Another 16S BMS (Basically the one Will did on Youtube)

[jasonhc73]

I'm all hooked up and it's working fine. I'm not sure at all about most settings, this app is and the display is for an e-bike, the little display and app have settings for KMh and wheel diameter, kind of neat.

How do I figure the AH for the capacity setting? I have 8 BYDs in 16s4p setup.
Any idea what number should go in the Internal resistance connection? the value is in mOhm. I never put my fluke on ohm for the cells, just voltage.

Any suggests or comments on what to actually use for the settings? I have some basic settings from about 2 weeks of using without a bms, just charging and discharging.

 

[simple2010]

I can't help you with specifics for a BYD setup but I have a xiaoxiang 100a version I've been tinkering with. On Android, there are lots of versions of the app floating around. The first one I used, so far, does what I wanted, changing settings and telling me status. I haven't tried other versions yet but this one has "enterprise" in the filename. The "e-bike" display is odd but after a few days is useful in a weird way. Quick info on volts and amps. The concept of capacity is something I'm still testing.

I'm hoping that a full charge and discharge cycle calibrates things but that may be too optimistic. You can tap that lock/unlock graphic to emulate a breaker which is great. I started with some conservative values on a sinopoly-based 12v setup and so far it has been great. The bluetooth module works a lot better if the antenna is given due attention.

 

[jasonhc73]

I can't help you with specifics for a BYD setup but I have a xiaoxiang 100a version I've been tinkering with. On Android, there are lots of versions of the app floating around. The first one I used, so far, does what I wanted, changing settings and telling me status. I haven't tried other versions yet but this one has "enterprise" in the filename. The "e-bike" display is odd but after a few days is useful in a weird way. Quick info on volts and amps. The concept of capacity is something I'm still testing.

I'm hoping that a full charge and discharge cycle calibrates things but that may be too optimistic. You can tap that lock/unlock graphic to emulate a breaker which is great. I started with some conservative values on a sinopoly-based 12v setup and so far it has been great. The bluetooth module works a lot better if the antenna is given due attention.

I totally agree about that Bluetooth, I've never seen one this far away work so well.

I am a little bummed, I have it going and the top end has settled, but my cells are just 0.002 V apart. So much for needing the balancer.

 

[nwillitts]

is that just one BMS on all your batts?

 

[jasonhc73]

is that just one BMS on all your batts?

For now, but I'll have the other 3 sets with the same by end of the week.

 

[nwillitts]

For now, but I'll have the other 3 sets with the same by end of the week.

what i mean is-one BMS per 48 volt pack in serial?

 

[jasonhc73]

what i mean is-one BMS per 48 volt pack in serial?

I am setup 16S4P.
I have one BMS installed so far.

I will have 4 total as soon as I get some other parts. I do have the other 3 BMS's ready to be installed. I have an experiment with the 8 wires in each of the packs that the External BMS connects to.

 

[FilterGuy]

A few questions:

• Do you know if the BMS app is the same as what Will just reviewed? (The one from Battery hook-up)
• I love the frankenstein switch, but do you have anything other than your checklist that prevents both the inverter and Grid from powering the house at the same time? Things could get 'exciting' if that ever happens.
• How big of a load can your house draw? A 50amp 250V plug seems undersized for a whole house.

 

[FilterGuy]

Hmmmmmm...a beer bottle opener right next to the power system... that could lead to some interesting stories

 

[jasonhc73]

A few questions:

• Do you know if the BMS app is the same as what Will just reviewed? (The one from Battery hook-up)
• I love the frankenstein switch, but do you have anything other than your checklist that prevents both the inverter and Grid from powering the house at the same time? Things could get 'exciting' if that ever happens.
• How big of a load can your house draw? A 50amp 250V plug seems undersized for a whole house.

1. Yes the same, or almost the same as what Will reviewed. Mine didn't have those oddball flat connectors on it. (Ebay Link)
2. The knife switch can be reverse powered if it is thrown.
   1. No, I do not have an interlock to keep this from happening.
   2. Yes, it will blow up my inverters if I have the Output from the House on the Output of the LV5048's.
   3. In the Army there is a saying, you have to be smarter than the equipment you are using.
3. I did a max load test last week running everything I could think of through the 50 amp breaker.
   • The batteries drew 92 amps DC, but the output from both inverters was only 8700 watts max, which is also 87%.
   • 92 amps DC inverted to AC results is 35 AMPS AC (240V split phase)
   • 10kW is just 40 AMPS AC

 

[nwillitts]

where did you get the adapters?

 

[jasonhc73]

where did you get the adapters?

DigiKey Electronics Home

Digi-Key offers 13.4M + products from 2,300+ manufacturers. Large in-stock quantities able to ship same day. Paypal accepted, order online today!

www.digikey.com

BYD BMS

A no cut, no splice, plug-n-play. It has a screen, Bluetooth, and more settings in the app than we know what to do with. Will already reviewed this BMS. The Inside wires: The Outside Wires: Conclusion: Molex Micro-Fit 3.0 430251000 10POS 3mm Molex Micro-Fit 3.0 or 1729521001...

diysolarforum.com

 

[nwillitts]

thanks buddy.

 

[jasonhc73]

Today I awoke to a very high cell V on one of my cells.

