Bestechpower BMS

dnmun

1 PW
Joined
Jun 9, 2008
Messages
16,186
Location
portland, or and loveland, co
my package finally arrived. 5 lipo BMSs and one balancing board. the D131 was $58 each and the D126 is $19.

henry decided to send me the D131A balancing board so i will have something to use for reference to diagnose problems on the D131 BMS when they arise.

the D126 has FB4310 mosfets from international rectifier.but they only rate the discharge through the 3 output mosfets at 15A. they should handle 150A without too much trouble. i am gonna hack the shunt with some 4 milli ohm shunt resistors to see if am able to increase the cutout current to about 70A.

i am gonna order some surface mount resistors, i think it will be either 68 ohms or 100 ohms, to the pcb on the big flats left on the underside and increase the shunt balancing current by 2X to 84mA with the 100R, or 104mA with the 68 ohm SMT 2512 resistors.
 

Attachments

  • DSC01605.JPG
    DSC01605.JPG
    51.7 KB · Views: 16,315
  • DSC01606.JPG
    DSC01606.JPG
    86.2 KB · Views: 16,315
Subscribed. I'm interested to know about the real world balancing of LiPo with these BMS'
 
i examined the D126 more. it is 15S lipo so i think it has the most promise. kinda the basis for my 15S lipo argument, using the limit of the 63V caps in the 36V controller most people have.

i found the gate circuit for both the charging mosfet and the output mosfets. they are driven by the first IC. the source voltage of the circuit current is from the 5th cell and i think it will be 21V on full charge. just what the ICs need for happy camping. so good gate voltage all the time.

the balancing is still done with that pchannel mosfet getting the signal from the 5 pin comparator directly above.

the cell voltages are direct inputs to the ICs so they are all ADC material for the processor. more reliable, exact measurements.

this is different from the balance board of all the analog BMSs which use the classic signal line driving the gate voltages of the mosfets.

i ordered 200 of the 100 ohm 2512 surface mount resistors to add on the back side, and i ordered some 4 milli ohm 2512 SMT resistor shunts to modify the current shunt which is 2 milli ohm down to about 1.33 milli ohm and that will raise the cutout current to about 75A peak.

i think this is gonna be a game changer. i should be able to make a 15S HK lipo pack of 63V DC, 15S 10Ah wide, with the 75A limit and the BMS for balancing and cell monitoring for a total of about $260. that's where i am now. haven't decided how to combine with the 48V ping yet. wondering how to do ideal diodes to combine two different voltage packs in parallel.
 
i finally got the pack wired up and put it on the charger. gonna be a few days before it balances. i think the lowest cell is 3.49V when i assembed the pack and the three cells on the top are fully charged, 4.2V. i am hanging some power resistors on the high cells and draining them down to match the others. but it may take a few days to get close. maybe 2-3Ah apart from high to low on 5Ah pouches.

this pack is from the three turnigy packs that silverheels donated a few months ago. there were three dead cells, two were puffed in that one 6S. so i ended up with 2+1 from that and the other two of the 6S packs make 15S.

i cut all the shrink wrap off and cut off the JST plugs and the 10AWG stubs and soldered some 12G solid wire to connect them together into 15S.

the gate voltage on this BMS is very low from what i am used to. only 3V on the gate, it is digital and not like the other classic BMS styles. this one has the thee chips for all 15 channels. on shorter versions of this same pcb they only have two of the chips up to 10S. some pictures and i will add more when i get it balanced. i should work on the icharger to see if i can make it run from parts to balance this thing. patience. this is the maxo test of this BMS. HVC cycles ffrom 4.15 to 4.35V. i am draining at about 400mA/hr.maybe 8 hours.
 

