Basics for Newbie///

[bobbill]

I have built three nice bicycles w. gas engines and decided to try elctrice for light and quiet, though spendy, wires and picky.
I built small, 36v 18650 10S3P pack... with BMS, very good batteries, all same voltage.

I had some problems, which I ironed most out myself and hoped I was home free, save the connectors--

Am not up on this stuff like gas motors.

Motor is front-hub geared Bafang kit from Grin. (I have weak knees and live on a hill.)

Long and short of it" I Finally charged battery pack, I think, as light went our on the 36V Luna charger. Went for a cruise around house...bike ran decent for about a mile, then motor stopped, which was expected...sort of a test ride.
Light on controller went out etc, lke connector separated etc. Just quit.

I now rechecked each cell for charge and seem to be all same, but now pack will not charge. (I am a bit anal and always follow instructions...

I checked serches here and found little on troubleshooting, so here I am. "Newbie Blank..."

My question is simple: Is there a check list to run to find out what the problem is? Just point me or chew on me.

One way or another, this rig is going to work...crazy thing, is frustrating.

Thanks in advance.

 

[dogman dan]

First thing to look at is the charger to battery plug connections.

I'd start looking at the connections between cells on your battery. if the cells have enough voltage for the charger to start, but they have a break somewhere, the charger thinks its not safe to start charging.

Pretty clear this is a battery problem IMO.

 

[bobbill]

First, Dogman thanks for reply.
I have checked wires and all register as having continuity...

Checked cell volts, all are 3.6(8cells) to 3.86(24 cells), rechecked wire to Grin meter etc.

Could be a connector, so decided to (maybe replace most connectors) But, must admit they are Andersons and have been flaky.

Outside of that am lost, unless BMS just shuts off as unbalanced.

No charger will not charge. Will recheck charge wires/connector again.

Maddening.

Again, thanks, hope springs eternal here.
Damn thing is mechancal, will make it run, I will.
Such a simple thing...and so much grief.

 

[amberwolf]

If they vary by that much voltage (0.2v) and are only in the 3.6-3.8v range, they are severely imbalanced, and at best half charged (assuming they are not LiFePO4; you dont' say what chemistry).

So your battery pack needs to be balanced, and it needs to be fully charged.


If the pack was built out of non-matched cells, especially if they are not from a known-good source (many places sell junk as if it was great), it's likely to get imbalanced a lot, and will have to be rebalanced every time you charge it.


There are many many threads about battery problems and battery builds that have troubleshooting information on this type of problem, most of them in the battery technology subforum.


There are also a number of threads about connectors, which have posts in them describing the problems people create for themselves with Andersons by misusing them, and how to not do that.

 

[bobbill]

Amber... thanks...it must be the cells (lithium cells, sorry - 18650s) as I checked every wire and recheck each battery.

I did aearch here for "connectors" and drew a blank, but will try searches using different words. KInd of logical othesr have ran into similar conundrums. Some heavy duty stull going on on these threads, I must say.

I was wondering if the BMS was bad also...but will try a second set of batteries...have another set...which tests each cell to 3.51 each...

Can I balance then charge this pack? - I mean it was fine for a short tiem then just stopped. Is that what happens when some cells are out of alance? (I am clueless.)

if one or two are off much perhaps I can add one whigh tests close?

Stuff is maddening, I must say, very picky.

Thanks again.

 

[bobbill]

I just remeasured all my 30Q cells, were not used, and all but one measure 3.44v and the outliar was 3.42.
The cells in the active pack are Panasonic, 3300s and measure as indicated earlier. (Italics now.)

All of which which brings up a question, if one cell can bring down the whole pack, isn't that wasteful, which is not the point of going electric?

If one or two cells are off, why not bring to up to par and have all about the same? No disrespect, just my logic.

 

[bobbill]

I just remeasured all my 30Q cells, were not used, and all but one measure 3.44v and the outliar was 3.42.
The cells in the active pack are Panasonic, 3300s and measure as indicated earlier.

Which brings up a question, if one cell can bring dowen the whole pack, isn't that wasteful, which is not the point?

If one or two cells are off, why not bring to up to par and have all about the same? No disrespect, just my logic.

Dogman, I recheck wires with connectors and all have continuity but one (Anderson) and will replace it in hour or so.

Thing is, this one connector/connection has nothing to do with the charge...and the charger is a Luna time and puts out 42volts.

Shoot me!

