Low side switching allows the circuit designer to use N-channel FETs (which have lower on-state resistance than the complimentary P-channel FET) with no need for a gate driver circuit that would be required if the N-channel FET were in the positive line. The bottom line is that low side switching is i little bit cheaper (It's about money, surprise, surprise).
It's also significantly more efficient. (and a gate driver circuit can't work for constant on without an isolated power supply that is added to the voltage unless you use a p-fet) And has less slam on shut off because of how real power works instead of the incorrect legacy of positive to negative. And the only reason to switch positive is legacy thinking because of negative grounded frames from cars. Literally NOTHING else.
@@jameshancockThere's other common negative reasons to switch positive, and they're about non isolated power supplies, multiple voltage supplies, and other legacy such as linear voltage regulators. It's not exclusive to the car stuff.
@@niktak1114 I concur in the case of isolated and un-grounded applications. But solar systems are generally required to be grounded. As Andy has already mentioned, communication lines between other components in a solar system could be subjected to undesirable current flow if the negative side of the battery is switched open. Perhaps there are ways to protect the comms circuits (opto-isolation or transformer coupling of low power signalling.) I'm not willing to condem negative side switching out of hand, but saving a small amount of money in one area of a design may easily be offset by cost s added elsewhere.
@@lua-nya non issolated power supplies should have shared negative. You can still switch the negative and the circuit is cut. Multiple voltage supplies shared negative switched no downside. Linear regulation is only for negative voltages and even then you can switch he negative shared rail without issues. It is literally negative as ground that is the only reason not to and that entirely bad decisions with cars. And 48v cars are isolated from the grounded frame so even that is being fixed and it’s literally for this reason.
I just love your passion! 😊 Trying to see why seplos v2 and v3 might have issue for balancing and came across this video. Asking for a friend on the forums.
Thanks Andy, as always, great work and very descriptive. It certainly convinces me that I think I’ll just stick with my JK BMS’s: simple, reliable, relatively cheap, easy to set desired behaviour and they 100% do the job that I need. KISS 💋(which of course means “keep it simple stupid” :-) I recently separated my packs and then kept one at a time at a fairly high voltage overnight with my power supply, for a good top balance with the JK BMS. Since then, PERFECT! 👍
I wish I’d stumbled upon your channel 6yrs ago when I built by first LFP pack. There are a lot of takeaway snippets of info that you give in your commentary that took me weeks to figure out.
Wow, thank you so much Andy!!! You are very welcome 😃 Maybe the active balancer actually works... because of the 50mV deviation threshold it should not balance in the flat part of the curve...it is not good, but at least it does not balance all the time in normal operation.
Thanks a lot for your generous donation. Hope you get some more subs and viewers on your channel. Yeah, I thought about their setup of the balancer and it could actually work that way, you describe it. You would end up with a 30-50mV deviation though even if fully charged. That works as well but could be better...
Great job Andy explaining what’s going on here. The more I see of these more complex BMSs with all the latest “features”, the more I think I’ll stick with my simple voltage based system with no closed loop communications. It works just fine without all the fuss.
At 45:21 it shows 3 modules online and available at 300AH BUT it does NOT show that 1 module is blocking Charge - the Victron value "Nr. of modules blocking charge/discharge".
Hi Andy - great video. I am learning a lot from your videos. BTW whilst the spec states the BMS start measuring at 0.7 A I have seen the BMS going down to 0.5 A. So the SW developers where right with the rest at 0.5 A - the guys doing the documentation where wrong
@@OffGridGarageAustralia but in a system like your shelf with 3 packs and each pack has a BMS. I Think It would be fine. The major issues I see is the negative not being switched and the lack of proper paralleling and ability to edit the settings while in parallel which every BMS fail at
@@OffGridGarageAustralia Hallo Andy, you told this several times. But i do not remember the reason for it. Could you please explain why you think JK is not good for parallel operation? Thanks!
@@Juergen_Miessmer pretty much as I explained here in the video at this position: ua-cam.com/video/AJLwxqHsHjk/v-deo.html In larger systems (2 or more batteries in parallel), no comms is a safety concern/risk. And also bad system design. If this is not clear, let me know and I'll make a dedicated video about it.
@@OffGridGarageAustralia Thanks ❤️ If this is the only reason, i feel save with my JKs, as they are monitored by 'battery safety controller' wich is configured to limit the charge current in this case. Even if the breaker popps (or is manually switched off)...!
Thanks Andy for all you do. I don't agree with your opinion about the balance settings of the Seplos balancer. Here in the summertime in Texas, I'm trying to run the air conditioner as much as possible. I'm off the grid. The range from 3.0 to 3.2 volt per cell, when the pack is nearly discharged, is where bottom balancing becomes important. My pack has one cell that is below average. For my pack, starting balancing at 50mV and stopping at 30mV would take care of top balancing AND bottom balancing. This would allow me to fully use my pack's total capacity instead of limiting the capacity to the lowest capacity cell. In the middle range from about 3.2 to 3.3 V, my pack typically stays balanced better than 30mV without any active balancing. To balance here would actually imbalance the pack. The 30mV stop setting would prevent this from happening, would it not? Contrast this with a NEEY smart balancer. The NEEY cannot do bottom balancing effectively without overbalancing in the flat range because when it starts balancing, it keeps going until it reaches 0mV difference. Most dumb balancers don't provide much current when the imbalance is in the 30-50mV range because they won't deliver their rated current until the cell voltage difference reaches nearly a volt. I'd be happy with a dumb balancer with no adjustments if it would deliver something close to its rated current (at a duty cycle approaching 50%) when the voltage difference is 30mV. Will the Seplos do this? Please keep up the good work. You are making great contributions to to our off-grid community.
I think some time in the past I mentioned I had an active balancer on all the time. I think it’s time I confess. After one year I’ve now changed to the muller bms for my 3 12v304ah batteries. My cells were always out of balance (even before the balancer - bad cells from aliexpress) but even worse after all the time balancing. From now on only balancing above 3.4 volts!
Wau! IMHO your best video so far! So well presented! Thank you again and again! Just wondering will we ever have any perfect BMS or should we finally develop it by ourself? I am pretty sure there are few guys with electronic backgrounds (including me) who can put our heads together and finally sort it out?
@@OffGridGarageAustralia amazing! One of the things I miss from the diybms is that you're can't make parallel packs with different relays. At the moment it has to be On or off, if one pack goes down for whatever and trips the alarm, the whole system goes off.
@@OffGridGarageAustralia hi Andy, has jk developed any of these industrial bms? I'm waiting to buy the right bms for my future victron system with multiple 48v battery packs
Hi! I discovered your channel a few weeks ago and I am truly amazed how much I already learned and how many small problems I solved thanks to you which were not really covered in the manual or other videos. I'm wondering, do you have a video available with a full set of recommended settings for this (or other) bms? Or maybe a spreadsheet? I found the "my settings" page but unfortunately for seplos there are no settings presented at all. Should I use the ones from jkbms that you posted and adjust just the settings mentioned in the linked videos?
Glad to see you have reviced your view on advantages of communication. The balancer to me is a HUGE dissapointment , as is the new version, also have been told it is still under development when i asked about it. They even failed to properly intergrate it, or make it manageble throught software/app Seems to me there arent any advantages to warrant an upgrade from my heap of 10e's.
@@OffGridGarageAustralia well, for a Victron system maybe, I personally like the 10e's the best as the work flawlessly with my sunsynk's ( deye/sol-ark), although I do run the 10c's I do feel seplos ( once again) missed the boat completely with this V3. Instead of improving their existing hardware platform by improving the software, they went to yet another OEM , and will start from scratch, and no backward compatibility..
In the software, can you set up one battery as you like it and then save that setting as a template which you could then hook up the other batteries one at a time and quickly load the settings to the new battery
Seplos just has too many options that can conflict and confuse. I’ll just hang on to my trio of JK’s that balance at full SOC(55.2) in minutes and no need to hold high SOC(stress) and no need to communicate with the AIO inverter (except for Solar Assistant). Simplicity that just works every day. Probably the most important thing I’ve learned from OGG is balance start voltage and the value of some type of smart balancer, so again thanks Andy!
I have 2 Seplos batteries and if they are well balanced at the beginning and the settings are correct, the system runs smoothly. My system runs without alarm and without problems for months and gives me every month 400kwh of battery power in bad weather or night times. Just because Andy tests everything, you do not have to adjust all parameters every day. I am satisfied with Seplos E10 BMS, has Bluetooth and works cleanly since installation. And again thanks Andy for show the shit Balancer from Seplos, i have a 1 A Balancer from JK in each Pack and all work fine and automatic.