Attachments

  • DSC01601.JPG
    DSC01601.JPG
    99.4 KB · Views: 16,058
  • DSC01602.JPG
    DSC01602.JPG
    78.4 KB · Views: 16,058
  • DSC01609.JPG
    DSC01609.JPG
    112.5 KB · Views: 16,058
  • DSC01610.JPG
    DSC01610.JPG
    66.8 KB · Views: 16,058
  • DSC01617.JPG
    DSC01617.JPG
    77.6 KB · Views: 16,058
  • DSC01620.JPG
    DSC01620.JPG
    80.3 KB · Views: 16,058
  • DSC01619.JPG
    DSC01619.JPG
    82.1 KB · Views: 16,058
i ended up looking at the pictures of their factory on their website. it reminded me of when i first moved to oregon in '72 and got a job working on an assembly line soldering together components for the Techtronics oscilliscopes. there is one picture where everyone on one side of the aisle is wearing blue lab coats, and the other side is wearing white. i bet they are competing against each other too. 3 assembly lines it appears. the SMT machines do everything, the girls are putting stuff like the screws in and mounting the heat sinks. and all the real assembly is done inside the three SMT machines.

http://bestechpower.com/factory/

if you look at the factory building you can see where the bosses offices are by the number of air conditioners.

i found this corporate creed from an article last year. it seemed to sum up how i see them too, after watching how BMS battery charges so much for the BMS when they sell it to us, and the people who manufacture it get nothing. this is what they said about it. something we should appreciate.
>>

Why trading companies can beat the manufacturers?
Published:2012-10-14
This article is talking about the manufacturers BesTech Power compete with the Broker Agent.
In Battery PCM/BMS/PCB business,there are many trading companies are quite famous,almost every battery business man knows about them instead of the end-source manufacturers?why?what happened ?
It is supposed to totally OPPOSITE! Manufacturers should be more famous than the borker,they have more advantages,from now on BesTech Power will turn this situations over,to ensure the customers get the end-source directly @ pretty competitive prices.
BesTech Power designs,develops and manufactures Li-ion/Li-Polymer/LiFePO4 battery PCM,PCB,BMS.
From us you can find almost all the battery controller/PCB/PCM/BMS you need,even you can not find the one you need,then we can develop one out in 3-5 days.
So Far we have 1S to 20S kinds of PCM,the current we can make it up to 200A.
Thanks.
 
Very interesting. I have been looking at the 12s 15A board as it is only 35mm wide. Such a bms will fit happily inside a frame tube. I need more like 25 amps though, and pondered on the use of large fet's rather that heatsinks on many boards. I have asked Henry if increasing the 15A rating might be possible with a heat-sink and other small mods, as the board size is ideal. However, the picture of the board is square rather than long and thin so I can't judge anything from either photo or dimensions. They don't match.

I really want 4s0p (I won't write 1p it seems like gibberish) hardpacks in a row within a 2x2 tube, along with bms. A goal I thought many would be working towards, but non seem to exist above 15 amps. All the other boards from wherever are 50mm wide or more, and a 2x2 is about 48.6mm internally. So close I have looked at shaving one. I could sit a board diagonally, but my battery tube must taper at the top of the triangle. FYI a 44.4v 5Ah pack fits in a 2x2 just 41.7cm long. My bottom tube offers a few more inch's to sit against, if I taper my battery stick down to an inch towards the head tube.

Sorry if my tube names are confusing. I'm still learning.

If he says it is possible, I will post

I got a couple of quotes from him.
D132 12S 25A $35 (double decker, 50x60mm)
D163 12S 35A $40
D167 12S 35A $40
 
you do not have to place the BMS in the battery. you can put it next to the controller and run the sense wires to it.

you can buy one of each and test them. then sell one to someone local to cover shipping.

i burned one of the three irfb 4310s on mine after pushing more than 30A through them. twice the spec, but it handled the 20A load.
 
