Electric Unicycle's BMS problem and solution

[joelren2k]

On 6/24/2015 at 3:46 AM, Fahrtwind said:

Hi,

thanks so much for the help, I shunted my generic Wheel as suggested in the pic above:

I already charged the Battery without smoke 

Quite easy to do. Especially in winter a plus for safety! Thanks again!

Manu

 

hi there,  

if you dont mind, may i know what brand of generic wheel you have?

thx.

[JamakinJoel]

I recently got a tg t3 from banggood and it shutoff on me yesterday going full speed.  Luckily I jumped off of it without falling.  My board looks like Fahrtwinds seen here: http://i.imgur.com/4Tl7SZN.jpg

But my board has a chip where he soldered. Seen here: http://i.imgur.com/VvNWaL9.jpg

Can anyone tell me where I should solder? Any advice is greatly appreciated.

Thanks, Joel

[Chriull]

10 hours ago, JamakinJoel said:

I recently got a tg t3 from banggood and it shutoff on me yesterday going full speed.  Luckily I jumped off of it without falling.  My board looks like Fahrtwinds seen here: http://i.imgur.com/4Tl7SZN.jpg

But my board has a chip where he soldered. Seen here: http://i.imgur.com/VvNWaL9.jpg

Can anyone tell me where I should solder? Any advice is greatly appreciated.

Thanks, Joel

You have a picture of "your" board shunted here: http://forum.electricunicycle.org/topic/459-electric-unicycles-bms-problem-and-solution/?do=findComment&comment=22027, or you can shunt it like seen on the first pictures (for TG BMS) here: http://forum.electricunicycle.org/topic/459-electric-unicycles-bms-problem-and-solution/?do=findComment&comment=4655 (The three mosfets seen here are the three mosfets on the right side of your picture).

Both ways to shunt are more or less the same - you can choose freely, which way you prefer.

Do not forget the comment from Neale Gray in your other thread " Do not under estimate the potential level of danger involved in dismantling Li battery packs, there are already too many serious injuries proving this fact. If it does go badly then the results can be extreme. " - The BMS is under voltage (connected to the high capacity batteries) and you may not shorten any other parts on it by accident or it can get very uncomfortable!

[JamakinJoel]

On 12/22/2015 at 1:15 AM, Chriull said:

You have a picture of "your" board shunted here: http://forum.electricunicycle.org/topic/459-electric-unicycles-bms-problem-and-solution/?do=findComment&comment=22027, or you can shunt it like seen on the first pictures (for TG BMS) here: http://forum.electricunicycle.org/topic/459-electric-unicycles-bms-problem-and-solution/?do=findComment&comment=4655 (The three mosfets seen here are the three mosfets on the right side of your picture).

Both ways to shunt are more or less the same - you can choose freely, which way you prefer.

Do not forget the comment from Neale Gray in your other thread " Do not under estimate the potential level of danger involved in dismantling Li battery packs, there are already too many serious injuries proving this fact. If it does go badly then the results can be extreme. " - The BMS is under voltage (connected to the high capacity batteries) and you may not shorten any other parts on it by accident or it can get very uncomfortable!

Thanks so much! I'll definitely keep all the warnings in mind.

[Torsten3008]

 Hey guys,  mine tg-t3 is also looking link this http://i.imgur.com/VvNWaL9.jpg. 

 

But how to shunt? There is not much space on the q1.

Any pics or hints? 

 

Is option in Green (2 or 3 mosfet?)  or blue better?

I did try the green marked solution...  Right?

Edit: i did a first test,  no shutdown,  green option is working fine. 

Thanks a lot  as this is my first post :-(

[uniler]

Ok, stupid question: My genetic wheel has a tilt back triggered by over speed, should i bother shutting? I'm gonna probably say not since it doesn't pose a big problem but would love your opinion. Also, is the tilt back a common feature?

 

 

[esaj]

16 hours ago, uniler said:

Ok, stupid question: My genetic wheel has a tilt back triggered by over speed, should i bother shutting? I'm gonna probably say not since it doesn't pose a big problem but would love your opinion. Also, is the tilt back a common feature?

Shunting won't affect the tilt-back behavior, it just prevents the BMS from cutting power if voltage drops too low or current goes too high (including short-circuit!). Most wheels have tilt-back, the ones I know that don't have tilt-back are (at least) older Gotways and Firewheel.

[joelren2k]

On July 28, 2015 at 1:19 AM, hobby16 said:

No, those wheels are X3 clones,  they all use the same battery with 16 cells in series so, mainboard, battery, all are interchangeable.

If you buy up to 170Wh capacity (should cost about $100), there won't be any space problem since they use 18650 cylindrical cells.

