Jump to content

Electric Unicycle's BMS problem and solution


hobby16

Recommended Posts

UPDATE on MoHoo:

The SHUNT of P- to B- seems to have worked.  I have ridden the MoHoo to only ONE bar of the FIVE illuminated and even after power down and power up the ONE bar is all that shows.  The riding seems strong with no studded or slow downs.  I am sort of afraid to ride the unit until no bars show.  Does anyone know if the MoHoo or the Huanxi EUC will give a message or sound when the motherboard decides you have reached the bottom of the riding range because BMS voltage is too low?  I think one of the YOUTUBE videos said that at low battery limit the MoHoo starts lifting up the petals and forces you to dismount.

When I finish riding MoHoo each time, I immediately disassemble battery pack and measure each cell voltage level.  I am now down the 3.57 for each cell.  Never seen more than 0.01 volts different between cells when doing the immediate voltage measurement.  But the balancing electronics is always trying to make them all get to the same voltage.  I am assuming that once I get to 3.00 volts per cell that the motherboard should "shut me down".  Is this correct?   Do they go lower than that? 

In two weeks my new BMS pack with Samsung good cells in it (2600maH per cell) should be here.  I will then put the two systems in parallel and ride the system until it stops.  Only a single quick release with two wires will be needed (will not parallel the charging leads).  Hope that cures all my problems.  Oh, my attempts to get circuit diagram from Samsung for this battery pack fell on deaf ears.  They said too many law suites are now in process, and they don't want any additional information floating in the public forum.

   tjcooper

Link to comment
Share on other sites

  • Replies 306
  • Created
  • Last Reply

@tjcooper, It is difficult to answer any of your above questions without intimate knowledge of both the mainboard firmware and BMS, which for that EUC I doubt anyone here has.

The actual voltage at which the mainboard, if it indeed does have the function, will tilt back will vary considerably. Do not forget it will be under load and not a static voltage as well. It could be anywhere between 2.5 and 3.4 volts per cell. Early KingSong's alarmed at 55 volts which is very nearly flat at rest but way way too high under load, I think, thanks to Jason McNeil, that is now down to a sensible 48V. It is possible that @electric_vehicle_lover May have a better idea of the low voltage behaviour of generic boards as he has studied them in great detail?

For the first 3 or 4 charges backs should be treated gently to help them condition so I really wouldn't recommend taking the new pack right down to cut off straight away. It is best not to take them that low at all if you can help it, it always ages the pack faster and ideally aiming for a pack with a range that is at least twice what you actually need so you don't often go below 50% makes for a better riding experience as well as less stressed packs.

Link to comment
Share on other sites

@Keith,

just got the answer to most of my questions. 

It took 5 days of riding after work to send the battery down to shutoff.  When the MoHoo maincircuit board reaches 3.00 volts DC per cell under load, its loud speaker says "under voltage" in some of the worst "chinglish" I have ever heard.  Then within 2 seconds it starts lifting up the petals so that you have to stop.  The first time it happened I did not know what the speaker was trying to say.  I still had "one bar" on the dial so I thought I could get back on and go really slowly (<2 mph) until either that bar went out or the message repeated itself.  I did not get 100 meters before the message repeated itself and the petals went up again.  I immediately opened the battery and measured every cell at 3.51 volts DC.  Since a 1C load on a LiPoly battery reduces the voltage at its end range by typically 0.5 volts DC, I surmised that the cut-off voltage must have been 3.00 voltsDC.   I will put an inductive DC current meter on the last cell connection in the battery pack.  I have recharged the system (now each cell is at 4.05 volts DC [I though it should have gone to 4.20 volts DC] but that is not what is happening after 18 hours of charge}.  I will again drive the pack until the mainboard gives me the "under voltage" message and will capture the final "under load voltage".  But I assume it will be very close to my 3.00 voltage I have imputed.

