Power cut again!

[dofbikes]

Hi everyone!
First of all I'm very sad because of the performance of GoldenMotor products. I bought it to escape from oil industry but 6 months gone and no success.
I've bought one more or less in the latest August.
Till now I've had no chance to run more that 7km with the motor running well. There's always something that does a power cut or is damaged.
I work in an production HVAC company that also uses motors attached to fans an we've no chance to fail no matter if there's or not controllers or any other coupled materials. It just cannot fail so often.
As you can read in the post http://goldenmotor.com/SMF/index.php?topic=4121.0 I've solved every problem I've had. The latest was to buy and solder an Hall Sensor to the twist throttle. After it it worked well but just 6 km. Then again power cuts.
I've checked the battery plug (GoldenMotor 48V 12Ah) and verified that there's an arc weld in the negative (-) side as you can see in the picture bellow:



Both connector come loose when supplied. After that I've changed the connectors to try to check if it was a connector problem but no way.
If I disconnect and connect it works again but I cannot accelerate too much. This was what happened just before the Hall Sensor gone (probably it was so stressed by the plugs and unplugs that failed).
If I accelerate slowly it runs and depending on something that is not constant it turns of. I've tried very very slowly to twist and sometimes it fails faster than others.
Please see the following video:

http://vimeo.com/34380748

After that I've discovered an amazing thing. If I accelerate an let it cut (twist throttle led lights turns off) then if I pedal a lot lights turns on and got stable, i.e. it does the same as connecting and disconnecting the plug. Now, that's strange. Please take a look at the following video:

http://vimeo.com/34381011

I've read this topic http://goldenmotor.com/SMF/index.php?topic=3704.0 and probably I've to do something similar. But I've to figure what it is in fact the switch that can be set to 1K ohm.

Thanks in advance,

Rui Santos

[Bikemad]

Hi Rui,

Total loss of battery power is usually caused by one of the following:

• A poor battery cable connection
• Burnt or failed contacts inside the battery switch
• A charging fault preventing the battery from charging correctly and the voltage eventually falling so low that it triggers the Low Voltage Cutoff (LVC) protection
• A faulty group of cells within the battery
• A faulty Battery Management System board (BMS) causing the power to be disconnected under normal load

Try monitoring the battery voltage with your digital voltmeter to see how low it drops just before the power cuts out altogether.

I'm guessing that your cutting out might be due to a BMS fault. I would want to try temporarily bypassing the BMS to see if it cures the fault, but as it's a recent battery, it might be better to contact GM to see what they suggest.

Alan
 

[dofbikes]

Till now I had no chance to get a new battery so I've tried to full charge the battery and it worked, but not for a long time.
Take a look into the following videos:

1st :  http://vimeo.com/35531945 (before any ride with some stress braking to check if it's failing write after a full charge. Conclusion, no it isn't)
2nd: http://vimeo.com/35532198 (after a 4km ride)
3rd: http://vimeo.com/35532382 (right after the 2nd)

Attached you can get the file of the data logger of my multimeter.

I suppose it's a group of cells that is damaged.

[dofbikes]

Hi everyone. Finally I've found the reason of power cuts. I'm happy for that but do not have good news about Golden Motor battery.

First of all I've tried to use 4 12V 7Ah SLA Batteries (picked from old UPS) and it performed perfectly. I've made an association of them and got 48V 7Ah.
I've runned about 8km's until it got completely uncharged.
You can see below some photos of the assembly:







The empty batteries:










After that I've tried again to full charge GoldenMotor LiFePo 48V 12Ah battery and ride some kilometers. No more that two and power cuts again.
The reason of failure? Some LifePo Cells died even without a big stress!
Since July 2011 I've ride the ebike no more than 10 times with that 48V 12Ah GoldenMotor battery. Since them I've troubleshooted many many problems with the whole system and step forward very slowly because many components failed during my "epic ebike journey". Everytime I use it I full charge it before an it gets around 53V. But then again rapidly it goes to 49V, 48V (I've already published in this forum a txt file with digital voltage reading during a power cut)  and a lot less if in stress and power cut happens

Take a look into the photos:



Please note that there are many cells have "FAIL" written because they're very low voltage!