With higher capacities, like 340Wh, you'll have 32 cells instead of 16 and the pack won't fit in your Weerda, forget it.

Up to 170 Wh.

If you love your lady, shunt the BMS. Don't shunt if you want her to go away.

Your other option to achieve 340wh is to attached the other 16 batt pack outside the shell. Just make sure the batt pack is secured with casing ang padding/foam to prevent burn/explotion should it hit other object during accident.  

 

[Neale Gray]

@esaj I read this post with interest especially the part about asking distributors to post here, Only a week ago I had a post removed for pointing out how dangerous these battery modifications are especially in the face of the current Li house fires.

The bottom line is that if you pay the absolute lowest price then you will receive the absolute lowest quality. When the device has a high power Li battery such as 130Wh (the smallest EU battery generally available) then your battery will cut out, be unstable and not do the job it was designed for, do forum members really think that batteries are the same and it is just manufacturers and resellers ripping you off.

There is a possibility of a thermal runaway happening with Li batteries and a lot of development has been undertaken to reduce this from happening.

Any and all modifications listed here are all highly NOT recommended. Buy a NEW battery if you are unhappy with the one you have, DO NOT modify in any way as all you do is loose any warranty and make yourself liable for any bad outcomes.

Do NOT treat Li like lead acid ie "you cant electrocute yourself with 60V", I can tell you if it goes bad (eg while soldering) a little DC shock will be the least of your worries.

So if you want a distributors opinion then for anyone who is unaware, EVERY SINGLE battery manufacturer, supplier or distributors follows the well known world wide recommendation: never ever modify or dismantle a lithium ion battery (unless dismantle/assembly/ is your profession).

[Jurgen]

I'm not into electronics, so it's probably a stupid question: has anybody thought about using a large (f.ex. 100V) capacitator to cover the voltage drop for like 1 second or maybe longer?

[esaj]

3 hours ago, Jurgen said:

I'm not into electronics, so it's probably a stupid question: has anybody thought about using a large (f.ex. 100V) capacitator to cover the voltage drop for like 1 second or maybe longer?

I think this has been discussed before somewhere in the forums... I believe you'd need a large supercapacitor (a few farads at least?) for the charge to last for any meaningful time with the amount of current the motor pulls. That large capacitors (especially ones that can withstand 100V) are heavy and big (maybe something like a coke bottle and maybe a kilo or two?). Not to mention they're probably not cheap either.

Edit: Taken from here:  http://www.mouser.com/pdfdocs/ELNACalcuDYNACAPDischargeTime_Jul2011.pdf

Calculation of necessary Capacitance ①For constant current discharge C　＝　I × t ／ （V0－V1)

t : Discharge time (sec.) V0 : Charge voltage (V) I : Discharge current (A) C : Capacitance (F) V1 : Discharge voltage (V)

 

Let's say the wheel is only pulling 250 watts (0.25 kilowatts) @60V => I = P/U  =>  250W/60V = 4,1666...A, and the voltage drop allowed would be 5V (V0-V1) over a period of 1 second (I'm not taking into account the increased current when the voltage drops that's required to keep the output at 250W, although there's the equation for constant power output in the pdf too). The capacitance would need to be

4.1666A * 1s / 5V  = around 0.833.. farads.

More extreme case, wheel is pulling 1000W (1 kilowatt) @ 60V, I = 1000 / 60 = 16.6666...A, same voltage drop and time period as above:

16.6666A * 1s / 5V = 3,333... farads.

Acceptable capacitance should probably be somewhat larger due to higher current needed as the voltage drops to keep the power output around the same. I took a quick look at Mouser, filtering out only supercapacitors that can handle 75V or more. I was left with 7 choices (capacitances between 5.8F / 160V ... 94F / 75V), the cheapest (12F / 80V) was at around 350€ per piece (apparently minimum order is 100 pieces or more), and sized around 24x25x7.4cm, no weight mentioned. The rest could be bought one piece at a time, price between 950€ - 6000€ / piece. Sizes were something like 20x20x7cm up to something like 40x70x25cm. They don't list weights, but I'm thinking several kilos.

EDIT2: I'm not saying it's a stupid idea or stupid question, actually it would be really useful in case of a sudden (short-lived) power cut, but the sizes, weighs and prices make it pretty much impossible (for our imperial unit-people: the smallest capacitor was around 8" x 8" x 3" and probably weights something like at least half a dozen pounds).

 

[Jurgen]

On Thursday, January 14, 2016 at 0:10 AM, esaj said:

I think this has been discussed before somewhere in the forums... I believe you'd need a large supercapacitor (a few farads at least?) for the charge to last for any meaningful time with the amount of current the motor pulls. That large capacitors (especially ones that can withstand 100V) are heavy and big (maybe something like a coke bottle and maybe a kilo or two?). Not to mention they're probably not cheap either.

