diy solar

diy solar

Sailboat going all Victron

Wow! So if I went with REC it looks like I would have to have 2 of the master BMS units ($330 ea), 2 contactors ($169 ea from REC, don't know about the one recommended by @ohthetrees), the PC software to configure the BMS's ($104), a couple of REC current shunts ($66 ea), and probably a couple of the Victron CANbus cables ($28 ea). I'm guessing that is over $1000 for the BMS. Am I missing something?
Someone is definitely missing something, not sure if it is you or me. You keeps saying you only have one house bank, but you keep talking about two BMSs, two Shunts, two of everything. You don't need all that stuff if you only have one house bank. I have one REC BMS, 1 contactor, one CANbus cable, and I already had the shunt. Sounds like you have a bunch of shunts piling up already, you can use one of them. You can even use the same shunt at the same time for the BMV and for the REC. I think it is all very fair pricing for its capability, and that it can talk to the Victron system, and with 4 relays, you can really control your electrical system very well. I do agree with you that it sucks to need to pay $ to have a way to configure the device. I went with the WIFI unit, but you do need either that or the PC software.
 
Someone is definitely missing something, not sure if it is you or me. You keeps saying you only have one house bank, but you keep talking about two BMSs, two Shunts, two of everything. You don't need all that stuff if you only have one house bank. I have one REC BMS, 1 contactor, one CANbus cable, and I already had the shunt. Sounds like you have a bunch of shunts piling up already, you can use one of them. You can even use the same shunt at the same time for the BMV and for the REC. I think it is all very fair pricing for its capability, and that it can talk to the Victron system, and with 4 relays, you can really control your electrical system very well. I do agree with you that it sucks to need to pay $ to have a way to configure the device. I went with the WIFI unit, but you do need either that or the PC software.
Ok, so let's start over.

I have 8 280Ah cells, and I want a nominal 12V bank. I could either build it as 2P4S (groups of two parallel cells wired together in series) or 4S2P (two groups of four serial cells wired together in parallel). I and many other people here believe pretty strongly that 4S2P is much better than 2P4S. Putting cells in parallel and then having the BMS see those cells as one is asking for trouble. The commodity cells most of us on this forum can buy will never be exactly matched together. One cell will almost certainly different than its partner cell in capacity, discharge / charge curve, and internal resistance. One cell will be adversely affecting the performance of the other cell for their entire life. The BMS isn't really managing each cell BMS's are intended to do, but rather manage 4 dysfunctional shotgun marriages. One cell could start to have problems, and you - and your BMS - won't know until the failing cell has wrecked its partner cell.

So my intent is to build a 4S2P bank. I don't want to call it a battery, because for all intents and purposes it is two parallel 4S batteries. Each has its own BMS. This has advantages besides countering the problems outlined in the previous paragraph. If one 4S pack hits LVD or HVD and its BMS takes it offline, you should still have an operational house bank because the other 4S pack is still going. If a cell fails and you lose that entire pack until you can replace it, you can limp along with half the capacity until you can order and receive a replacement cell.

So I have a "bank" that is just like people have been building battery banks for boats, RVs, and off-grid homes for a long time: parallel batteries to provide increased capacity. It is made up of two separate batteries, each 4S. It is a much better implementation than 2P4S.

I'll admit that if I was facing the purchase costs of the REC BMS, I would be very tempted to throw together a 2P4S single battery. In my mind, I would be left with a really good BMS watching over a poorly designed battery.

Sounds like you have a bunch of shunts piling up already, you can use one of them.
You mean two? I guess that's a bunch. :unsure:

I'm still learning about the Cerbo GX. Having the BMS help manage the charge and discharge currents does seem like a pretty good feature to have. I'm just not sure I could convince my friend it is worth it. I'll keep studying.
 
That’s the missing piece for me. I didn’t catch that you are going 4S2P, rather than 2P4S. If you don’t want to parallel cells, you do need multiple BMSs, and I guess I wouldn’t recommend a centralized REC style BMS to you. I fall into the opposite camp. I think the pros of a 4P4S outweigh the cons, but this isn’t the thread to debate that, there are plenty of other threads for that. All I will say is that it is not settled that 4S4P is the way to go, and you will find people hold pretty strong opinions on both sides of the debate. I’m not aware of how to achieve good victron integration with the topology you are proposing, so I’ll bow out of the thread. Good luck to you, hope your build goes well.
 
That’s the missing piece for me. I didn’t catch that you are going 4S2P, rather than 2P4S. If you don’t want to parallel cells, you do need multiple BMSs, and I guess I wouldn’t recommend a centralized REC style BMS to you. I fall into the opposite camp. I think the pros of a 4P4S outweigh the cons, but this isn’t the thread to debate that, there are plenty of other threads for that. All I will say is that it is not settled that 4S4P is the way to go, and you will find people hold pretty strong opinions on both sides of the debate. I’m not aware of how to achieve good victron integration with the topology you are proposing, so I’ll bow out of the thread. Good luck to you, hope your build goes well.
Fair enough. Thank you though, as you've helped me understand what REC does. I still need to understand the "managing" the charge and discharge, as I'm not used to that being a thing. In my mind I would have thought that the Cerbo GX would do that, based on inputs like the sense voltage of the bank, and it would send control information to the MPPTs and Inverter. My offgrid place does that via Schneider equipment, but without some of the cool stuff Victron has done.

I do know that the Cerbo GX can happily take in information from multiple shunts or battery monitors, but I guess I don't know what it does with that data. I get the feeling that the Cerbo does not - or people believe that it does not - combine the information in a useful way.
 
Back
Top