One thought, try measuring between a LOWER point on the stud and the base. The nut makes contact lower than just the top, the lower threads make a firm contact along the full length of the nut, not at just the top of the stud.
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
You are absolutely correct about the current through the welded studs. The current through each of the two paths will have a ratio that is inverse of the ratio of the resistance between the two paths. By the time you get the total resistance through the stud and nut it will be a lot more than from the pad to the busbar.
The most simple BMS, with no balancer or balancer turned off, just a voltage cutoff basically high/low on each cell. Leave it running 6 months and compare capacity before/after and see how balanced it stays without a balancer - and compare to those you are OCD balancing :)
Well, we have tested this already with the QUCC BMS a few times here. I rarely charge my battery full and if so, the cells are usually all over the place. SO I need a strong balancer which can handle that. So far only the JK could excel here. This is also more a test of how good your cells are matched.
@@stupidmonkeykev All depends on your system setup and what cells you use. I have seen set ups without balancer but they usually keep a top and bottom buffer and run the cells only in the middle part.
@@OffGridGarageAustralia i think your 3 delivery's in 1 battery may not be ideal. New setup maybe more stable. However i also think you panic at 0.01% capacity imbalance (100mv @3.5v) be good to see you leave 1 of your 4 48v batteries "unbalanced for months and see if that effects capacity. (Spoiler, it wont, but prove it :))
05:29 Why do the Chinese make small protrusions on the battery cell terminals instead of making them flat? If the contact surface of the busbar is wide, there will be less resistance and better current flow.
I would love to see the Batrium, I have seen lots of people using it, but I feel like I never really get any indepth information on how the system works, or how well it actually works, all of which I know you would test very thoroughly
This would be my suggestion too! Does the cost truly justify the performance would be the "Why"? The other question would be since you can have multiple batteries connected to a Batrium...can it keep all the cells in multiple batteries all in balance?
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
Hi Andy. I would like to see how the Batrium Core works. There should be at lot of parameters and monitoring possibilities for the battery. But the question are how good are they to the job.
Well the batrium is pretty pricey, and I don’t understand how it works. But would like to learn more about it. Other channels don’t cover it that well.
would really enjoy a seplos bms review. seplos seems very interesting and feature rich. i particularly like the idea of having multiple bms, connected in master slave configurations where only the master communicates with the inverter and the slaves only communicate with the master. this way, you don't have a situation of having more than 1 bms potentially telling the inverter different things....either by way of canbus or by sending a quick signal to see if charging current is available. also, the seplos can function with 16 bms in parallel, plus they also have that niffty lcd screen. the seplos platform seems to bring a professional features to the dyi market.
Yesterday I got my Seplos BMS. I'll test them tomorrow :D I got one 150A and two 100A versions. That'll be a lot of fun! So far I am using a REC-BMS...it works well but sometimes the SOC jumps from like 80% to 100%...I'm sure, that this has something to do with software and could be improved, if I tried harder ;)
@@OffGridGarageAustralia When one cell reaches low voltage disconnect, the inverter will switch off. When one cell reaches high voltage disconnect, the solar charge controller will reduce the charge current. You'll need a Cerbo GX or a other victron-device, that is capable of CAN-commucation. Also the BMS will basically replace your battery monitor.
@@OffGridGarageAustralia I could be wrong about all of this but from my understanding, where the value of the Seplos platform is, is the fact that you can have multiple batteries, each with their own BMS all linked to a master BMS that communicates to the inverter. So let's say, you have BMS1, 2 and 3, each with its own 48v battery where BMS1 is the master and BMS2 and 3 are slaves. If BMS1 has a full battery but BMS2 and 3's batteries still need charging, BMS1 will simply cut off charging to it's own battery but will not communicate to the inverter to stop sending charging current until BMS 2 and 3 reports back to BMS1 that they are also now full. It is at that point, that BMS1 tells the inverter to stop charging. I am not sure how that same scenario will pan out if again, 3 batteries each with their own BMS and neither BMS is aware of the other but 1 BMS has a fully charged battery and the other 2 don't. Would appreciate your feedback on all of this Andy....thanks in advance.
Just got 17 cells from Qishou for 148usd a piece. In stock in a warehouse in Poland, shipping time 5-7 days. Apparently pretty pricey but still a steal compared to what german resellers want for what is essentially the same cell. (They charge around 200€/210usd per Cell). Communication was great and straight forward, no issues at all. Even got the order and tracking numbers done on a sunday :) I think the matching comes from NMC cells where there is a direct correlation between the IR and the remaining capacity (Relative for each model of cell of course). Andreas Schmitz here on youtube has created a database that can predict your remaining capacitance based on the IR and can be used to evaluate salvaged NMC cells.
You should try assembling one of the batteries without top balancing to test how matched the cells are. Then try using a balancer bms to see how long it would end up balancing it.
That's what I'm going to do with the new battery packs. I'm not doing an initial top balancing. They go in production as they are and the BMS has to sort it out.
I would vote for Batrium BMS, especially because of it looks like perfect expandable setup for more batteries as you are using, offers management of them all together with direct communication with victron. Also balancing performace looks not bad based on specs. I woud say perfect device for your detailed testing and playing around :)
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
You've tested many of the more affordable BMS from Asia but I'd really like to know if they can compare to the higher end or higher priced solutions such as those offered by Batrium (from there in Australia), orionbms (US) or REC-BMS (Slovenia). What especially interests me though is the ability these BMS have of connecting to the Victron inverter and charge controller via CANbus. I want to know if that integration is worth the extra $$$$?
About the BMS integration? What would that do apart from the Victron system being able to read single cell voltages? I still don't see any benefit of having this connection. One viewer wrote, he's got the BMS linked to his solar charge controller and slows down charging speed if the cells get too hot during charging. That is actually a good reason! Apart from that 🤷♂️
Andy. I have a brand new ANT BMS for you. Never used. It’s yours to say thanks for all your great work if you want it. Just need to figure out shipping from Canada. Cheers.
Isolate a busbar on the bottom with tape and look at the temperature could be a nice test. The stud->nut->busbar touch surface compared with the resistance measurement is much bigger and the resistance will be a lot lower.
Use an ANTBMS as the 3rd one. It comes with a massive lcd display, so it should be interesting. It can tell you the individual cell voltages, lowest, highest and average cell voltage and more. It has a large state of charge indicator on the display as well. And it can be pretty serious, I have a 400A one...
So you can simply test it by putting a rubber ring under the busbars and let the current flow only via the rods and the knots to the top of the busbar?
Get a couple of the new JKBMS with the screen. The screen is cool, not great, but I think the JKBMS is probably the best on the market, despite their small gauge wires.
great testing Andy. I've now installed several packs using the jkbms (no screen) without any issues. 304ah cells. 16s. No problem at 200ah discharge 80% and the standard cables. I add a 9v battery with a N.O. push button switch for the bms boot. As well the battery paste works great. I noticed price increases for batteries but 230ah are at a good price Grade A from a USA supplier free shipping. I'm in sunny southern California. 7day delivery. I have 45v 9a panels and I use a boost converter to 56v for my solar charge controller to the bms. BMS controls all the charging. Thanks for your testing.
Great ides with the pushbutton. Why is the external voltage needed when starting the JKBMS? It seems to me like it could just steal the voltage from the batteries via the BMS cables.
@@porter5976 I thought the same but the bms start needs to be more than the battery pack voltage so it can't be borrowed from the cells. Just the workings of the bms. Once everything is set up its only needed if the B+ is disconnected. The B+ is the negative of the pack. The B+ needs to be more positive for boot. Using the cells would short the pack minus to a cell positive.
