Top Balancing, Still Worth It? - Part 8 - Big Battery Build! (Ep. 45)
Вставка
- Опубліковано 27 лип 2024
- LiFePO4 - (M)SDS Documents -
Relion - www.continentalbattery.com/as...
Canbat - www.canbat.com/lithium-iron-p...
PowerSonic - www.power-sonic.com/wp-conten...
Original iCharger X8 "Learn with Me" video - • Learn with me; iCharge...
0:00 - Mea Culpa
0:50 - This Videos Question
1:15 - Your Help
2:15 - Your Help - Stik Dragon
3:43 - Your Help - Mike Gray
4:43 - Your Help - W. Lou Little
6:41 - Your Help - Up North And Personal
7:23 - Your Help - Like Farmer
9:01 - Your Help - Paul Freeman
10:58 - How I charged Before Top Balancing
12:15 - Charging oddity with Pack A and F
17:19 - Last Cycle, Showing Pack Deviation
19:05 - Pulling the Packs Apart
19:48 - A Mistake is Found...
22:06 - A Tatters Interlude
22:26 - Another Mistake is Found!
24:02 - Checking Polarity
24:40 - Balancing Begins!
27:30 - Remembering How to Use the iCharger X8
30:03 - Charging Voltage Too High?
32:20 - Trust Your Tools!
33:02 - Letting Banks Equalise
37:33 - Balancing Again to a Lower Absorption Current
40:08 - Final Thoughts - Наука та технологія
The authenticity of showing us your errors is much appreciated Madi... anyone can be 'momentarily' distracted no matter how careful they are! Making detailed final check lists looks advisable perhaps? Sure is a 'motherlode' of 'nuts' to keep checking on though! Crack on!
I have so much battery envy right now.
Thank you so much for taking the time to film your process, for me it was very helpful and useful.
Thank you for sharing and as always keep building 👍
Heh, then I see people who have twice what I have... lol. It comes down to "do you have enough?" Once yes, then all good. :)
I've not top balanced a large 48V pack ever since I've started using the JK BMS for my batteries. I still do for smaller 12V packs since it doesn't take a lot of time and a passive balancing BMS like the JBD just doesn't perform as well for this. In all other cases, just hook up the BMS and start using the battery - it will balance just fine over time.
Bingo. 143kWh. Zero top balanced.
By the time I finished, I realized the answer was clear, which is exactly why I opened the video with the answer. I saw no reason to try to leave the question open.
really, well you will figure JK only balance the highest to the lowest cell but if BMS does not know the 100% reference state, you will figure over time that you screwed up.
@@chrisr819 It's based on voltage: you only balance at the upper knee. It doesn't need to know the capacity at all, or what the 100% state of charge is. I power a house with this set-up, in the third year now, and no issues with balancing. Maybe once a year I take the pack to 3.6V/cell, otherwise it doesn't go past 3.5/cell. Balancing with the JK works perfectly fine.
@@upnorthandpersonal In my opinion, people are way too tense when it comes to balancing, I use 8 cells (Eve LF280K) with an SBMS0 that balances with only 150/200mA and have also no issues with the balancing, 10mV delta with a small increase in the upper knee to 40mV, after charging stops at 3,55V cell the cells fall back to the resting voltage of 3.37V and the delta is 10mV again, no initial charging done.
It’s ok..
Wow... so many of the things you encountered in this video, bringing back memories of the things I have learned (the hard way) over the past couple of years. It's so good to see things like this, as it reinforces the need to check our own work. Well done for presenting it so well.
A snippet (don't ask for the long version) of my own procedure for building batteries (because whether you see it or not, you're now manufacturing batteries)...
I get the raw cells, check the voltages and physical appearance.
Then they get packaged into the 4S pack and BMS attached. (one of my new rules, is that NOTHING other than a cat in distress interrupts the use of a torque wrench on each cell nut).
My next step is to put a 2kW load on the pack, this is used to a) calibrate the BMS internal shunt against a calibrated Victron shunt while setting my own parameters via the PC App, and b) show up any potential issues. (we'll visit this again soon).
Then it goes to a Bulk Charge bench where 8 batteries are all in parallel, on a 60A charger. They sit there for about 48hrs. The first 24hrs is putting the bulk charge in, and the second 24hrs is allowing the Heltec Active Balancer to balance the cells down to 0.002v delta or better. This has taken many 'people hours' off the charging setup, with no bolting and unbolting of cells.
