June 11, 20214 yr So the only real danger I can see with paralleling with individual BMSes is the chance for measurement inaccuracies between them. So with two batteries in parallel, for example, one could decide that 98A is too much before it shuts down while the other measures slightly higher and shuts down at 100A. When the first shuts down early, suddenly the second sees the full 198A before it shuts down. Maybe the BMS won't like that and cause a failure? Same could go for pack voltage at low cut-off. These are fringe cases but still something a manufacturer has to be prepared for... These are my speculations at least. The only way Hubble or Vision could get around this is to have very accurate measurements of current/cell voltage or to communicate between BMSes. I'm pretty sure none of them have some secret Comms protocol so it must be the accuracy? Either way, if you're just looking for more capacity then I see no issues in paralleling if you're staying away from low cut off. If you're wanting to parallel to draw more than 100A and closer to 200A then maybe think twice.
June 11, 20214 yr 2 minutes ago, Igubu said: From the hubble spec sheet: prismatic I cannot say this enough: dont buy a battery based on capacity on paper, with lithium especially- there are many more components that decide if its something that will last or a cheapie… capacity/ volts not everything- you can build 8-cell or 16-cell batteries, both seem “similar” but the latter a lot more actual capacity.. Honestly it is worth (for those who have not purchased yet) buying something like this rather than the slightly cheaper variants- they are also local in Jhb and Ctn - so support is at hand- I have seen some hubble talk on the forum as well- unsure if there are resellers - I only know of a CT based installer/reseller, perhaps you can buy rrp from the direct (no idea)? That's an old spec sheet, The older Hubble S-100 (with serial # prefixes starting with DS) can be configured in series or parallel, but not both. The newer Hubble S-100 batteries (with serial # prefixes starting with HL) can be configured up to 4 batteries in series with 4 batteries in parallel. New spec sheet is here: https://www.hubblelithium.co.za/uploads/1/3/5/0/135079623/hubble_lithium_s-100_a4_pamphet_v2.2.pdf
June 11, 20214 yr When this thread started several moons ago who ever thought it would result in such enriching discussions. Thanks to @Igubu @JaseZA @Rooney Hat @YellowTapemeasure and everyone else that's been contributing so much knowledge, sharing their experience, and providing wisdom. Together we'll beat the power woes!
June 11, 20214 yr 3 minutes ago, Igubu said: From the hubble spec sheet: prismatic I don't trust "Prismatic" in spec sheets as it seems some people use it to refer to hardcase cells while others use it to mean 'not cylindrical' but possibly still pouch. That might not be the case here though... Anyone want to buy one and open it up? 😂
June 11, 20214 yr 41 minutes ago, JaseZA said: So the only real danger I can see with paralleling with individual BMSes is the chance for measurement inaccuracies between them. So with two batteries in parallel, for example, one could decide that 98A is too much before it shuts down while the other measures slightly higher and shuts down at 100A. When the first shuts down early, suddenly the second sees the full 198A before it shuts down. Maybe the BMS won't like that and cause a failure? Same could go for pack voltage at low cut-off. These are fringe cases but still something a manufacturer has to be prepared for... These are my speculations at least. The only way Hubble or Vision could get around this is to have very accurate measurements of current/cell voltage or to communicate between BMSes. I'm pretty sure none of them have some secret Comms protocol so it must be the accuracy? Either way, if you're just looking for more capacity then I see no issues in paralleling if you're staying away from low cut off. If you're wanting to parallel to draw more than 100A and closer to 200A then maybe think twice. Thanks, that's the plan - I wanted to buy one 200ah which would suit my current needs, but then have the option of adding another 200ah later if my needs change for whatever reason.
June 11, 20214 yr 43 minutes ago, YellowTapemeasure said: That's an old spec sheet, The older Hubble S-100 (with serial # prefixes starting with DS) can be configured in series or parallel, but not both. The newer Hubble S-100 batteries (with serial # prefixes starting with HL) can be configured up to 4 batteries in series with 4 batteries in parallel. New spec sheet is here: https://www.hubblelithium.co.za/uploads/1/3/5/0/135079623/hubble_lithium_s-100_a4_pamphet_v2.2.pdf Thanks! Was only using it to demonstrate prismatic cells Edited June 11, 20214 yr by Igubu *tx
June 11, 20214 yr On 2021/05/02 at 8:45 PM, nembudziya said: The installer came today and seems this was just about right. He recommended using the AGM setting on the inverter and I wasn't sure if that was right but he's the "expert" so I was kinda stuck. Any thoughts? Vision SP 12-200 LiFeP04 User Manual.pdf 717.66 kB · 9 downloads Late to the party, not sure if anyone has replied on this. But you should not be using AGM settings for these batteries. AGMs charging and float voltages are way higher than Lithium. I believe the max you should be using is 53.2v float and charge. So it will likely be USE setting.
