Best 16s BMS for Headway Pack

[comradegerry]

After reading as much as I can o the subject of BMS, their pros and cons, I still cannot seem to find a definitive "best model".
I am planning 2 x 10ah 16s Headways, I figure I need 2 good quality BMS to give the packs a long happy life. Which is the best model/brand to buy for reliabilty and where are they available? Any point in the right general direction would be greatly appreciated.

Gerard

 

[Spacey]

I can tell you for certain to STAY AWAY from any 'Capacitor' based auto balancing BMS, and also stay away from Evassemble.com.

I have been happy with the BMS that came from Linda at Headway, It worked first time and has continued working as far as I know. The red coloured BMS's have just been nothing but trouble for me. The capacitor ones can drain your cell pretty quickly. There is someone on this forum from USA that is designing his own BMS based on cutting the throtlle rather than the power and not running the battery through the BMS. I like this idea very much, it's a HUGE thread though so be warned.
http://endless-sphere.com/forums/viewtopic.php?f=14&t=5416

 

[dnmun]

why build 2 packs? why not just one pack 20Ah wide? headway makes a nice BMS, and ping sells the signalab too, or build gary's newest, bestest.

 

[comradegerry]

Thank you for the replies. I did see the monster BMS thread, but it seems the end product is still in trial and not open for sale quite yet, though I would be happy to purchase and guinea pig one.

The reason for 2 x 10ah packs was mounting the packs in the frame trinagle.I did not see a BMS that could balance 32 cells if I had the packs combined into 1 big unit. I thought each cell had to be connected individually to be monitored, hence I was thinking 2 16 cell packs, 2 BMS and then the packs in Paralell for 48volts?

This will be my first build so trying to learn as I go...

Gerard

 

[pgt400]

Pings V2 BMS works well. I've used a 16S version for about 1 year. Works perfect.

 

[dnmun]

if you are building a 48V pack then you only need one 16S BMS. if you have to split it up into two groups to find a spot to mount it you can make 2 separate 16S packs which have the serial connection each pack and are connected between the two packs at each series node by the balancing size wire. then connect the BMS sense wire in the middle of the wire connecting the 2 cells in parallel. just one BMS is needed. collect all the current from the two packs together and connect the bottom of both packs to the B- of the BMS. connect the two reds together out to the controller.

or you can make 2 packs with the cells connected in parallel in each pack which would be 8S2P and connected in series between the top of 8S and the bottom of 9S with a very large conductor. the problem of having this kinda connection is that the voltage drop on the series connector between 8S and 9S can throw off the BMS LVC detection so that the BMS would cut out early.

the series connection is how the current is carried for the controller, and the parallel connection is what keeps the 2P cells balanced at the same voltage.

 

[comradegerry]

Thanks Dnmun for the explanation, it is good to know I only need to deal with one BMS. At the moment it looks like I can fit 22 cells in one block and 10 in another close by each other. I will pair the cells in parralell for the the balance wires and then series the 2 packs as you sugest to make one pack of 20ah. Now all I need is to go shopping for the parts...

Gerard

 

[dnmun]

in a case like that it would be easiest to build the two sections as 11S2P then have a large cable to jumper to the other end of 5S2P.

the voltage drop across the connection of the two packs when added to the sag of the cells during discharge causes the BMS to shut down for LVC prematurely on the top cell of the lower pack. so go maxo there, maybe double/triple 10G or 8G if you can for 20Ah wide. all imho

edit, even better would be to attach the sense wire for the top node of the lower pack right in the middle of the jumper between the two sections, that would split the voltage drop between the two adjacent cells there, reducing the impact on either one, but doubling the number of nodes susceptible.