Below you can see a picture illustrating all the results of the voltage readings that I've made using digital multimeter. I hope that the experts understand:



It's evident that 3 triplets of batteries "died" or are very close to "death". I don't know how GoldenMotor can justify that.

Please comment it if you can because something has to be done.

Thanks in advance.

Rui Santos

[Bikemad]

Hi Rui, it looks like you have at least six cells that have definitely failed, and three more that are not as good as they should be.

Send an email to David at GM (wyh@goldenmotor.com) and explain what you have found, and if you attach the photos and the diagram with your results, it should be easier for him to understand the fault.

Incidentally, your battery is not a LiFePO₄ pack, as it has the older type Lithium Manganese cells.

Anyway, please let us know what happens.

Alan
 

[Enpro]

Are you absolutely sure these are not lifepo4 cells Bikemad?

[Bikemad]

Hi Rui, did you get a reply from GM?

Are you absolutely sure these are not lifepo4 cells Bikemad?

GM no longer advertise the 48V 12Ah LiMn packs, as they have been superseded by the 48V 10Ah LiFePO₄ packs.

The LiMn cell has a capacity of 4Ah with a nominal voltage of 3.7V and is charged to 4.2V Max.
Your pack consists of 13 groups of 3 paralleled cells connected in series (13S3P), which produces a 12Ah pack with a nominal voltage of 48.1V (13 x 3.7 = 41) and a capacity of 12Ah (3 x 4Ah).
The fully charge pack voltage would be around 54.6V (13 x 4.2V) if the cells are charged right up to their maximum, and I suspect that this is the voltage stated on your charger.

LiFePO₄ cells are typically 3.2V nominal and 3.7V max, and 13 of these would only give 41.6V nominal and 48.1V fully charged, which is a lot lower than the 53V that you mentioned in one of your previous posts.

The new GM LiFePO4 packs have (LT) on the label after the battery size and a green paper label that stains easily as you can see. Why paper

This is one of my personal 48V10AH LiFePO4 packs



Gary

I think your pack's item code will be: LFP-4812S.

I hope this helps to clarify things.

Alan
 

[Morgen 3Eman]

Hi Alan,

I find your analysis very interesting, as I bought what was supposed to be a LiFePO 48 Volt battery from GM, and it is putting out about 56 volts.  So does this use more cells to achieve the higher voltage?



TTFN,
Dennis

[Bikemad]

I bought what was supposed to be a LiFePO 48 Volt battery from GM, and it is putting out about 56 volts.  So does this use more cells to achieve the higher voltage?

The LiFePO₄ pack does have more cells, check out this post for further details.

Alan
 

[dofbikes]

Hi BikeMad.

I've replied GoldenMotor and they told me that the most they can do is to sell me a new pack with a discount...
For me it's not acceptable.

Thanks again,

Rui Santos

[Leslie]

They may be able to replace the 6 cells, try that.  But if I was getting a replacement I would disassemble that 3.77v P string into seperate cells and see if one of the 3 cells drops considerably over a day..

Cut the tabs and cells appart, do not try to unsolder tabs and resolder.  to replace you just have to resolder cut tabs and this does less damage to new cells.



If no luck getting replacment cells. I would remove the dud cells short the BMS channel and ride on.  Shorting the BMS channel on a missing cell should make the LVC correct. Then I would design a crowbar charger cut off, and use this as my full charge off switch.  Yep Id run the charger 2s short and it should draw more amps from the charger.  If thats a problem, Id find a way of reducing the voltage of the charger ouput.


I found some that maybe compatible.

http://www.goodluckbuy.com/imr-26650-lithium-3-7v-4000mah-rechargable-battery.html



^^^Are these the correct cells ^^^



http://www.batteryspace.com/limnni-26650-battery-3.7v-8ah-29.8wh-20a-rate-w/o-pcb-for-diy.aspx



These might do, but a little expensive. but you can buy 3 pairs, cut the tab between 2 Parallel cells leaving the remaining tab on, and solder more easily the cut cell onto to two joined for 2 * 12ah P strings.




http://www.everbuying.com/product71400.html



These would be cool too.  Id pick up 12 of em just incase that 3.77 cell has a dud cell along the parallel..



Just look around.  Reparing this pack may be cheaper than a whole new pack, you look like y9ou can manage a repair.  Just look into how to solder tabs.  Whether you need lead tin or some other means to solder. 

The cells already connected should prove easier to manage because you can buy 3 pairs cut solder the tabs of one and join it to the two that already joined. Also I would guess your cells failed due to too much heat occuring on the jointing, one bad connection and that cell gets drilled by the heat of the resistance over the bad connection..
.



LiMn has one ore two advantages over LiFepo4.  If you get LiMn going good, they are lighter and more powerfull and can run for over 3 years if you don't require 100% dod.  They can live as long as they are rated at higher C level and 20% more capacity than the LiFePo4 pack..

[Supatrike]

Hi All,

I'm having the same problem...  see this video:  https://vimeo.com/39948336

I am so incredibly pissed because this battery is a replacement for another that GM messed up on.  I paid $80 USD to have them ship me a new battery that doesn't work!!! If it weren't for Alan and Gary I'd probably go ballistic.

Anyway, I really love the GM package, but they have to (HAVE TO!) work out some of these problems. 

So here's where I am at.  I think I am going to try to take apart the battery and swap out the cells for new ones.  I have two of these batteries (a 24V and a 36V - the 36V is bad).  I have opened both the 24 and 36 batteries, but the 36V is different inside - it's all wrapped in plastic and hard to remove from the shell.  I can post photos and such later, but here's my initial question:

Do you guys recommend that I take the battery apart and try to rebuild it?  I have some electronics skills, but not great (I can read schematics and solder).  Can someone help walk me through it?

I'm not wasting any more time trying to deal with China.

Thanks.

[Leslie]

So here's where I am at.  I think I am going to try to take apart the battery and swap out the cells for new ones.

Hey maybe you can give or sell dofbikes "Rui" some cells cheap @ cost from your old/new pack to help make up the cost.

We are a community here and it seems Rui has the rawest deal out of bad things we read in this thread.. 

When you buy pre assembled packs there is no way telling if someone has messed up the assembly on them until they get used.  Sometimes it takes heavy use for bad connections or cells to show their ugly head.  This is why many people build their own packs. 

If you do load testing always test the volts in between the welded tabs and not the cell tops.   Always make sure the weld are not failing before you ride.

You may see a volatge drop between the tab and welds.  If you get a 1v drop accross the weld and try to draw 20 amps through it this is 20 watts of concentrated heat sitting on your cell.   The cell will gladly suck the heat up and overheat at it connection points inside.

[dofbikes]

Les thanks a lot for the links for replacement batts.
Supatrike, it can be a good idea to sell me some of your cells if this is not a problem to you. Also you can take a look into my videos here http://vimeo.com/user9420783/videos. They are about troubleshooting MP motor and Goldenmotor battery pack.
Right now I'm making some tests with SLA's and sooner or later I'll publish also videos about that.

Les can you tell me if it's possible to retrofit the battery pack using less batteries ie less voltage and even tough the BMS works good?

Thanks,

Rui

[Supatrike]

Hi Rui,

I have watched your videos - thanks for posting the info. I have a LiFePO4 battery pack, and you have the LiMn pack... so, I don't know why I would sell you my cells - besides, how am I gonna ride? 

My pack did work fairly well for a few days, and hasn't worked since. 

Is there anywhere I could buy replacement cells for the LiFePO4?  Anyone want to buy my 36V LiFePO4 (not working)

Bart