Just receive a packet from DPD containing HCX-PCB-D167A BMS from Henry. He also included 5 mosfets that i blew on my other Besttech BMS. This one looks like 16S BMS done for 10S. The soldering is better than on previous one and it has temp sensor for thermal overload. Here are some photos:

20130618_115859


20130618_115909


20130618_115912


20130618_115919


Data sheet:
20130618_115928
 
i agree. they use the 16S board and then just do not populate the channels if you order a BMS with less than the 16 needed.

they use the same board to go up to 20s by putting a little daughter board on pins raised above the last 4 channels to make a total of 20 channels.

i am gonna order one very similar to yours, a D166. also i am gonna order a D107 which is the simpler, cheaper one with no heat sink but still they rate it at 35A so will see if i can break it too like it did to the D126 by sending twice the spec current through the mosfets.

i have figured out how to hack these BMSs down in series. i was able to modify a D131 from 24S lipo down to 20S to test it. after i use this D166 20S lipo i will hack it down to 18S because oi wanted someone else who runs on 18S to test it also to see if we can break it.

so now i order the max series available in each model i am testing since i can hack to whatever number of channels i need. i cannot hack the D126 yet though because it uses that digital controller and i have not figured out how to hack it to spoof the processor. but 15S is what i wanted for that so that is ok.

you really did a great service by finding these guys and posting up about them way back when. i still want to buy one of those three layered D132 BMS for 16S lifepo4, to see if i can break it too. that was the one you showed originally.

they have good quality BMSs and they are so cheap, so cheap when you think about the junk you gotta buy from BMS battery and they charge so much for just putting on the normal number of mosfets, and you can only buy the 12S or 16S from them, and cannot hack them down in series because they have the black goop over the signal lines so i cannot hack the signal lines. plus Bestechpower makes the biggest of all the BMSs i have ever seen too.

the D170 they offer for 16S lifepo4 has 40 of those big TO-247 mosfets for the output mosfets. even though they say 100A max, i don't think there would be any problem to run those up to 200A+ with the addition of some buss wire to the traces to spread out he delta voltage across the array of mosfets so the load is shared more effectively. from looking at the shunt i know that it will not cut out until about 180A or maybe more if the shunt is hacked a little bit. plus i thought of how to doubledecker the output mosfets to double the power rating.

the mother of all BMSs: http://www.bestekpower.com/768v24spcmbmspcbforlifepo4batterypack/BMS-D170.html
 
You get so much of good and valuable information on this forum, that you are obliged in some ways to return at leas small part :)
I was looking into other companies that manufacture BMS'es, this time capacity transfer ones. i don't know why ir is not widely available? It makes sense to transfer charge from high cells to low ones meaning no wasted energy but this one company quoted me 132USD+shipping for 24S 8/15A one. That was a shock cause even Founding Power programmable one costs less or the same with LCD/OLED unit. I want to work out a way to bypass discharge fets all together, this is another part i don't understand. you choose a controller which limits your current and then you get a second player in limiting discharge and that's like fifth leg for a dog. Just does not make sense to me.
I wander if you can get normally closed high voltage relay to drive that ignition wire on the controller for LVC disconnect?
 
OK, i knocked up a little drawing of how i would like to run these BMS'es:
Untitled-1.png

I dont know if it is clear enough, never the less, when key switch is disengaged relay is in NORMALY OPEN state. When key switch is engaged it engages relay and we get connection between conroller positive ad ignition wire (orange) so the controller is live. When a cell reaches LVS, BMS cuts off P- and there fore disengages relay braking connection of ignition wire to positive.

Am i thinking the right direction? What sort of relay/relay contactor i would need to suit all the voltages, say up to 100vdc in?
 
if what you wanna do is to drag down the throttle line or the ebrake signal you could use an optotransistor and drive it with the gate signal for the output mosfets. or just a regular small mosfet would work too.

i am also concerned about the risk of short circuits when no one is present. but you can use the thermal breaker lead and put a switch in that to turn the battery off altogether to prevent shorts.

but for me the ability to charge without having to sit with the lipo is crucial. not everyone sits with their battery for 2-6 hours while it charges. jmho
 
No, not the throttle line. There is a thin ignition wire on a controller. Only when this wire is connected to positive controller is active. The drawing is incomplete but there is a fuse to prevent shorts.
Here is which one:
471258463_544.jpg

Note plug 8, there are 3 wires, battery -, battery + and third orange wire which needs to be connected to battery + too to enable the controller
 
ok, i just realized i never posted the picture of my 20S hack here. converted a 24S lipo BMS to 20S.

this is a D131 24S lipo BMS with the signal lines opened above the 20S level to prevent false HVC and LVC signals from the 'ghost' cells above. i was able to restore this BMS to 24S by resoldering the resistor on the LVC line and resoldering that resistor on the HVC line on top back onto the pads to reconnect the signal line to the bottom of 21S.
 