Yeah the JK sems obvious. It's the best I have seen and tested so far. I have never turned my one off after the initial testing, so the start procedure seems less important...
hello from Arizona. Great video sir! I appreciate the attention to detail. I just received 16 of the EVE 280AH cells last week and am now going down the rabbit hole of BMS and Battery Balancing. I will watch some of your other videos now...
Looks like some type of ablation when they laser weld the bolts. I would be interesting if they could put a hex bolt on there so a wrench could be used on the bolt to prevent it being torn off if over torqued.
Many types of stainless are magnetic, we called them chrome iron as a general description but i think they are called ferritic stainless steels, non magnetic stainless can gain magnetic properties during machining or work hardening which may explain the slight pull on the nuts (insert joke here). Question - what key features would you like to see in an active balancer ? Challenge - could you make your own balancer?
Well done Andy, After seeing these measurements on the screw terminal, I understand why the measurements in the Vlog " All these different busbars... and why they don't matter!" showed huge wrong results, in which the internal resistance did not match the measured voltage and current of 160Ampere according ohms law. Because in that video you measured the transition resistance of the bus bar on top of the screwterminal so, including the resistant of the screw.
Andy, have you ever had issues with pricing from Qishou? Every time I reach out for a quote, they wana charge me more per cell than what’s listed on their Alibaba page.
Prices are going up at the moment. I usually get a quote and lock it in, they then create the invoices through the Alibaba website with exactly that price.
I would recommend the Seplos BMS or another brand which can be individually addressed by setting DIP switches. On the negative side, I believe that they are PASSIVE balancing rather than ACTIVE and therefore I would recommend adding a high current active balancer to the pack.
See, I would not do that, buying a BMS plus adding a balancer. If I already had a Daly, yes, that would be the way to go. But buying a new BMS and already knowing I need an active balancer, well, that would then be a JK-BMS. The Seplos does not excite me at all. I cannot find any feature where I would say, wow, that is great and I wand this. Communciation? All other BMS can do the same these days.
@@OffGridGarageAustralia I fully agree with you which is why I mentioned adding an active balancer. I truly wish the Seplos would have an active balancer rather than a passive one.
@@danielglang6312 Guys, passive balancing on Seplos still does the job but a bit slower. It is a MAC vs PC argument in my opinion. You have to read more into how it works and in my opinion is more efficient but slower. At the end the result is what matters and not the personal preferences. IMHO A
What materials the bas bar made of and the washer on the battery are they the same material ? In order not to scratch the bas bars with the nut you can use a washer as said no current is issued there
I'm very interested in the "Seplos" BMS. Seems to be quite high-end to me. It has a Can bus and should be compatible with Victron SMA inverters and can be combined via RS485. I guess it's used in some of the rack batteries Will Prowse likes so much. They also come with a display. I have one lying around but didn't have a chance to test it yet. I'd love to hear your opinion.
Apparently balancing sucks with the Seplos and no Bluetooth? All BMS's have now communication ports which can talk to Victron. But for that reason? It is not necessary.
What a great introduction I would usually have missed - I mean the studs resistance measurements. I was looking for the next bms and not expecting a basic introduction to the EVE LF280K battery. Great content, high quality and on point. 4 NM is pretty low and a reminder to me to get a torquemeter / Drehmomentschlüssel.
Since you are going to have 3 banks of cells in parallel, there is a benefit in keeping the total resistance of each bank of cells the same. If you use different BMSs on each bank, the resistance will be different and because of that, the current that is drawn from each bank will be different. (There are a few reports on the DIY forum where different BMSs cause a significant difference in current). This means the banks will be used at different rates and worn at different rates. However, as Will Prowse likes to point out, the cycle life of the cells is so long you may be hit by calendar life failures first. Having said all that: I love your reviews of equipment so seeing what you think of some of the other BMSs would be fun!!
Three banks with different resistance will not necessarily end up with different currents. The one with the lowest resistant will deliver the most current, so will also have the lowest voltage. Hence the other two banks deliver more current to supplement as they try to keep the same voltage. I thought about this 'problem' but think it isn't one. We will see, test and learn ;)
@@OffGridGarageAustralia Yes, The current will start to equalize AFTER the one that has the lower resistance has discharged some. The one with lower resistance will always be discharched to a lower SOC than the others.
Yes I pointed this out on a previous video, as few people are aware that stainless steel has terrible electrical conductive properties! Most 12V batteries (yep LifePo4 cells too) are supplied with stainless steel terminal bolts, and most people do not understand that the only sound electrical contact is through the the terminal post it's self, and next to nothing through the bolt or stud & nut. This is one of the reasons why it is so important to stack battery lugs in the correct order, i.e. largest at the bottom and then the next size etc etc 🙂
@@jmaus2k yes, but only if there is aluminum on steel in the thread, otherwise crevice corrosion will occur. In this case, this does not occur because the connection is aluminum to steel welding.
@@jmaus2k I will be using Zinc coated Steel bolts/studs & nuts from now on, as it seems to have the best non-corrosive conductivity option of all I have looked at when used with aluminium, and also with the bigger 12V batteries that mostly have brass inserts.
I have a question about the JK BMS and maybe someone can help me out. On the description of the product it says that it doesn't support series Connection and parallel connection. What does it mean exactly? If I have two identical batteries using two JK BMS, I can't connect those batteries together in parallel for instance?
I spotted that too, but ordered one anyway. You clearly can have 2 identical systems in parallel, as Andy has done. I assume that they mean you can't have 2 banks in parallel & have one BMS trying to balance those cells.
I also think its plausible that the 280 and 304 ah are identically physically, i think its possible that by derating to 280ah, this may be where the extra life cycles data comes from, i will be interested to see when you do your top and bottom balance if the 280ah pull anywhere near 304ah when brand new. If you owned 304ah rated to 3000 cycles and only cycled them to 280ah, you will probably get the 6000 cycles, maybe, maybe not. keen to see your tests
Uh, no, that definitely not the case. That would mean all 280Ah cells are in fact 304 cells but didn't pass the capacity test. That would be really bad and they would fail far earlier. The LF280K have a different chemistry, hence can deliver more cycles.
I would love to see you utilise the new Batrium K9. I realise it’s $$$$, but…. Remote cell level monitoring. Victron intergration. Shunt CB activation. Australian made. Thank you.
Hmmmm... We can now connect pretty much any BMS to the Victron system and monitor the cell levels. JK will also come out with two-way comms between Victron and BMS soon, so that's another level again.
Hello ANDY! I would test one of the early jk bmses with a 400 milli amp balancing current. They are cheap but I don't know if they can keep such large cells under control. But we are trying to get the most out the least always, right?
With the "slot" in the busbar the area that is open in the slot that is not contacting the flange is not part of the circuit. Only the horseshoe shape of the busbar is passing current. So it might be best to tug the bar all the way over to the stud and tighten it. If not then current is coming through only the sides that connect the plate.
REC BMS told me I can use the one BMS for three parallel 16 cell series (so about 45kWh) but I've yet to order my 32 other cells. Again, this is for the SMA Sunny Island, which is a bit particular.
@@showmequick2245 it's the REC-BMS-SI and in their words "You can put the battery cells in parallel first and then in series to increase capacity and still use only one BMS. You can set the parameters accordingly with Wi-Fi module you ordered."
Hi Andy. You could actually measure the resistance of the stud, through the nut to the busbar, if you but something isolating under the busbar, for example a plastic washer. So that it does not have direct contact with the upper side of the pole. Torque with 4Nm and then compare the resistance to a normal properly connected busbar. Wouldn't that be an interesting experiment?
I would test one of the early jk bmses with a 400 milli amp balancing current. They are cheap but I don't know if they can keep such large cells under control. But we are trying to get the most out the least always, right?