 

[AngryBob]

Charge the battery to full. Check voltage on outputs. Should be 42V. Wait one hour, test, wait one day, test. Should STAY 42V. Maybe 41.9. Ride one-quarter mile, test, check each quarter-mile till shutoff. If possible, check individual cells or parallel strings at the BMS each test.

BMS, will have a cutoff voltage for each parallel string, controller for whole battery. Varies per unit and usually not changeable. Also, poor batteries will drop voltage rapidly, as yours certainly appeared to do, will trigger a cutoff, and then very quickly rebound back up after load is removed (shutoff). 3.6 not usually low enough to trigger an LVC, but if they dropped that low in one mile they almost certainly dropped a lot lower before rebounding.

Knowing the source of the cells might provide some useful diagnostic information, also the connection method used, as it is highly probable there is a serious problem here. Either worthless cells, horrible connection method, or defective charger.

NWIH decent batteries will drop from 4.2 down to 3.6 in one single mile. You built the pack, and your charger is from Luna. Those are the two places you need to look the hardest.

 

[bobbill]

Bob, Thanks, I think I understand.
Will replace batteries with those noted and give it another go.
Much thanks Re BMS also.

 

[amberwolf]

Can I balance then charge this pack?

If all your cells are good and safe, and your pack wiring is good, and your BMS is working and is the type that can balance cells, then you can leave the pack on the charger continously, as long as it is not in use, and the BMS will balance the cells for you over time. This may take hours, days, or weeks, depending on how badly imbalanced the pack is (how different the state of charge is between cells), and how much balancing current the BMS shunts handle. (usually not very much, so it usually takes a long time).

But I would recommend only doing this while you are physically present at the pack, or if you put it in a fire-safe location outside away from buildings, etc. (because any pack with a problem is a potential fire hazard, especially when you don't know what the problem is or why it happened).

There are other ways to do this faster, if you read some of the other battery troubleshooting threads, most of which are easily found in teh Battery Technology (and ebike technical/general) sections by glancing thru the titles of the threads, describing a battery with a problem of some type.

I mean it was fine for a short tiem then just stopped. Is that what happens when some cells are out of alance? (I am clueless.)

It can. The BMS is there to protect the cells so you dont have a fire. So it is designed to prevent:
discharge when a cell drops below it's LVC, whatever that is for that BMS.
charge when a cell drops far enough below that LVC that it could be damaged.
charge when a cell rises above it's HVC, wahtever taht is for that BMS.

There's lots of threads about people's battery problems and battery builds which will show you how this all works, and how to troubleshoot the problem(s) you are having. It's a lot of reading...but you need to do that anyway:

I would highly recommend at least reading some of the successful build threads, and some of the failed build threads, as well as the do's and don't's threads, and FAQ and Sticky Index threads, over in the battery technology subforum, before continuing, because one of the most dangerous things you can do is build your own batteries without knowing all the potential problems, causes, fixes, etc. You can burn your house or apartment building down, very easily.

 

[bobbill]

Amber...thanks.

Charger (Luna 4amp) seems work...red and green are on and no fan, so not charging.

Before, one light and and fan working to point where both lights were same (green) and fan quit. It is set to 100% for first three charges, and now appears BMS prohibits charge.

Maybe I should switch BMS boards? See what happens.

BTW, FWIW, have perused much here and other sites, re builds just have not encountered same situreation. Seems like all are bigger and run well, save a connection change or two.

Oh yes, redid a past search and got 25K responses..must have had typol....will try some more.

Your responses add to knowledge bank.
Thing is, the processs is frustrating as it seems I get no where wtih lots of research and testing, so must be missing something.

The 18650s in pack now are NCRs (2900mA) and when I last measured, were all same voltage. Now some are low, some high, as indicated. All very confusing.

I also have 30 18650s, been sitting for a bit, Samsung 3000mA, and spare BMS I could plug into current pack to see what happens but leery.

Thanks for assistance. Will persist searching builds.

Ichecked notificaations, but do not get, so if replying I may be tardy reader.

 

[bobbill]

I was about to begin to add a different BMS and wondered.

How does a BMS alter volts from similar batteries with same volts while running, and then quit?

Or, will BMS with different battery set charge or will BMS not charge, because it used old data?

 

[AngryBob]

Are these cells new, or heavily used?

If new, there MUST be some severe connection issues for them to get so majorly unbalanced so quickly.

BMS - what these SHOULD do, but do not always, and sometimes are defective, is, at the end of charge, bleed down fully-charged parallel strings in order to allow lower-voltage strings to "catch up".