Attachments

  • DSC01630.JPG
    DSC01630.JPG
    98.7 KB · Views: 15,664
subscribed !!!

Tommy L sends.......
 
got some more of the Bestechpower BMSs in sent out a bunch to people who joined in the shipping. most of them were the D167 since they are the best deal i can see from these guys. 35A continuous and 70A cutoff. i cannot create a 70A load so i cannot test that limit but from looking at the heat sink and number of mosfets i think the D167 can handle 50-60A with not much problem.

i got the D167 in 16S, 18S, 20S, lipo and lifepo4 and i also bot some more of the D131 24S lipo and lifepo4, both 80A continuous and 170A cut out. i got the D122 12S because it looked interesting, tiny tiny, but it is rated 15A and i suspect it can handle 25A since it has the irfb3607 mosfets. the rest were the D126 12S and 14S for lipo. $755 total. $50 air freight on 21 BMSs.

also i built the first 14S pack using the D126. i posted up the build on Lee's thread http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=54024&p=804676#p804676

i finally got the pack together and discharged a few Ah and then charged it up the first time to about 58.4-58.5V so that the highest cells were right at 4.21-4.20 and the lowest cells were 4.15V. did a discharge test to determine capacity of the 5Ah turnigy and got 4.6Ah from charged to cutoff.

so i charged it up again, and this time i used a higher voltage, tweaked it up to 59V to force the pack to fully charge and to test it to see how high the high cells would go when the BMS was subject to overcharge level voltages. the highest any one cell got was about 4.23V, with several above the 4.20-4.21V level but eventually i was able to pull the lowest cells that were around 4.15V up to full charge at 4.20V so then i terminated charge and let the pack balance at the 4.2V level/cell.

then i did a second discharge and got about 5Ah from the same turnigy pack, about 6% increase in capacity, which is expected since 4.15V is about 6% below full charge.

i had limited the discharge to 15A on the 5Ah pack, 3C. today i did another discharge through that D126 and pushed the current up to 27A to see how hot the mosfets would get, 1500W of power.

i did that for about one minute and the current shunt resistors got very hot and the mosfets were getting hotter and hotter, really only warm to my touch but i had the thermal breaker taped on top of the mosfets to see if it would open from reaching the 90oC level that the thermal breaker opens at.

i decided not to push it for very long so after a minute at the 26A level, down to 25A rapidly, i cut off some of the current back to 12A for few minutes and then pushed it back up to 17A for the duration of the discharge and was not able to burn up the mosfets under this load. so even though they are rate 15A, and 25A cutout, i am comfortable thinking it can run at the 1000W level when fully charged without too much trouble but running above that level for longer periods i think would damage the mosfets from the heat.

the shunt resistors are next to the mosfet on the end so i think it is the most liable to thermal runaway from the heat input from the shunts. i also think maybe that is the better place to put the thermal breaker to protect the BMS from over current overheating since the response to the current is immediate.

so i am gonna recommend sticking to something under 20A for long periods, 17A i know is ok, but restricting higher current above 20-22A to relatively short periods, say less than one minute max. this is all just my opinion, and i think richard will come up with solution to allow current to bypass the mosfets on discharge and use a signal to the controller to shut down the controller at LVC. that would protect the battery from over discharge if the controller is left on by accident and drains the pack.
 

Attachments

  • DSC01686.JPG
    DSC01686.JPG
    58 KB · Views: 15,288
  • DSC01687.JPG
    DSC01687.JPG
    57.1 KB · Views: 15,288
  • DSC01688.JPG
    DSC01688.JPG
    69.3 KB · Views: 15,288
  • DSC01689.JPG
    DSC01689.JPG
    55.3 KB · Views: 15,288
  • DSC01692.JPG
    DSC01692.JPG
    60.8 KB · Views: 15,288
Hi,

i just bought the D167 20s 35A. Cabled it up and the FETs conduct. I want to test if it really shuts down if one cell goes below 3V. So i just cut 4 cells while appling little load (lamp with ~20W).

with the lower and upper cells, it shouts down withing >1sec.