June 11, 20214 yr Apologies, I just saw the manual, and it states the charging voltage is 13.6-13.8v, that's pretty much the same as my old AGMs... seems odd, but ok, but it doesn't indicate float voltage. I assume that should be the same as charge v, and that is higher in AGM standard settings. So it might still be best to have it as USE setting with float and charging at the same. So with two of them, it should be 27.2v - 27.6v. But I am in no way an expert, so I would get an expert opinion. Edited June 11, 20214 yr by Rooney Hat
June 11, 20214 yr On 2021/05/27 at 4:29 PM, nembudziya said: Hello everyone just thought to post an update on the 3kVa/24V backup system that I've been working on. Last week the power utility informed us that they would be working on our power line and we would be switched off the grid from 7am on Monday (24 May 2021) to Tuesday (25 May 2021) at 4pm. They switched us off at about 11:40am on Monday and power was only restored for a few minutes this afternoon (Thursday 27 May 2021). During that time, with regulated usage of our fridge and freezer, we were able to run our lights, power laptops and charge phones from 11:40am on Monday up to approximately 3pm on Wednesday. By regulated usage of our fridge and freezer I mean we ran them for about 4 hours on battery power on Monday and 5 hours on Tuesday (3 in the morning, 2 in the late afternoon). As for the inverter settings, the cutoff voltage is 24V and following the advise of someone on this forum we used the following battery settings: Battery type: USER Settings 26: 27.6V Setting 27: 26.8V Hopefully I'll be able to add PV power soon and that should save us the frustrations of dealing with poor service from the utility. Hey @Rooney Hat thanks for chiming in. I got help from someone on this forum and I'm currently using the settings indicated in a previous post (quoted above)
June 11, 20214 yr 58 minutes ago, Rooney Hat said: Apologies, I just saw the manual, and it states the charging voltage is 13.6-13.8v, that's pretty much the same as my old AGMs... seems odd, but ok, but it doesn't indicate float voltage. I assume that should be the same as charge v, and that is higher in AGM standard settings. So it might still be best to have it as USE setting with float and charging at the same. So with two of them, it should be 27.2v - 27.6v. But I am in no way an expert, so I would get an expert opinion. Rooney The manual I have on line says charge voltage of 14.6 v ? Of course the manual in the box could be correct
June 11, 20214 yr This is where lithium replacement batteries require more understanding.. without CAN you have to set them correctly- never equalise a lithium battery. So for axpert type inverters without comms, or these batteries without comms just a basic bms, you have to follow the manufacturers specs. There should be a bulk or float or both V value, then perhaps something called absorbtion and also equalisation. Float is max the manufacturer says what is termed as “full” bulk is sometimes a little more. Equal and absorb must be the same, sometimes these could be “bulk”, and any setting around de sulphate/equalise every x days” should be disabled or set to 0. As an example on a custom 5kW 48V lithium and a sunsynk- the float for the specific battery and it’s bms and amount of cells are: 54V. Absorb and Equal both 57.6V, and days set to 0. The axperts name them differently- and your values will be different- but it’s important to set them correct. Remember a difference of just 2V on the top For lithium (so 50V vs 52V) is almoat 50% capacity! Same applies to a float of 13.5V might be 100% and 12.5V 40%…. this is why it’s important to understand why an inverter with CAN, and a battery with CAN Is worth it- you loose at least 30% usable capacity without it- something to consider perhaps when a “replacement” costs only 20% less … Edited June 11, 20214 yr by Igubu *
June 11, 20214 yr 11 minutes ago, Igubu said: this is why it’s important to understand why an inverter with CAN, and a battery with CAN Is worth it- you loose at least 30% usable capacity without it- something to consider perhaps when a “replacement” costs only 20% less … https://pclinkshop.co.za/ups-inverters-and-powerbank/inverters/bms-communication-box-for-using-lithium-batteries-on-rct-axpert-inverters/ I'm assuming this would solve that problem?
June 11, 20214 yr Unfortunately not:( cant find much detail on that device, but have a feeling it’s a RS485-serial/usb converter. So might “talk” to your axpert, but the intelligence comes from the bms side- and doubt the said bms will have the same level of detail to feed and control the inverter. There is no quick fix or easy way to fake “can/rs485” comms unfortunately. It is Very complicated to get right- the reason so much money is spent on certification for any specific brand of battery to be compatibly with the different inverter brands. The guys building batteries will tell you the same- it’s months and months of work/coding/testing etc If you want to future proof- get rid of the axpert, Buy something decent like a 5 or 8kW sunsynk/deye (full hybrid inverter) - and start with a battery like the am-2 hubble or similar.. efficiencies of hybrids vs - lets be honest and call the axperts what they are : “cheap and nasty” - is a long other topic…. Of course this all costs money- but if you look at backup power as a requirement for the next couple of years, rather invest in the correct kit now that will last 5-10years, have 5 & 10 Y guarantees- for 30% more upfront - or go as cheap as you can and in 2 years have to replace it all again… Whats that saying “goedkoop koop Is duur koop” - thats why these forums exist I suppose- for free advice from users who's already done it all (a few times)
June 13, 20214 yr From my limited understanding this is is a 4s BMS @ 12v 200amp ? Technically it should be possible to replace the BMS with a 16s BMS @ 48v 200amp which offers canbus, 485, monitoring etc. (daly or equivalent), it would also be possible to keep the pouch cells in their existing plastic "containers" effectively building a 9.6/10 kw battery. Alternatively one could just replace the 4s BMS with a parallel capable bms or one which provides canbus. On 2021/06/10 at 9:09 PM, JaseZA said: Looking from the side I could now see how the battery was put together - 8 x 100ah pouch cells stacked on top of each other in pairs - each pair flipped to connect in series with the next pair. Nice strong connections between cells and proper balance leads to each pair. Even a temperature sensor! I checked the balance between the cells and with this battery close to fully charged I had exactly 3.401V on each cell! Very impressive! BMS looks beefy as well with decent soldering and connections. I guess heat might be the only concern in a sealed battery like this but there's a decent bit of metal to dissipate it. I'm now quite undecided on what I want to do with my setup. I'm much less keen on mixing my hard case 100ah cells with these pouch cells. So I could just put it back together and sell it on and stick with what I have. The other option is to try and find two more and create one big 48V 10kwh battery with new BMS and then retire my old bank... decisions decisions 😅 I haven't closed it up yet, if anyone has any requests or questions about it.
June 13, 20214 yr 18 hours ago, xcooling said: Technically it should be possible to replace the BMS with a 16s BMS @ 48v 200amp So this is now exactly my plan. I'm going to combine 4 of these into one big box with my 8-16S, 320A Smart BMS that I currently run. The layout works nicely in that I can put them all side by side and have easy connections using the existing cables and even use the current cell monitoring cables adapted into the BMS. Will be quite a big, heavy box though, so might have to rethink that once I've got them all together. Will definitely post some pictures of it once I'm done!
June 13, 20214 yr 27 minutes ago, JaseZA said: Will definitely post some pictures of it once I'm done! This would be fun to see, I'm looking forward to this project. All the best!
June 14, 20214 yr Do you have CAN working on your bms? I’d be very surprised- but good luck with it. It’s very tricky to get CAN working properly- as mentioned before if your bms has it- unfortunately doesn’t mean it is programmed and talking to an inverter- you do need the software and all programmable parameters of your cells (there are almost 100 programmable paramaters on some Of the bms models!) They are notoriously easy to blow- and please please be careful- you are working with high currents - lethal. If you dont have experience or are qualified in electricity- you are playing with dangerous things. last bit of advice is for 300+ amp bank and single inverter you need the correct inline fuses - probably 400 amps- and at least 50MM battery cables- these are expensive. Thats part of the reason the 3 - 5kW lithium batteries are usually built as 100A- you can then use 35mm cables and 160A fuses- which are very common and affordable. Large high current banks are not recommended for home use- rather parallel smaller amp banks..
June 14, 20214 yr 16 minutes ago, Igubu said: Do you have CAN working on your bms? I haven't got direct CAN interface between the BMS and inverter but I have a raspberrypi collecting all the BMS data, reporting that to my Linux server running Home Assistant. This can then make all the decisions the inverter would and I can send commands through to the inverter via the same RaspberryPi. Been working well for the last year or so - but yes, definitely not the easiest way of doing things! Hard to find a CAN compatible DIY BMS for a reasonable price... But the beauty of this system is I have access to all the data and can manipulate it as far as my imagination will take me! Nothing like getting a Whatsapp from your House telling you that cell 3 on your battery is getting a bit low and you should check the balancing 😆 I run 120A fuses in the main battery lines - proper high current disconnect type fuses. The BMS is also set to cut at 120A. Drawing 5kw at 48V = 104A so I should never see currents that high anyway. I run doubled 35mm2 cables for positive and negative to the battery. I initially ran just one but the voltage drop across that at 100A is something to behold! If I recall I calculated the loss at around 100W to heat!