If I was going to add a washer for more contact area, I'd use a washer that just fit over the stud to give the most contact to the flat area, not the weld area, and wide enough to give more contact to the bus bar. Would two slightly-better contacts in series be better than the original contact? Who knows? As you said, it may not matter since it's working well enough as is.
@@OffGridGarageAustralia Bad idea in my opinion. You are adding more points of failure, every extra connection is extra resistance and at the end it doesn't matter how many washers you stack on it as the first one only has the original surface area but every consecutive one will add to resistance so I fail to see the benefit here. A
I've sent JK a couple of inquiries now to run 12s LiFEPo4 36v system with the EVE 280K cells (dc motor draw up to 300A, maybe add a bit for potential overhead). In looking at the data sheet on their site, I don't see this configuration supported. Hopefully, they will respond soon. DALY seems to have options to support this, but JK seems like a better option with the 2A active balancing built right in.
I see the Watchman Core with the Cellmate K9 has been suggested. I would be very interested to see you set it up. I would love to see a BMS that be configured and monitored over the network. WiFi or even better by cable.
I'm sorry for the basic and stupid question I'm going to ask: Can I lay the battery down? (stacked one on top of the other - LifePo4). I have enough height space to put batteries but I don't have wide space
That's a good question. I believe you can do this for some time but not permanently due to the fluid inside the cells. I'll confirm this with the manufacturer and supplier...
@@OffGridGarageAustralia I appreciate your kindness and will be waiting. I really haven't found anything RELIABLE about it. Chinese sellers say YES to everything.
I personally think you should have combined the two battery banks (of 16), with the differing aH ratings and then tested a BMS (such as the JK unit) across all 32 cells, wired in to 16S (which would then be 16S2P). You could then describe a situation with mixed batteries and how the BMS would adjust to the scenario. That would have been educational... :)
Well, if we parallel the cells first, the BMS would not care as it sees these cells as one cell only. I don't like this at all because we don't have access to all individual cells. If I parallel 3 cells for example, how would I know if one of them is faulty? I would not recommend such a setup with these large capacity cells. It can be quite dangerous. It is far better to have single 16s1p strings of batteries with their own BMS each and then parallel them.
Are you sure about this? Would 2p really be any different than 1p. 2 cells become one. Does parallelling 2 banks bring any complexity? Do you treat the set of 2 parallel banks just as one bank?
@@HobbyFinn2 if all the cells are perfectly matched then 2P isn't an issue. I rarely have seen 8 batteries matched much less 16 or 32. HOWEVER, 2P with active balancer does not require all cells to be matched. When my new BMS arrives in a month, I am trying the 2P thing. It is best to match your best cell with your least best as a pair when paralleling.
Hi Andy, I have and like the JK BMS, but had thought about getting a Seplos BMS as many server rack battery manufacturers use them. Another one is the Batrium, and I would like to see how they both compare to the JK BMS. I have an interest in getting the data from RS 485 and believe it's possible with these. I look forward to watching your videos, Thanks
Hey Andy test the seplos bms 200a in combindig with an heltec and / or neey aktiv balancer. I have ordered this combination and will give it a try. The seplos has rs485 and canbus so with an canbus had on the pi it should come up as pylontech. Thanks for all your work! Stay charhed!
Hi. Love your testing. About the active balancer; do you think igs possible/smart to have this atatched to the batteries together with the bms all 24/7 all year round? Or do you thing its a "maintenence" thing to turn it on only when needed? When connected, and the battery is balansed, what is the power consuption from the balancer? Have a great easter hollyday. Runar from Norway
Hi Runar, the balancer should only be running from 3.45V onwards. It creates a unbalance otherwise as we have see with the earlier built of the NEEY balancer. If it balances all the time, it will destroy your top balance. This may work with Li-ion but not with LFP and the flat curves we see.
Hi Andy, have you seen the Elerix BMS (actually callled Cell Performance Monitor). It is a BMS which do not have live current in it, the principle is interesting. Thanks for all the great videos which you share. Cheers
Thanks for a pile of great videos. I've just spent a few hours watching many of them on 2x speed & have now ordered a JK BMS as a result. Having said that, my vote goes for DIY-BMS because I want to have decent reporting/alarming/integration ultimately, for when the novelty wears off...
Regarding the IR of the threaded area, the only issue is that the nut will have an almost infinitely larger contact area than the probes. That will cause the functional resistance through the nut to be lower than you're measuring. Remember, resistance on a surface has to do with area and aspect ratio. It's per square unit of area. The thread surface has a very wide contact patch, creating a very very low resistance for the connection. The nut likely doesn't carry the majority of the current, but I'll bet a fancy coffee it's well over the 5% you claim.
Batrium, you'll see all cells on one screen, they all ready in Australia, I just ordered 2 batrium systems with the k9, should be here in tn another week
Yeah, they got in contact with me a while back and offered something. I wanted to stay low though and rather test simple BMS's and not such a huge system. I find it a bit overkill to install a Batrium one just one bank and leave the others with different BMS. To upgrade them all, well... $$$
Check out DiyBMS. It’s a lot more involved than the others. Still very much a work in progress. But it’s coming along. RS485 CANbus 2 onboard relays 2 onboard solid state relays 4 gpio SD card logging Support for writing to influxDB and MQTT Touch screen (interface still in development) Uses an ESP32 as the controller. Each cell has its own module based on an attiny841 Can talk with Victron GX devices and DVCC over VE.Can Cell modules support 850ma resistive balance current. I am using a relay to activate a Heltec 12-16s 5A active balancer.
Balancing sucks with the DIYBMS, setup with relays is complex with lots of crimping and connections. I looked into it a few times but never got excited to be honest.
I have been happy with the overkill solar BMS. It’s only rated at 100A, but I have built 2 battery banks in parallel with them and they work well. Very rare I pull over 40A through them at the same time.
I have the Overkill BMS as well, just totally forgot about it. Madi has sent me hers from Canada. 100A would be OK for me as each bank will only be discharged with 50A max. So that could be an another option we already have in stock.
Good video, you had to put the batteries in parallel and measure the resistance and in series and measure the resistance. To balance the batteries I made an idea that for 4 batteries in series I had a resistance of 1Kohm parallel across each battery during charging, it now only an idea and will probably not work in practice.
QUESTION ON THE LENGTH of the Red and Black. I understand that it needed to be the same. But do u guys calculate the Blue - out of the (-) of the BMS out of - battery? and the black - from the BMS which is like 10 inches in blue and black on the bMS ? my battery got high voltage at cell 15 , 16 but low volt at 1234 little higher than 56789. The 15 n 16 cells got to 3.5-3.67 while other are still in the 3.3. Thanks (i understand from busbar to inverter black and red are the same length.)
I been interested in the seplos, with all the features, the current limiting on charging, I think maybe a heater function, plus communication, I was thinking about getting one. But my system is 7s 24v, 3.7/4.2v, I also have a backup system LiFePo 12v system. I would like a larger smarter BMS,
@@OffGridGarageAustralia Depending on the temperature. In cold climates Seplos will regulate the charging current based on the cell temp as it has to be lower to avoid the damage. Also there is a 10Amp safety feature too... Their manual is written in bad english but it covers all the scenarios and describes all the features.
Hi Andy, why don't you try out ANT bms. It is a popular type of the not-so-expensive, but (seems to be) smart enough mbses, an alternative to DALI. It would be nice to know its pros and cons, I think many folks would like to know, whether it worth to buy, or it is better to spare further for a more expensive bms.
Hi Andy I’d like to see a test of the Batrium as it doesn’t run the battery through it and balances the cells which is potentially causing more losses than Double FETs but does it in real life? Some feedback says no difference. It seems simpler to install and expand but is that worth the price penalty?
