pip-MGX 64v max charge 14s or 16s

M

MasterCATZ

New member
Joined
Jul 13, 2018
Messages
18
http://www.mppsolar.com/v3/catalogs/PIP-MGX.pdf


max charge 64v

14s 54~56v , bottom 3.4v top 4v (then Ican still use my regular48v 500ah solar battery bank )

16s 64v , bottom 3v top 4v
(can not use any other battery bank here unless I cheat with a DC-DC step up buck converter)


I am just trying to decide how I should make a bank that will be made from scrap 18650 cells with capacities under 1000Mahwithout using balancers (I will still fuse every cell)


I was thinking of just making small packs of either 14~16s 1p cells or 28~32s 2p cells then just parallelthem all on the main bus bar (makes it easier for me to find matching sets)

I have been finding it hard getting 18650 cells as just about every laptop that passes through the store lately are not 18650

currently I have70KwH worth of solar power going to waste (offgrid) and use 120KwH daily
and need to have more storage while18560 cell qtytrickling in
6 years and only 1/4 of my goal
I will not pay more than 50 cents per 1000mah , I normally buy 12v batteries for $50 / 100ah
 
Personally - I'm very happy with 14s (48v noninal) - gives a wide range of options compared to 16s.

For cells -not sure where you live - but these 18650 cells are 50c / cell for 3000mah / cell andappear brand-new -https://www.batteryclearinghouse.co...lls-in-ring-packs?_pos=1&_sid=36c595d0b&_ss=r I've made several purchases of these RING packsover the last few monthsand had 2 of 1,000 that were 0v and the other 998 all tested 3000mah (100% of new) and appear perfect:)
 
Will still need to use balancing, aka bms, with "scrap" cells. There's no way around it. Well, unless you do a 1s configuration.

You can still make a 14s2p setup with multiple strings. You'd just want to include a parallel connection between each series connection. This would electrically make it 14s4p (for 2 strings), 14s8p (for 4 strings), etc. This gets away from having a separate bms for each string.

Then if you'd need to disconnect a string for any reason, you just disconnect the parallel balance leads and pull the string.
This could be done easily by having a jst like connector for each string with a jumper wire that has 12 wires (only connects across the strings from between cells as absolute pos/neg are already connected by the buss).
 
Thanks for reply

really interested in thoughts of if its worth while using more cells in series like 14 / 15 / 16 as these could be limited to max charge 4v and could all be setup for minimum 3v discharge

in AU country , always seem to miss out on the good deals here hbpowerwall seems to snap them all up
tesla batteries pop up for $2k but I feel they are only worth $600 compared what I pay for deep cycle batteries

you have me lost @ 1s configuration did you mean 1p ?
I thought he idea is to match capacity in every-bank and not mix ?

I was planing on doing something like

14~16s 500mah (14-16 cells) 42~56/48~64v
14~16s 1500mah (14-16 cells) 42~56/48~64v
14~16s 2500mah (14-16 cells) 42~56/48~64v

= 14~16s3+p ~4500mah
and just keep adding 14~16s packs as I found matching sets ?



just trying to find something expandable making use of minimum matched cells and giving some of these 100mah cells a reason to live , just about all the cells I have bought from ebay / gumtree are absolute crap

even the brand new $800 13s4p battery packs I started off with at the start ~ 2014 did not last long paid premium price $15 a cell and still got shafted

I was just going to use fuses to kill off the entire 14~16s1p packs if they became unbalanced
 
I was making a reference to the fact that you seem to not want to balance. The only way you can get away from that is to have a 1s string. In other words, only 1 cell in series can be used without the need/warrant of using a bms, or balancing. If you go with 2s or higher, you will need a bms for balancing and monitoring. There is even a bms for single cells, but that's only limited to over/under voltage and over current protections.

You do want all series packs to have as equal as possible capacity as the others. The closer, the better. Even having 1 pack be 100mAh different than another could cause problems if you do full cycle charge/discharge often. Especially if not having some sort of bms to monitor and balance the packs.

Fuses do not protect packs from each other. Fuses only protect a cell from within a pack. A pack in this instance is a group of cells connected in parallel. A string is a bunch of packs connected in series. Just so we have the right terminology here.

Yields of a range of 3V - 4V /cell
14s: 42V - 56V
15s: 45V - 60V
16s: 48V - 64V

14s fits "most" applications. Most charge controllers and inverters operate in this voltage range. Remember, these devices were originally designed around lead acid batteries. A PbA battery voltage range is about 11.1V - 15.0V. So at 4s for a 48V system, that range is 44.4V - 60V.