Hey Andy & community, I can confirm from my experience that there is HARM TO DAMAGE the BMS when the POSITIVE is switched! I am using the Gobel PC200 BMS and now it is DAMAGED. 👉Pls. add new tests to the comparison sheet (see below). Here's my setup (unfortunately I can't share the schematic as a picture here): - the Gobel Power BMS switches the negative path - a 200 amps fuse is in the positive path On a discharge test with only 120 amps over several hours, the fuse blew due to overheating. The BMS was still running but showed a SHORT CIRCUIT. But there was definitely no short circuit on any of the cabling as ALL cables are protected by 6 amps fuses and all are intact. And here the drama begins: the SMD parts next to the fourth connection/terminals (for cells B13+, B14+, B15+, B16+) for the voltage probes melted. Unfortunately the BMS is damaged and no longer reports correct cell voltages. And guess which BMS I have used? It is the Gobel Power PC200. I am still in contact with one of the Gobel Power engineers for further examination... I can keep you posted if of interest. Let me know and feel free to propose a better forum/blog (maybe under the comments section of Andys' Gobel Power PC200 video?). As a proposal: + it would be beneficial to add another test category for "short circuit safe" to the BMS Comparison Spreadsheet: what happens with the BMS if the short circuit protection fuse is triggered (by it's test button)? @Andy: Did any vendor provide you with spare/test BMS to run such a test? + in case test equipment is available, doing an actual short circuit (with an nearly empty battery) would also bring some light into which BMS can safely handle such an extreme situation. Thanks, Bernd (from GER)
The 55,18 and 55,06 difference is for sure voltage drop in you’re system it’s not just the cable loss it’s every connector shunt fuse and other connections it all adds up even at 4 amps do the calculation
Is it just a coincidence that Seplos BMS v3.0 hardware works just like Pylontech with regard to how the batteries must be connected in parallel (RS485-2 to RS485-1) and how the master battery is determined by which battery has RS485-1 open. Is there a particular sequence to connecting a PC to RS485-1 without interrupting communications with the inverter, since RS485-1 is not empty any more?
Souldn't the top battery's settings be pushed to the lower batteries from the master battery?, Isn't that the standard for parallel battreies? If the main master battery doesn't push the setting to the other batteries, can you export the setting from the main battery you setup and import into the other batteries?
Nope, that does not work in any parallel setup. They are all solo fighters and can have different settings. They only use the master BMS as a reporting device but no settings are being replicated.
@OffGridGarageAustralia Hi Andy, I was in the process of ordering materials to build my solar battery when your new video about the Seplos V3 BMS landed. I am now very confused about what BMS to purchase for my system. I have a 6kw Solis Hybrid inverter, I am going to build a battery pack with 16 300a 3.2v batteries. What is the best BMS for my setup, JK BMS or the Seplos V2 of the Seplos V3? Any advice would be greatly appreciated.
@@gerardbduff great question, the JK can communicate via CAN and RS485, so there is a good chance it will work. But you never know and it's not guaranteed.
very nice video again. I an not sure if you have covered this in a previous video but you where showing how the BMS restricts the output of the solar controller. I was just looking to confirm what would happen if you have mutable solar controllers like you have on your main system, when it switches to the 10A absorption I assume this would be divided between all your solar controllers? I am currently running my system with lead acid batteries and have been watching closely to decide which BMS to chose when I do eventually invest in LiFePO4. but as you have demonstrated there is no BMS that hits all the marks. i have found your testing a fantastic source of information. keep the videos coming.
Hi, I also have two batteries with BMS 3.0 but I notice a big problem namely the inconsistency of the Soc between the two batteries, which being read by the inverter creates charging and discharging problems, in fact my storage with 2 5kWh batteries one had a 100% SOC and the other 70% with the same voltage, so 3kW was missing in charging and discharging because the batteries' SOC intervened and the inverter blocked everything
@@OffGridGarageAustralia sure both reached 100 with the same voltage, this morning I found one at 36 and the other at 44 but with the same voltage, the soc gets high while discharging
I'm a happy JK BMS User so far. But will upgrade next year to a Deye hybrid Inverter. As it lacks a canbus I will keep my DIY Battery and choose Battery voltage to communicate with the inverter. After watching this video I'm still looking into changing to Seplos. But wait for some improvement before I buy.
They probably still use 20Ohm resistor for balancing like in V1 and V2. And if you assume that you balance at 3.45V+ that should be a balance current of 170mA. I fear that measuring currents that small with a current clamp meter designed for 100+A is not the best idea. Measurement error is way too high. You need to use more precise instrument when checking the balance currents.
If you set the battery monitor to a smart shunt, does the CCL work correctly? In a non-BMS controlled environment the "maximum charge current" setting under DVCC does a similar thing to CCL, and still allows maximum current for loads from connected MPPT controllers (and controlled chargers like a multiplus), it just ensures that the current flowing into the battery is managed by that setting. Its also possible its not working correctly because of the "has DC system" setting, maybe with Victron Inverters delivering the load it would work.
Hi Andi, it would be interesting to understand if the batteries managed by these BMS can be put in series to obtain voltages higher than 48v. Or if there are BMS for high voltage, I know that the Deye Inverter high voltage battery version also accepts batteries with manual parameters.
Heya, that is real nice exteanded test. wel it still doesn't do what we hope it should do for our setup. well maybe something for the future to upgrade
@@OffGridGarageAustralia Thanks for reply. ❤️ A little more comfort for a single setup process, payed by adding less reliable functionality for ever... :-( Not a good idea i think...
Hi. Just found your vids regarding these Bms systems. I was looking at the 16s 200 amp version with active balancing Have you had any experiance with this model. ? I think the ones you are using are the passive model. I have a number of 110 amp/h. Cells. My plan was to make 3 batteries each with 32 cells in a s/p. Config With the Bms looking after two cells linked with 32 amp fuse each string will have 200 amp circuit breaker. The 32 amp fuse is to stop the other cell discharging into its twin if a short occurs in a cell. I need the Bms. To be able to talk to the sofar power inverter. Any tips. Or links to vids that may help appreciated Bill
My Mason 280L that was delivered last week from our UK supplier came with the v3 bms, the active balance board mounted into the case, but no active balance harness…. All a bit odd. And the bus bars don’t fit the B-, P- terminals as the holes are drilled wrong.
This happened to me as well, call Fogstar to send you the right accessory kit, you probably received a V2 kit. I called them and got a V3 accessory kit, containing the balancing harness, and the comms cable from balancer to bms. Also you need a power leads separate for the balancing board.
Switching negative is likely a design decision. N-channel power MOSFETs would require more complex gate control circuitry for "high-side" switching. It is both cost and reliability/stability issue to ensure the MOSFET´s never are driven into linear mode. I would say that a majority of power battery control designs use low-side swiching, particularly for higher DC system voltages. However, for generic products to be used outside of an enclosed design as a battery storage unit, positive/high-side switching is required.
I have only used Daly BMS so far, I bought a Seplos BMS 3.0 unit for 16S 200Amp LFP. However, after watching your videos, which were great, thank you, I am not sure I want to swap the Seplos in. What is it going to do for me that the Daly doesn't do? Seems like you have more reservations than praise. My system is a 2P16S 560Ah 48V prismatic battery into 8kva Multiplus II, no solar yet, just peak shaving.
The weird communication port behavior makes kind of sense if you look at how Modbus over RS485 works. Every client on the bus needs to have a unique ID in a very limited number range and it's rather tricky to assign these numbers dynamically in a system of unknown size, particularly if you want those IDs to survive a system restart. It's not impossible but adding something like DHCP on top of it would make the whole thing unnecessarily complex and prone to failure.That's why you usually have those DIP switches, which set the communication IDs of every node on the bus. What they did here to make it "self configuring" is to have two ports on each box. Port 1 is a modbus slave device and port 2 is a modbus master. The computer is also a master device and that's why it only works when it is connected to port 1. That's also why you have to connect port 1 on one box to port 2 on the second box (slave
Dear sir. How can i contact you. I need your advise with my battery
7 місяців тому
Andy, I've one question for you. I've assembled a Victron system (Multiplus II based) with Cerbo GX and SEPLOS battery package (304Ah) but I've noticed that as the battery gains voltage, charge is slower. Of course I assume this is correct for high SOCs but I've limited voltage via DVCC (53.3V) and as soon as it enters in 53V range is really slow. But if I charge from grid (via charge schedule) this does not happen. So, I'm injecting a lot of energy into the grid with my battery charging really slooooow at 75%. Any hints? (I think I've properly set Multi via VEConfigure for LiFePO4 batteries and linear charge curve). Thank you for your tips!
Excellent video Andy very informative for those inquisitive on BMS technical settings. I wonder though, as all the LFP manufacturers say to use the battery between 80 - 20% DOD for best cycle life is the BMS at 100% even relevant? Surely you just use the battery between 80 - 20% and the BMS takes care to balance any cells that go outside of 50mv from the average or can you only truly balance the battery by charging/ discharging to 100 - 0% occasionally?
The issue seems to be victron not being integrated problem, i bet that in any integrated system like voltronic or a deye inverter, if the bms tells not to charge, it will not prevent the system to use solar, because it will use the energy before reaching the charger. In the victron all the energy goes througt the charging part even if is not trying to charge..