Question, I’ve went through a lot of your videos, over a year ago when you were building the battery box, you asked the question, should you put the shunt before or after the BMS and was waiting for comments. You never said after that what way you would mount it. So my question is what way did you decide? Also on the new rack battery setup, which way do you plan on connecting the shunt? And are you going to use a shunt for each bank? I’m adding more batteries and would like your input. Right now I have the bms after the shunt but was wondering how it works before the shunt. Thanks
Batrium! 1) they are in AUS so you can just go knock on their door and 2) So many people swear by them, with their price you should see if it's really worth it.
I would like to see the JBD 7-20s 200a BMS. Why? Because here in America, I was able to order my cells, with this BMS,s, and, get them within a week. No waiting 3 months. Here's an idea, do a "Best set up practices/method", for each of the BMS's, not everyone is going to be able to get they're ideals, so, a do the best with what you got would help a lot of people I think.......
I don't think it's a disadvantage to be missing the cables of the Heltec BMS. On your other rack battery you had to use those other connection points anyway. So it's a little sweat to make a number of perfectly sized cables that match your system and work without added bus studs. Perhaps use some scrap copper bar across the four attachment points to make it perfect connection.
seplos bms , a few of the commercially built rack mount battery packs use it, if i was building an home big bank, i would look in to it, i seen it mentioned on a vlog that it is a couple hundred US.
I have narrowed my search to just the JK and Seplos, maybe you can do a comparison, see if the Seplos Balancer is any good? Pros for the JK are the 2A active balancer, robust case CONS No CAN bus support for inverters standards seems their own standard? Pros for the Seplos CAN bus support for many inverters, Cons Balancer is passive 150mA(almost useless) no case.
Hi Andy great video, in regard to BMS solution I would raccomend the JK BMS, because it as an Active Cell Balancer so it will balance the battery much quickly than a normal BMS; and if you are intrested in something similar you could use the Daly BMS but: you have to keep in mind that at high SOC and above 3.45 you need to use an Active Cell Balancer connected to Voltage Comparator so, that it will switch on the balancer from a voltage above 3.5V like what I did.
Hi Andy, I am requesting an individual BMS because it would be good to compare and see if it has any advantages because when people buy a off the shelf LifePo battery many come with a BMS inside. Can you show in real life how this affects usage, temp and charging? André in Sydney
I'm about to build a 2P16S 48v pack out of new LifePo4 280ah cells (CATL) from Battery Hookup. I asked the Support team over there if it was okay to do a 2P16S configuration using a single BMS, or if I am better off building 2 x 16S packs with each having a BMS. They told me that a 2P16S configuration is perfectly fine because the cells are new. If they were used, he said he would recommend two separate packs/BMS's, but because they are new cells, a 2P16S configuration is good. DO you agree with this?
I would not be happy with any parallel batteries. You can never be sure what happens between the two cells. They may be new and matched now but what about in 5 years time? How good are they still matched by then? I would always recommend using two separate banks with a BMS each (yes, more expensive, but you have spend thousands of Dollars on the batteries, a second BMS is not much in addition. Full control over all the cells at any time. Redundancy: if one BMS or bank goes down, the other one keeps going.
Allez par curiosité je continu les vidéos pour avoir la suite du feuilleton du BMS de HELTEK et de ces balanceurs actifs. une chose est vrai que je n'avais jamais encore vu un BMS sans cable positif et neutre de connection... pourquoi pas. Merci ANDY pour la suite.
One of the best bms is the DIY BMS, it is a bms that costs a little to manufacture, but has a great capacity for configuration and communication. The balancing is a bit lacking, but it could be combined with a capacitor balancer or a jk. I think the seplos is also a good option.
I had a look at the DIY BMS and found it quite geeky with all the programming and flashing you need to do. It's not beginner friendly or easy to setup. I'm also not a big fan of the Seplos BMS so far. It has potential but is not there yet.
Is there any harm in having a capacitive active balancer connected to a battery all the time - not just when over 3.4V per cell? If so, what is the issue, please Andy?
Below 3.4V gives no information on SOC. So your balancer will be actively unbalancing your pack based off bus bar voltage drops or inaccuracy in the measurement.
I purchased the Daly back in November for 16s 48v programed by Daly for the 280 Ah EVE. Unfortunately Life gets in the way and have yet to purchase the remaining batteries. So have you had any bad experience with the Daly brand?
One thought, try measuring between a LOWER point on the stud and the base. The nut makes contact lower than just the top, the lower threads make a firm contact along the full length of the nut, not at just the top of the stud.
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
You are absolutely correct about the current through the welded studs. The current through each of the two paths will have a ratio that is inverse of the ratio of the resistance between the two paths. By the time you get the total resistance through the stud and nut it will be a lot more than from the pad to the busbar.
The most simple BMS, with no balancer or balancer turned off, just a voltage cutoff basically high/low on each cell. Leave it running 6 months and compare capacity before/after and see how balanced it stays without a balancer - and compare to those you are OCD balancing :)
Well, we have tested this already with the QUCC BMS a few times here. I rarely charge my battery full and if so, the cells are usually all over the place. SO I need a strong balancer which can handle that. So far only the JK could excel here.
This is also more a test of how good your cells are matched.
@@OffGridGarageAustralia I'm 3 years in and no issue with no balancer ....
@@stupidmonkeykev All depends on your system setup and what cells you use. I have seen set ups without balancer but they usually keep a top and bottom buffer and run the cells only in the middle part.
@@OffGridGarageAustralia i think your 3 delivery's in 1 battery may not be ideal. New setup maybe more stable.
However i also think you panic at 0.01% capacity imbalance (100mv @3.5v) be good to see you leave 1 of your 4 48v batteries "unbalanced for months and see if that effects capacity. (Spoiler, it wont, but prove it :))
05:29
Why do the Chinese make small protrusions on the battery cell terminals instead of making them flat? If the contact surface of the busbar is wide, there will be less resistance and better current flow.
I would love to see the Batrium, I have seen lots of people using it, but I feel like I never really get any indepth information on how the system works, or how well it actually works, all of which I know you would test very thoroughly
This would be my suggestion too! Does the cost truly justify the performance would be the "Why"? The other question would be since you can have multiple batteries connected to a Batrium...can it keep all the cells in multiple batteries all in balance?
+1
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
+1 on Batrium!
A good thing with Batrium is it can be configured to monitor every cell in a 4S2P battery pack.
Hi Andy. I would like to see how the Batrium Core works. There should be at lot of parameters and monitoring possibilities for the battery. But the question are how good are they to the job.
I would like to see it Batrium myself, I bought 128x 300Ah Batteries and planning to use Batrium…
Well the batrium is pretty pricey, and I don’t understand how it works. But would like to learn more about it. Other channels don’t cover it that well.
would really enjoy a seplos bms review. seplos seems very interesting and feature rich. i particularly like the idea of having multiple bms, connected in master slave configurations where only the master communicates with the inverter and the slaves only communicate with the master. this way, you don't have a situation of having more than 1 bms potentially telling the inverter different things....either by way of canbus or by sending a quick signal to see if charging current is available. also, the seplos can function with 16 bms in parallel, plus they also have that niffty lcd screen. the seplos platform seems to bring a professional features to the dyi market.
Yesterday I got my Seplos BMS. I'll test them tomorrow :D
I got one 150A and two 100A versions. That'll be a lot of fun!
So far I am using a REC-BMS...it works well but sometimes the SOC jumps from like 80% to 100%...I'm sure, that this has something to do with software and could be improved, if I tried harder ;)
@@Raphael_Hofmann ok nice. if you have the time, let me know how your testing goes.
What would the Seplos BMS tell my inverter?