Edit: Taken from here:  http://www.mouser.com/pdfdocs/ELNACalcuDYNACAPDischargeTime_Jul2011.pdf

Calculation of necessary Capacitance ①For constant current discharge C　＝　I × t ／ （V0－V1)

t : Discharge time (sec.) V0 : Charge voltage (V) I : Discharge current (A) C : Capacitance (F) V1 : Discharge voltage (V)

 

Let's say the wheel is only pulling 250 watts (0.25 kilowatts) @60V => I = P/U  =>  250W/60V = 4,1666...A, and the voltage drop allowed would be 5V (V0-V1) over a period of 1 second (I'm not taking into account the increased current when the voltage drops that's required to keep the output at 250W, although there's the equation for constant power output in the pdf too). The capacitance would need to be

4.1666A * 1s / 5V  = around 0.833.. farads.

More extreme case, wheel is pulling 1000W (1 kilowatt) @ 60V, I = 1000 / 60 = 16.6666...A, same voltage drop and time period as above:

16.6666A * 1s / 5V = 3,333... farads.

Acceptable capacitance should probably be somewhat larger due to higher current needed as the voltage drops to keep the power output around the same. I took a quick look at Mouser, filtering out only supercapacitors that can handle 75V or more. I was left with 7 choices (capacitances between 5.8F / 160V ... 94F / 75V), the cheapest (12F / 80V) was at around 350€ per piece (apparently minimum order is 100 pieces or more), and sized around 24x25x7.4cm, no weight mentioned. The rest could be bought one piece at a time, price between 950€ - 6000€ / piece. Sizes were something like 20x20x7cm up to something like 40x70x25cm. They don't list weights, but I'm thinking several kilos.

EDIT2: I'm not saying it's a stupid idea or stupid question, actually it would be really useful in case of a sudden (short-lived) power cut, but the sizes, weighs and prices make it pretty much impossible (for our imperial unit-people: the smallest capacitor was around 8" x 8" x 3" and probably weights something like at least half a dozen pounds).

 

Thanks for the calculations, I figured this would be the limiting factors. 1sec is actually quite long, what if you would use it to delay the shutdown with just 0.5sec or .25sec to give you some extra reaction time, increasing the probability to run off the wheel safely (not to push the limits of the wheel)?

[Simon Herbert]

HELP!   

I made the the mod as suggested to the mosfets for the battery that looked most like mine. I testing it by turning the device on and all was ok, I then testing charging it from a partial discharge and again it charged up nicely. 

As i have been ejected from my EU twice ( in cold conditions) I was keen to test it in a controlled way I.e. Whilst holding onto parallel bars whilst going back and forth for a while after leaving my EU outside in 5C temp for an hour. Again all seemed ok for a while until I got a beep and then total cut out after a few minutes into the test. Initially I could turn the unit back on ok but after a couple of times of this happening the unit would just beep with all four  the battery indicator battery lights flashing just once and then cutting off. 

Sadly when the unit is now plugged in to recharge the charger just show the green light and then when you try and start the EU it beeps and dies every time. 

I have tried installing a battery from a spare EU and then all is ok. 

So in shunting the mosfets I have killed the battery somehow.

I have tested the voltage of the connectors one is 59.9 and the other is 2.2. As a fully charged battery that I tried without modification measured 67.2 and near 0v. So I'm guessing something has gone wrong with the BMS During testing.

Is this fixable or should I just recycle the battery?

the pictures show the battery label and mosfets before modification but for the record I shunted the two middle ones.

[Chriull]

Just a short comment here, since i helped some (mostly TG3 owners) to apply the BMS shunting. I assume one of the main reasons for the shut-offs were just due to bad quality cells (maybe in combination with bad designed BMS boards). Shunting is just removing the symptom and not the cause - the battery cells still are bad quality, and bad quality cells can get really dangerous - especially without proper safety measures applied (well designed BMS, regular manual checks etc.)

So imho the best way to stay happy, healthy and enjoy the wheels is to get high quality battery packs!

Last but not least: As it seems (or better i got the feeling here over the time), unfortionately many shunters invest no time and are not anyhow interested in what they are doing or have any real experience in doing this. Imho neither they have any clue what could happen if one makes a wrong connection... ;( Thats also the reason why i quite stopped supporting the new posters in this thread - it is already everything clearily and in every detail written here whats necessary to shunt a BMS. So everyone who is experienced enough has no problem to find the information here, and everyone who is not should stick imho to the advice "Don't try this at home"

ps.: I hope no-one feels offended by my last paragraph - i do not know how to write it in a more polite way...