GOOD NEWS:  no face plants.  The P- to B- connection has stopped the shut offs and the whole system seems to work as designed.  I now feel that if I add the second battery pack in parallel I will have a good and safe reliable machine.   It has taken over 9 months to start learning the EUC, feel the pain of many face plants,  tear apart the system,  seek help in getting a SHUNT to work,  testing the result.  With the great help from the EUC group, I have finally over come the problems and can feel relatively safe in riding my first machine.   MORAL OF STORY:  buy the good machine (Ninebot in this case) first and spare yourself the 9 months of pain and uncertainty.  Again, thanks.

    tjcooper

Link to comment
Share on other sites

  • 3 weeks later...

I have just modified MoHoo to have the B+ line come outside the unit.  I put on an accurate DC ampmeter with clamp-on.  When riding the MoHoo at around 3-5 mph, I read typical of 1.3 amps DC and a max of 2.0 amps.

This is a 16S system with 132 W-hr battery and a 350 watt motor.  Does that level of current seem reasonable?  The typical power is from 70 to 100 watts.  Seems like current draw is very low (still have 4 out of 5 bars on battery).  Kind of neat to see the AMPS while you ride but mounting of clamp-on meter is rather crude just using transparent packing tape up near the handle.

 

QUESTION: today I did a search on "tjcooper" and I only got 3 returns.  I must have 60 posts that I have done this year.  Has anyone else had their posts go missing?  Actually the posts are still there, just the search does not find them like they used to.

    tjcooper

Link to comment
Share on other sites

  • 1 month later...

Hello Everyone,

1st post on this forum, I hope I post in the right topic.
I believe my problem is related to what is described in this thread but not sure. Feel free to move this post  anywhere else if more relevant.

New French wheeler since July 2016, I now feel comfortable for more or less all rides but cannot perform any tricks yet (reverse, one leg on more than 2-3 meters etc...). I own a Kingsong 14C 680 Wh, bought new, weight 95kg with equipment and I face the following issue :

I live on a small hill (2-3% slope nothing extraordinary). Wherever I need to go when I start from home i have to start by going down. When the battery is fully charged, it simply does not brake if i go over a certain speed (estimated at 10 km/h) going down. The wheel beeps, and is falling backward. In summer it happened also but at higher speed, in winter I cannot even start riding more than 5-7 km/h because the engine cannot cope with the brake power needs to balance the wheel and my weight. Even on a flat surface, if i start full battery charge, go to 25km/h and needs to make an emergency brake, the wheel do not follow and is leaning backward to put me on the ground. Happened to me yesterday - it never happened in Summer.

When the battery is partly discharged, I can go downhill at 20km/h or even higher and brake without any issue.

The technical service from KINGSONG in France let me know that this was linked to the fact that when going down the baterry was charging a little, since it was full, it cannot accept any "surcharge" and to protect battery, it cuts off. Really ? 

I made several trials : the tire pressure has an influence (the lower pressure, the worst is the behavior), the weight of the driver is obviously critical (my son has not this issue), but temperature seems to have a great influence. In winter problems occurs much more than in Summer

I am a complete noob when it come to technical discussions and knowledge around electrical and electronic stuff but the discussion and reading this topic make me think that my problem is in a way similar to the one describe here. Can anyone confirm it ?

Has anyone experienced this behavior ? Can someoone try to explain what is happening if my explanations are not correct ? Except unmounting the wheel and do some "hardware" stuff which I am absolutely uncapable doing, would you have any recommendation or easy fix that i could potentially use ?

I do not expect any miracles but if I can put an understanding on my issue and get some further info if not ths start of a solution I would be already happy.

Thanks a lot in advance

Link to comment
Share on other sites

@djerr.  I'm no expert and I'm newer than you to EUC but I spent some time reading about this yesterday and I believe your dealer is telling the truth. See recent thread by fat unicycles.  the Post  about math problem. Your answer is in that thread.  I.e go uphill first for a certain distance to free up some battery capacity, or stop charging battery at 80 or 90 % by either setting timer, pulling plug manually or get a charge doctor from member of this forum, also in France I think.  I'm getting one as soon as I can figure out where to have it sent.  I'm renting a place in Spain with a very small mail box.