Some chargers simply cut off when battery as a whole reaches full voltage, again what they should do is continue to output at a very low amperage level when reaching the tail end of the charge.

I personally have a near-zero confidence level in Luna cycles equipment. This is based on a LOT of research and reading, of which a truly amazing amount of previously posted information is no longer available.

You should post pictures of your pack construction, folks here can spot connection issues fairly quickly. Not using a Vruzend kit, by any chance? Main problem with these is inconsistent and variable cell connections due to the basic design and its implementation, both of which are seriously flawed. Even if you use the recommended assembly method of whacking it repeatedly with a large wooden mallet.

Also, you are making changes with no apparent system to determine exactly what is causing the problem. Information on cell ages and pack construction is important. Some of the probable causes and severe dangers of battery fires result from very, very similar conditions to what you seem to be experiencing, and resolving the underling causes should be priority #1.

Edit - re-read your BMS question, and, NO, it does not. All cells SHOULD draw down evenly, IF repeat IF they are carefully matched in both capacity and resistance. Only time a BMS varies input or output to parallel strings, is when charging. IF Cells at same charge level become different, especially in a short time, then either their connections vary dramatically, or their capacity and/or resistance is very unmatched. SFAIK output is never affected by BMS, unless there is some sort of severe defect, such as the aforementioned bleed-down is stuck on. That bleed-down is extremely minimal and would not result in the dramatic differences you have stated, though the precise time frame involved would need to be specified. Should take many hours or days for that level of variation.

 

[bobbill]

First. A thanks for the responses, my logins needed some help...and did not receive email notifications...'

Second. I replaced the cells and the BMS, and the connectors (Anderson 45A Pwr Poles). Will charge when I can watch or be around so quench heat, if a problem arises.

Angry Bob, it is a Vruzend (mistake) and used nickel-coated copper bus bars for series and for balance. (No steel and used magnet to be sure.

The rig or motor is Grin thing with V4 and the thing ran, until it did not.

All cells were new, and measured 3.52 or 3.51 in first group and were Panasonic 2200mAH, the second are Samsung 3000mA - All 3.44, save one is 3.42. All checked before and after packing.

Pack was 33.3...have not tried to charge yet.

Redid Anderson connectors...used other BMS I had for second 10S pack, figuring first one is locked-up.

Will try some way to reset later. )I like Power Poles but PIA to do. Have alternates covered plug types...but hate idea of clipping working connectors...)

Checked all balance wire connections (+ terminals, crimped to rings).

All wires and buss show continuity.

Have not tried to charge yet...

 

[AngryBob]

Your cells are equal and should be relatively well matched. The problem is your connection method. The connections are not all equal. They vary considerably. This is the nature of the Vruzend kit. Nearly everyone using anything over an extraordinarily low power draw with these are experiencing similar problems. It just does not work, ESPECIALLY in a vibration-prone environment. Your "parallel" groups are not truly parallel. They should all have exactly the same level of connection, and they don't.

30-40 cells, that means 60-80 seperate connections, and no two are identical. This is guaranteed to cause the type of variation you are seeing, which will cause lowered performance, safety issues which can become severe, and eventually destroy one or many more of your cells.

Either get a spotwelder or look into the NESE modules, ditch the Vruzend crap.

 

[bobbill]

I must admit, I found a not good Anderson to the BMS so maybe that is problem. I switched to new BMS.

Connections, as in the Anderson's yes, I agree, but maybe not the Vruzend crate thingies?

Bob, are you saying the pack needs (parallel) stability for balance drawdown? Seems to me, logically, should really not matter...I mean it ran, and road way is like glass. And I am new. Meaning of "SFAIK" is?

Besides, cannot get Ness items. And, I switched BMSs as would not even charge, even though charger output was 42v at 4 amps..,per the meter thingy I used.

Really do not want to go with welder. Costly and hard to remove bad cell, if necessary. Right? I already have the Vruzends (2 10s/early and late versions). Must be better way. And, I do thank you for your help...so I should consider AnnPower item or something from BatteryBloc with more solid basese or holders? Whatever product, would like to keep in 10S3P size as it fits and have container for it on bike.

 

[AngryBob]

Paralleled cells SHOULD draw down evenly. HOWEVER, if one of a 3p string has a good connection, and the other two do not, then most to all of the load lands on that one cell. Dropping it's voltage dramatically, possibly causing permanent damage as it is being forced to provide several times the amperage it is designed for.

