18650 10s3p battery

it is not for you, it is for the other people who need to figure out what is wrong with the BMS. or charger, or controller.

other people learn from seeing how things get fixed.

again, i feel you were misled into thinking a BMS needs a fuse or that it failed because of some over current and i know that is not the case just from looking at the pictures.
 
20050525g.gif


check out that image above, that gives a good diagram and picture of what dnmun is talking about.


Preface: I'm an EE. I really like to know what's going on with my hardware, and if something breaks....I'm finding and fixing the problem, so I see where a lot of this is coming from. However if I buy something as a finished product I expect help on the other end. Where this all meets in the middle however is that what we're doing on this forum is a lot of hands on development and while we're starting to have finalized products come out we're still very much so a DIY community. So now that warranties and such have been enacted, why not figure out what's wrong with the bad bms instead of arguing about this stuff?

curiosity on my end: what IC is handling the BMS? probably a TI chip im guessing right?
 
my unsolicited .02

Onloop - great customer service! It's great to troubleshoot something, but at the end of the day you backed up the customer to make sure they are taken care of. Nicely done sir.

cmatson - great attitude at any age, let alone 15 y/o! You will figure this stuff out and will continue to learn tons with that kind of positive attitude.

While i'm curious and would want to troubleshoot something to fix it (especially something $$ like this pack), for a RETAIL item like this i think there is a clear "best effort" and then just get it replaced. Then figuring it out is just a fun thing to learn from.

A possible good option for you cmatson - skip the bms part and configure so you can charge it simply w/ a standard charger? use it, recharge it, etc. You can still tinker with the BMS later if you want the single simple plug for charging.

dnmun - i get your point, but i also feel like you are being a bit harsh. It's a great skill, and will be useful the rest of your life, to learn scientific and troubleshooting methodologies, but as a customer i think Onloop did the right thing to get him working. It seems like he has the old unit to continue tinkering with and that will give him plenty of time to learn how to diagnose and troubleshoot. I agree that showing those steps to the community help others when they search or find those steps to troubleshoot, but as a paying customer i think he has the right to expect it be fixed or replaced quickly. Us DIY'ers however building a similar product would just have the community to help troubleshoot until fixed.
 
no, i was not harsh. i never called him names or insulted him. i never talked down to him like he was too young to understand so i had to make up some fake reason the BMS was bad so he would not be bothered. i never told him to go waste his money buying a new BMS on ebay that he would never know how to install. i feel i was entirely even handed.

if i had any criticism it would be for the people who led him on this wild goose chase making him think there was something wrong with his BMS so that he should go buy another one. those people should have done the same thing i did and ask him to test it and offer assistance in diagnosing what was wrong.

there is nothing wrong with that BMS. if he had measured the gate voltages it would have been clear what to examine next and in a few minutes we could have isolated the disconnected sense wire and he would be able to use the battery.

he never asked me what parts of the mosfet were the gate and source and it is pretty common knowledge among anyone who uses this equipment imo. i have explained this to dozens of people here when they asked but he never asked because he got a free battery instead so he has no reason now to learn.

this guy who sold him the battery is only involved here because he wants to make it clear he is supportive of the customer experience. he is a salesman. i feel he could have offered the same advice on how to diagnose the BMS turning off the output that i did but he did not. so the guy learned nothing and people paint me as some evil child abuser because it is obvious to me that there is nothing wrong with the battery except for a loose and disconnected sense wire.

these 5 pin count digital ICs on the BMS are very common now and are indicative of a better and more robust design. richard has actually figured out how the addressing works on them so the serial cell count can be adjusted in the addressing of the IC. it is used on several of the smaller BMSs that bestechpower makes.
 
dnmun said:
there is nothing wrong with that BMS. if he had measured the gate voltages it would have been clear what to examine next and in a few minutes we could have isolated the disconnected sense wire and he would be able to use the battery.