The other thing this setup also does, is highlight if a certain cell is misbehaving, because in a 4P (or 16P in your case) setup a misbehaving cell is concealed amongst the others. This has been VERY valuable.
Next they go to the test rigs, where I again put a 2000w load, and after 6Ah has been drawn down (approx 5%) I take a voltage reading at the battery terminal. While this reading is not great on its own, I've found it very helpful to show up a potential issue within a batch. A low voltage can highlight a bad connection.
Then straight after that, I check the IR with the same meter that you and Andy use. A high IR (relative to others in the batch) can also show up a bad connection or issue.
Only then, providing everything checks out, do we continue with a full 2 cycle discharge-charge test to rate the individual battery.
Once 2 cycles are done, and the 2 results are similar, we take the average and that is the "actual" capacity of the battery. Anything 1% lower than the expected capacity is deemed a reject or NQR battery, and will either be sold off cheap as is, or the offending cell swapped and the entire process repeated.
So a good way to check for potential bad connections, is to put a high current charge or discharge, and look for differences between each battery pack.
Every setup will have variations, but if you can see consistency within your 6 battery packs, you know they are done right. Anomalies are easy to spot under high load. If the room also changes temperature a lot, then you will also notice significant voltage differences between hot and cold. In my setup, it can range from around 15 deg C up to over 40 deg C, so it's very clear on my datalogging spreadsheet when the room is hot or cold.
Oh and thanks for the mention too. I think people focus too much on absolutes lately, and neglect to consider that there are huge ranges in everything we do. There will be good and bad of anything, and there will be better or worse than both of those moments after we discover them. So we all need to make educated decisions about product choices according to our needs.
Even with the low current, I could see the differences. My mistake was thinking the problem was inherent to the cells or BMS, and not re-checking my work. Lesson learned...
I will be sure to create an installation check-list. I need to think of a way to make the front of the packs, BMSes and whatnot accessible without a disassembly of the packs once installed. Periodic checks will be a pain otherwise.
Glad i could be helpful :)
When charging the reason you see the difference between charge voltage and cell voltage is because the charger is attempting to force a set current through the pack, and the only way it can do that is to overcome the resistance of the both the cells themselves *and* the wiring. The cell voltage is measured at the cell, and the charge voltage is measured at the charger. From this you can work out what the effective resistance it's seeing is by applying Ohm's law, which is V=IR. You know that between the two measurement points there's a voltage of about V=0.25v when a current of I=30A is flowing, so R must be 0.008ohm. This is a relatively tiny resistance, but at such high currents it starts to become significant, nonetheless. You can also approach this from the other direction: that looks like about 12awg wire? A quick search suggests it should have a resistance per meter of 0.00521ohm, and i guess you've got about a meter or so of wire in total (plus connections, etc), which correlates.
As the charge current drops, so does the voltage drop in your cables and bus bars, and so the two reported voltages from the charger start to approach each other.
You can calculate power loss in your cables this way too. Watt's law is P=VI, so in this case your cabling was producing 7.5W of heat. This will happen when you're using the battery too, the current flowing will heat the cables. The fatter the cable, the less resistance, less voltage drop, and so less heat. Wire ratings are based off this.
One thing i periodically do with my lead-acid pack is measure the voltage drop between each cell's battery terminal while it's under heavy load (i turn on the electric cooker which gets about 100A flowing) and this lets me estimate the resistance of the interconnects. I'm not worried about the absolute value, but if i see one that's much higher than the others then i'll know i've got a bad connection there. In your case you've got an awful lot more of them, so this might not be as practical to do! The important thing is that you jam the probes against the metal of the battery posts themselves, so that you see the voltage drop across the interface between the metal surfaces too. Your FLIR camera might be good for this too, as you'll get a slightly warmer interconnect if there's a high resistance somewhere
Great comment!
Once I saw that it wasn't a problem, I realized what was going on. That's why I said that I was impatient (and perhaps a little paranoid). The wire was 10awg, but it wasn't marine stuff, it came with my solar panels and had few, large strands. I didn't think much of it, which is why I used it up for temporary applications like the charger. I'd not be surprised if the resistance was a bit higher than expected.