 

I don't think your problem is related to this BMS thread, but you weren't to know, otherwise you wouldn't have needed to post the question.

Link to comment
Share on other sites

@djerr: Welcome to the forum!

you could "solve" your prob by shunting the BMS - but for your case i would definitely _not_ recommend this! Li ion cells are quite sensitive to overcharge!

the more or less only solution to your prob is to don't drive down a decline, when the wheel is fully charged - you'll have to try, but already quit "small" discharge before going down could help (going up the hill before going down).

you also could adjust your charger to a bit lower max voltage (there should be 3 trim potentiometers inside - one for max current, one for max voltage and one for "idle" current). Could be that your chargers is adjusted to a bit too high max voltage already? You have th possibility to measure the no load output voltage of your charger?

or you use @hobby16's charge doctor - this can be set up to partly charge your wheel.

Link to comment
Share on other sites

@djerr, this is a well known problem on almost all, probably absolutely all, EUC's and certainly is one of the very few failure mechanisms on the KingSong (indeed, in a year of riding my KS-14C, it is the only problem I have ever had).

1 hour ago, Chriull said:

or you use @hobby16's charge doctor - this can be set up to partly charge your wheel.

@Chriull is correct, the best thing is to less than fully charge by using charge doctor, I now do that and have never had another failure. 

Basically, EUC's have only one possible way of braking and that is regenerative braking (like KERS on F1 cars) to slow you down they have to put charge back into the battery, if it is full the cannot do that and, BTW I do not think any modification of the BMS would improve that situation. 

The charge doctor will allow you to stop charging at a fixed point below fully charged, not only leaving you room for braking power but also improving the lifespan of the batteries, so if you don't need to be travelling the maximum distance a full battery gives you it is a win win situation and the charge doctor works very well.

However, battery cell balancing only occurs once the battery is full, so every (say) 10 charges the battery should be topped right up and left on for an hour or so after the green light comes on to ensure the pack is properly balanced. Obviously pick a day when you won't be starting down hill for the occasional full charge ?

Link to comment
Share on other sites

@Chriull

I got the new battery from DHgate.com out of China.  Good Samsung cells and about 155Whr.

The price was like $78 delivered to USA (16S Samsung cells at 2600maH).  But the delivery had to come through the Netherlands from China to get them ground delivered to California.  Go figure the logic in that shipment.

That battery has exactly the same Deans and Xt-60 connectors that the original MoHoo battery had.  I have now ridden with the new pack until only 1 bar of 5 shows.  No cut off and shut downs.  This battery pack seems to work well without needed a SHUNT.  But since I am going to give the unit to my son, I want to SHUNT this new unit also.  When I opened the pack, it is different from the old Samsung pack,  It has C+ and C- going to Deans charge line, and the B+ and B- going to XT-60 power plug.   There are P+ and P- designations on the PCB but no wire connections.

So for the old pack I shunted P- to B-.   So for the new pack I should shunt C- to B- ????

Is there some simple multimeter test I can run to be sure this is the right hook-up?  I will start my shunt with a 1 amp fuse in line with the wire.  If that does not blow the fuse then I remove the fuse from the circuit and just short C- to B-.  After that it is field test time.  Any comments would be appreciated.

I can include pictures of the new PCB but I have run out of image space on this forum.  Maybe have to put them on YouTube.

     tjcooper

Link to comment
Share on other sites

10 hours ago, tjcooper said:

...  Any comments would be appreciated.

As you know, the shunt is a short circuit of the drain and source (either including the current sensoring resistor or not) of the "overdischarge protection" mosfets. So without knowing your specific BMS or any possibility to compare the PCB with @hobby16's schematics of the protection part of a BMS i could not think of any reasonable comments. Maybe just that P+/-, C+/-, B+/- is by no way any standardized nomenclature - also maybe it evolved to some kind of quasi standard in the BMS world? Imho they are just a perfect startpoint to check the PCB and verify the right points to short circuit.