That is a problem of off-grid or grid connected and how you charge (MPPT, inverter, or both). The solar charging has to stop in off-grid. Where else should the power go? This has nothing to to with Victron and Seplos in particular but I don't know any other system where the MPPTs are also controlled by the inverter or a controller. Usually they are just dumb chargers which in a Victron system the Venus OS controls them as well as charging coming from an inverter.
if there is a connector called "active balancer costumize" in BMS 3.0 that is connected to active balancer "dry contact port" in active balancer board, with 4 wires, (assuming VCC, GND, RX and TX) wouldnt that allow the BMS to control de balancer? BMS software configuration, should at least allow On/Off of the balancer. I assume "Separately or compatible seplos BMS 3.0" means that if you use BMS 3.0 then you should have some control to active balancer. Let me know if my thoughts make sense and if you can test this. Thanks for all the information your share with your videos :) Regards from Portugal
Hi Andy, I'm following you with great interest from Italy and I'm studying with you to build my own solar system. You are number 1! Seplos v3.0 doesn't seem like a complete bms to me. In your opinion, knowing the Tiangbanda password, is it a valid bms? do you recommend it? I have to integrate it into the Victron system. what do you recommend for good integration with the victron system and that it works well as a bms? gobel, paceBms or something else? Thank you
Hello Andy, I have two Seplos 280ah upright cases with V2 & V3 with Balancer. My V3 case BMS board is missing the BMS to Balance board connection port. So my Balance board is strictly MANUAL on / off only. It can not be controlled via the BMS / Software or bluetooth. You can only control it via the switch on the rear. This is terrible, they advertise as "active". Worst part is that V2 & V3 can NOT be used in PARALLEL . I wish they had a true Active Balancer like JK-BMS
You should have a dry contact port on bms, that’s where the balancer board plugs into the bms
7 місяців тому
This is just how I assembled my pack...however...i have to press button in order balance to start doing it's job. Should I unplug manual switch and leave only the dry contact?
I can confirm, that V2 does not talk to V3 BMS. V2 uses a different PC software and Android APK. You need a separate piece of software and a different APK for V3. This is not ideal because you need two pieces of software on your PC and phone for these batteries. Stick to either V2 or the near V3. There's already a V4
Hi, I have a Growatt SPH6000 inverter and two BMS V3 batteries. I've selected the Growatt protocol, but for some reason, the charge stops at 49%-55% and doesn't reach 100%. The SOC seems to level off at 51V. Do you know why this might be happening or what I can do to fix it?
I've been watching your videos ever since they started popping up on my UA-cam feed. After watching a few that focus on solar generation and how the output of your panels vary throughout the day in different seasons, I'm curious if you have answered the question about why you don't trim any of the trees which are causing shade on your panels? have you answered this question before? It seems like you spend a lot of time and attention on tweaking your system components for the most efficiency, but you never talk about addressing the trees. Thanks!
I'm a bit confused, in my simple experience panels supply solar charger, charger charges battery & shuts down when fully charged, but inverters are on the charging bus so they become a dc load even while the battery is charging. The battery should charge and stop and all excess should run inverters and therefore house. The battery should kick in when the panels can't supply enough for the house. The situation you describe with the batteries charging, then blocking the charge controller is a nightmare. Just keep the charger dumb, connect to the gx for monitoring but don't let the BMS control the charger. the BMS should shut down charge when it reaches full and won't stop the charger supplying power to load. Am I missing something fundamental here?
I will be building a 16s battery for use with multiplus2 GX in my AC coupled ESS. Would you advise Seplos V3 with active balancer or is there a better option? i'm confused lol
The 10mA Balance current is actually 0mA...you probably measure the change of magnetig field of the earth by moving the clamps orientation. That is the problem, when the meter is so sensitive... And yes...on this tiny as keyboard, I messed up and typed a "r" instead of a "t"...I should have done a peer review first 🤪
28:00 this section regarding "Battery Full, Solar Off" frustrates alot of people with "My Inverter takes power from Battery instead of Solar Panel" . . . sigh... Thanks to Andy. I believe this is your 3rd? or 4th? time explaining it.
This is a Victron problem not a BMS problem. Victron should be able to take the BMS limits and add the current needed to supply the loads on top of that number. Other integrated hybrid inverters don't have this problem, I use Deye, and the current limits from Seplos only affect battery charging.
Hi, andy The studio software only works with V3 bms as it was confirmed today by seplos. I dont know why it won't work with V2. Also iam expermanting with apxium bms whis seems identical to seplos, have you tested them? Thanks
@@joseo.cordero7286 Definitely not the software. They may put together the hardware from components they buy elsewhere. This could be the origin of the BMS as per our research: www.enfsolar.com/cn-energy and Seplos is rebranding it.
@@OffGridGarageAustraliaThis is very interesting if true. I am a little bit familiar with this company because LVFU uses their BMS. There are a few things odd about this company being the source of Seplos hardware and software. First, the documentation is a subset of Seplos user manual, and poorly written. Second, they say absolutely nothing on their website about making BMS hardware. Next, they use Seplos Battery Monitor for monitoring and configuration but they say nothing about firmware updates. I’ve long suspected some relationship between all of the various companies selling batteries using some version of the same BMS hardware. Different port configurations but Solar Assistant is compatible with all of them. I’ve suspected there’s a common supplier, but it’s been impossible to dig any deeper. It will be ironic if I rejected LVFU because they don’t make their own BMS, in favor of Seplos, if Seplos doesn’t make their own hardware or software either. Finally, at present only Seplos has BMS 3.0 as far as I know. Lvfu is definitely on a previous version.
Hi And I am from Germany, living in the Philippines for many years . I have learned a lot from your channel, also building Lithium batteries and learned a lot from you. I have now a Seplos Smart BMS 3.0 but i can not start it. I topple check all the connection, measured all the voltage all is normal. I also contact Seplos but there have no answer what the problem could be. Do you my have any Ide how i can start this Seplos Smart BMS 3.0. Regards Michael
I had the same problem - there is a B+ screw "connector" on the BMS itself which needs to be connected to the + of your battery. I could reset the BMS - it came up for 5 seconds and switched off again - with the wiring described above it stood on for 48 hours (parameter in the spelos config - if battery is not in use)
Well, sobering result with the V3. The only advantage is the different installation systems. It will probably take version 20 until Seplos understands how to build a BMS that meets the requirements.
the alarm trigger resetting SOC to 100% after 30m > pack voltage? seplos bms manual doesnt say anywhere! seplos manual not even say that alarm warning limits current to 10amp! it says almost nothing!! im from italy ,excuse me but i speak very little ,thank you so much for testing!
Victron, please stop giving the BMS direct control over the MPPTs! Have the communication up, get the BMS values but treat it for what it is, the BATTERY REQUEST and not a total system setting! No BMS will ever be a system controller, its not called a SMS for a reason. If the BMS says max 10A charging, perfect, use that data and honor it, but add any loads to that value and set the total as the charge current in the MPPT. I say it again, this is a Victron problem, not a BMS problem. Victron could fix all these issues if they just wanted to, but for some reason they don't. Maybe they just want you to buy a "better" (aka Victron) battery and BMS instead... In the off-grid scenario its just the loads that is not being powered by solar, in an ESS setup the export to grid would also be halted because the battery is full, yea makes total sense Victron, good job.
With the newest Venus OS 3.01 installed, I did not have this kind of issues in a ESS. I did not try it in a OFF-GRID installation jet. I think, that most of the BMS manufacturers do not implemet the CAN-protocoll correctly. That is also, because it is not available to the public because of some disclosure agreements with pylontech... Usually, I just have the smart shunt installed and select: No BMS controll in the DVCC and the SmartShunt as the main battery monitor in the System Setup. I have two Seplos 10C BMS in the ESS...wokrs fine without a active balancer, when you have the right settings set.
@@Raphael_Hofmann Thanks! I just upgraded to v3 a few days ago and had missed the new setting to disable BMS control. Now i will play some more with DVCC on, i had it turned off on v2.9 as it quite often shut down my MPPT and therefore stoped exporting my excess solar. Also big thanks for the UNI-T mods, now i need a new E-model of the 210! :P
@@tamazerd After the changes in DVCC, make sure, you go into the settings of the MulitPlus-II into Advanced and restart the VE.Bus-System. Also go to the solar charge controller and make sure, that it is not controlled by the BMS anymore. If it does not work emediately, just restart the Cerbo GX...that often fixes the issue.
This exactly... Its the Victron issue. The solar charger should take current limits of all BMSs and add a value of current needed by the DC/AC converter to supply the loads. That way you could easily supply your loads from the solar. All other Hybrid inverters do it like that. I can also confirm from my experience with Deye inverter, there is no problems with charge current limits, because those limits only affect the battery, not the system like in Victron case. Seplos works exactly right in this situation.
That separate active balancer with capacitors is also a bad idea. It can be turned on by BMS, but highest cell will turn it on, yet it will be on for all cells and start balancing even cells that are still in flat area of the curve. Smart balancer, that can turn on balancing per cell, not all or nothing, is needed.
It should not balance any cells under the start voltage. It discharges cells above and charges cells which are low. That's how it should work. We will see, I guess, when this thing comes out.
@@OffGridGarageAustraliaI meant that dumb active balancer with capacitors, that can be added to other BMSes and then turned on/off based on some logic in that BMS. The issue is that it indiscriminately balances all cells once it's turned on. Even those that should not be balanced yet. It would work with Li-ion cells, but for LFP it's bad.
Hello, does someone know something about Seplos V3 200A V10 vs V11? I like to do a FW update to 0.6 but there are two versions inside the folder like in the 100A version. The pdf from Andy how we can differ these versions with the shunt position does not fit to the 200A version.
This problem sounds familiar. When reading Victroc documentation, they state, multiple times, that solar production on the DC Side (OffGrid) should always obey the 1:1 rule im still not sure how the 1:1 rule works completely..but its very important regarding victron documenation not 100% sure; but it could be the 'the problem' is actualy a realy smart feature. cfr; its not a bug; its a feature. maybe
The Victron 1:1 Rule only applies where there is a PV Inverter providing AC to the AC-Out of the Victron inverter. Designed so that current coming in to the AC-Out of Victron inverter does not exceed its capacity.