@@OffGridGarageAustralia When one cell reaches low voltage disconnect, the inverter will switch off. When one cell reaches high voltage disconnect, the solar charge controller will reduce the charge current. You'll need a Cerbo GX or a other victron-device, that is capable of CAN-commucation.
Also the BMS will basically replace your battery monitor.
@@OffGridGarageAustralia I could be wrong about all of this but from my understanding, where the value of the Seplos platform is, is the fact that you can have multiple batteries, each with their own BMS all linked to a master BMS that communicates to the inverter. So let's say, you have BMS1, 2 and 3, each with its own 48v battery where BMS1 is the master and BMS2 and 3 are slaves. If BMS1 has a full battery but BMS2 and 3's batteries still need charging, BMS1 will simply cut off charging to it's own battery but will not communicate to the inverter to stop sending charging current until BMS 2 and 3 reports back to BMS1 that they are also now full. It is at that point, that BMS1 tells the inverter to stop charging. I am not sure how that same scenario will pan out if again, 3 batteries each with their own BMS and neither BMS is aware of the other but 1 BMS has a fully charged battery and the other 2 don't. Would appreciate your feedback on all of this Andy....thanks in advance.
It is good that the welded stud is working as is, but I still don't understand why they don't use a larger base on the welded stud.
I think you are right. Most people won't have a dinamometer wrench at hand, and things can get rough...
Just got 17 cells from Qishou for 148usd a piece. In stock in a warehouse in Poland, shipping time 5-7 days. Apparently pretty pricey but still a steal compared to what german resellers want for what is essentially the same cell. (They charge around 200€/210usd per Cell).
Communication was great and straight forward, no issues at all. Even got the order and tracking numbers done on a sunday :)
I think the matching comes from NMC cells where there is a direct correlation between the IR and the remaining capacity (Relative for each model of cell of course). Andreas Schmitz here on youtube has created a database that can predict your remaining capacitance based on the IR and can be used to evaluate salvaged NMC cells.
You should try assembling one of the batteries without top balancing to test how matched the cells are. Then try using a balancer bms to see how long it would end up balancing it.
That is actually a good idea. I like that!
That's what I'm going to do with the new battery packs. I'm not doing an initial top balancing. They go in production as they are and the BMS has to sort it out.
@@OffGridGarageAustralia You are awesome Andy. Keep up the good work! Regards from sunny Argentina.
I would vote for Batrium BMS, especially because of it looks like perfect expandable setup for more batteries as you are using, offers management of them all together with direct communication with victron. Also balancing performace looks not bad based on specs. I woud say perfect device for your detailed testing and playing around :)
GOOD MORNING GUYS!!!! My question is. How many of the: over kill BMS: 16s 48v 100 amp load charge BMS to be able to handle Load about 10,000 watts 240v continue for 6-10 hours use/ charge car. Any one please help. if you have the Math show how to calculate will be great. I will buy -30 - 40 kwh battery bank thanks a lot. Solar 5kwh, INverter: 3x 5kwh Growatt inverter.
You've tested many of the more affordable BMS from Asia but I'd really like to know if they can compare to the higher end or higher priced solutions such as those offered by Batrium (from there in Australia), orionbms (US) or REC-BMS (Slovenia).
What especially interests me though is the ability these BMS have of connecting to the Victron inverter and charge controller via CANbus. I want to know if that integration is worth the extra $$$$?
About the BMS integration? What would that do apart from the Victron system being able to read single cell voltages? I still don't see any benefit of having this connection. One viewer wrote, he's got the BMS linked to his solar charge controller and slows down charging speed if the cells get too hot during charging. That is actually a good reason! Apart from that 🤷♂️
@@OffGridGarageAustralia with this integration you can turn off the Victron under certain conditions instead of relying on a shunt.
Andy. I have a brand new ANT BMS for you. Never used. It’s yours to say thanks for all your great work if you want it. Just need to figure out shipping from Canada. Cheers.
Thanks James, that is a very generous offer.
Just sent you an email with more information. Cheers!
Isolate a busbar on the bottom with tape and look at the temperature could be a nice test. The stud->nut->busbar touch surface compared with the resistance measurement is much bigger and the resistance will be a lot lower.
Use an ANTBMS as the 3rd one. It comes with a massive lcd display, so it should be interesting. It can tell you the individual cell voltages, lowest, highest and average cell voltage and more. It has a large state of charge indicator on the display as well. And it can be pretty serious, I have a 400A one...
No screen recording for us to follow, but more than two screens on our devices will be hard to follow for us to see at the same time.
So you can simply test it by putting a rubber ring under the busbars and let the current flow only via the rods and the knots to the top of the busbar?
That would be a great interesting Test
@@berndtwagner189 I think so too :)
at this point just invest in a batrium with so mutch capacity and growing on ur setup u need a decent bms to protect it.
The JK is not decent?
I would rather stay 'low' with the BMS so other can follow the same path tbh.
Get a couple of the new JKBMS with the screen. The screen is cool, not great, but I think the JKBMS is probably the best on the market, despite their small gauge wires.
great testing Andy. I've now installed several packs using the jkbms (no screen) without any issues. 304ah cells. 16s. No problem at 200ah discharge 80% and the standard cables. I add a 9v battery with a N.O. push button switch for the bms boot. As well the battery paste works great. I noticed price increases for batteries but 230ah are at a good price Grade A from a USA supplier free shipping. I'm in sunny southern California. 7day delivery. I have 45v 9a panels and I use a boost converter to 56v for my solar charge controller to the bms. BMS controls all the charging. Thanks for your testing.
Great ides with the pushbutton. Why is the external voltage needed when starting the JKBMS? It seems to me like it could just steal the voltage from the batteries via the BMS cables.
@@porter5976 I thought the same but the bms start needs to be more than the battery pack voltage so it can't be borrowed from the cells. Just the workings of the bms. Once everything is set up its only needed if the B+ is disconnected. The B+ is the negative of the pack. The B+ needs to be more positive for boot. Using the cells would short the pack minus to a cell positive.
Yeah the JK sems obvious. It's the best I have seen and tested so far. I have never turned my one off after the initial testing, so the start procedure seems less important...
I'd say v2 jk bms, as it had a screen and I want a jk and your making it so tempting to get them once the gravy train comes
hello from Arizona. Great video sir! I appreciate the attention to detail. I just received 16 of the EVE 280AH cells last week and am now going down the rabbit hole of BMS and Battery Balancing. I will watch some of your other videos now...
Did you see the EVE cells LithiumSolar got with the double bolt terminals? video yesterday.
Those look awesome...
I just bought 12 cells with those terminals. A lot more surface area for 2 inch flat copper bus bars I'm working on.
I was going to say the same thing. But didn't want to Andy to have buyers remorse.
Yes, saw his video...
Looks like some type of ablation when they laser weld the bolts.
I would be interesting if they could put a hex bolt on there so a wrench could be used on the bolt to prevent it being torn off if over torqued.
Many types of stainless are magnetic, we called them chrome iron as a general description but i think they are called ferritic stainless steels, non magnetic stainless can gain magnetic properties during machining or work hardening which may explain the slight pull on the nuts (insert joke here).
Question - what key features would you like to see in an active balancer ?
Challenge - could you make your own balancer?
Any suggestion on how to attach terminals that have come off the battery?
Well done Andy, After seeing these measurements on the screw terminal, I understand why the measurements in the Vlog " All these different busbars... and why they don't matter!" showed huge wrong results, in which the internal resistance did not match the measured voltage and current of 160Ampere according ohms law. Because in that video you measured the transition resistance of the bus bar on top of the screwterminal so, including the resistant of the screw.
Andy, have you ever had issues with pricing from Qishou? Every time I reach out for a quote, they wana charge me more per cell than what’s listed on their Alibaba page.