Your bike SHUT DOWN in one mile. That means that either the BMS, or the controller, detected a voltage that was too low for safe operation. Controller only detects pack as a whole. BMS detects each individual parallel string, if two cells are at 4V and one (of 3p) is down to 2V, that averages to 3.3V which is below safety level. Note that the 2V cell has likely sagged way low due to high amp draw, when that draw is removed, will likely rebound to a significantly higher voltage, as in, by the time you measure it, back to 3V, which would average out to 3.66 for the three cells.

I am greatly simplifying the numbers to make a scenario which is understandable. Note that a connection which is poor could well read a correct voltage but not pass a significant amount of amperage.

The bad Anderson connector would not cause this situation. If really bad, no work at all, slightly bad will just heat up and draw more power, but not shut down in one mile.

A bad Anderson is unusual, a bad Vruzend connection, at least one, is a virtual guarantee for an entire pack, as one for each end are all of the same very poor connection method.

If you are not familiar with the potential explosive fire risk you are almost certainly looking at, do some googling for 18650 pack fires, almost all come from charging cells which have dropped too low for safety. Like what you have. Think multiple roman candles, with each setting off the surrounding cells, all 30 of them.

What problem do you have with obtaining the NESE modules? They ship everywhere, 'S'o 'F'ar 'A's 'I' 'K'now. These were designed by a long time member with some proven design and engineering skills, not somebody attempting to meet safety standards for a podunk town in Pakistan, as in, no standards at all.

 

[bobbill]

Bob, Thanks.

Believe it or not, this stuff is beginning to sink in...

Mechanically, I thought the pack was solid, but the facts say no, even if I tightened down every connection, and hammer the crapped out of it to move cells and even pack...so might simply junk the Vrusend stuff and go with another kit...still want to not weld up cells to bus bars. I may have sufficient coils of material to do a new pack, using unused Samsung cells

Will try Nishi again, Seems I had the spelleing wrong, and also tried to bring site up and so on, but nada.

Also, I do know about dangers of these cells. I am anal and do not charged at night. And replaced cells in the used pack with the unused Samsungs, figuring BMS would not like, which reminds me, I should remeasure the cells I just took.

Will look for NIshi again. I got spelleing from Electric Bike review by "Spinnina Magnets."
The AnnPower site was not specifi enought for me...of course I am not so smart, obviously.

Wil let you know how things are going...FWIW, I check the email notify, and get nothing re activigty here. So I have to remember to come back.

Still have not perfected Anderson 45 crimping either...such a klutz I am.

Robert

 

[amberwolf]

Will try Nishi again, Seems I had the spelleing wrong, and also tried to bring site up and so on, but nada.

If you want to bring up the right product, you have to read the actual words posted by Angrybob.

There is no "nishi" anywhere in his posts, though there is a word that starts with the same letter and has one other common with it.

I have a feeling that a significant part of your problems is caused by getting wrong information from some form of reading issue, where you are seeing different words than are actually present. I recommend checking that out to see if it is something you can work around, like with dyslexia, where there are a number of things different people do to help them read what's actually there despite the dyslexia.

 

[bobbill]

Amberwolf, while you are correct, I knew my spelling of name was incorrect, however, found the site and appreciate your concern.
So you know. I do tend to skim...but always anal on the doing parts. Logic is a biggie for me.

 

[bobbill]

Amberwolf, while you are correct, your take is well founded.
I also knew my spelling of name was incorrect and let it ride, (lazy) however, I did locate the site and appreciate your concern.
So you know. I do tend to skim...but always anal on the doing parts. Logic is a biggie for me.

Bob, shipping time for the items is 10 weeks, so have to try to make do with the stuff I have. (Am learning and apprciative)

In my lurching cam across battery packs with BMSs attahed and might redo and make a series of them into packs, just have to find source. And Must be some sort of secret to getting 45amp Andersons crimped. But like I note, mechanical, figure to prevail.

Must get done. Thanks.

 

[agniusm]

Amberwolf, while you are correct, your take is well founded.
I also knew my spelling of name was incorrect and let it ride, (lazy) however, I did locate the site and appreciate your concern.
So you know. I do tend to skim...but always anal on the doing parts. Logic is a biggie for me.

Bob, shipping time for the items is 10 weeks, so have to try to make do with the stuff I have. (Am learning and apprciative)

In my lurching cam across battery packs with BMSs attahed and might redo and make a series of them into packs, just have to find source. And Must be some sort of secret to getting 45amp Andersons crimped. But like I note, mechanical, figure to prevail.