It was "verschlimmbessert" ! How can you do something like that? Reading the new specifications, it sounds like the theme is that they have listened to their customers. But what do you have to conclude here? It actually only gets worse. What kind of drugs are they using? Well, the BMS is not coming into my house. I'll stick to my 10E. I've got that under control to a certain extent now. But one more question: can the new BMS Studio be used with the old BMS versions 10C and 10E? Because the software looks well structured now.
No, the 10C and 10E do not work with the new software...They just do not have the competence to do their one development. And a lot on information is lost in translation...the actual BMS engineers do not speak english and have no idea about the actual reqirements...they do not know about the whole system...they only see the Battery and the BMS. But even there they kind of suck 😅
@@Raphael_Hofmann Yes, unfortunately you are right. The Chinese are sometimes quite narrow-minded when it comes to their work. They rarely look to the left and right, and most of the time they can't see the forest for the trees. They have a hard time dealing with complex issues and dependencies. Although they are extremely good in many areas. The only thing that makes them change is when it comes to money. Then they pay attention. As already written, the specs read really well and raise a lot of hopes, but the implementation unfortunately failed. Therefore, the only way to build up pressure is through negative content. The approaches are good and have all the potential to become a truly ingenious system, but they have once again died on the way. They simply don't go the last mile. A pity really.
Hi Andy,and Thanks again for this video and your great explanation,but the problem it's always the Same BALANCING!! ridiculous and not acceptable, you have a Cell at 3.65v OVP and many others at only 3.35v?! I don't understand why they don't react because they have the tools to do it? why not couple an active balancer which will follow the balancing parameters already set at the BMS. incomprehensible?? And in addition, each time you charge these battery, you will always have these same cells with such a difference?! and I even think that in long use, the phenomenon will be amplified. Sometimes it's incomprehensible, they let it work like that, without worries?? I find it completely wrong! Since all this time he has been making battery and BMS? Anyway, thanks again Andy! (maybe 1 day they will correct this balancing problem??)
Thank you for this heads-up! In all this BMS seems to become a big disappointment, where I get the impression a lot of the choices are based upon pure cost (cutting). I'm sure if you dig deeper you will find more issues. I rather would opt for the Gobel Power BMS at any time, from the looks of it. I'm most disappointed that SEPLOS made some very odd choices. Maybe I call them mistakes.
Yeah, especially that I have advised them what features we need. They could have updates the V1 BMS with a new software and it would have been 95% perfect (except the balancing).
@@OffGridGarageAustralia That is truly sad! I work for a (totally non-related!) vendor and I strongly believe that your customers (via feedback, support, partners etc) are building 80% (or more) of your products. Especially the negatives as these tend to be remembered longer. I do like one can have the BMS with separated modules and integrated/piggy-back modules, version A, B and C, depending on your own build & needs. But that is a nice feature, not a requirement for me. The fact they fail on so many levels on the operational level already as you have shown, and the feedback & responses (or rather: not) you got so far, does not give me high hopes that they will fix the issues with a firmware-update. Especially when they are so stubborn, they could have been Dutch! ;-) I guess at its current state this BMS won't get many from Andy's Frogs. (only 2x I see in the spreadsheet, I believe that is generous haha)
Awesome channel Andy. Suggestion: Most of us are small battery users. 12v . Can you do a video showing how to troubleshoot a 12v battery with a built-in BMS.
Seplos be careful When they reach this state they start to build the v3 ☝️I know I got one sitting on the couch And two half me and half here’s .. and the make v4’s
Such a shame that v3 is unfortunately another flop. And as Andy said, this could be already solved twenty times by just doing software updates on V1 or V2. Even the 170mA balance current is actually completely fine in the long run (20Ohm discharge resistor and 3.4V cell voltage -> 170mA). The larger balance currents are anyway needed only for initial setup of the pack. Rarely you will encounter a imbalance in the system that could not be fixed by 170mA. And if you do a proper top balance before assembling the battery pack, even that is a non-issue. Also they are correct by limiting the balancing on adjacent cells, its about 0.6W power loss per resistor, and those resistors are quite densely packed on the PCB... But on the other hand, gluing a small heatsink on the bottom of the PCB below the balancing circuit, would solve most of those problems. Oh, and that active balancer... what a joke... 😂
any updates when neey will develope a BMS ? Seplos gets boring - I will not buy one anymore, one is enough. And I do not like this stupid mentality. I had a lot of troubles ordering one frome them cause they have a terrible shipping agent which had called me to pay 60€ on top for customs and fees. When they tried to deliver I recjected it and it went back to China. Why ? I had ordered meanwhile from aliexpress from the minerals shop a new one and this way I had to pay 3€ customs and 6€ dhl fees, so 51€ saved on top of a cheaper price, so 80€ saved in total. I am happy with the neey 4 balancer and the old seplos. If there will be a pace or any one else supporting the growatt can bus protocol for the sph series I will replace the seplos. I am tired and disappointed after 2 years and already 2 seplos bms.
Coming soon... Seplos is done, in my eyes. I will a few more tests with the batteries in parallel and then we move on. They keep developing new BMS (or better, let them develop) but there is no progress. I can only see the same mistakes over and over again. Once the new JK BMS comes out, it will be game over for many other BMS. Like they did before...
In addition, this relay balancer certainly has an horrible lifespan. We all know that any relay that is doing clic-clac all the day will die after a few month. Of course this is the reason why it is a separate board : it is engineered to be a consumable, ans this is the reason why the parameters are setup the way they are : they want to be sure it is acting all the time to be sure that you will change it just after the end of warranty period....
The reason why victron system is garbage, all in one inverter is much better from this system.. because it will supply the loads when battery is full, and no need communication with the battery... Andy, change your system for another people to understand this... Victron is the problem, not battery or BMS...
Low side switching allows the circuit designer to use N-channel FETs (which have lower on-state resistance than the complimentary P-channel FET) with no need for a gate driver circuit that would be required if the N-channel FET were in the positive line. The bottom line is that low side switching is i little bit cheaper (It's about money, surprise, surprise).
It's also significantly more efficient. (and a gate driver circuit can't work for constant on without an isolated power supply that is added to the voltage unless you use a p-fet)
And has less slam on shut off because of how real power works instead of the incorrect legacy of positive to negative.
And the only reason to switch positive is legacy thinking because of negative grounded frames from cars. Literally NOTHING else.
That's what I was thinking. I can't think of any reason to switch the positive on an ungrounded system.
@@jameshancockThere's other common negative reasons to switch positive, and they're about non isolated power supplies, multiple voltage supplies, and other legacy such as linear voltage regulators. It's not exclusive to the car stuff.
@@niktak1114 I concur in the case of isolated and un-grounded applications. But solar systems are generally required to be grounded. As Andy has already mentioned, communication lines between other components in a solar system could be subjected to undesirable current flow if the negative side of the battery is switched open. Perhaps there are ways to protect the comms circuits (opto-isolation or transformer coupling of low power signalling.) I'm not willing to condem negative side switching out of hand, but saving a small amount of money in one area of a design may easily be offset by cost s added elsewhere.
@@lua-nya non issolated power supplies should have shared negative. You can still switch the negative and the circuit is cut. Multiple voltage supplies shared negative switched no downside.
Linear regulation is only for negative voltages and even then you can switch he negative shared rail without issues.
It is literally negative as ground that is the only reason not to and that entirely bad decisions with cars. And 48v cars are isolated from the grounded frame so even that is being fixed and it’s literally for this reason.
I just love your passion! 😊 Trying to see why seplos v2 and v3 might have issue for balancing and came across this video. Asking for a friend on the forums.
Thanks Andy, as always, great work and very descriptive.
It certainly convinces me that I think I’ll just stick with my JK BMS’s: simple, reliable, relatively cheap, easy to set desired behaviour and they 100% do the job that I need. KISS 💋(which of course means “keep it simple stupid” :-)
I recently separated my packs and then kept one at a time at a fairly high voltage overnight with my power supply, for a good top balance with the JK BMS. Since then, PERFECT! 👍
We just need an updated version of the JK BMS to be able to limit current and add communication port with built-in Victron support.
Watch this space...
I wish I’d stumbled upon your channel 6yrs ago when I built by first LFP pack. There are a lot of takeaway snippets of info that you give in your commentary that took me weeks to figure out.
great video. lots of great and useful information.👍
Thanks a lot.
you are good engineer and i love victron engery and seplos
Wow, thank you so much Andy!!!
You are very welcome 😃
Maybe the active balancer actually works... because of the 50mV deviation threshold it should not balance in the flat part of the curve...it is not good, but at least it does not balance all the time in normal operation.
Thanks a lot for your generous donation. Hope you get some more subs and viewers on your channel.
Yeah, I thought about their setup of the balancer and it could actually work that way, you describe it. You would end up with a 30-50mV deviation though even if fully charged. That works as well but could be better...
@@OffGridGarageAustralia Absolutly...I got more than 20 new subscribers during the last night...I might have to do more videos for them 😂
Great job Andy explaining what’s going on here. The more I see of these more complex BMSs with all the latest “features”, the more I think I’ll stick with my simple voltage based system with no closed loop communications. It works just fine without all the fuss.