BUT: If I disconnect some in the middle (like cell 12..16) it does not shutdown :( Any Idea what could be wrong here? Why should undervoltage detection not work like this? I'll try to repeat the test by connecting a potentiometer in between one cell to be able to feed it with any voltage i want to try.
 
izeman also found this happened to him when he disconnected the sense wires.

the reason it doesn't turn off is because of the way the comparators on each channel take their power from the top and bottom of the channel.

when there is not a voltage across the channel then the comparator cannot function and so it cannot turn on the transistor that delivers the voltage from the channel to the signal line.

izeman feels like this is a big risk, and i think it is a concern too when charging lipo especially but i use soldered sense wire leads so i don't have an expectation the sense wires will somehow become disconnected.

edit, also if this theory is correct then it will lose the ability to signal LVC or HVC on the channel above the one with the missing sense wire as well as the channel where the sense wire is removed.
 
I'm thinking of getting a BMS from Bestechpower. I've never used one before. I think I understand batteries, and motors, and controllers well enough to be able to choose which one to get, but not BMSes.

I run with 8S Lipos, using the Hobbyking 4S 5AH 20C hardcase packs. I charge them with an iMax Lipo charger, so I'm not really worried about overcharging, and I run them with a battery monitor on each battery, that beeps when any cell goes under 3.2 volts. My wattmeter says that my max amps draw was 35 amps. So I think I'm already covered.

But everyone here seems to be advising "get a BMS". And Besttechpower sound good, from this thread.

After reading this thread, I had a look at their web site, and there's a huge number of products, although I think not as many as it seems because they're using essentially the same board for different S-numbers.

So I emailed them, saying I wanted maximum discharge 40 amps, and I'd be charging at 7 amps or less. They suggested a HCX-131 http://www.bestechpower.com/296v8spcmbmspcbforli-ionli-polymerbatterypack/PCM-D131.html

Minimum quantity 2 pieces at $27/piece (that's OK, I can make up two batteries), shipping etc $29 (sounds good). A few dollars plus or minus aren't an issue.

I don't know why they suggested that one; maybe because I said I'd be charging and discharging at different amps. So, I have two questions.

1) Do you think that's the right BMS for me, and if not, what would be better?

2) How do I power it? It wants 33.6 volts. What I'm using now, is a PC power supply 12 volts to the iMax charger. Does it want exactly 33.6 volts, or will it be happy with something in the range 40-50 volts, for example?

Thanks.
 
why is your battery only 8S? that is a cheap price since i paid $58 for my D131 and i got a big discount for buying a buncha BMS at the same time.

for 40A max you could use the D167 and you can buy the 16S version and use it for 8S now and then finally go to 14-16S later.
 
My battery is only 8S because my motor is 24v. I tried it at 12S, and it's faster than I need it to be.

I'm not really convinced that I need a BMS. I'm using iMax chargers, they don't let any cell charge over its limit. And I'm using a battery monitor on each cell, that beeps to alert me when it's gone down to 3.2v.

DO you think I should still be looking at getting a BMS?
 
if you use a balancing charger it will balance your pack so you don't need a BMS for that.that is the primary protection a BMS provides, to prevent overcharging.

most people have a good idea when the battery is exhausted from loss of power and pack voltage but that is not why there is an LVC built into the BMS. the LVC is there in case the pack is left with a load like the controller or lights left on, and the battery discharges itself to nothing if unattended so the LVC will turn off the battery when it gets too low.

the other protection a BMS provides is when the output of the battery is short circuited and then the BMS will turn off the battery to prevent a fire.

the monitors on the battery will cause it to become unbalanced if left on constantly, but your balancing charger can compensate if it gets recharged, but if you leave them on for long periods without recharging they can drain just a few of the cells in the pack and then if you use the pack without recharging those low cells could drop too low. but it is something people manage all the time.

i have shorted a few batteries by accident so i appreciate having the overcurrent shut off.
 
Back
Top