I suppose if you specifically operate at 3V - 4V range, you could use a 15s arrangement. There's nothing "wrong" with using this, so as long as your equipment can handle the voltage range.

However, a LiIon cell actually has a better life at no lower than 3.2V. So this changes the equation a bit.
14s: 44.8V
15s: 48.0V
16s: 51.2V

As you can see, 14s is now within the lower range of most "48V" systems very nicely. And then, if you went to 4.1V as the upper end, that would make 14s be 57.4V. But, this is starting to push the cells. So I would suggest 4.0V as you mentioned earlier. Just make sure your charger will terminate at 56V
If you went with 15s anyways, then 48V - 60V (3.2 - 4.0V range) would still work as most chargers/inverters can handle 60V just fine.

As you know, more voltage means less amp load. But, more cells in parallel also means lower amp load per cell. So both designs will work to lower per cell amp load.
 
MasterCATZ said:
really interested in thoughts of if its worth while using more cells in series like 14 / 15 / 16 as these could be limited to max charge 4v and could all be setup for minimum 3v discharge
Click to expand...
Typically you set 'max charge'thru charge controller absorb/float voltagesettings. Its not a 14s vs 16s 'thing' fora properly matched charge controller - its just a setting that you can bump up or down as you wish. There isn't any power advantage between16s vs 14s - you can just make 14s have more cells in parallel for same power.

I would (gently) suggest youthink more in terms of 3.4-3.5v / cell low voltage instead of 3.0v. Typically there is a sharp discharge curve after 3.4v and notmuch power between 3.4 and 3.0v. You'll find that if you try to discharge down to 3.0v you'll likely have 'wider' balance differences - not worth it.
 
thank you both for chiming in again
your right 3v is the dead side , been a while since I played with li-Ion , 3.7v used to be my cutoff 4.1v was my max

https://github.com/stuartpittaway/diyBMSv4 was what I was going to use then found out they had a 64 module limit and the modules only do 1 cell

just trying to find a cheap way to balance every cell individually because ideally I want the packs to only be 1 string because I can not seem to hoard enough cells to ever complete large packs and 4x5 cell holder 100 packs seem to be the cheapest option to use = 1000 cells for $60
all other sizes cost up to 30 cents a cell ...
https://www.ebay.com.au/itm/4x5-Cel...-Shell-Plastic-Holder-Bracket-AU/153186367393

mainreason I was looking towards 16s was it was an even number to make use of a cell holder better and allowed + - terminals to be where I wanted them for mounting

anyone else here done single string packs? or do you all do large parallel backs then string those together on the bus bar

also not wanting to loose a lot of capacity if I have to remove a failed string
I think the best I could possible build would be 15s4p packs I only have a few matched sets with over 70 cells and that took me years to collect

originally I was doing 13s4p packs and their efficiency dropped to 50% after a few months (brand new cells) and died after 1 year
mean while I am getting over a decade out of my x telco deep cycle batteries and can still get $$ for them selling as scrap
only difference is the charge time
 
tbh, doing a lot of single cell strings will be harder to keep balanced than a large parallel pack.
No to mention the amount of wiring and upkeep to do it.

I think you'd be ok to do a ~100cell strings. If you went with 15s, that'd be 6 or 7p (90 or 105). That should be doable, even with limited cells. Then you can build as many parallel strings as you want/need.

As far as stuarts bms, each string is it's own entity. So if you had a 15s, that'd use 15 units, plus the main controller. Then another 15s would have another 15 units plus a controller. You don't have to hook them all up to the same controller.
Granted, the more you have, the more difficult it is to keep track of everything across controllers. I suppose you could have some scripts written that'd sync them up.
I state this because you want to make sure they "all" balance at the same voltages, and "all" have the same upper/lower voltage limits as well. Otherwise you could have 1 battery bank trying to feed another, or you may end up loosing part of your battery capacity because several strings cut disconnected because the bms did what it was programmed to do.