The FLIR is how I think I'll go about checking things... Once the packs are assembled in the engine bay, before I close them up, I'll find a way to load the pack up for 10~15 minutes and check for hot spots. Only if all looks good will I close them up and slide them into their final position. Once in place, accessing them again would be a right PITA, so anything I can do to avoid opening them again will be worthwhile.
"its a good job I have no pride" - love your attitude, thanks for being real and helping everyone learn.
Scary potential current there!
16P
Drop a bus bar or a tool and it could be fireworks.
Looks like a lot of fun!!!
What if you had 15 charged cells and one half charged... Put them in parallel, and there is no limit to charging current!
Oh dear Maddy I feel your pain. Thanks for being so honest- so people many aren’t!
We’re all human and we all make mistakes- which is why we say check, check, and cross check again.
I could have told you right at the beginning with those voltages flipping around (up and down) so quickly that there were a few high resistance joints. (I was silently telling you that as I watched the last video.)
I could have also told you that the JK BMS will balance them just fine in time thanks to the great 2A balance current.
I guess that is a lesson that you will never forget. Stay safe and keep smiling, and g’day from Brissie 🇦🇺.
A short post for you. A FLIR thermal camera would be a fine safety addition for the build and the boat, you can pick up one that plugs into you phone on eBay for a few hundred pounds.
Very useful, I've got one I've shown a few times in past videos. :)
@@TheDigitalMermaid 👍🏻
@@TheDigitalMermaid I remember that you used the IR camera to see the temps of the BMSs and screens - did you ever happen to point it at the tops of the cells? I am curious if it would have noticed the flaky connections. Might be a fun experiment to undo one of the nuts once you're all back together and see what happens!
@@BillCarlson I had the thermal camera, not out tops of the cells were hidden behind the cardboard. I'm about to do a capacity test, let me see if I can screw up the courage to loosen a bolt... :)
For large setups like this it seems a necessity to me.
As an old systems engineer for key production processes I can not stress enough that the next job just begins when everything is working. Soon it will show an anomaly, degraded performance or total system down.
What are you going to do when that happens? Be prepared and have your maintenance procedures and fault checking already thought though.
Especially on a boat. You need to perform when you are weakened, tired and the elements beat you up. What can you do and what can't you do when at seastate. Stick to those plans. Your life depends on it. A short will end things rather dramatically.
Good amping you pioneers !
Love to see more. These are exciting times at the brink of energy source turnover. Hope to see as much as I can from this transition.
Good job.
So now you know how to troubleshoot your packs by paying attention to your cell voltages during cycling. The two trouble cells were obvious places to double check connections and crimps. You could have saved yourself hours of rework. It also proves that your initial setup was correct and would have worked flawlessly once the two bugs were fixed.
HOWdy D-M,
Thanks for showing us the mistakes that can OCCUR when doing multiple - Multiple - MULTIPLES ( of anything ) !!!
&
Although you state "SAFETY GLASSES" ... it scares me seeing you tightening SOLID / RIGID Bus-Bars with your socket & Ratchet in Reading GLASSES only
Thanks for taking us along on the DIS-assembly + RE-assembly
the WiSeNhEiMeR from Richmond, INDIANA
COOP
...
The voltage on the charger is "Cell voltage + voltage drop". 30Amps through those XT90? connectors, solder joints etc. etc. That could well get you a 0.2V drop and it's why it has the sense leads.
You mentioned 'life critical' in regard to the BMS. I understand wanting a reliable battery but you have so many options on your boat. If the batteries are working but the BMS craps out you can bypass it. You'll lose the protection and fine balancing but your chargers will still work just fine. Of course you'd need to spot the BMS has failed before any damage is done to your battery.
Nice video, thanks for the hard work. I have also experienced that creating a large and reliable battery without mistakes is extremely difficult and tedious. Keep up the great work and I look forward to seeing the next video.
Your assessment of how the BMS works and the brief voltage spikes on the cells is correct, because it measures the voltage across the cell and then activate mosfets to discharge the cell which takes some time.
The main thing that occurs with cells is they overheat, which destroys the plastic separator and melts at 250° f if it melts it creates an internal short, which causes the cell to continuously draw current.