Either way i would not thrust my physical integrity or my flat/house on if and how someone used this labels...

You could compare your BMS with the different pictures of shunts of BMS's in this thread or on @hobby16's homepage - should be quite a reasonable change to find a succesfully shunted BMS correponding to yours.

10 hours ago, tjcooper said:

I can include pictures of the new PCB but I have run out of image space on this forum.  Maybe have to put them on YouTube.

Space on this forum is very limited. But one can easily link here pics uploaded to i.e. Imgur by pasting the link here.

Link to comment
Share on other sites

13 hours ago, tjcooper said:

 When I opened the pack, it is different from the old Samsung pack,  It has C+ and C- going to Deans charge line, and the B+ and B- going to XT-60 power plug.   There are P+ and P- designations on the PCB but no wire connections.

 

hmm, this probably means that the battery (B+ & B-) supplies directly the wheel, without passing through the BMS protection circuit (P+ & +-), meaning that the pack is in fact already shunted. The BMS' protection circuitry is just there for window dressing (and wired-on only for regulatory purpose when needed). So you do NOT need to shunt.

BTW, Tishawn, don't you have a Ninebot One ? If yes, since it is powered with 15S packs, putting a 16S  battery means you made an overdrive mod. Wouldn't that be nice ? :D

 

Link to comment
Share on other sites

  • 7 months later...

After alot of time researching and learning about BMS systems I can not understand why anyone would want a BMS. Ultimately they seem to cause more problems than they fix. Could this BMS simply be a way to save production cost on chargers? After I changed my batterypack I don't really know if my BMS is still working properly and part of the circuitboard is drenched in some thick tar-like substance that prevents me from checking the circuit with a multimeter. Why not get rid of the onboard BMS and make a balancing charger? An arduino hooked up to a recycled powersupply from a few old cellphone chargers and scrap components should make that possible for a low price. Would this cause any problems`?

Link to comment
Share on other sites

53 minutes ago, Cryptonitor said:

Why not get rid of the onboard BMS and make a balancing charger? An arduino hooked up to a recycled powersupply from a few old cellphone chargers and scrap components should make that possible for a low price. Would this cause any problems`?

Well having to plug in a plug with at least 17 connectors on it (21 for the 20 cell wheels) capable of taking the individual cell charge currents or plus two more thick wires for charge and 17/21 wires capable of taking balance currents might be a bit of a problem. You also lose all the protection capability within the pack, admittedly most of that is better off in the wheels main controller but maybe there are legal issues with selling a consumer device of this power without electronic safeguards?

If a BMS works correctly, I.e. is designed for a self balancing device and reports individual cell problems to the controller, instead of taking action itself then surely a BMS is a much better option than bringing all the balance leads external.

I.e. Surely the answer is not do away with the BMS, but to implement it properly and safely for both the rider and the battery.

Could this BMS simply be a way to save production cost on chargers?

One BMS in a charger or one in every battery pack, how would having the BMS in the battery pack instead of the charger save money other, of course, in the cost of all the extra wires and connectors coming out of the wheel.

Link to comment
Share on other sites

  • 3 weeks later...

Hi all,

 

I'm a proud owner of a Gotway Msuper V3S 1300Wh 84v, and i decided to create an extra battery pack to have some extra range... (should had bought the 1600wh version).

Since i never did anything similar i have many doubts and i'm spending lots of time reading about everything. Already build the battery pack 20S2P, and now i'm just waiting for the BMS... generic Li Ion bms 20S 50A... now i read here that i should shunt the B- with the P-, is this information correct?

Thank you for your guidance on this topic.

Link to comment
Share on other sites

23 minutes ago, Noillek said:

Hi all,

 

I'm a proud owner of a Gotway Msuper V3S 1300Wh 84v, and i decided to create an extra battery pack to have some extra range... (should had bought the 1600wh version).