Again, I would not do 1p anything. Waaaay to much hassle. I suggest doing 10p minimum.
 
well what would your thoughts be on spring loaded holders ?
I do not mind paying $1 a cell holder if it means I could build something expandable that I could start running now

https://www.ebay.com.au/itm/18650-C...IY-Battery-Golf-Cart-Fool-Proof-/333452699433

then I could pop in cells as I get them ? maybe setup banks based on mah
I could set them up laying on shelve/slideout drawer to rule out batteries falling out
and could allow denser packing then just hanging on the wall
I could do a few parallel sets with bypassswitches to disable a group to swap cells out with
or would the total mah still need to be matched up with other banks ?

Bank 1
1 2 3 .. 14s =~48v~1000mah capacity
500 1000 250 600 mah
500 250 400 mah
................250
................250

Isolation switch

Bank 2
1 2 3 .. 14s=~48v~450mah capacity ?
100 800 200 200 mah
500 250 400 mah


I would work something out to spread the 250A charge current around
(15 kw solar inverter but the deep cycle should take most of that load and Li-Ion can take the edge of the inrush peaks)

maybe 4x banks / shelves~50p with BMS's =$280 if I use thehttps://github.com/stuartpittaway/diyBMSv4
or see if I can gethttps://github.com/simat/BatteryMonitorworking

~2800 cells that I doubt i would ever reach unless I import from other countries


if I could get my hands on a few thousand cells I would have done a heap in parallel years ago ,
but for now the 100kg deep cycle battery's have worked out easier to source on the cheap buying them as scrap then reselling them as scrap latter on

drives me nuts how I can see band new cells for sale over seas for $1 / 1000mah and 10x the price here


anyone using these ? or have any idea what they plug into
https://au.banggood.com/13S-48V-5A-...4.html?rmmds=search&ID=49920&cur_warehouse=CN
 
14s all the way.its the most nominal layout.
 
MasterCATZ said:
well what would your thoughts be on spring loaded holders ?
I do not mind paying $1 a cell holder if it means I could build something expandable that I could start running now

https://www.ebay.com.au/itm/18650-C...IY-Battery-Golf-Cart-Fool-Proof-/333452699433
Click to expand...

honestly, those cell holders would be trash to use in this application. They are only good for small applications/projects, portable even. And definitely nothing that required a decent amount of current flow, anything more than 500mA (even that is pushing it). The tabs are nickel coated steel. Not very good coating at that. There's a lot of resistance build up at moderate to high amp loads. There's a video/images somewhere on here that shows the tabs heating up a LOT under 1A load.

Also, $1 per holder is expensive and would eat up all your savings you got from reclaimed cells. You'd be better off taking all the $$ and using it for brand new cells.

If you insist on cell holders, use ones that use beryllium copper alloy. Not only do they handle the current better, they also don't loose their tension over time.

Also, you do not want to run used cells to 4.2V. You have no idea how many cycles they already have and by going to 4.2V, you are assuring them a short life. Just by dropping to 4.1V you can add 100's of cycles. Dropping to 4.0V you can add over 1000 cycles.
A cycle doesn't necessarily mean from full->empty->full. A partial discharge to about half and then back to full would be considered a cycle as well.
 
that was why I was wanting to go for the higher 16S so upper cell voltage was lower and would be impossible for the mpp to over charge
now just waiting on my diybms boards to arrive
I am hoping 1 controller can do 4x 16s banks , just unsure if the controller can only do 1 bank at a time or not
min order was 5 controllers anyhow so no biggi just sucked having to order 30 batches at a time

as I have 1.2m heated bed I decided to 3d print my own cell holders inspired by https://www.thingiverse.com/thing:456900
how ever I have my positive and negative opposite of their design
used just under 50m of filament per 40 cell pack ~2hr each to print
the negative contact fingers are made directly from 2mm x 40mm x 800mm copper flat bar (should handle 300amps )
by cutting slits every 10mm and rolling back as fingers and have positives with spot welded fuse fed through the printed spring hole , this allows tension on the negatives bus bar

16s40p x4 banks

allowed me to do slide out shelves using my old server rack rails and be able to swap out single cells easily
using flexible welder cord to link the shelves ,
ended up with ~2k 3000mah cells
and a heap of others around half of that
what peaved me off the most is the cells that worked out ~1500mah cost me $15 a cell (new $800 e-bike battery packs ) while the 3000mah ones cost me $1.5 a cell (used tesla modules)

how ever now that I think about it spotting a blown fuse will be hard ... next time I will run them over the plastic spring
can only melt through it
I was going to do threaded bolt / nuts but then thought knowing my luck I would over tighten and explode
 
You must log in or register to reply here.
Back
Top