When discharging cells connected in series, I have discovered recently by doing it that if one cell reaches 0 volts it has a large increase in its internal resistance and shorts internally because it heats up very hot. It then becomes a resistor, and can no longer hold charge. It is extremely dangerous because shorting large amounts of current causes large amounts of heat which will destroy adjacent cells by heating them up.
well she thrown away the advantage when you build your own big battery by having multiple BMS effectifly coping drop in battery setup. all EV makers with billions in R&D parallel the cells till they get the capacity they need and then put these packs into series to get the desired voltage and use 1 BMS to control all. There is not a single one who is using multiple BMS to control each pack. Multiple BMS are peanuts in costs compared to the cells, it's not their costs. And EV makers top goal is operating security and its wise to take this into account. propulsion setup is equal to EV. so is it really the right setup. for me definitely not plus JK BMS is made for storage setup, not boat propulsion setup and 'issuing a lot external safety features needed....eg what happens if the 11kw motor gets locked from full speed to 0 because of a fisher net gets into the prop...reverse current will destroy the mosfets of the JK BMS and all BMSs are fried...just one use case of nearly hundred you need to take care in propulsion LFP setup
I guess you now have a new appreciation for the surgeons putting robocop together and keeping the central nervous system intact (balance lead spaghetti!).
May not be a bad thing to add to your annual-ish maintenance schedule, go around and using a torque meter, re torque all your connections, enough waves on the hull will translate force and over a long enough time line, everything on boat comes loose...
You could create a simple visual indicator with a little paint stroke spanning the nut and the busbar on every one so you can quickly visually identify any connections working loose without needing to get hands on to every nut with a torque wrench :)
Great lessons learned and I'm sure Andy appreciates the nod :)
Stay safe!
Torquing with a witness mark is absolutely part of the plan. I'm not sure if I'd retorque each year though, as it would require a major deconstruction of the engine bay. Torquing and testing before sliding into place though, that will certainly happen.
" am I going to watch my own Video " so funny :)
Nice video. After many different scinarios. I too decided to top balance at 1S, 16P. Threw 3V62 into them with at up to 50A worked a treat. The reason I did this was simple, the recovery time charging at 16S, 1P with a JK BMS (when 1cell hit 3V65) was extending the time significantly. Nice background music during the pack strip as well🙂
What did you try to convey to us diy people?
@@sharyncomstock7066 Sorry if I did'nt put this across too well! I find the majority of us "top ballance" via a BMS. i.e. using 12V or 24V etc. into multiple cell packs (4, 8 or 16). When doing this, one of the cells would reach 3V65 as programmed into the BMS, this would close off the entire pack charging until this cell recovered to the lower set limit. This temprorary shut down impacts on the time it takes to "top ballance" the pack. So I found "top ballancing" at 1S 16P (for a 48V pack) saved time in the long run as no shut down occured until reaching the "set point" i.e. 3V65. Thanks for taking the time to read my comment!
i really like your honesty
Thank you :)
My 2 bits after top balancing a couple of large packs is definitely don't bother bolting them together as an xP1S battery. Every time you screw with the bolts you risk stripping one out. To actually do the top balancing use the JK with the balancing enabled at 54V and above with a maximum cell voltage of 3.6V (not 3.65V) and then use a LAB power supply set to 57V and limited to 0.5A. NOT some monster multi-amp high rate charger. This way the 2A balancing is just about the same as the 1/2A charging rate. The BMS will then only rarely disconnect on "cell overvoltage". Leave it all sit until it's balanced. That can take up to about a week. No excitement and nothing moving fast. Works well.
Oh, and you can use a higher charge rate to do most the charging just switch to the above once a 'cell overvoltage' occurs so a 1/2A charging rate doesn't take 3 months LOL. .
When reaching 90% SOC, reduce charging current. That will save the cells and improve the balancing.
I laughed when you said “not good” After watching Andy I find myself saying it even if things are good lol😂
There are so many iconic sayings... Another one I catch myself using is Big Clive's "One moment please..."
@@TheDigitalMermaid lol
Hello, I think you need to learn that the voltmeter is your friend. You need a good multimeter you can trust. Then, if something doesn't seem right to you, just take it and check it. If you had done that when the weird readings started appearing on the screens you would have found the loose or missing screw right away. As they say, measure twice, cut once....
yay sort of missed your videos
I have set a lab power supply for the first charge with tail current of .8amps to charge to 3.55 per cell. ... With jkbms of course.
It works flawless the whole season. Minor differences of .01* volts during solar charging and using an electric oven draining the battery to 20% it stays that way
Obviously it is highly dependent on the cell s er receive and their ohm resistance.