Prices are going up at the moment. I usually get a quote and lock it in, they then create the invoices through the Alibaba website with exactly that price.
Hi andy, Have you try Batrium BMS ?
I would recommend the Seplos BMS or another brand which can be individually addressed by setting DIP switches. On the negative side, I believe that they are PASSIVE balancing rather than ACTIVE and therefore I would recommend adding a high current active balancer to the pack.
See, I would not do that, buying a BMS plus adding a balancer. If I already had a Daly, yes, that would be the way to go. But buying a new BMS and already knowing I need an active balancer, well, that would then be a JK-BMS.
The Seplos does not excite me at all. I cannot find any feature where I would say, wow, that is great and I wand this. Communciation? All other BMS can do the same these days.
@@OffGridGarageAustralia I fully agree with you which is why I mentioned adding an active balancer. I truly wish the Seplos would have an active balancer rather than a passive one.
@@danielglang6312 Guys, passive balancing on Seplos still does the job but a bit slower. It is a MAC vs PC argument in my opinion. You have to read more into how it works and in my opinion is more efficient but slower. At the end the result is what matters and not the personal preferences. IMHO
A
What materials the bas bar made of and the washer on the battery are they the same material ?
In order not to scratch the bas bars with the nut you can use a washer as said no current is issued there
I'm very interested in the "Seplos" BMS. Seems to be quite high-end to me. It has a Can bus and should be compatible with Victron SMA inverters and can be combined via RS485. I guess it's used in some of the rack batteries Will Prowse likes so much. They also come with a display. I have one lying around but didn't have a chance to test it yet. I'd love to hear your opinion.
Yes, Seplos pls! 👍
@@Ericcando81 Hi Andy I also like the look of the Seplos BMS just a shame no bluetooth. Thanks keep up the good work
Apparently balancing sucks with the Seplos and no Bluetooth?
All BMS's have now communication ports which can talk to Victron. But for that reason? It is not necessary.
@@OffGridGarageAustralia Come on Andy, give it a try, balancing is not that bad and you can you use the monitoring software they supply.
@@gerritgroothuis Let's see...
What a great introduction I would usually have missed - I mean the studs resistance measurements.
I was looking for the next bms and not expecting a basic introduction to the EVE LF280K battery.
Great content, high quality and on point.
4 NM is pretty low and a reminder to me to get a torquemeter / Drehmomentschlüssel.
Since you are going to have 3 banks of cells in parallel, there is a benefit in keeping the total resistance of each bank of cells the same. If you use different BMSs on each bank, the resistance will be different and because of that, the current that is drawn from each bank will be different. (There are a few reports on the DIY forum where different BMSs cause a significant difference in current).
This means the banks will be used at different rates and worn at different rates. However, as Will Prowse likes to point out, the cycle life of the cells is so long you may be hit by calendar life failures first.
Having said all that: I love your reviews of equipment so seeing what you think of some of the other BMSs would be fun!!
This is why I am blue in the face for saying cells should be paralleled and not putting banks in parallel
Three banks with different resistance will not necessarily end up with different currents. The one with the lowest resistant will deliver the most current, so will also have the lowest voltage. Hence the other two banks deliver more current to supplement as they try to keep the same voltage.
I thought about this 'problem' but think it isn't one. We will see, test and learn ;)
@@OffGridGarageAustralia Yes, The current will start to equalize AFTER the one that has the lower resistance has discharged some. The one with lower resistance will always be discharched to a lower SOC than the others.
I'm only posting for the people who might believe this. You are incorrect sir. On all counts.
@@patshannon5107 I guess we will agree to disagree.
stainless steel stuts would have much more resistance because the specific resistance of stainless steel is higher than from steel.
Yes I pointed this out on a previous video, as few people are aware that stainless steel has terrible electrical conductive properties!
Most 12V batteries (yep LifePo4 cells too) are supplied with stainless steel terminal bolts, and most people do not understand that the only sound electrical contact is through the the terminal post it's self, and next to nothing through the bolt or stud & nut.
This is one of the reasons why it is so important to stack battery lugs in the correct order, i.e. largest at the bottom and then the next size etc etc 🙂
@@ValRigoli Yes, but aluminum requires stainless bolts to avoid corrosion.
If conductance was a concert brass studs would have been used. Carbon steel is much stronger and won't have nuts get stuck on it like stainless does.
@@jmaus2k yes, but only if there is aluminum on steel in the thread, otherwise crevice corrosion will occur. In this case, this does not occur because the connection is aluminum to steel welding.
@@jmaus2k I will be using Zinc coated Steel bolts/studs & nuts from now on, as it seems to have the best non-corrosive conductivity option of all I have looked at when used with aluminium, and also with the bigger 12V batteries that mostly have brass inserts.
Hi Andy
I also like the look of the Seplos BMS just a shame no bluetooth but looks a well made bit of kit.
Thanks keep up the good work
I have a question about the JK BMS and maybe someone can help me out. On the description of the product it says that it doesn't support series Connection and parallel connection. What does it mean exactly? If I have two identical batteries using two JK BMS, I can't connect those batteries together in parallel for instance?
I spotted that too, but ordered one anyway. You clearly can have 2 identical systems in parallel, as Andy has done. I assume that they mean you can't have 2 banks in parallel & have one BMS trying to balance those cells.
I also think its plausible that the 280 and 304 ah are identically physically, i think its possible that by derating to 280ah, this may be where the extra life cycles data comes from, i will be interested to see when you do your top and bottom balance if the 280ah pull anywhere near 304ah when brand new.
If you owned 304ah rated to 3000 cycles and only cycled them to 280ah, you will probably get the 6000 cycles, maybe, maybe not. keen to see your tests
Uh, no, that definitely not the case. That would mean all 280Ah cells are in fact 304 cells but didn't pass the capacity test. That would be really bad and they would fail far earlier.
The LF280K have a different chemistry, hence can deliver more cycles.
I would love to see you utilise the new Batrium K9.
I realise it’s $$$$, but….
Remote cell level monitoring.
Victron intergration.
Shunt CB activation.
Australian made.
Thank you.
Hmmmm...
We can now connect pretty much any BMS to the Victron system and monitor the cell levels. JK will also come out with two-way comms between Victron and BMS soon, so that's another level again.
@@OffGridGarageAustralia maybe I don’t need to be saving up for a Batrium then!
Thank you
What do you think about ELERIX Cell Performance?
Hello ANDY! I would test one of the early jk bmses with a 400 milli amp balancing current. They are cheap but I don't know if they can keep such large cells under control. But we are trying to get the most out the least always, right?
With the "slot" in the busbar the area that is open in the slot that is not contacting the flange is not part of the circuit. Only the horseshoe shape of the busbar is passing current. So it might be best to tug the bar all the way over to the stud and tighten it. If not then current is coming through only the sides that connect the plate.
REC BMS told me I can use the one BMS for three parallel 16 cell series (so about 45kWh) but I've yet to order my 32 other cells.
Again, this is for the SMA Sunny Island, which is a bit particular.
This is interesting. Do you know what model they’re talking about ? Thx
@@showmequick2245 it's the REC-BMS-SI and in their words "You can put the battery cells in parallel first and then in series to increase capacity and still use only one BMS. You can set the parameters accordingly with Wi-Fi module you ordered."
Hi Andy. You could actually measure the resistance of the stud, through the nut to the busbar, if you but something isolating under the busbar, for example a plastic washer. So that it does not have direct contact with the upper side of the pole. Torque with 4Nm and then compare the resistance to a normal properly connected busbar. Wouldn't that be an interesting experiment?
4:23 With 50A, each stud causes a voltage drop of 10mV, Heat = 0.5W
With 100A, Voltage drop = 20mV, Heat = 2W
That would be for the full length of stud.