Since i never did anything similar i have many doubts and i'm spending lots of time reading about everything. Already build the battery pack 20S2P, and now i'm just waiting for the BMS... generic Li Ion bms 20S 50A... now i read here that i should shunt the B- with the P-, is this information correct?

Thank you for your guidance on this topic.

Depends on your BMS but generally this should be for about 99,9% of the BMS's correct.

Link to comment
Share on other sites

39 minutes ago, Noillek said:

Hi all,

 

I'm a proud owner of a Gotway Msuper V3S 1300Wh 84v, and i decided to create an extra battery pack to have some extra range... (should had bought the 1600wh version).

Since i never did anything similar i have many doubts and i'm spending lots of time reading about everything. Already build the battery pack 20S2P, and now i'm just waiting for the BMS... generic Li Ion bms 20S 50A... now i read here that i should shunt the B- with the P-, is this information correct?

Thank you for your guidance on this topic.

As you use a 10Amp continuous cell and that in a 2 Parallel config....even with some Peaks there should never run more than 30Amps throught the complete BMS/package.

So i would see no Need for a shunt on a 50Amp BMS.

Link to comment
Share on other sites

8 minutes ago, KingSong69 said:

As you use a 10Amp continuous cell and that in a 2 Parallel config....even with some Peaks there should never run more than 30Amps throught the complete BMS/package.

So i would see no Need for a shunt on a 50Amp BMS.

50 minutes ago, Noillek said:

Hi all,

 

I'm a proud owner of a Gotway Msuper V3S 1300Wh 84v, and i decided to create an extra battery pack to have some extra range... (should had bought the 1600wh version).

Since i never did anything similar i have many doubts and i'm spending lots of time reading about everything. Already build the battery pack 20S2P, and now i'm just waiting for the BMS... generic Li Ion bms 20S 50A... now i read here that i should shunt the B- with the P-, is this information correct?

Thank you for your guidance on this topic.

Do you use this pack as sole pack or in parallel with the original packs?

if in parallel then you should shunt it or the current will flow over the charge protection circuits and charge wires if this bms activates the overcurrent protection .

Link to comment
Share on other sites

21 minutes ago, Chriull said:

Do you use this pack as sole pack or in parallel with the original packs?

if in parallel then you should shunt it or the current will flow over the charge protection circuits and charge wires if this bms activates the overcurrent protection .

As i have recommended to him on FB i would do it in "daisy chain" Position between chargeport and charging of the original packs (because, as the GW V3s+ original packs have balancing/communication between the packs BMS's which for sure you can not implement easy in a 3rd Party BMS)

Can you explain the sentence "will flow over the charge protection circuits and charge wires"???

I can't follow...

 

Otherwise: if this kit is only able to handle 3,5 Amp....i would not do it at all :-)

Link to comment
Share on other sites

28 minutes ago, KingSong69 said:

 

Can you explain the sentence "will flow over the charge protection circuits and charge wires"???

I can't follow...

If the packs were in parallel ( discharge wires connected and charge wires connected) and one pack engages the overcurrent protection the charge side is still connected and (without reverse voltage protection diodes) the current will flow over the charge wires and the other pack to the mb.

28 minutes ago, KingSong69 said:

 

Otherwise: if this kit is only able to handle 3,5 Amp....i would not do it at all :-)

Me neither :ph34r:

Link to comment
Share on other sites

Totally, agree with @KingSong69, there cannot be any need to muck about with the BMS on an additional add on parallel pack.

For a start off, there is a risk of accidentally shorting it if it is external to the wheel so the BMS protection circuitry is needed.

Secondly the main pack will not, or at least should not shut down so if the secondary packs BMS goes into protect mode it will not result in the main pack shutting down as well.

Thirdly, the current will be shared so the total current being supplied by the additional pack will be lower than if it was on its own so the BMS protecting the pack shouldn't happen except in a low voltage situation, which an additional pack will also make less likely to happen

A diode will prevent flow back down the charge wires between packs if the BMS should switch on output protection.