Yes I am a salesperson, i am also the warehouse guy, the product designer, the marketing department, the customer service manager, the cleaner, the stay at home dad, the teacher & the student.... and I want to learn more.

I am here willing to learn.

So please, if you are willing, explain your diagnosis. It seems that from a few pictures you can confidently rule out BMS failure. I am curious to know hiw/why. Is because you cannot see any scorch mark? Somthing else?

you think it is a sense wire, what could cause this? Maybe the solder joint overheated & melted?

I am hoping to learn more about this... so please if we can delve in further I would be very appreciative.

Also caden if you need some more specific instructions please list your questions in point form & include pictures to help others learn from this.
 
first you have to determine if only the output mosfets are turned off or if only the charging mosfet is turned off. in this case we know the output mosfet is turned off since he has no voltage on the output of the battery.

if you measure the gate voltage on the mosfets, both the charging mosfet and the output mosfet, then you can determine if both of them are turned off at the same time. in that case it is likely that the circuit current is turned off.

that could be caused by an open circuit in the thermal breaker or there is no power to one or the other of those two ICs on the pcb that control the balancing and the HVC and LVC functions of the BMS.

you can check the thermal breaker by measuring the resistance across the breaker. this has to be done when the BMS is disconnected at the sense wires. if the breaker is still closed as it should be in normal operation then the resistance measured will be 0 ohms. if you measure open circuit here then that would immediately identify the problem. it can be solved by placing a jumper across the thermal breaker since the breaker really provides no protection to the BMS or the battery. it is essentially useless.

continuing when there is no voltage on both charging and output mosfet gates. if you look at the wiring that provides power to those two ICs, every 5th sense wire delivers power to each of the ICs and the ground of the second IC is from the 5th sense wire. you can see that in the traces on the pcb. so if the B0 sense wire, or the B5 sense wire or the B10 sense wire is disconnected then there is no power to one of those ICs so the BMS will keep both of the mosfets turned off.

it is simple enuff to just use the DVM, 20 V DC scale and measure the voltage on each of those two sets of traces to see if there is power to the ICs. if both are powered up then it is not shut off because of those sense wires being disconnected.

if both of the mosfet gates are turned off then there is somehting else going on that would require more examination but i will continue on the main and simplest path where only the output mosfet is turned off.

if you take your DVM, 20V DC scale, and measure the cell voltage of each cell where the sense wire plug socket is soldered to the pcb, you should be able to measure a voltage between each adjacent pin. that voltage would have to be above the LVC spec for the BMS, usually 2.7V. if there is one cell below that voltage or if there is no voltage present then the BMS will turn off the output mosfet to prevent damage to the battery. i think this is the most likely case since it is the most common case.

if the mosfets had been subjected to excessive current then the BMS would shut off the current before it could damage any of those 4 output mosfets. the turn off of the mosfets is orders of magnitude faster turning off than a fuse blowing.

there is no damage visible on the mosfets. the shellac that covers the pcb is still intact on top of the mosfets and it would have scorched and melted if the mosfet overheated. if the mosfet had shorted as they normally do on failure then the BMS would still conduct current and the mosfets would overheat and melt the solder underneath them.

so i know the mosfets are still functional, from looking at his picture. so i am certain that the BMS is not only not broken, it is functioning normally but has inputs that cause it to think it should be turned off. if there is no gate voltage on the output mosfet.

if there is gate voltage on the output mosfets then the cause of the battery not expressing voltage at the terminals is because of the wiring or the the output plug itself, not the BMS. so then just even a simple examination of the wires from the B+ terminal of the pack and the P- terminal of the BMS would show where the open circuit is located.

usually open circuits in the wiring are right inside the connectors. bent pins or clips in an XLR type connector or dislodged contact in the anderson connectors. or the solder could be broken that connects the wire to the connector. all very easy to locate and identify.