I would test one of the early jk bmses with a 400 milli amp balancing current. They are cheap but I don't know if they can keep such large cells under control. But we are trying to get the most out the least always, right?
If I was going to add a washer for more contact area, I'd use a washer that just fit over the stud to give the most contact to the flat area, not the weld area, and wide enough to give more contact to the bus bar. Would two slightly-better contacts in series be better than the original contact? Who knows? As you said, it may not matter since it's working well enough as is.
Ah, right, OK, that makes sense to place the washer just on the top ring... I didn't think about that option...
@@OffGridGarageAustralia Bad idea in my opinion. You are adding more points of failure, every extra connection is extra resistance and at the end it doesn't matter how many washers you stack on it as the first one only has the original surface area but every consecutive one will add to resistance so I fail to see the benefit here.
A
@@HG-Pilot All these extra connections... absolutely right.
Hi there, on one of your videos you mentioned a Digital Torque Adapter with LED,been trying to find that video to see the one you used
But you need to measure the resistance in the screw just from the bottom to the height of the busbar, which is potentially quite low.
I've sent JK a couple of inquiries now to run 12s LiFEPo4 36v system with the EVE 280K cells (dc motor draw up to 300A, maybe add a bit for potential overhead). In looking at the data sheet on their site, I don't see this configuration supported. Hopefully, they will respond soon. DALY seems to have options to support this, but JK seems like a better option with the 2A active balancing built right in.
Never read about someone building a 36V system. Is there any gear and devices available for this option?
@@OffGridGarageAustralia It's not a custom build, other than the battery, but meant to power a golfcart that's stock at 36v. Yes, DALY has an option.
2:30 Hmmm.. now I am seeing the big problem... tough luck getting this batch
I see the Watchman Core with the Cellmate K9 has been suggested. I would be very interested to see you set it up. I would love to see a BMS that be configured and monitored over the network. WiFi or even better by cable.
I'm sorry for the basic and stupid question I'm going to ask: Can I lay the battery down? (stacked one on top of the other - LifePo4).
I have enough height space to put batteries but I don't have wide space
That's a good question. I believe you can do this for some time but not permanently due to the fluid inside the cells. I'll confirm this with the manufacturer and supplier...
@@OffGridGarageAustralia I appreciate your kindness and will be waiting. I really haven't found anything RELIABLE about it. Chinese sellers say YES to everything.
@@OffGridGarageAustralia I would be keen to see a proper response on cell orientation too...
Adding washer will add the resistance which in parallel reduce the overall resistance.
Prebuilt batteries have screens and interfaces. How to add that? Which BMS’s have the output of that? Or is it all Bluetooth?
I personally think you should have combined the two battery banks (of 16), with the differing aH ratings and then tested a BMS (such as the JK unit) across all 32 cells, wired in to 16S (which would then be 16S2P). You could then describe a situation with mixed batteries and how the BMS would adjust to the scenario. That would have been educational... :)
Well, if we parallel the cells first, the BMS would not care as it sees these cells as one cell only.
I don't like this at all because we don't have access to all individual cells. If I parallel 3 cells for example, how would I know if one of them is faulty? I would not recommend such a setup with these large capacity cells. It can be quite dangerous.
It is far better to have single 16s1p strings of batteries with their own BMS each and then parallel them.
Are you sure about this? Would 2p really be any different than 1p. 2 cells become one. Does parallelling 2 banks bring any complexity? Do you treat the set of 2 parallel banks just as one bank?
@@HobbyFinn2 if all the cells are perfectly matched then 2P isn't an issue.
I rarely have seen 8 batteries matched much less 16 or 32.
HOWEVER, 2P with active balancer does not require all cells to be matched.
When my new BMS arrives in a month, I am trying the 2P thing.
It is best to match your best cell with your least best as a pair when paralleling.
Hi Andy, I have and like the JK BMS, but had thought about getting a Seplos BMS as many server rack battery manufacturers use them. Another one is the Batrium, and I would like to see how they both compare to the JK BMS. I have an interest in getting the data from RS 485 and believe it's possible with these. I look forward to watching your videos, Thanks
Hey Andy test the seplos bms 200a in combindig with an heltec and / or neey aktiv balancer.
I have ordered this combination and will give it a try.
The seplos has rs485 and canbus so with an canbus had on the pi it should come up as pylontech.
Thanks for all your work! Stay charhed!
Hi. Love your testing. About the active balancer; do you think igs possible/smart to have this atatched to the batteries together with the bms all 24/7 all year round? Or do you thing its a "maintenence" thing to turn it on only when needed? When connected, and the battery is balansed, what is the power consuption from the balancer?
Have a great easter hollyday.
Runar from Norway
Hi Runar, the balancer should only be running from 3.45V onwards. It creates a unbalance otherwise as we have see with the earlier built of the NEEY balancer. If it balances all the time, it will destroy your top balance. This may work with Li-ion but not with LFP and the flat curves we see.
Hi Andy, have you seen the Elerix BMS (actually callled Cell Performance Monitor). It is a BMS which do not have live current in it, the principle is interesting.
Thanks for all the great videos which you share. Cheers
I seemingly come across tutorials in the right order to understand all I need to know
I just looked to see the different types of stainless ; there are some types that are ferrous but I would think you want the non-ferrous
Definitely the best solar channel.
Thanks a lot!
Thanks for a pile of great videos. I've just spent a few hours watching many of them on 2x speed & have now ordered a JK BMS as a result. Having said that, my vote goes for DIY-BMS because I want to have decent reporting/alarming/integration ultimately, for when the novelty wears off...
Regarding the IR of the threaded area, the only issue is that the nut will have an almost infinitely larger contact area than the probes. That will cause the functional resistance through the nut to be lower than you're measuring. Remember, resistance on a surface has to do with area and aspect ratio. It's per square unit of area. The thread surface has a very wide contact patch, creating a very very low resistance for the connection. The nut likely doesn't carry the majority of the current, but I'll bet a fancy coffee it's well over the 5% you claim.
So, If I try a nylon nut, what would you expect will happen?
@@OffGridGarageAustralia Rhetorical question. You already know the answer.
Batrium, you'll see all cells on one screen, they all ready in Australia, I just ordered 2 batrium systems with the k9, should be here in tn another week
Yeah, they got in contact with me a while back and offered something. I wanted to stay low though and rather test simple BMS's and not such a huge system. I find it a bit overkill to install a Batrium one just one bank and leave the others with different BMS. To upgrade them all, well... $$$
Check out DiyBMS. It’s a lot more involved than the others. Still very much a work in progress. But it’s coming along.
RS485
CANbus
2 onboard relays
2 onboard solid state relays
4 gpio
SD card logging
Support for writing to influxDB and MQTT
Touch screen (interface still in development)
Uses an ESP32 as the controller.
Each cell has its own module based on an attiny841
Can talk with Victron GX devices and DVCC over VE.Can
Cell modules support 850ma resistive balance current.
I am using a relay to activate a Heltec 12-16s 5A active balancer.
Balancing sucks with the DIYBMS, setup with relays is complex with lots of crimping and connections. I looked into it a few times but never got excited to be honest.
"Seplos" BMS...would be nice a test.And also will be usefull for your multiple battery pack...can connect in parallel
Batrium BMS, because it is bloody expensive :D
Is there a BMS for a string of 19s? The charger can drive up to 66V = 3,4xV per cell which is enough for me.
Carbon paste may well reduce the resistance at the base of the studs in the polished area. Nice Video thanks.
I have been happy with the overkill solar BMS. It’s only rated at 100A, but I have built 2 battery banks in parallel with them and they work well. Very rare I pull over 40A through them at the same time.