@Noillek, above all else, do not lose sight of the fact that parallel packs  must always be at exactly the same state of charge (same voltage) when connected or huge, damaging currents will flow between them.

Link to comment
Share on other sites

33 minutes ago, KingSong69 said:

As i have recommended to him on FB i would do it in "daisy chain" Position between chargeport and charging of the original packs (because, as the GW V3s+ original packs have balancing/communication between the packs BMS's which for sure you can not implement easy in a 3rd Party BMS)

Good Idea!

just two points to regard 

- the original packs should not have a reverse voltage protection diode (or circuit)

- the charge protection circuits of the original packs have to deal now with the discharge current of the new pack (each with about half of it)!

Link to comment
Share on other sites

2 hours ago, Keith said:

Totally, agree with @KingSong69, there cannot be any need to muck about with the BMS on an additional add on parallel pack.

For a start off, there is a risk of accidentally shorting it if it is external to the wheel so the BMS protection circuitry is needed.

Secondly the main pack will not, or at least should not shut down so if the secondary packs BMS goes into protect mode it will not result in the main pack shutting down as well.

Thirdly, the current will be shared so the total current being supplied by the additional pack will be lower than if it was on its own so the BMS protecting the pack shouldn't happen except in a low voltage situation, which an additional pack will also make less likely to happen

A diode will prevent flow back down the charge wires between packs if the BMS should switch on output protection.

@Noillek, above all else, do not lose sight of the fact that parallel packs  must always be at exactly the same state of charge (same voltage) when connected or huge, damaging currents will flow between them.

Thank you, I'm planning to connect only... and leave it connected at 100% charge, so i guess that the voltage will match.

 

3 hours ago, KingSong69 said:

As i have recommended to him on FB i would do it in "daisy chain" Position between chargeport and charging of the original packs (because, as the GW V3s+ original packs have balancing/communication between the packs BMS's which for sure you can not implement easy in a 3rd Party BMS)

Can you explain the sentence "will flow over the charge protection circuits and charge wires"???

I can't follow...

 

Otherwise: if this kit is only able to handle 3,5 Amp....i would not do it at all :-)

Thank you KingSong69, do you by any chance have a draft of the connections? I was planning to just connect to the original charge cables and the with a (Y) XT60 connector plug it to the PCB (control board)...does it make any sense? (for some reason i see it very simple... might be missing some steps here.

Thanks once again.

Link to comment
Share on other sites

2 minutes ago, Noillek said:

Thank you, I'm planning to connect only... and leave it connected at 100% charge, so i guess that the voltage will match.

 

Thank you KingSong69, do you by any chance have a draft of the connections? I was planning to just connect to the original charge cables and the with a (Y) XT60 connector plug it to the PCB (control board)...does it make any sense? (for some reason i see it very simple... might be missing some steps here.

Thanks once again.

"Daisy Chaining" is as follows:

The Chargeport wires goes to the charging wires of your Pack. Then the power Output from your pack goes to the Charge Input from ALL internal packs.

So it is as easiest when you use the Connection/wires directly after the chargeport. No Y-cable needed!

Like said you should only establish this Connection, when voltage of your pack and the voltage of internal packs is the same!

Link to comment
Share on other sites

5 hours ago, KingSong69 said:

"Daisy Chaining" is as follows:

The Chargeport wires goes to the charging wires of your Pack. Then the power Output from your pack goes to the Charge Input from ALL internal packs.

So it is as easiest when you use the Connection/wires directly after the chargeport. No Y-cable needed!

Like said you should only establish this Connection, when voltage of your pack and the voltage of internal packs is the same!

I did a drawing of what my plan was... but just found out that the GX16 4pin can only handle 5 amps... drawing is still valid but need to change the connectors to XT30 for charging and XT60 for power output to the pcb.

File_000.thumb.jpeg.80ec89df818c1d945e4bb79e8382b72a.jpeg

i still want to remove the battery pack to upgrade and or maintenance.

Thanks

Link to comment
Share on other sites

  • esaj unpinned this topic
  • esaj unfeatured this topic

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...