Note I used a digital torque wrench and drilled holes for the balance leads on the original busbar from China.
I noticed one pole was very loose! In october I moved the battery. Would never have found out until too late. Wise lesson for me to do proactive maintenance.
Checks and maintenance are key, for sure.
Torque wrench or adapter on each connection. Mark each after torqued.
Consider the REC BMS. It isn’t cheap but neither is your safety. You’ll be pushing a lot of watts each way when you get this in the boat. Having canbus to manage the system is a huge advantage
Great content!
REC looks solid, no doubt, but the lack of active balancing is why I originally didn't consider them. They also don't seem to interrupt charge/discharge, I don't see FETs or relays in their design. I assume they command loads / chargers to disconnect? That makes me nervous, if so.
@@TheDigitalMermaid it’s all through the canbus and with external contactors. Instead of breaking the circuit thru the negative side, the REC external BMS sends signals to contactors (large relays) for allow to charge and allow to discharge (ATC & ATD). This, 500A (or more😁) can be controlled safely. The internal (FET based) have practical limits of 200a or so.
Full disclosure: I went with Victron smart Lithium (Lynx BMS) on my boat. I couldn’t get past the science/engineer lab look/risk of the diy battery bank. I looked at the $ vs $$$$$ and thought about it from a buyer’s perspective: do I trust what the mad scientist (me!) built or would I rather buy something with a lot of blue and trust Victron’s engineering 🤔. It was hard to part with the $$$$, but insurance isn’t free…
That said, your build is awesome. It’s tailored for your boat and systems. VERY impressive.
Another big advantage is that they integrate into the Victron ecosystem very well. The BMS becomes the “master”’of your chargers (Quattro and mppts) so the batteries get what the BMS wants and the charge sources can be prioritized (eg, solar, hydro, wind, shore, genset).
If active balancing is a major “must have,” you can add it as a switchable maintenance accessory. Should you find a troublesome cell or two, pull that bank offline, turn on the active balancing, and return the bank to service once you are satisfied. As you have demonstrated, there is usually a reason for a cell to be out of balance (corrosion, loose connection(s), or a faulty/mismatched cell itself. I’m not convinced ($.02) active balancing is needed if there isn’t an underlying reason for the imbalance in the first place. In fact, active balancing may mask a latent problem.
From the aviation world, don't walk away from a bolt or nut unless you torque it all the way. If you're not going to tighten it, don't start it.
@@carman3-700 Aviation is even more high caution than sailing. I'll be torquing and putting witness marks on all bolts during the final install.
@@TheDigitalMermaid the remote functions of eg a Victron Multiplus or Quattro is tested by Victron to work 100000times and will work more reliable then any other solution. You exactly don't wanna have mosfets in a propulsion setup as reverse current of a motor that get lock will kill these mosfets in the BMS instantly, the remote switch off not.
I balanced mine top balanced mine, in service one at a time with a power supply. I only have 4 cells though. Basically going around each with 5A and 3.65V and leaving it until it dropped below 0.5A, the move to the next and repeat a few times. After that it's MinMax has been around 0.007V. Except when the solar panel is spanking it at 14.5V then they can be out by 30 or more mV. Still need to configure that absorption phase down a little. I'd prefer 14.40V, but I think it's set for 14.5V because "AGM". Anyway, it boosts for 2 hours, I might shorten that to 30 minutes.
I top-balanced my 16s pack without taking it apart. First charged it up with solar to voltage cutoff. Then charged each cell full with my iCharger moving it from cell to cell. Took a couple of hours, perhaps not as good as parallel them, but i didn't need to disassemble the pack at all :-)
That works of course... Not sure if I would have saved time though, with 96 cells, haha.
Having built (and operating) a bunch of these packs myself, I'm starting to question the use of all these upfront balancing effort .. why ? .. simply it does' nt change the fact that some cells might have a slightly smaller capacity than others. What I do instead now is to "heal" such a pack, by first locating those cells (typically they are always the same that get 'full' as the first ones triggering OVP .. and putting 32700 cells with 7Ah (or smaller/bigger ones) parallel! to the affected cell(s) ... This really helped me on 2 packs already. Btw: Putting those cells in parallel is not without danger: Imagine you have 'one' bad cell that will eventually come up with an internal short during that time.. there is no fuse no nothing preventing the entire pack from "melting" ... !