Must get done. Thanks.

Hi. I would advise against making Vruzend work. I was polite most of the time when i was talking about vruzend but their approach and design is total garbage. I say this cause someone might get hurt badly. For presure kit you want your cell to be constrained at both ends so it sits firm. Vruzend gambles on this, cells float. Not even that but the current path is through threads which is a no go, basic electrician knows. Cable ties holding it all together is just a gimmic. If you want even presure holding everything you need trapezoidal structure for cabled tension to work, look up A123 AMP20 packs.
Now, 10 week shipping is due to COVID and is advertised by our post office as possible delay. It might not happen but it also could. I found out that most of the time the fault is at US border, they could not cope with increased shipments.
I am releasing my V2 of the modules and i am selling out my V1 stock. V2 are reduced in size but V1 is 100% reliable system. People use it on skateboards, down on the deck, unprotected, with over 1000 miles clocked.
If you want to take a chance V1 are discounted 40% so you might as well get yourself a bargain. They do have a resell value too.
I am out of the series bus bars but they should be available in a weeks time. I also support my customers over email, phone/whatsapp/FB messengee or video chat if there is a way to align time if timezones are far appart.

 

[bobbill]

I have no reason to disagree, save I have them and the lead time for competitive items is 10 weeks.
I also wish to point out (to Amberfox) there are similar makers: "Nishi" which seems to be Indian and Nesi which is European and Mr. Agniusm (Nese) responded here, which I apprciate.) must slso credit the review by Ron ("Soinningmangnets"), which I look at frequently.

I have noted-well stability and portabiliy and like them.

I also look favorably on the latter's rigid cases (V1s-?) and am thinking-out how to use with current BMS and battery products to combine to form 10S3P, 36v packs. Still, will have to consider transit time as the season wanes.

As you assume, I went into this blindly and am working it out wih trials and errors. Will have to decide re items and delays.

The we packs look to be the ticket, I must say. And appreciate the input. Learning curves are what they are and mine is too high, I know. Obviously, "newbie is me, as nooby," but we learn.

 

[bobbill]

I am back and confused...

History, as above and latest events are below.

Have assembled Vruzend 36v 10S3P 18650 pack, charged and installed on bicycle...(Bike is powered by ((GrinCA) 36v Bafang geared front hub) ran for 3 blocks or so and quit. Inspection showed cells, which were all the same initially, out of balance. BMS seemed to not want to charge so,

Replaced cells with Samsung NPR m3000 cells and a second 36v BMS I had. All cells were 3.53v. Charged pack to 40vOa. All these cells were 3.52 v.

Now some parallel sets of cells (three sets, or 9 cells measure only 2.71v) while most are up to 4.01v and pack is now 38V and seems to go lower daily by a volt. Was 40v now 36v...?

Have not put on bike or wrappedetc. Seems so pointless.

I have checked each conntection...all have continuity and as above...

I am at wits-end and ready to just buy a 36v pack and be done with it.

BTW, I checked "notify..." and nothing, so my return here is sporadic

Comments???

 

[AngryBob]

Well, Bill - I'm going to call you Bill because calling you Bob would just be too confusing - I Have actually been laughing myself silly for several days thinking about this, because I have come to a conclusion about your problem, and the best currently available solution for it.

You need a large, wooden mallet, take the whole battery apart, and whack each end of each cell repeatedly, with measured, even strokes. Ensuring even pressure for the connection across all cells is the fundamental problem with the Vruzend kit. Ideally you would then add rigid side plates and many long bolts to maintain the pressure over time, with a layer of poron foam in between the cell caps and side plates to act as a long-lasting spring. At some point in here the Vruzend cell caps actually become a negative and you would be better off removing the friction-fit portion and just keeping the contacts.

What is likely happening is that at least one, or more, of the 6 end caps in each low parallel group has a weak connection and is not consistent with the rest of the group, and is not equally connected for either charge or discharge. Simple continuity test is not good enough to detect this.

It is also possible that you have several bad cells, or that your BMS is screwed, or even both of these AND bad connections.

I would disconnect the BMS and see if the voltage drain continues. That should not be happening and could be dangerous. You will have your answer fairly quickly. Self-discharging cells need to be removed asap. The BMS could be the source of the drain.

I would then isolate the bad parallel groups and give them a few mallet wacks individually, charge and test again. After all cells equal, do nothing and check for voltage drop over days.