Yeah, don't change it if it runs well.
At 45:21 it shows 3 modules online and available at 300AH BUT it does NOT show that 1 module is blocking Charge - the Victron value "Nr. of modules blocking charge/discharge".
Yeah, good point. It is a bit of a definition how you see it. It also is never 100% compatible with inverters.
Hi Andy - great video. I am learning a lot from your videos. BTW whilst the spec states the BMS start measuring at 0.7 A I have seen the BMS going down to 0.5 A. So the SW developers where right with the rest at 0.5 A - the guys doing the documentation where wrong
Great video, I'm going with JK BMS 2A Balancer its it shown the ability to balancing with no dramas
For balancing purposes, yes, for parallel safety in larger systems, I think it is too risky.
@@OffGridGarageAustralia but in a system like your shelf with 3 packs and each pack has a BMS. I Think It would be fine. The major issues I see is the negative not being switched and the lack of proper paralleling and ability to edit the settings while in parallel which every BMS fail at
@@OffGridGarageAustralia
Hallo Andy, you told this several times. But i do not remember the reason for it.
Could you please explain why you think JK is not good for parallel operation?
Thanks!
@@Juergen_Miessmer pretty much as I explained here in the video at this position:
ua-cam.com/video/AJLwxqHsHjk/v-deo.html
In larger systems (2 or more batteries in parallel), no comms is a safety concern/risk. And also bad system design. If this is not clear, let me know and I'll make a dedicated video about it.
@@OffGridGarageAustralia
Thanks ❤️
If this is the only reason, i feel save with my JKs, as they are monitored by 'battery safety controller' wich is configured to limit the charge current in this case. Even if the breaker popps (or is manually switched off)...!
Thanks for the overview Andy! Very helpful. Seems like I can skip the V3 BMS for now.
Very cool video, thank you!
Thanks.
Hi Andy, a long but very nice video 👍😎
Thank you. Yes, I didn't want to make several videos again about one single BMS.
Great video Andy!
Sweet tool, and setup 😂 with a newspaper to boot 😎
Thanks Andy for all you do. I don't agree with your opinion about the balance settings of the Seplos balancer. Here in the summertime in Texas, I'm trying to run the air conditioner as much as possible. I'm off the grid. The range from 3.0 to 3.2 volt per cell, when the pack is nearly discharged, is where bottom balancing becomes important. My pack has one cell that is below average. For my pack, starting balancing at 50mV and stopping at 30mV would take care of top balancing AND bottom balancing. This would allow me to fully use my pack's total capacity instead of limiting the capacity to the lowest capacity cell. In the middle range from about 3.2 to 3.3 V, my pack typically stays balanced better than 30mV without any active balancing. To balance here would actually imbalance the pack. The 30mV stop setting would prevent this from happening, would it not? Contrast this with a NEEY smart balancer. The NEEY cannot do bottom balancing effectively without overbalancing in the flat range because when it starts balancing, it keeps going until it reaches 0mV difference. Most dumb balancers don't provide much current when the imbalance is in the 30-50mV range because they won't deliver their rated current until the cell voltage difference reaches nearly a volt. I'd be happy with a dumb balancer with no adjustments if it would deliver something close to its rated current (at a duty cycle approaching 50%) when the voltage difference is 30mV. Will the Seplos do this? Please keep up the good work. You are making great contributions to to our off-grid community.
I think some time in the past I mentioned I had an active balancer on all the time. I think it’s time I confess. After one year I’ve now changed to the muller bms for my 3 12v304ah batteries. My cells were always out of balance (even before the balancer - bad cells from aliexpress) but even worse after all the time balancing. From now on only balancing above 3.4 volts!
66.666 subscribers - congrats 2/3 of the magic 100k achieved.
It went past that 66.7 now
Oha, I missed, wow, unbelievable! 🤦♂️
Wau! IMHO your best video so far! So well presented! Thank you again and again! Just wondering will we ever have any perfect BMS or should we finally develop it by ourself? I am pretty sure there are few guys with electronic backgrounds (including me) who can put our heads together and finally sort it out?
Thank you for your feedback,
There is one BMS coming ins soon which I have huge hope for... From a company we all like so much...
Just keep in mind the diybms, cheers
@@garbi8526I am in contact with Stuart...
@@OffGridGarageAustralia amazing! One of the things I miss from the diybms is that you're can't make parallel packs with different relays. At the moment it has to be On or off, if one pack goes down for whatever and trips the alarm, the whole system goes off.
@@OffGridGarageAustralia hi Andy, has jk developed any of these industrial bms? I'm waiting to buy the right bms for my future victron system with multiple 48v battery packs
Hi! I discovered your channel a few weeks ago and I am truly amazed how much I already learned and how many small problems I solved thanks to you which were not really covered in the manual or other videos.
I'm wondering, do you have a video available with a full set of recommended settings for this (or other) bms? Or maybe a spreadsheet? I found the "my settings" page but unfortunately for seplos there are no settings presented at all. Should I use the ones from jkbms that you posted and adjust just the settings mentioned in the linked videos?
Glad to see you have reviced your view on advantages of communication.
The balancer to me is a HUGE dissapointment , as is the new version, also have been told it is still under development when i asked about it.
They even failed to properly intergrate it, or make it manageble throught software/app
Seems to me there arent any advantages to warrant an upgrade from my heap of 10e's.
The Seplos BMS V1 with FW2.8 was the best so far. Did everything and worked like a charme. Except balancing of course).
@@OffGridGarageAustralia well, for a Victron system maybe, I personally like the 10e's the best as the work flawlessly with my sunsynk's ( deye/sol-ark), although I do run the 10c's
I do feel seplos ( once again) missed the boat completely with this V3.
Instead of improving their existing hardware platform by improving the software, they went to yet another OEM , and will start from scratch, and no backward compatibility..
Does the 10e turn off the mppt when the batteries at 100% on the 10e?@@houseofancients
In the software, can you set up one battery as you like it and then save that setting as a template which you could then hook up the other batteries one at a time and quickly load the settings to the new battery
Yes, correct. You can use the export settings and import settings function. You still need to connect to all batteries individually though.
Hi. Do you have a link for the seplos battery monitor software?
Seplos just has too many options that can conflict and confuse. I’ll just hang on to my trio of JK’s that balance at full SOC(55.2) in minutes and no need to hold high SOC(stress) and no need to communicate with the AIO inverter (except for Solar Assistant). Simplicity that just works every day. Probably the most important thing I’ve learned from OGG is balance start voltage and the value of some type of smart balancer, so again thanks Andy!
Thanks Karl. I absolutely agree. 0 problems with my battery shelf since installation.
I have 2 Seplos batteries and if they are well balanced at the beginning and the settings are correct, the system runs smoothly. My system runs without alarm and without problems for months and gives me every month 400kwh of battery power in bad weather or night times. Just because Andy tests everything, you do not have to adjust all parameters every day. I am satisfied with Seplos E10 BMS, has Bluetooth and works cleanly since installation. And again thanks Andy for show the shit Balancer from Seplos, i have a 1 A Balancer from JK in each Pack and all work fine and automatic.
Hey Andy & community,
I can confirm from my experience that there is HARM TO DAMAGE the BMS when the POSITIVE is switched! I am using the Gobel PC200 BMS and now it is DAMAGED. 👉Pls. add new tests to the comparison sheet (see below).
Here's my setup (unfortunately I can't share the schematic as a picture here):
- the Gobel Power BMS switches the negative path
- a 200 amps fuse is in the positive path
On a discharge test with only 120 amps over several hours, the fuse blew due to overheating.
The BMS was still running but showed a SHORT CIRCUIT. But there was definitely no short circuit on any of the cabling as ALL cables are protected by 6 amps fuses and all are intact.
And here the drama begins: the SMD parts next to the fourth connection/terminals (for cells B13+, B14+, B15+, B16+) for the voltage probes melted.
Unfortunately the BMS is damaged and no longer reports correct cell voltages.
And guess which BMS I have used? It is the Gobel Power PC200.
I am still in contact with one of the Gobel Power engineers for further examination... I can keep you posted if of interest. Let me know and feel free to propose a better forum/blog (maybe under the comments section of Andys' Gobel Power PC200 video?).
As a proposal:
+ it would be beneficial to add another test category for "short circuit safe" to the BMS Comparison Spreadsheet: what happens with the BMS if the short circuit protection fuse is triggered (by it's test button)?
@Andy: Did any vendor provide you with spare/test BMS to run such a test?
+ in case test equipment is available, doing an actual short circuit (with an nearly empty battery) would also bring some light into which BMS can safely handle such an extreme situation.
Thanks,
Bernd (from GER)
The 55,18 and 55,06 difference is for sure voltage drop in you’re system it’s not just the cable loss it’s every connector shunt fuse and other connections it all adds up even at 4 amps do the calculation
It's not voltage drop, it's also if 0A are going in. It's measurement tolerance between devices. SVS fixes that.
Is it just a coincidence that Seplos BMS v3.0 hardware works just like Pylontech with regard to how the batteries must be connected in parallel (RS485-2 to RS485-1) and how the master battery is determined by which battery has RS485-1 open.