Anything can break no matter what the cost which is why resilience is designed into critical systems . Resilience can be as simple as carrying a couple of spares on the boat up to manual or automatic failover .
Exactly why I'm building them as six independent packs. Any failure means the loss of only 1/6th capacity.
Great video Dear, if you don't mind me asking; What books are you reading and/or have dead?
Thanks for sharing and Happy Thanksgiving🤝💯🌹🌹
I love the intro, reminds me of Leo's intros.... but he calls himself a boat builder, not a dumbass haha.
I really like the comment reviews... I'd love to see you do more of that (Maybe you do, I am still catching up on your videos)
Hehe. I reference comments when it makes sense... When the channel was smaller, it was easier to go over comments, but unfortunately (in a good problem kind of way) now there are too many. So I generally only respond to comments in the video if there's a strong reason to do so. That said, I do try to reply to as many as I can, if there's something to say. :)
@@TheDigitalMermaidI really appreciate that you communicate in the comments! This series has helped me understand so much more than the must faster paced videos you see that have a lot of production and assumption of a person's knowledge. Having this be your first time, and my first time really makes it easier to learn. Yes, I know electrical stuff, but I don't know much about BMS and JK and Lifepo4... again, thank you for this great channel. It's really building my confidence, especially when you mentioned JK is good for a home setup. LOL, yes I want a great BMS, but my home battery just needs to not burn my house down.
It happens to the Best of us! God bless! :)
I think with as many cells as you have, find cells that charge fastest and make a battery of then, down to the slowest cells and make a battery of them. That should make each bms not have to do as much to balance and keep charging without bms having to stop charging because some cells would otherwise overcharge.
They all charged at about the same rate, save for the two I messed up the terminals on.
@@TheDigitalMermaid Yep, saw that after I wrote that, but if you do see slower and faster cells and keep track, when you move them or take apart again, you can regroup into more naturaly balanced batteries, though I hope they don't show any in the future.
Just a suggestion in the future if you have problems with balancing instead of paralleling the whole pack for a balance charge use your 8s icharger to balance half of the pack at a time without disassembling it. Question how did you decide you needed 86kwh of batteries and how big is your solar array?
Hi Maddy, maybe a stupid question, did you verify the real voltage of the overvoltage cells manually? It might be a loose (or high resistance) connection the the battery + and -. Also when you top balance you might opt to set the Victron to first 3.45 per cell, then 3.55 and at last 3.6 per cell to not have crazy to make your life safer. Thanks for sharing all, this is very helpful.
I did not check the cell, which I absolutely should have, in hind sight. The Victron does pack voltages, I can't dial it in per cell, which is why I've got the iCharger. Cheers
@@TheDigitalMermaid what I mean was 16x 3.45 to start etc 👍🏻
hmm need to watch this again, lots of information here.....
You like pulling out spines huh? I thought you had a FLIR camera. so I would think you would have seen the two bolt problems with that. Curious if you did check out the connectors with the FLIR camera and if it didn't show anything.
Feels silly in hindsight, but loose connections didn't even cross my mind to check. Well, lesson learned.
@@TheDigitalMermaid I don't have a FLIR camera as they are pricey. It would be interesting to see if a loose or an unbolted connection even shows up as hot on a thermal camera. If you get a chance, would you be able to give that a try and talk about it. Might give me a reason to try to save up for one.
Excellent video Mermaid, it just goes to show how one poor connection can affect your whole battery. It never hurts to double check your work , i just put a cabinet face frame on upside down and i did not catch it until my wife pointed it out ? So you are not the only dumbass around .lol.
no way for a active balancer to balance the cells if the balance cables are not connected properly! Double check everything and double test I guess. I like watching others build packs so I can learn from their mistakes rather than my own!
Glad to be of service is showing things to avoid. lol
mistakes are bound to happen when you do so much... maybe should of had a checklist for each pack you did so you wont make that mistake again ... when people ask me to build them cables they think its a quick slap job when i have to like triple check things so i dont send out cables that are not on the right balance ends and when they plug it in the charger it wont read the cells right ... take your time and take breaks and make a checklist.... the most basic things are the ones that always gets messed up and miss look, but that's life live and learn from mistakes
Checklists are excellent, and something I need to make before the final install, for sure.
a trick to tightening ; take a crayon or chalk and mark the ones that are done. (old mechanic trick)
"Witness marks", I think they're called. Ya, I'll get a grease pencil for that when it's time to install on the boat.