[Chriull]

I feel this should be regarded by any shunter - not only ninebot users:

Imho dead bridges producing a short circuit were reported also with other models - at least once i think...

Just from the "small" (regarding battery) EUCs like the Tg3 and nonames i cannot remember reading anything like this - maybe there is not enough power available to fry the mosfets in the bridge?

[Rotator]

14 hours ago, Chriull said:

I feel this should be regarded by any shunter - not only ninebot users:

I agree, but I would add

"... or if shunted, it must be done with a fuse instead a thick wire."

[joelren2k]

18 hours ago, Rotator said:

I agree, but I would add

"... or if shunted, it must be done with a fuse instead a thick wire."

How to use the fuse?

 

[Chriull]

41 minutes ago, joelren2k said:

How to use the fuse?

 

From the fotos of the different BMS boards i could not find a nice spot to implement the fuse directly at the board.

The best way should be to use an appropriate fuse holder and insert it into one of the discharge cables from the BMS to the Mainboard.

 

[EUC Extreme]

I have a problem with RockWheel.
Strenuous exercise, it turns itself off, and will not start again unless more quickly use the charger in place.
A chore.
I do not have enough information about BMS that I could say whether something can be done.
While the battery is full, the shutdown occurs.
Apparently, the voltage drops so much that the shutdown occurs.

I have this little off-road driving, but now I tried.
Driving is impossible Apparently my driving this is too much

 

[HEC]

Do you have voltmeter to measure the batter pack voltage? Per you description it looks like one (or more) of the cells might have died. How old is the battery pack? Can you measure voltage on the individual cells or at least groups of them so see any differences?

[EUC Extreme]

All battery cells are good, and exactly the same voltage.
So I do not think that the individual cells is defective. Of course, the battery is already a few years old, but not very much used.

[Chriull]

23 hours ago, EUC Extreme said:

I have a problem with RockWheel.
Strenuous exercise, it turns itself off, and will not start again unless more quickly use the charger in place.
A chore.
I do not have enough information about BMS that I could say whether something can be done.
While the battery is full, the shutdown occurs.
Apparently, the voltage drops so much that the shutdown occurs.

I have this little off-road driving, but now I tried.
Driving is impossible Apparently my driving this is too much

 

 

13 hours ago, EUC Extreme said:

All battery cells are good, and exactly the same voltage.
So I do not think that the individual cells is defective. Of course, the battery is already a few years old, but not very much used.

So if your batteries are all "ok" (same voltage) you could be succesfull with shunting the overcurrent side. And i assume you are aware about the possible risks.

so if you send detailed pictures from the lower half of the BMS ( from the left side with the power cables up to the right side with the mosfets) i could help you identifying possible shunt points. Identifying the mosfets would be nice too (seems that there is someting written on them) - but it should be just a little risk assuming standard pins...

[esaj]

1 hour ago, Chriull said:

 

So if your batteries are all "ok" (same voltage) you could be succesfull with shunting the overcurrent side. And i assume you are aware about the possible risks.

so if you send detailed pictures from the lower half of the BMS ( from the left side with the power cables up to the right side with the mosfets) i could help you identifying possible shunt points.

I was going to write something earlier about this, but then I couldn't figure which mosfets are for discharge and which are for charge side  Looking at the picture showing the two boards, it would seem that the two left-side mosfets' sources connect to the "C-", but not sure...

Quote

Identifying the mosfets would be nice too (seems that there is someting written on them) - but it should be just a little risk assuming standard pins...

IRFB3607, typical "G D S" -pin order (when watching from the text-side with legs down):  http://www.infineon.com/dgdl/irfs3607pbf.pdf?fileId=5546d462533600a401535636a42b2176

[Chriull]

28 minutes ago, esaj said:

I was going to write something earlier about this, but then I couldn't figure which mosfets are for discharge and which are for charge side  Looking at the picture showing the two boards, it would seem that the two left-side mosfets' sources connect to the "C-", but not sure...

I would also assume that the two left mosfets are the "overvoltage" and the two right side mosfets are the overcurrent. But since the ?drain? is just connected by the heatsink the "normal" shunt will not work easily so one should identify "B-" and "P-" for shunting...

Quote

IRFB3607, typical "G D S" -pin order (when watching from the text-side with legs down):  http://www.infineon.com/dgdl/irfs3607pbf.pdf?fileId=5546d462533600a401535636a42b2176

I just realized, that the pictures are already detailed and should show about everything essential (one just need to click on them ).

Imho the shunt should be (shorting drain-source of the overdischarge protection Mosfets) between the heatsink and the source pin. Could be easily made as shown by the "faint" red line in the following picture:

http://imgur.com/n5SP3ho

 

[EUC Extreme]

OK. Thanks everybody.  I do so both batteries.