Is there a particular sequence to connecting a PC to RS485-1 without interrupting communications with the inverter, since RS485-1 is not empty any more?
hi Andy. would it be possible for you to export your settings so we can download and load in our bms?
Souldn't the top battery's settings be pushed to the lower batteries from the master battery?, Isn't that the standard for parallel battreies?
If the main master battery doesn't push the setting to the other batteries, can you export the setting from the main battery you setup and import into the other batteries?
Nope, that does not work in any parallel setup. They are all solo fighters and can have different settings. They only use the master BMS as a reporting device but no settings are being replicated.
@OffGridGarageAustralia Hi Andy, I was in the process of ordering materials to build my solar battery when your new video about the Seplos V3 BMS landed. I am now very confused about what BMS to purchase for my system. I have a 6kw Solis Hybrid inverter, I am going to build a battery pack with 16 300a 3.2v batteries. What is the best BMS for my setup, JK BMS or the Seplos V2 of the Seplos V3? Any advice would be greatly appreciated.
If you have only one battery banks, go with the JK-BMS
Thanks Andy, will the JK BMS communicate with the Solis inverter?
@@OffGridGarageAustralia Hello, and for 2 battery banks ? 😙
@@gerardbduff great question, the JK can communicate via CAN and RS485, so there is a good chance it will work. But you never know and it's not guaranteed.
Hi Andy.. I have a similar question to this. But what BMS do you advise for 2-3 (16x280ah) banks in parallel @@OffGridGarageAustralia (16x280ah)
Hi Andy, can I ask you if I can install the Neey 4A with the Seplos 3? thank you
very nice video again. I an not sure if you have covered this in a previous video but you where showing how the BMS restricts the output of the solar controller. I was just looking to confirm what would happen if you have mutable solar controllers like you have on your main system, when it switches to the 10A absorption I assume this would be divided between all your solar controllers?
I am currently running my system with lead acid batteries and have been watching closely to decide which BMS to chose when I do eventually invest in LiFePO4. but as you have demonstrated there is no BMS that hits all the marks. i have found your testing a fantastic source of information. keep the videos coming.
Hi, I also have two batteries with BMS 3.0 but I notice a big problem namely the inconsistency of the Soc between the two batteries, which being read by the inverter creates charging and discharging problems, in fact my storage with 2 5kWh batteries one had a 100% SOC and the other 70% with the same voltage, so 3kW was missing in charging and discharging because the batteries' SOC intervened and the inverter blocked everything
Have they both ever shown 100%?
@@OffGridGarageAustralia sure both reached 100 with the same voltage, this morning I found one at 36 and the other at 44 but with the same voltage, the soc gets high while discharging
Most excellent video Andy. Does anyone know whether the new BMS-Studio software is compatible with the older version 2 Seplos BMS hardware?
I'm a happy JK BMS User so far. But will upgrade next year to a Deye hybrid Inverter. As it lacks a canbus I will keep my DIY Battery and choose Battery voltage to communicate with the inverter. After watching this video I'm still looking into changing to Seplos. But wait for some improvement before I buy.
Hello Andy, is there a way to make the make the seplos v3 communicates with the older v3. Thank you
Hi, I think that clamps are not very precise. How could we assure that it is really 10mA ? Could you test the modified clamp with your powerbench ?
Oh, yeah, I thought the same and tested it with my power supply. It works a treat.
They probably still use 20Ohm resistor for balancing like in V1 and V2.
And if you assume that you balance at 3.45V+ that should be a balance current of 170mA.
I fear that measuring currents that small with a current clamp meter designed for 100+A is not the best idea.
Measurement error is way too high. You need to use more precise instrument when checking the balance currents.
If you set the battery monitor to a smart shunt, does the CCL work correctly? In a non-BMS controlled environment the "maximum charge current" setting under DVCC does a similar thing to CCL, and still allows maximum current for loads from connected MPPT controllers (and controlled chargers like a multiplus), it just ensures that the current flowing into the battery is managed by that setting.
Its also possible its not working correctly because of the "has DC system" setting, maybe with Victron Inverters delivering the load it would work.
Hi Andi, it would be interesting to understand if the batteries managed by these BMS can be put in series to obtain voltages higher than 48v. Or if there are BMS for high voltage, I know that the Deye Inverter high voltage battery version also accepts batteries with manual parameters.
That would be a question to the Seplos engineers. Seplos has HV DC batteries just not sure if they use the same sort of BMS.
Heya, that is real nice exteanded test. wel it still doesn't do what we hope it should do for our setup. well maybe something for the future to upgrade
Hello Andy,
what happens if you switch off the BMS in the middle? Will it still transmit the data from the last one?
No, it does not. If a BMS is off, the communication to other BMS after that is lost. Could be a safety problem.
@@OffGridGarageAustralia Thanks for reply. ❤️
A little more comfort for a single setup process, payed by adding less reliable functionality for ever... :-(
Not a good idea i think...
Hi Andy....been waiting
Hi. Just found your vids regarding these Bms systems. I was looking at the 16s 200 amp version with active balancing Have you had any experiance with this model. ? I think the ones you are using are the passive model.
I have a number of 110 amp/h. Cells. My plan was to make 3 batteries each with 32 cells in a s/p. Config With the Bms looking after two cells linked with 32 amp fuse each string will have 200 amp circuit breaker. The 32 amp fuse is to stop the other cell discharging into its twin if a short occurs in a cell. I need the Bms. To be able to talk to the sofar power inverter. Any tips. Or links to vids that may help appreciated Bill
My Mason 280L that was delivered last week from our UK supplier came with the v3 bms, the active balance board mounted into the case, but no active balance harness…. All a bit odd.
And the bus bars don’t fit the B-, P- terminals as the holes are drilled wrong.
Who was your UK supplier?
@@blacko3539 fogstar
This happened to me as well, call Fogstar to send you the right accessory kit, you probably received a V2 kit. I called them and got a V3 accessory kit, containing the balancing harness, and the comms cable from balancer to bms. Also you need a power leads separate for the balancing board.
Switching negative is likely a design decision. N-channel power MOSFETs would require more complex gate control circuitry for "high-side" switching. It is both cost and reliability/stability issue to ensure the MOSFET´s never are driven into linear mode. I would say that a majority of power battery control designs use low-side swiching, particularly for higher DC system voltages. However, for generic products to be used outside of an enclosed design as a battery storage unit, positive/high-side switching is required.
I have only used Daly BMS so far, I bought a Seplos BMS 3.0 unit for 16S 200Amp LFP. However, after watching your videos, which were great, thank you, I am not sure I want to swap the Seplos in. What is it going to do for me that the Daly doesn't do? Seems like you have more reservations than praise. My system is a 2P16S 560Ah 48V prismatic battery into 8kva Multiplus II, no solar yet, just peak shaving.
There is a new FW for the Seplos BMS3.0. Please watch this video here with all the info ua-cam.com/video/mE_J-lGppHg/v-deo.html
100% Soc reset ist not working for me with that bms 😢
The weird communication port behavior makes kind of sense if you look at how Modbus over RS485 works. Every client on the bus needs to have a unique ID in a very limited number range and it's rather tricky to assign these numbers dynamically in a system of unknown size, particularly if you want those IDs to survive a system restart. It's not impossible but adding something like DHCP on top of it would make the whole thing unnecessarily complex and prone to failure.That's why you usually have those DIP switches, which set the communication IDs of every node on the bus.
What they did here to make it "self configuring" is to have two ports on each box. Port 1 is a modbus slave device and port 2 is a modbus master. The computer is also a master device and that's why it only works when it is connected to port 1. That's also why you have to connect port 1 on one box to port 2 on the second box (slave
Hello Andy,
what happens if the breaker popps? Will the max charge current also be reduced then?
Dear sir. How can i contact you. I need your advise with my battery
Andy, I've one question for you. I've assembled a Victron system (Multiplus II based) with Cerbo GX and SEPLOS battery package (304Ah) but I've noticed that as the battery gains voltage, charge is slower. Of course I assume this is correct for high SOCs but I've limited voltage via DVCC (53.3V) and as soon as it enters in 53V range is really slow. But if I charge from grid (via charge schedule) this does not happen. So, I'm injecting a lot of energy into the grid with my battery charging really slooooow at 75%. Any hints? (I think I've properly set Multi via VEConfigure for LiFePO4 batteries and linear charge curve). Thank you for your tips!
Excellent video Andy very informative for those inquisitive on BMS technical settings. I wonder though, as all the LFP manufacturers say to use the battery between 80 - 20% DOD for best cycle life is the BMS at 100% even relevant? Surely you just use the battery between 80 - 20% and the BMS takes care to balance any cells that go outside of 50mv from the average or can you only truly balance the battery by charging/ discharging to 100 - 0% occasionally?
The issue seems to be victron not being integrated problem, i bet that in any integrated system like voltronic or a deye inverter, if the bms tells not to charge, it will not prevent the system to use solar, because it will use the energy before reaching the charger. In the victron all the energy goes througt the charging part even if is not trying to charge..
That is a problem of off-grid or grid connected and how you charge (MPPT, inverter, or both). The solar charging has to stop in off-grid. Where else should the power go?