June 15, 20214 yr On 2021/06/14 at 8:07 AM, JaseZA said: I haven't got direct CAN interface between the BMS and inverter but I have a raspberrypi collecting all the BMS data, reporting that to my Linux server running Home Assistant. This can then make all the decisions the inverter would and I can send commands through to the inverter via the same RaspberryPi. Been working well for the last year or so - but yes, definitely not the easiest way of doing things! Hard to find a CAN compatible DIY BMS for a reasonable price... But the beauty of this system is I have access to all the data and can manipulate it as far as my imagination will take me! Nothing like getting a Whatsapp from your House telling you that cell 3 on your battery is getting a bit low and you should check the balancing 😆 I run 120A fuses in the main battery lines - proper high current disconnect type fuses. The BMS is also set to cut at 120A. Drawing 5kw at 48V = 104A so I should never see currents that high anyway. I run doubled 35mm2 cables for positive and negative to the battery. I initially ran just one but the voltage drop across that at 100A is something to behold! If I recall I calculated the loss at around 100W to heat! Amazing definitely then an “advanced”’user if you are looking for a can compatible bms- seplos in china has a 16s 100A and 200A unit… but unsure about single imports- last I heard you were looking at R4500 for the 100A … which also explains why proper batteries with CAN are a few grand more you really do pay for a proper bms… Edited June 15, 20214 yr by Igubu *
June 18, 20214 yr Hi Thanks for all the help and advice offered in this thread, I have concluded that the Mecer would suit my needs, so I've ordered and it's arrived yesterday. It was sitting on 13.27v for what it's worth. I'd just like to confirm that these settings are correct? The inverter's charge current is unfortunately set in increments of 10, so it's either 10 or 20. For the time being(under stage 1 &2) I should probably set it to 10 and see how that goes? Thanks Edited June 18, 20214 yr by johanp
June 18, 20214 yr 49 minutes ago, johanp said: The inverter's charge current is unfortunately set in increments of 10, so it's either 10 or 20. I would happily leave it at 20A - that's only 0.1C for a 200ah battery. I intend to charge mine at 50A or 0.25C. Most lithium recommendations I've seen are around 0.5C max for continuous and 1C max intermittent. I generally like having the inverter DC low cut off higher than the battery BMS. This way the inverter cuts on it's own terms rather than there being a chance the BMS might cut first. This is unlikely though as there is almost always some voltage drop across the cables from battery to inverter, so the inverter tends to read lower. Raising this value will reduce your total capacity but increase the life of your battery.
June 18, 20214 yr 4 minutes ago, JaseZA said: I would happily leave it at 20A - that's only 0.1C for a 200ah battery. I intend to charge mine at 50A or 0.25C. Most lithium recommendations I've seen are around 0.5C max for continuous and 1C max intermittent. I generally like having the inverter DC low cut off higher than the battery BMS. This way the inverter cuts on it's own terms rather than there being a chance the BMS might cut first. This is unlikely though as there is almost always some voltage drop across the cables from battery to inverter, so the inverter tends to read lower. Raising this value will reduce your total capacity but increase the life of your battery. That is great advice, thanks. I'll set it to 20A, and cutoff at 10.9
July 3, 20214 yr On 2020/03/16 at 7:42 PM, Igubu said: Also saw the " cannot parallel" part - which seems very strange to me? Assuming each one has it's own BMS, not sure why parallel of 8x (4x series) of these are not possible? There is no intelligence between them, like many lithium " gel replacement" options, no can bus, but don't see how anything will prevent it to be connected that way? Will be glad if some clever person explains As long as your cable lengths from the positive/negative side of each string matches in length, and of course is thick enough to allow 400Ah (19.2kW! - not many small inverters allow that kind of current anyway) - you should be fine? They look very similar to another battery sold on some sites a month ago - and I know of a few field installs where they seem to work very well. The other ones were listed as 180 as well - the new ones also 200 - Attached a picture for those interested You can see the 4S3P packing. (4x Serial 3 Parallel) Hi and thanks for this! I’m using a Mercer 720w inverter(model: LOBO 1200/VR-1200LBKS)with this battery(no solar/wind hookup just battery backup) but it seems to always want to be on floating charge. I heard that LiFePO4 batteries don’t like that because they could be over charged. The battery therefore never gets to 14.6V(fully charged). Is there a way to change the settings on the inverter to make it know that the battery connected is a LifePO4 instead of an AGM? Thank you.
July 7, 20214 yr On 2021/07/03 at 7:10 AM, Proninety said: Hi and thanks for this! I’m using a Mercer 720w inverter(model: LOBO 1200/VR-1200LBKS)with this battery(no solar/wind hookup just battery backup) but it seems to always want to be on floating charge. I heard that LiFePO4 batteries don’t like that because they could be over charged. The battery therefore never gets to 14.6V(fully charged). Is there a way to change the settings on the inverter to make it know that the battery connected is a LifePO4 instead of an AGM? Thank you. I had the 24v version of that one. It had no user configurable options unfortunately. Edited July 7, 20214 yr by johanp
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.