the bother.

Edit: never mind, I found, using the opposite side bridges :)

For the general info in the thread: the case clips are where the bumps are, + one set on the sides.

Edited by Mimolette

Share this post


Link to post
Share on other sites
Posted (edited)

Good to hear you managed to measure it @Mimolette! Getting to some of them is pretty awkward without removing tape, etc.

I thought I'd say as well, I've been riding on these batteries every day for like a year since (hitting 1500km+) with no issues. Worse things have happened since then like dropping the entire unit in a huge puddle and having to replace MOSFETs but it's still going!

Edited by Knifa

Share this post


Link to post
Share on other sites

Thanks for the comment @Knifa, and good to hear about the followup!
Mine had very different charge on all cells (everything between 4,2 and 3,6v), but no remarkably so, luckily a couple of overnight charges seems to have balanced them out. Sorry for tagging on your thread but as it was linked to measuring individual cells.

Cheers.

Share this post


Link to post
Share on other sites

Nice little how-to here. I stopped by to hopefully get advice. My battries are somehow weak now. I can ride approx 15-18km with fully charged packs. In the app all is very even. No bigger discrepances between both packs. But where has capacity gone? I was able to ride 25km+ on one charge at minimum?

I Pulled the packs out at about 65% Charge (app) and measured the cells:

Pack 1:
long row: 3,80 3,81 3,80 3,81 3,79 3,87 3,86 3,87 3,87  3,86
short row: 3,88 3,87 3,88 3,88 3,88 (V)

Pack 2:
long row: 3,76 3,77 3,76 3,77 3,77 3,87 3,86 3,86 3,86 3,87
short row: 3,88 3,88 3,89 3,88 3,89 (V)

both read the same data via the app, but Pack 2 has more of the weaker cells, whilst in both packs the short row has the strongest cells.

No cell seams to be really odd.

Any suggestions?

Share this post


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...