What are your thoughts and experience with BMS boards like this one for use during charging for overvoltage protection and balancing (will not use the discharging function, seems the discharge current is too low for the VX1)?
Will getting 3 of these in series do the trick for the 18 pack Leaf conversion (36 cells to balance to 4.2V each)?
Hi,
Did you look at Foundingpower? They sell programmable BMS's like this;
http://www.foundingpower.com/en/Products/Smart_BMS/20121202100.html
I used a 24S version of this one with a display in a 60V E-scooter. Works excellent.
Price was about 150 euro complete. I can't find the (old) 24S anymore..
Kocho
I am at a loss as to why you would bother with any BMS with a LEAF conversion when all the evidence to date would suggest that provided the cells are finely balance charged and equalised prior to fitting that they hold their balance extremely closely in use!! Proper charger firmware will deal with any charging issues.
Hi Peter,
IMO: Without BMS, overcharging is a major danger when a charger push the cells to the sum of there subsequent charging voltage. If one of the cells reaches this voltage earlier than others, it's internal resistance will grow rapidly, causing higher voltage on the cell. When the voltage frequently exceeds the maximum charging voltage, the cell will damage.
There is no guarantee the cells will come balanced (although they were promissed to come balanced). I am not aware of any mechanism in the bulk-charging approach in our bikes without a BMS that would re-balance them if they do get unbalanced.
Without a means for equalization charging and balancing, any disbalances will have to be dealth with by limiting the useful SoC (to avoid over-charging or over-discharging individual cells).
Also, the ESD charger is known for a particular failure mode where it does not stop charging during the CP mode (I have observed it several times with the modified firmware personally). Regardless of the reasons and how can minimize the chance of this happening, it could happen. The ESD can't recognise a few cells being slightly out of balance - it will overcharge the rest by the time these catch-up.
A nice top-balace will ensure that the bulk-charging with the ESD does not overcharge individual cells (the only value the charger can reasonably accurately measure is the total pack voltage, and it will hide small imbalances of individual cells). The BMS does not have to stay on after a charge and does not need to be used at every charge if I determine the cells remain in balance. However, I don't have a balance charger (yet) and a basic charging-only BMS with a balance function seems to cost about the same as a basic R/C charger (which some here use to initially balance and pre-charge each cell) but will take care of the entire pack in one shot - less hassle, faster (all cells at once, up to 15A charging vs. perhaps 3-7 of the cheap R/C chargers), less chance for error (wire it once and forget it).
A nice bottom-balance (while there is some debate about wether one can actually reliably do that or not) will help ensure that all cells remain balanced when near empty. This is good if one needs to discharge the pack near emoty (and rely on the bulk voltage protection that the MC offers, which again can't regognize individual cells being out of balance/empty/ready to reverse-charge/get damaged). But, just like a top-balance does not ensure balance when discharged near fully, a bottom-balance does not ensure balance near fully charged. Both due to potentially unequal cell capacity. So one has to leave room for error at the top (when bottom balancing) or room for error at the bottom (when top balancing). Without a BMS it is a choice between 2 evils we have to make and chose between bottom and top balance based on how we plan to use our bike (top-balance will give better performance and more high-current use but one will have to stop using the bike before the pack is fully discharged; bottom balance will lose some top-end performance as it would bet dangerous to attempt to charge fully, but will give more peace of mind when using the bike near empty).
So a BMS + per-cell voltage monitor with alarm = peace of mind and eas of use.
I do hope that the Leaf cells will indeed remain reasonably well balanced over time for a long time, so that I won't need a BMS. That would be best. I will see if that is the case or not. I can live with having to rebalance the pack once a year or so manually, but if I have to do it more often than that, I will probably buy the BMS just for that. And if the pack does go out of balance more often, I will buy a BMS to keep it balanced automatically. So I have not yet bought one, yet. I'll use the bike a while and see how the pack does.
What I did buy (and have yet to receive) is a per-cell voltage monitor with harnesses to plug in each cell. I bought half a dozen small voltage monitors (each monitoring from 1 to 8 cells). I have seen these sell for as low as $1.99 a piece on ebay (with free shipping!). I only need 5 for the 36 cells I need to monitor, got one extra just in case one of the batch is bad. They have adjustable low-voltage alarm and are easy to use. Also iff eBay, I got enough cables with couplers (for 8 cells each) for these meters ($12 or so for the lot). I will have these leads on one end connected to the Leaf half-cells, on the other the connectors will be under the seat, easily accessible. There I could plug in these tiny voltmeters during charge and/or use them to monitor the voltave on each individual cell during riding and/or at rest. This way I will know reasonably well if any individual cell gets out of balance.
Here are links to the v-meters and leads, FYI:
http://www.ebay.com/itm/171141996803
http://www.ebay.com/itm/131150798718
What I don't know yet is how these v-meters work in terms of using some energy from the battery - some of these things suck power from one cell while monitoring many cells, which itself causes disbalance of that one cell. If that is the case, I will not keep them plugged in for long. But if I determine they use power from all cells equally, I might as well let them stay connected permanently (or at least while the bike is on/charging).
Thanks, I'll check these out. I would need 2 of them, so $300E... If I decide to go for a BMS eventually (not going with one at first), for more than a couple of hundred $ I'd probably go with a ful-featured brand-name system where I can hope for some support/warranty (even if it costs more).
Are there BMS system known that offers 36S to 42S? I can't find so quick on the Internet.
KOCHO
Have you used the latest Laird firmware on/for Leaf cells to validate your comments above or are you referring to your experience with NiMh?
I have seen the same fault occur when using ESD chargers
the fault is not likely with the firmware loaded, or if it is, it's a bug that has yet to be found. (I have seen it happen with several different firmware varieties)
There are a few theories, relating to possible flaws in the firmware, I have one blaming the hardware, but nothing really concrete.
If using an ESD charger, you must have an automatic means to disconnect the charger if it doesn't stop charging when the battery is full
it just happens that a BMS will do this when any one cell is full, but there are other options too
Daily Ride:
2007 Vectrix, modified with 42 x Thundersky 60Ah in July 2010. Done 194'000km
Which comments? I made several :)
If about the failure to stop charging due to voltage cutoff, I refer to the use of Laird's software on 2 separate NiMh bikes. It does stop due to time limit. It usually works, but occasionally hangs. I've witnessed it several times. For my Li conversion I am not using Laird's firmware, so have no observations about his 60Ah version for the Leaf cells.
If about the balancing, I do not have observations about disbalance/rebalance yet. I am saying the Li Leaf cells are not intended to self-balance through overcharging the way the NiMh cells are.
One observation/analysis I do have on the Leaf pack is that 150V charging is perhaps unnecessarily high for 18 Leaf cells - that puts them in the steep section of the charge/discharge curve, with individual cells at 4.2V - that is near 100% SoC and, presumably this is not ideal for the longevity of the cells. A 148V total pack voltage would be better, with minimum loss of capacity in use (perhaps 5%).
Another ESD observations I have is that the ESD charger voltage (which is displayed during charging) is one to three volts higher than the MC bus voltage (which is displayed with then kill switch/left brake). And (at least on my last ESD charger) the measured voltage during charging is half a volt to a volt higher than the voltage I measure with my multimeter at the battery terminals (and I have no idea how accurate is my multimeter, but I think it is probably more accurate than the ESD). Folks should be mindful of these in estimating what the pack voltage actually is.