This has nothing to to with Victron and Seplos in particular but I don't know any other system where the MPPTs are also controlled by the inverter or a controller. Usually they are just dumb chargers which in a Victron system the Venus OS controls them as well as charging coming from an inverter.
if there is a connector called "active balancer costumize" in BMS 3.0 that is connected to active balancer "dry contact port" in active balancer board, with 4 wires, (assuming VCC, GND, RX and TX) wouldnt that allow the BMS to control de balancer?
BMS software configuration, should at least allow On/Off of the balancer.
I assume "Separately or compatible seplos BMS 3.0" means that if you use BMS 3.0 then you should have some control to active balancer.
Let me know if my thoughts make sense and if you can test this.
Thanks for all the information your share with your videos :)
Regards from Portugal
Hi Andy, I'm following you with great interest from Italy and I'm studying with you to build my own solar system. You are number 1! Seplos v3.0 doesn't seem like a complete bms to me. In your opinion, knowing the Tiangbanda password, is it a valid bms? do you recommend it? I have to integrate it into the Victron system. what do you recommend for good integration with the victron system and that it works well as a bms? gobel, paceBms or something else? Thank you
Hello Andy, I have two Seplos 280ah upright cases with V2 & V3 with Balancer. My V3 case BMS board is missing the BMS to Balance board connection port. So my Balance board is strictly MANUAL on / off only. It can not be controlled via the BMS / Software or bluetooth. You can only control it via the switch on the rear. This is terrible, they advertise as "active". Worst part is that V2 & V3 can NOT be used in PARALLEL . I wish they had a true Active Balancer like JK-BMS
You should have a dry contact port on bms, that’s where the balancer board plugs into the bms
This is just how I assembled my pack...however...i have to press button in order balance to start doing it's job. Should I unplug manual switch and leave only the dry contact?
Is it confirmed that V2 (10e) and V3 cannot be paralleled together?
I can confirm, that V2 does not talk to V3 BMS. V2 uses a different PC software and Android APK. You need a separate piece of software and a different APK for V3. This is not ideal because you need two pieces of software on your PC and phone for these batteries. Stick to either V2 or the near V3. There's already a V4
I can confirm, my V3 BMS has a manual button to do the balancing. Meaning there is no active balancing.
Hello, I have one question : I need to use 2 SEPOS batteries in parallel. BMS are 10E version. How do I have to setup micro switch on BMS ?
Battery1: Switch5 on, all others off (Master)
Battery2: Switch1 on, all others off (Slave)
Hi, I have a Growatt SPH6000 inverter and two BMS V3 batteries. I've selected the Growatt protocol, but for some reason, the charge stops at 49%-55% and doesn't reach 100%. The SOC seems to level off at 51V. Do you know why this might be happening or what I can do to fix it?
I've been watching your videos ever since they started popping up on my UA-cam feed. After watching a few that focus on solar generation and how the output of your panels vary throughout the day in different seasons, I'm curious if you have answered the question about why you don't trim any of the trees which are causing shade on your panels? have you answered this question before? It seems like you spend a lot of time and attention on tweaking your system components for the most efficiency, but you never talk about addressing the trees. Thanks!
I'm a bit confused, in my simple experience panels supply solar charger, charger charges battery & shuts down when fully charged, but inverters are on the charging bus so they become a dc load even while the battery is charging. The battery should charge and stop and all excess should run inverters and therefore house. The battery should kick in when the panels can't supply enough for the house.
The situation you describe with the batteries charging, then blocking the charge controller is a nightmare. Just keep the charger dumb, connect to the gx for monitoring but don't let the BMS control the charger. the BMS should shut down charge when it reaches full and won't stop the charger supplying power to load. Am I missing something fundamental here?
I will be building a 16s battery for use with multiplus2 GX in my AC coupled ESS. Would you advise Seplos V3 with active balancer or is there a better option? i'm confused lol
Seems to me that we need to not bother with seplos can and just program the inverter/mppt and leave the BMS as a safety backup.
The 10mA Balance current is actually 0mA...you probably measure the change of magnetig field of the earth by moving the clamps orientation. That is the problem, when the meter is so sensitive...
And yes...on this tiny as keyboard, I messed up and typed a "r" instead of a "t"...I should have done a peer review first 🤪
Thanks for the Multimeter, Raphael. It is a unique version! I love it.
@@OffGridGarageAustralia I'm glad you enjoy it! Thanks for promoting my channel 😁
28:00 this section regarding "Battery Full, Solar Off" frustrates alot of people with "My Inverter takes power from Battery instead of Solar Panel" . . . sigh...
Thanks to Andy. I believe this is your 3rd? or 4th? time explaining it.
Yeah, it pops up more and more often. Only a few BMS can handle this.
This is a Victron problem not a BMS problem.
Victron should be able to take the BMS limits and add the current needed to supply the loads on top of that number.
Other integrated hybrid inverters don't have this problem, I use Deye, and the current limits from Seplos only affect battery charging.
Hi, andy
The studio software only works with V3 bms as it was confirmed today by seplos. I dont know why it won't work with V2.
Also iam expermanting with apxium bms whis seems identical to seplos, have you tested them?
Thanks
They are a rebadge of the Mason 280 kit, with a 10e bms
It's not their own development, unfortunately. Neither BMS nor software.
@@OffGridGarageAustralia You mean Seplos does not develop their BMS and the BMS software? I was under the impresion that they did.
@@joseo.cordero7286 Definitely not the software. They may put together the hardware from components they buy elsewhere. This could be the origin of the BMS as per our research: www.enfsolar.com/cn-energy
and Seplos is rebranding it.
@@OffGridGarageAustraliaThis is very interesting if true. I am a little bit familiar with this company because LVFU uses their BMS. There are a few things odd about this company being the source of Seplos hardware and software. First, the documentation is a subset of Seplos user manual, and poorly written. Second, they say absolutely nothing on their website about making BMS hardware. Next, they use Seplos Battery Monitor for monitoring and configuration but they say nothing about firmware updates.
I’ve long suspected some relationship between all of the various companies selling batteries using some version of the same BMS hardware. Different port configurations but Solar Assistant is compatible with all of them. I’ve suspected there’s a common supplier, but it’s been impossible to dig any deeper.
It will be ironic if I rejected LVFU because they don’t make their own BMS, in favor of Seplos, if Seplos doesn’t make their own hardware or software either.
Finally, at present only Seplos has BMS 3.0 as far as I know. Lvfu is definitely on a previous version.
Hi And
I am from Germany, living in the Philippines for many years .
I have learned a lot from your channel, also building Lithium batteries and learned
a lot from you. I have now a Seplos Smart BMS 3.0 but i can not start it.
I topple check all the connection, measured all the voltage all is normal.
I also contact Seplos but there have no answer what the problem could be.
Do you my have any Ide how i can start this Seplos Smart BMS 3.0.
Regards Michael
I had the same problem - there is a B+ screw "connector" on the BMS itself which needs to be connected to the + of your battery. I could reset the BMS - it came up for 5 seconds and switched off again - with the wiring described above it stood on for 48 hours (parameter in the spelos config - if battery is not in use)
I Think we need a seplos BMS V 4.0 ( Branding Andy)
Haha, if they would just listen to us...
Like# 4 - Woot, From Saigon today!!!
Well, sobering result with the V3. The only advantage is the different installation systems. It will probably take version 20 until Seplos understands how to build a BMS that meets the requirements.
Yeah, it seems, they make the same mistakes over and over again. A new version of a BMS won't help if the understanding is not there.
the alarm trigger resetting SOC to 100% after 30m > pack voltage? seplos bms manual doesnt say anywhere!
seplos manual not even say that alarm warning limits current to 10amp!
it says almost nothing!!
im from italy ,excuse me but i speak very little ,thank you so much for testing!
Victron, please stop giving the BMS direct control over the MPPTs! Have the communication up, get the BMS values but treat it for what it is, the BATTERY REQUEST and not a total system setting! No BMS will ever be a system controller, its not called a SMS for a reason. If the BMS says max 10A charging, perfect, use that data and honor it, but add any loads to that value and set the total as the charge current in the MPPT.
I say it again, this is a Victron problem, not a BMS problem. Victron could fix all these issues if they just wanted to, but for some reason they don't. Maybe they just want you to buy a "better" (aka Victron) battery and BMS instead...
In the off-grid scenario its just the loads that is not being powered by solar, in an ESS setup the export to grid would also be halted because the battery is full, yea makes total sense Victron, good job.
With the newest Venus OS 3.01 installed, I did not have this kind of issues in a ESS. I did not try it in a OFF-GRID installation jet.
I think, that most of the BMS manufacturers do not implemet the CAN-protocoll correctly. That is also, because it is not available to the public because of some disclosure agreements with pylontech...
Usually, I just have the smart shunt installed and select: No BMS controll in the DVCC and the SmartShunt as the main battery monitor in the System Setup.
I have two Seplos 10C BMS in the ESS...wokrs fine without a active balancer, when you have the right settings set.
@@Raphael_Hofmann Thanks! I just upgraded to v3 a few days ago and had missed the new setting to disable BMS control. Now i will play some more with DVCC on, i had it turned off on v2.9 as it quite often shut down my MPPT and therefore stoped exporting my excess solar.
Also big thanks for the UNI-T mods, now i need a new E-model of the 210! :P
@@tamazerd After the changes in DVCC, make sure, you go into the settings of the MulitPlus-II into Advanced and restart the VE.Bus-System. Also go to the solar charge controller and make sure, that it is not controlled by the BMS anymore. If it does not work emediately, just restart the Cerbo GX...that often fixes the issue.