I'm a bit late to your channel, but am enjoying it - I've a faily ,assize home ESS using EV cells - you've spent a lot of money (very sensibly) on quality Victron equipment - please throw that BMS away and treat yourself to a "proper" BMS such as something from Batrium or REC - they are so much more accurate, reliable and packed with features - what you get for the money is fine, but, given what you've invested in Victron kit, and no doubt you boat - the BMS you currently have is an expensive incident waiting to happen.
I've not yet found a better quality BMS that can match the specs of the JK... Particularly the active balancing feature on 16s. I'm ears if you've got specific recommendations.
@@TheDigitalMermaid hey - what sold me on ether Batrium or REC is their total (CAN) integration with Venus - the need for active balancing (which in my world is the ability to shuffle charge pack to pack within a battery) is a highly debated topic, if you've got a quality battery you won't need to active balance. It's hard to describe here how feature rich either of the systems I've mentioned are - you'll likely cringe when you see the cost, but you'll only need one BMS not the 6? per system - you have Rolls Royce victron equipment and good quality cells being managed by a less than ideal BMS that's ignorant of your system as an entity.
@@SeanMoore2008 I'm working on an alternative BMS, I'm not happy with the jk quality. However, for various reasons, none of the other options on the table appeal to me. I want six separate packs, for redundancy, and as the cells age (from any vendor), they'll fall further or if sync so in my opinion, decent balance current is a must.
@@SeanMoore2008 I went REC as they are one of the few that communicates with SMA. And they seem like a quality setup. I am afraid of magic smoke and fire and I’m on land. Even scarier on the water.
143kWh. None top balanced. If you buy a good cells, JK 16S 200A 2A balance does the job just fine.
I counted 96 batteries, are you going to connect them all in series ? they will be 300 volts or more 🐸😱
6 different 16S/51.2v packs in parallel.
Oh gosh Madi, that was painful to watch >.< I can bearly imagine how much worse it was to actually do it! Useful and interesting experiment though. Is a 99.11% success rate good enough? (2 out of 224 bolts for the 7 batteries built so far)
#TheyDidTheMath. haha! It doesn't sound so bad when put that way, but still, it was a rookie mistake.
@@TheDigitalMermaid maths is awesome xD you got to make a lot of rookie mistakes before you become a master.
@@MattWells0 TRUTH!
1 minute into the video i'm commenting: NO - you don't need to top balance!! the bms will do it on its own. over time, of course! BUT, it's not like it'll take YEARS to do... more like a few days! and, let's be honest, we're gonna have these batteries for YEARS. so a few days for them to 'top balance' themselves is fine. IM(not so)HO! LOL!
yup. that's what i thought. especially with the JK and it's 2A balance. i had the same thinking when building my 6 packs - build them and charge them at 56-ish volts, (to get the most energy in the quickest), then rebuild them and top balance them at 3.6-ish to finish. THEN i had the epiphany that the JK will just do the work for me over time anyway. no need to be that meticulous. and let me tell you - I'M METICULOUS!
How much is the capacity of 1 cell?
3.2v @ 280Ah = 896wh per cell
Some kind of Vent
You may want to check out electrodcus. He cuts off charging at 3.65 by default. If I were you I would start with a smaller battery just as sort of training wheels. And the DIY BMS seems good.
Check my older videos; I started with a single pack. :)
@@TheDigitalMermaid ok I'll have a look. Don't feel bad redoing. Yesterday my sensor cable acted up. Not knowing how to correlate the error my BMS was giving with which cell to look at, I just tore out the old one and redid it using new wires. Your channel is inspiring me to convert a truck into EV. EVE cells are so cheap!
@@surfreadjumpsleep Ooooh, what kind of truck? What's the plan?
Parkside x20 to Bosch@@TheDigitalMermaid hmm thought i'd already replied here.
It's a dodge dakota 1987 truck. It was my Dad's truck. It get's horrible gas mileage... so bad that i never drive it. If I could put some batteries in... It'd be perfect for groceries errands and trips to the lake. Wouldn't even need to big a huge battery as I have a toyota for longer drives.
I don't understand why car batteries need to be such high voltage. Is it because the motors are high voltage? Or is it because with that much power required, you don't want too thick cables running around?