I have the Overkill BMS as well, just totally forgot about it. Madi has sent me hers from Canada. 100A would be OK for me as each bank will only be discharged with 50A max. So that could be an another option we already have in stock.
I would also want to see a test of the Seplos Bms. It has my vote.
Doesn't connecting them all in parallel for a few hours do the same thing?
Good video, you had to put the batteries in parallel and measure the resistance and in series and measure the resistance. To balance the batteries I made an idea that for 4 batteries in series I had a resistance of 1Kohm parallel across each battery during charging, it now only an idea and will probably not work in practice.
QUESTION ON THE LENGTH of the Red and Black. I understand that it needed to be the same. But do u guys calculate the Blue - out of the (-) of the BMS out of - battery? and the black - from the BMS which is like 10 inches in blue and black on the bMS ? my battery got high voltage at cell 15 , 16 but low volt at 1234 little higher than 56789. The 15 n 16 cells got to 3.5-3.67 while other are still in the 3.3. Thanks (i understand from busbar to inverter black and red are the same length.)
I been interested in the seplos, with all the features, the current limiting on charging, I think maybe a heater function, plus communication, I was thinking about getting one. But my system is 7s 24v, 3.7/4.2v, I also have a backup system LiFePo 12v system. I would like a larger smarter BMS,
Why would you want to limit the current while charging?
@@OffGridGarageAustralia Depending on the temperature. In cold climates Seplos will regulate the charging current based on the cell temp as it has to be lower to avoid the damage. Also there is a 10Amp safety feature too...
Their manual is written in bad english but it covers all the scenarios and describes all the features.
@@HG-Pilot So the actual BMS will limit the current? Or does it report the numbers to the SCC and they then throttle down?
Hi Andy, why don't you try out ANT bms. It is a popular type of the not-so-expensive, but (seems to be) smart enough mbses, an alternative to DALI. It would be nice to know its pros and cons, I think many folks would like to know, whether it worth to buy, or it is better to spare further for a more expensive bms.
Have you ever tested the 123 smart BMS, because I would love to hear what you think of it
Hi Andy I’d like to see a test of the Batrium as it doesn’t run the battery through it and balances the cells which is potentially causing more losses than Double FETs but does it in real life? Some feedback says no difference. It seems simpler to install and expand but is that worth the price penalty?
Hey Andy, how many beers does it take for you to look into Batrium BMS? I would love to see your thoughts and experiments with it!
baterium of 123 smart BMS , what are the differences and possibilities,
why I can't figure out which one to take with the next battery bank
Batrium would be a whole new BMS system as such an not a single BMS.
Question, I’ve went through a lot of your videos, over a year ago when you were building the battery box, you asked the question, should you put the shunt before or after the BMS and was waiting for comments. You never said after that what way you would mount it. So my question is what way did you decide? Also on the new rack battery setup, which way do you plan on connecting the shunt? And are you going to use a shunt for each bank? I’m adding more batteries and would like your input. Right now I have the bms after the shunt but was wondering how it works before the shunt. Thanks
Batrium! 1) they are in AUS so you can just go knock on their door and 2) So many people swear by them, with their price you should see if it's really worth it.
I would like to see the JBD 7-20s 200a BMS. Why? Because here in America, I was able to order my cells, with this BMS,s, and, get them within a week. No waiting 3 months. Here's an idea, do a "Best set up practices/method", for each of the BMS's, not everyone is going to be able to get they're ideals, so, a do the best with what you got would help a lot of people I think.......
I don't think it's a disadvantage to be missing the cables of the Heltec BMS. On your other rack battery you had to use those other connection points anyway. So it's a little sweat to make a number of perfectly sized cables that match your system and work without added bus studs. Perhaps use some scrap copper bar across the four attachment points to make it perfect connection.
I'm using a bank of those "flying capacitor" balancers...
They work well. I like them very much.
seplos bms , a few of the commercially built rack mount battery packs use it, if i was building an home big bank, i would look in to it, i seen it mentioned on a vlog that it is a couple hundred US.
I have narrowed my search to just the JK and Seplos, maybe you can do a comparison, see if the Seplos Balancer is any good? Pros for the JK are the 2A active balancer, robust case CONS No CAN bus support for inverters standards seems their own standard?
Pros for the Seplos CAN bus support for many inverters, Cons Balancer is passive 150mA(almost useless) no case.
The Seplos won't be as good as the JK... I will get one at some stage to test it but just look at the specs. The JK-BMS is far superior in many ways.
good afternoon a question you will have a coupon code to buy the inverter vitron
Hi Andy great video, in regard to BMS solution I would raccomend the JK BMS, because it as an Active Cell Balancer so it will balance the battery much quickly than a normal BMS; and if you are intrested in something similar you could use the Daly BMS but: you have to keep in mind that at high SOC and above 3.45 you need to use an Active Cell Balancer connected to Voltage Comparator so, that it will switch on the balancer from a voltage above 3.5V like what I did.
Seplos should come with cables connected, I believe that the heatsinks are more chunky.
Hi Andy, I am requesting an individual BMS because it would be good to compare and see if it has any advantages because when people buy a off the shelf LifePo battery many come with a BMS inside. Can you show in real life how this affects usage, temp and charging? André in Sydney
So far, nothing beats the JK-BMS. Soon it will be available as an 4s-8s BMS.
I'm about to build a 2P16S 48v pack out of new LifePo4 280ah cells (CATL) from Battery Hookup. I asked the Support team over there if it was okay to do a 2P16S configuration using a single BMS, or if I am better off building 2 x 16S packs with each having a BMS. They told me that a 2P16S configuration is perfectly fine because the cells are new. If they were used, he said he would recommend two separate packs/BMS's, but because they are new cells, a 2P16S configuration is good. DO you agree with this?
I would not be happy with any parallel batteries. You can never be sure what happens between the two cells. They may be new and matched now but what about in 5 years time? How good are they still matched by then?
I would always recommend using two separate banks with a BMS each (yes, more expensive, but you have spend thousands of Dollars on the batteries, a second BMS is not much in addition. Full control over all the cells at any time. Redundancy: if one BMS or bank goes down, the other one keeps going.
Allez par curiosité je continu les vidéos pour avoir la suite du feuilleton du BMS de HELTEK et de ces balanceurs actifs. une chose est vrai que je n'avais jamais encore vu un BMS sans cable positif et neutre de connection... pourquoi pas. Merci ANDY pour la suite.
GOOD VIDEO, there is also such a problem with metal studs. Metal has worse conductivity than aluminum.
Hi Andy, may I suggest that you test a REC BMS from Slovenia
It's a European brand, 2A balancer. It's often used in Marine and RV applications
One of the best bms is the DIY BMS, it is a bms that costs a little to manufacture, but has a great capacity for configuration and communication. The balancing is a bit lacking, but it could be combined with a capacitor balancer or a jk. I think the seplos is also a good option.
I had a look at the DIY BMS and found it quite geeky with all the programming and flashing you need to do. It's not beginner friendly or easy to setup.
I'm also not a big fan of the Seplos BMS so far. It has potential but is not there yet.
Is it possible to communicate with victron system and control the charge current?
Is there any harm in having a capacitive active balancer connected to a battery all the time - not just when over 3.4V per cell? If so, what is the issue, please Andy?
Below 3.4V gives no information on SOC. So your balancer will be actively unbalancing your pack based off bus bar voltage drops or inaccuracy in the measurement.
I purchased the Daly back in November for 16s 48v programed by Daly for the 280 Ah EVE. Unfortunately Life gets in the way and have yet to purchase the remaining batteries. So have you had any bad experience with the Daly brand?
Yes, I made a few videos about DALY. That's really the last BMS I would consider, to be honest. The balancing totally sucks.