the same could be true of the sense wire connections. the sense wire could be torn off the top of the cell in the pack. it could be cut by accident in use or even in manufacturing and did not fail until use. i looked at the soldered connections of the sense wire plug socket and i could not see cold solder joints and i would not expect that either on a new pack in any case and not the sort of thing that would happen since the sense wires do not carry much current which is usually what creates cold solder joints. but the sense wire is connected to tiny clip that fits inside the sense wire plug and those connections could be open or the clip itself is not seated fully into the plug so the sense wire does not make contact inside the plug to the pins on the pcb.

that would be obvious when you measure the cell voltages on the pcb in the solder at the base of that sense wire plug socket.

this is more complicated than most people can grasp when first examining the BMS so i find it easier to ask people to examine the device themselves in a sensible progressive manner to isolate the failure. it does not require an EE degree. i do not have an EE degree. i did take the junior and senior level classes in EE when i got my masters in physics along with my physics coursework but that was 40 years ago and has no bearing on the procedures to follow in examining the BMS which anyone can do with a simple $4 DVM. i did go to grad school in EE for year before getting a job but i did not study any more EE then, i just took courses in electron transport theory and properties of semiconductor materials and plasma physics. maybe this guy will use this as way to develop an interest in learning how to do this stuff but that is not the american way. now everything is handed on a platter so there is no need to strain to learn new stuff. so i get annoyed with it i guess when i see that all the science projects are done by indian and chinese students and not the all american types that used to be common when i was young. now there is no need since there is always money to just buy new and throw away the old. we were expected to beat the russians and just by accident we invented the integrated circuit and everything that has followed since. facebook and the iphone and the mind that is never quiet enuff to see how the transport equation works or the solve the scattering matrix. yep it is sore spot.
 
Wow... that some nice detail...

I really need to study this topic more. I get the gist of what you are saying..

But i have always been a visual learner... and im finding it hard to put your words in to reality. its probably pushing the friendship to ask for you to label everything onto the pictures that caden provided?

If you are right caden could easily fix this. Then he has two batteries and also saves money on replacement BMS.

So is there a reason this happens, you said it's common?

How can I build better packs?.
 
not common but the usually failure mechanism is a disconnected sense wire that causes the BMS to turn off the output mosfets.

most lipo BMSs have a HVC shut off during charging that requires that the cells be discharged down to 4.05V in order for the charging mosfet to turn back on. but that is the charging mosfet.

i don't see anything wrong with the way your pack is built.

i assume each of those mosfets can handle about 8A continuous so the BMS should have no problem putting out 32A and 40+ in surge. more than the cans will push imo. your BMS should have that in the spec already.

i don't do photo shopping but someone else may wanna do it but there is already a diagram of a mosfet in the thread so it should be obvious how to measure the gate to source voltage.

everybody has to do the same thing when they have a problem with the BMS turning off on them. measure the cell voltages on the BMS pcb, not at the cells. measure all 10S.
 
Thanks for the long explanation dnmun.

When I get back home (we are at the beach until sunday) I will attempt to figure out what went wrong, based on your steps. If I have any more questions, I'll post them up here, along with a picture that has an arrow pointing to what I'm talking about on the BMS.
 
ok, so great news: I got the space cell working!

culprit: one of the sensor wires wasn't all the way in the plug, which was a 2 minute fix after cutting away some of the apoxy stuff to re-insert it.

Sorry it took so long, my voltage meter was acting up and I didn't trust anything it told me... just sitting there it would give me random numbers without touching anything. Had to wait until the new one I ordered came, before I could test the battery, but it turned out to be really simple.

well, now that I have a battery and no board, it is very lonely. It seems that it needs a VESC and shinny blue 190kv motor to play with... just maybe

Thanks again guys for pushing me to get to the bottom of this, even though it turned out to not be anything complicated
 
That's great news! I'm glad you figured it out.

Thanks to everyone on this thread who gave some input.

This is the beauty of ES and the DIY maker movement, nothing is impossible if we are all willing to learn and co-operated and be patient with each other....
 
Back
Top