This BMS has so many settings and no documentation. Some of the settings make clear sense as to what they do. And some just have no meaning at all.

I had the balancing set to start when the cells are 3.199 V and the CellDiff Value of 0.008 (this is also the default).

LiFePO4 are max V 3.65. I have the BMS set to stop at 3.599. (The CellHigh alarm kicked in several hundred times from last night to this morning)

This is only cycle 4.

One setting has yet to make any sence to me until now.

• (System Reference Voltage)
  • The number was 65536 (default)
  • The acceptable values are 2.990=<X<=3.010
    • This didn't seem to make any sense to me either.

The active balancing on the BMS was only running for about 1 second, then the system would bounce around for a few seconds. Next the cell alarm kicked in.

I set the (System Reference Voltage) to 3.000. and would get an error, and then the 65536 would show again. I saw a video where they had to press set 3 times and then the setting would set. So I did just that. Now the 3.000 set.

The Cellbalancing started right up and that high cell, cut off at 3.598, is now coming down to the rest average 3.355.

The Active balancing only ran for about 5 minutes, and now the high cell is passively balancing itself.
I do not understand what is actually happening, but the CellDiff was 258 millivolts, and now 20 minutes later it's at 145 and quickly getting closer to each other. That 3.598V is discharging to the other cells and now at 3.476V

 

[nwillitts]

interesting.
well,looks like you are making your own documentation for it.
looking forward to getting mine and following your lead.

 

[Craig]

With these used batteries I am concerned traditional active balancing will not only not work but actually make things worse. I have been trying to make a post about this but am having a hard time explaining my theory. In short all batteries I have tested have different power curves so if you balance all cells to X volts they may not all be at same state of charge some could be 50% charged others could be 60% charged while others could be less. So if you have all batteries at X volts and you put same energy into each the 60% battery will reach high limit much faster than the rest. I'm thinking the answer is to only balance at top and bottom of range. This was also eluded to in Will's battle born interview. I will try and make a better post about this tonight with real testing data.

 

[jasonhc73]

With these used batteries I am concerned traditional active balancing will not only not work but actually make things worse. I have been trying to make a post about this but am having a hard time explaining my theory. In short all batteries I have tested have different power curves so if you balance all cells to X volts they may not all be at same state of charge some could be 50% charged others could be 60% charged while others could be less. So if you have all batteries at X volts and you put same energy into each the 60% battery will reach high limit much faster than the rest. I'm thinking the answer is to only balance at top and bottom of range. This was also eluded to in Will's battle born interview. I will try and make a better post about this tonight with real testing data.

I get what you are saying.

I also think you are over-analyzing it.

This BMS does not balance until charging the entire pack is done charging.
This BMS does not balance while discharging either.

The "active" balancing that I have seen it do only happened way after the bulk charging completed. The active balancing that I saw it do only lasted about 20 minutes, and it only did it from 248 millivolts cell difference to 165 millivolts cell difference, after that the "balancing" feature never turned on again. I have it set to balance when greater than 8 millivolts, but it doesn't actually do anything active. At least it doesn't show that anything is being done on purpose. The first 4 cycles I did with it, it didn't balance anything ever. At the end of the 4th cycle, something unusual happened and that seemed to cause what appears to be a runaway cell. But this BMS has an overvolt limit that I set, and it kicked in and appears to be what saved the cell.

So to be concerned about SOC is kind of a non-issue, charge the battery as a battery, then let the cell balancing happen. I ran these batteries 30 cycles before I installed the one BMS (3 more to install when more parts get here). I have the "float" set to the nominal voltage, so basically I bulk charge and then don't charge, just like the Denis mentioned in the interview.

I had float charging for about 10 cycles then I read this page: https://www.solacity.com/how-to-keep-lifepo4-lithium-ion-batteries-happy/ scroll down to the "Charging an LFP Battery"

1. Keep the battery temperature under 45 Centigrade (under 30C if possible) – This is by far the most important!!
2. Keep charge and discharge currents under 0.5C (0.2C preferred)
3. Keep battery temperature above 0 Centigrade when discharging if possible – This, and everything below, is nowhere near as important as the first two
4. Do not cycle below 10% – 15% SOC unless you really need to
5. Do not float the battery at 100% SOC if possible
6. Do not charge to 100% SOC if you do not need it

 

[Craig]

I'm confused I thought you said cell balancing started at 3.199V.

Another thought Is it self balancing or is it settling or is that what you mean by passive balancing.

 

[jasonhc73]

I'm confused I thought you said cell balancing started at 3.199V.

Another thought Is it self balancing or is it settling or is that what you mean by passive balancing.

I honestly don't have any idea how it works. I mean I set it to 3.199V and 0.008, but nothing happened, until the BMS was way way wayyyyyyy past the 0.008.

Right now, after the bulk charge it shows 0.007 CellDiff. Yesterday's problem cell is sitting happy with all the others right next to it with the exact same 3.281 V.

I ran the pack(s) for 10 hours today, and it only took 2 hours to recharge it. That's charging at 18 amps, to 64 cells. I bulked at 56.5V and the pack is sitting there cozy relaxed at 52.5V, reporting 95% SOC (I still do not know within the settings where to really define it)

This is what the BMS shown after 10 hours use:

And this is 2 hr 40 min later.

It's just a setting to configure: