Yose Power kit strange battery behaviour

[Jammyb]

Morning,

First real problem I've had this morning - or is this normal behaviour for a flat battery:

Usually charge battery after 2 commutes (14 mile ish ) but thought i'd see how far the battery would last before being empty so today tried my third commute without charging. Battery was reading 36-37 so thought that should be ok. Soon as i started pedaling motor cut in and out with no real power. Monitoring real time voltage on the KT-LCD3 it was dropping to 25v then springing back up again. Is that just what a flat battery would do?

I turn pedal assist to 0 so the motor didnt kick in and noticed the voltage still occasionally dipped below 30 before returning to stabilize around 37v again. this happened periodically during my commute when there was no load on the battery.

Have my charger with me at work so hoping a full charge is all that required ( cross's fingers )

Battery is a 14.5ah samsung 29E 10s5p that was part of the kit from ymsbattery seller ( kits no longer available with battery included ) http://amzn.eu/44KRVH7

 

[Nealh]

Not right the battery bms should be cutting out well before you see 25v on the lcd, 11v sag is awful.
Cause could be a few factors.
1. A battery cell issue with either one or more poor cells.
2. A bad connection probably a parallel one.
3. A faulty bms/ lvc set to low not working correctly.
Even the controller should have a 31v lvc, with no load a 7v sag should not occur.

I have two 29E dolphin batteries both over 3 years old, although capacity has waned. On both they still give me 18 -25 miles range so still very usable. At 36/37v they sag at full amps or throttle use to the LVC then bounce back after 5 -10 secs, so I amble along in a lower pas and they are ok.
When they cut out I only see a min of 33.5 - 34v and all 10 strings are in perfect balance condition at both ends of the voltage scale.

 

[Nealh]

We need to see voltage readings, after a full charge and 3 or 4 hrs after a full charge once the battery has been left to sit.

Also it might help if could see what the 10 parallel strings are reading at low voltage to ascertain if there is an issue with any of them or the bms.
Other then check the voltage charge after charging as mentioned at the top, the battery needs to be opened to get more info following a discharge. My feeling is the bms is faulty.

http://www.pedelecs.co.uk/forum/threads/help-bms-board-dead-1-4v-low-battery-output.28570/
This thread shows a 48v Hailong battery opened and the jst / sense lead/connector where you can measure the parallel strings, a 36 v battery will be much the same but have a couple wires less to take readings.

 

[Jammyb]

Thanks Nealh

I didn’t think it was normal behaviour [emoji53].
I have my watt meter plugged in inline on the charger at work at moment. I know it over reads a little but seems to be charging ok, it was reading the battery voltage as 37 when plugged in at about 7:45 this morn and have been steadily rising as it charges to now read 41.34( been on charge for about 2h 25 mins)

According to the power meter under no load the battery voltage seems stable. I’m wondering if there’s a problem with the controller that’s causing the strange readings, I also noticed at various points this morning the battery indicator logo was a some point showing full and other points flashing empty.

I have added an xt60 connection between the controller and battery so I can plug in the gt power meter but didn’t have this connected on the way in. Will try on my commute home with it connected and see if the voltage sag is registered there or just from the controller lcd.

If it doesn’t work once fully charged I’ll open the battery case when I get home to check individual strings a suggested.

Should be fully charged soon, now reading 41.59 so battery shouldn’t of been that flat if it’s taken less than 3 hours to reach 42v. Watt meter says it’s put 181wh / 4.6ah back in so far

Sent from my iPhone using Tapatalk

 

[Nealh]

Don't rely on the Watt meter alone back up any voltage reading with a reliable working DVM. I had one of those black WM's and it was well out reading wise.

 

[Nealh]

Controller could be faulty, I had a KT from a well known UK supplier that was faulty and failed, it also had the same julet connectors so I replaced with a dual voltage model form PSWPower. UK supplier never replied to my emails so in the end I went direct to China, the UK supplier I have never used again and now don't give them a plug at all due to poor CS.

My Oxydrive Lishui integrated controller went tits up and I got lvc error messages on two near fully charged batteries and bike just kept cutting out. A KT controller swap was used and all fine again.

Problem with integrated controllers is they are small and compact and heat builds up with out much escape despite the ali heat sink and being at the base of a battery with airflow. All so being in cased in silicone rubber leaves little for heat to dissipate.

Getting a replacement might not be easy, might be better stripping out the old controller keep the fitting as you need it for the battery. Extend the supply leads and fit a dual voltage small KT in a under seat tool bag and have 48v/13s capability up your sleeve for another option.
With my Yose kit I use both 36v &48v batteries on a ride, adding an extra on/off switch to my 36v battery to cut supply before connecting 48v soft packs up allows this.

 

[Deleted member 4366]

If the battery's BMS switches off, the LCD would switch off and you'd have to switch it back on again. If the LCD doesn't go off, the problem is most likely elsewhere. Maybe a wire shorting or something like that.

 

[anotherkiwi]

Don't rely on the Watt meter alone back up any voltage reading with a reliable working DVM. I had one of those black WM's and it was well out reading wise.

And you can get lucky... All of my stuff gives the same reading to within 0.0X v including my Turnigy Wattmeter and my LiPo alarms!

So compare the reading of that Wattmeter with a known reliable value would be a great idea.

 

[Jammyb]

Woohoo bike is all working again, problem was all down to yours truly!

A while ago extended the cable from battery holder terminals to controller to put an xt60 connector inline so I could connect my watt meter and remove again when not in use. A bad solder joint on the xt60 connector had come loose! At least I’d used heat-shrink and self amalgamating tape to wrap all connections up so thankfully no shorting out damage.

Feel a bit of a fool, always the simple things hu!

Thanks for all the suggestions and replies you guys are really helpful [emoji106]

So glad it was nothing else, commuting felt so slow today, got used to the speed

Here’s what a video of what the controller was doing

Sent from my iPad using Tapatalk