With hindsight being 20/10, just kidding. I would highly recommend that you get in the habit of first checking torque whenever you get a flaky reading on a BMS. Retorquing a connection has fixed a lot of read errors I have seen in the early days of building battery banks. Sometimes if that doesn't improved it, the act of taking the connection apart, inspecting, cleaning and reassembling fixes the problem.
It is a good thing you are learning all of the ins and outs now and can work through the issues if need be before it ends up in the boat where it can be less noticable to see these kinds of issues unless you know to look for them. All in all, I think showing your errors and mistakes makes you more credible and knowledgeable since it shows you have taken the time to gain more experence and any opinions you have in the future on battery building are far more likely grounded in that experence.
I could tell you the last mistake I made on my system was forgetting to cool the Solid State Relays I was using for the 48V water heater I was converting. I had heatsinks attached but did not realize just how fast the 2 relays would heat up during testing with 20A going through them even though they are rated for 60A. I only had the power on for a minute and then realized that one relay was not shutting off and had instead failed short. After that I finished wireing up the cooling fan so that whenever the system shunt voltage allows the water heater to run it turns on the 12v signal to the water heater control and thus turns on the fan to keep the relays cool. Since I added the fan it has been rock solid. I still want to add a second safety that opens a contactor if the battery voltage ever drops way to low to kill all nonessential things on the bus bar.
Ya, I'm kicking myself that I didn't immediately check the connections on the odd cells. That was a seriously obvious over-sight on my part.
Doing all this learning on land is part of my plan. When the boat gets here, I want to be able to focus purely on the install, and get back in the water as soon as I can. The more I learn before hand, the better the chances of a quick(ish) repower will be.
I seen that over protection cell voltage happen the first week a lot. I didn't see my cells go over 3.68(3.65 is my setting). This week almost every cell is top balanced. I have 2 cells in 4 banks (2 out of 64 cells) that have not caught completely up. I bet next week they will be. mainly my cells are 3.5. I do see after morning charging I have a difference of .85 on average. 4-5 hours later balancing stops as they are all .004 or less. If I wanted to be anal I could take every low cell from each bank and build them that way. I do not see the need. After a couple months they will all be even closer. Sounds like you found your loose and missing nuts. I would have just did the 2 packs having issues. You will notice that the cells really like 3.45-3.47 to sit at once the bms balances them until my solar doesn't have sun anymore then they start running the house for the night. Please get a nano meter torque reader. I got the same one Andy has and I love it. I also checked my nuts and all are good after 2 weeks of powering my house. I do love what you are doing. I would also in the boat try to get rubber mounts for the boxes. that will help keep vibration from loosing them.
I've got a small mechanical torque meter (I like the click over the beep :) ) that I'll use for the actual install.
Just hook them up as a 16s 6p and be done with it.
That's a bad idea in my use case for several reasons. I need more peak power than a single BMS can provide, I'd be running the FETs harder even if I was in the range, I'd have no redundancy, etc.
3.8v or and 3.5v ... 0.3v difference becouse you have voltage drop ...
Have you seen diybms? Open source design that you can have made for you.
Doesn't that max out at 8S?
@@adon8672 does it? The designer of DIY BMS just put together a big battery I don't know how big.
That looks like an interesting project. When I first looked at the github readme, I thought it was for NMC cylinder cells, but I found the video where he built a 16s LFP pack. Huh...
@@adon8672 maybe you are thinking of SBMS from electrodacus? That's what I use now and it does max at 8S. Unlucky me to buy 90ah Winston cells about 1 year before eve cell price cliff.
@@surfreadjumpsleep Yes, I was thinking about the electrodacus SBMS. I'm sorry for the mix up.
Hi maddy, maybe you might be acting too quick sometimes (not a fault but a human behavior), sometimes having a drink and sit before ripping things apart might save you a lot of work. You might look at some of Andy's videos, he will first check the cell and current mostly to deduce the issue. Awesome that you manage to get all of this done! B.r. Edward
I absolutely should have done checks, that was a mistake on my part.
@@TheDigitalMermaid you'll learn from this, so when you are sailing with the batteries you already covered a lot of mistakes which helps then. Take care Maddy!
@@edwardvanhazendonk Yup, learning on the bench is a lot easier and safer than learning on the water.
Yes video akb 🔋96 LiFePO4...?!🔋👍 ⚡🔌💡🇺🇦
Yup, 96 cells in 6x 48v (16s) + 8 cells in 2x 12v (4s).