This exactly... Its the Victron issue.
The solar charger should take current limits of all BMSs and add a value of current needed by the DC/AC converter to supply the loads.
That way you could easily supply your loads from the solar. All other Hybrid inverters do it like that.
I can also confirm from my experience with Deye inverter, there is no problems with charge current limits, because those limits only affect the battery, not the system like in Victron case.
Seplos works exactly right in this situation.
🐸🐸🐸
That separate active balancer with capacitors is also a bad idea. It can be turned on by BMS, but highest cell will turn it on, yet it will be on for all cells and start balancing even cells that are still in flat area of the curve. Smart balancer, that can turn on balancing per cell, not all or nothing, is needed.
It should not balance any cells under the start voltage. It discharges cells above and charges cells which are low. That's how it should work. We will see, I guess, when this thing comes out.
@@OffGridGarageAustraliaI meant that dumb active balancer with capacitors, that can be added to other BMSes and then turned on/off based on some logic in that BMS. The issue is that it indiscriminately balances all cells once it's turned on. Even those that should not be balanced yet. It would work with Li-ion cells, but for LFP it's bad.
Hello, does someone know something about Seplos V3 200A V10 vs V11? I like to do a FW update to 0.6 but there are two versions inside the folder like in the 100A version. The pdf from Andy how we can differ these versions with the shunt position does not fit to the 200A version.
0.6 is for V10
😎
This problem sounds familiar.
When reading Victroc documentation, they state, multiple times, that solar production on the DC Side (OffGrid) should always obey the 1:1 rule
im still not sure how the 1:1 rule works completely..but its very important regarding victron documenation
not 100% sure; but it could be the 'the problem' is actualy a realy smart feature. cfr; its not a bug; its a feature. maybe
The Victron 1:1 Rule only applies where there is a PV Inverter providing AC to the AC-Out of the Victron inverter. Designed so that current coming in to the AC-Out of Victron inverter does not exceed its capacity.
@@bipedalprimate8109 Thanks for the info; this seems logic indeed.
That is for AC coupled system only...
There's no BMS that is accaptable. Even JK. I don't like solded wires. And chinese apps for handling.Thank you for testing all these stuff for us!
It was "verschlimmbessert" !
How can you do something like that? Reading the new specifications, it sounds like the theme is that they have listened to their customers. But what do you have to conclude here? It actually only gets worse. What kind of drugs are they using? Well, the BMS is not coming into my house. I'll stick to my 10E. I've got that under control to a certain extent now.
But one more question: can the new BMS Studio be used with the old BMS versions 10C and 10E? Because the software looks well structured now.
No, the 10C and 10E do not work with the new software...They just do not have the competence to do their one development. And a lot on information is lost in translation...the actual BMS engineers do not speak english and have no idea about the actual reqirements...they do not know about the whole system...they only see the Battery and the BMS. But even there they kind of suck 😅
@@Raphael_Hofmann
Yes, unfortunately you are right. The Chinese are sometimes quite narrow-minded when it comes to their work. They rarely look to the left and right, and most of the time they can't see the forest for the trees. They have a hard time dealing with complex issues and dependencies. Although they are extremely good in many areas. The only thing that makes them change is when it comes to money. Then they pay attention. As already written, the specs read really well and raise a lot of hopes, but the implementation unfortunately failed. Therefore, the only way to build up pressure is through negative content. The approaches are good and have all the potential to become a truly ingenious system, but they have once again died on the way. They simply don't go the last mile. A pity really.
Hi Andy,and Thanks again for this video and your great explanation,but the problem it's always the Same BALANCING!! ridiculous and not acceptable, you have a Cell at 3.65v OVP and many others at only 3.35v?! I don't understand why they don't react because they have the tools to do it? why not couple an active balancer which will follow the balancing parameters already set at the BMS. incomprehensible??
And in addition, each time you charge these battery, you will always have these same cells with such a difference?! and I even think that in long use, the phenomenon will be amplified.
Sometimes it's incomprehensible, they let it work like that, without worries??
I find it completely wrong! Since all this time he has been making battery and BMS?
Anyway, thanks again Andy! (maybe 1 day they will correct this balancing problem??)
my bms only works with the protection numbers. The alarm numbers i believe could be anything. Lion
Thank you for this heads-up!
In all this BMS seems to become a big disappointment, where I get the impression a lot of the choices are based upon pure cost (cutting). I'm sure if you dig deeper you will find more issues.
I rather would opt for the Gobel Power BMS at any time, from the looks of it.
I'm most disappointed that SEPLOS made some very odd choices.
Maybe I call them mistakes.
Yeah, especially that I have advised them what features we need. They could have updates the V1 BMS with a new software and it would have been 95% perfect (except the balancing).
@@OffGridGarageAustralia That is truly sad! I work for a (totally non-related!) vendor and I strongly believe that your customers (via feedback, support, partners etc) are building 80% (or more) of your products. Especially the negatives as these tend to be remembered longer.
I do like one can have the BMS with separated modules and integrated/piggy-back modules, version A, B and C, depending on your own build & needs. But that is a nice feature, not a requirement for me. The fact they fail on so many levels on the operational level already as you have shown, and the feedback & responses (or rather: not) you got so far, does not give me high hopes that they will fix the issues with a firmware-update. Especially when they are so stubborn, they could have been Dutch! ;-)
I guess at its current state this BMS won't get many from Andy's Frogs. (only 2x I see in the spreadsheet, I believe that is generous haha)
Folks you don’t need an active balancer if you are buying a pre-assembled rack battery, or you have decent cells that you have properly top balanced.
Though if this only has 10mA balance current, that really isn’t good enough!
Awesome channel Andy.
Suggestion: Most of us are small battery users. 12v .
Can you do a video showing how to troubleshoot a 12v battery with a built-in BMS.
Those chinese from Seplos needs some German engineering skills !!! Love from Belgium ! 😄
Funny because 0A is < 0.5A. But "IF charging AND chg_current
No, they haven't tested it properly. Very frustrating having to do all the work for them.
Just give me a simple port i can hook my Neey up too :)
Exactly, that's all we would need.
Can i disconnect passive balancing and use a Neey?
This clamp metter its easy change the settings you need only I2C to USB
Sounds like this whole scheme is remarkably sophisticated but not really ready for use by the general public.
Seplos be careful
When they reach this state they start to build the v3
☝️I know I got one sitting on the couch
And two half me and half here’s ..
and the make v4’s
The V1 was almost perfect and needed only a few software tweaks over time. A shame, they don't build them themselves...
0:32 "Srarts" 😆😆😅😅🤣🤣🤣😂😂😂😂😂
8:50 Balancing at 3.000V ??? that aint good. it looks like a Neeey version 0.1
Such a shame that v3 is unfortunately another flop.
And as Andy said, this could be already solved twenty times by just doing software updates on V1 or V2.
Even the 170mA balance current is actually completely fine in the long run (20Ohm discharge resistor and 3.4V cell voltage -> 170mA).
The larger balance currents are anyway needed only for initial setup of the pack.
Rarely you will encounter a imbalance in the system that could not be fixed by 170mA.
And if you do a proper top balance before assembling the battery pack, even that is a non-issue.
Also they are correct by limiting the balancing on adjacent cells, its about 0.6W power loss per resistor, and those resistors are quite densely packed on the PCB...
But on the other hand, gluing a small heatsink on the bottom of the PCB below the balancing circuit, would solve most of those problems.
Oh, and that active balancer... what a joke... 😂
I'm sooo glad I'm running the newest JK which already has a good functioning active balancer than this crapfest.
any updates when neey will develope a BMS ?
Seplos gets boring - I will not buy one anymore, one is enough. And I do not like this stupid mentality.
I had a lot of troubles ordering one frome them cause they have a terrible shipping agent which had called me to pay 60€ on top for customs and fees. When they tried to deliver I recjected it and it went back to China. Why ?
I had ordered meanwhile from aliexpress from the minerals shop a new one and this way I had to pay 3€ customs and 6€ dhl fees, so 51€ saved on top of a cheaper price, so 80€ saved in total.
I am happy with the neey 4 balancer and the old seplos.
If there will be a pace or any one else supporting the growatt can bus protocol for the sph series I will replace the seplos. I am tired and disappointed after 2 years and already 2 seplos bms.
Coming soon...
Seplos is done, in my eyes. I will a few more tests with the batteries in parallel and then we move on. They keep developing new BMS (or better, let them develop) but there is no progress. I can only see the same mistakes over and over again.
Once the new JK BMS comes out, it will be game over for many other BMS. Like they did before...
Ouch, we can't generalise like that on the open interwebs...
In addition, this relay balancer certainly has an horrible lifespan. We all know that any relay that is doing clic-clac all the day will die after a few month. Of course this is the reason why it is a separate board : it is engineered to be a consumable, ans this is the reason why the parameters are setup the way they are : they want to be sure it is acting all the time to be sure that you will change it just after the end of warranty period....
4th!
The reason why victron system is garbage, all in one inverter is much better from this system.. because it will supply the loads when battery is full, and no need communication with the battery...
Andy, change your system for another people to understand this...
Victron is the problem, not battery or BMS...
10 mA no no no serious?? Lol