The Cerbo can do everything the BMV-712 does.
The Orion-TR is pretty much limited to relay input control.
don't understand Lifepo charging, seems like voltage/time-based charging is flawed?
- Thread starter blutow
- Start date
The Orion-TR is pretty much limited to relay input control.
I got some of my victron components today, including the cerbo GX and had some time to dig into some features. I found a few system glitches, but this stuff is really plug and play and I'm pretty impressed.
It turns out the Cerbo does have something called "DVCC" - distributed voltage and current control that coordinates current and voltage across mulitipel charge sources. As I understand it, you can use the mulitplus inverter as the "master" and it will coordinate charging between the multiplus and one or more solar chargers. The muliplus will sync up charge voltage and you can also set a max charge current across all chargers. The system will take inverter load into account and increase max current limits accordingly. However, it does not currently take DC loads into effect and it can't control the Orion TR chargers (yet). Even if you don't use the inverter to control the solar chargers, you can still set a max total system charge current with the solar chargers taking priority and you can also configure the system so that all chargers use the same source for temp and voltage reference.
I still need to do more research to fully understand whether this feature even makes sense for lifepo batteries, but it sounds like a good idea. This cerbo thing can do a bunch of stuff I don't fully understand, but trying to learn.
It turns out the Cerbo does have something called "DVCC" - distributed voltage and current control that coordinates current and voltage across mulitipel charge sources. As I understand it, you can use the mulitplus inverter as the "master" and it will coordinate charging between the multiplus and one or more solar chargers. The muliplus will sync up charge voltage and you can also set a max charge current across all chargers. The system will take inverter load into account and increase max current limits accordingly. However, it does not currently take DC loads into effect and it can't control the Orion TR chargers (yet). Even if you don't use the inverter to control the solar chargers, you can still set a max total system charge current with the solar chargers taking priority and you can also configure the system so that all chargers use the same source for temp and voltage reference.
I still need to do more research to fully understand whether this feature even makes sense for lifepo batteries, but it sounds like a good idea. This cerbo thing can do a bunch of stuff I don't fully understand, but trying to learn.
One correction to a prior post - the cerbo has 4 temp inputs and 2 relays, but it appears that you can't currently trigger a relay based on a temp excursion. I just assumed that was possible. I did some reasearch on the victron community forum and there seems to be an "advanced" work-around that requires repurposing a tank level input as a temp sensor to trigger the relay, but it can't be set up through the standard functionality using the temp inputs. It sounds like it's a feature that they will get to, just kind of silly to have temp sensors and relays and not be able to control the relays via temp.
Does anyone know what the SOC would be for measured Voltages at e.g. 0.0C, 0.01C, 0.1C, 0.2C etc. Discharge Rates and Rest Times? Is there a Table of this data specifically for Battle Born 100AH? To be even more precise, some statements about tolerances/ranges regarding Age, Charge Time, Rest Time, Temperature and Manufacturing variations would be included for those of us who find these details interesting and wanting to understand a battery's physical SOC in addition to a charge controller's definition of SOC. It may not be practical, but it can be interesting.
You can't do that with voltage. The only way to know state of charge with LiFePO4 is to use a Coulomb counter.
I was just re-reading through my original post and some of it was misguided by my lack of understanding how these charge cycles work. Since then, I've built four 272Ahr DIY batteries and have been running the system since early summer. I'm still far from an expert, but have much more real-world expertise on the system and Lifepo charging dynamics. The system has worked very well. It took quite a bit of tweaking to get everything configured and running well. In particular, the default settings on the smart shunt are really dumb for resetting SOC to 100%. I originally thought the system was resetting randomly before I understood the logic. It was incorrectly resetting when it saw extended voltage drops from solar charging with swing up and down on partially cloudy days (it was assuming charging was complete when it saw the extended drops). I can't remember what settings I had to tweak, but the default settings seemed totally illogical to me and it's been working great once I understood it and changed them.
I still think Victron could/should give the option to incorporate SOC in it's charging logic to make it even better. The system works perfect when it's in actual use. When it's just sitting in my driveway (fridge and toilet fan running), It basically tops off the battery every day with solar even if the battery is almost full (98-99%). I guess it's not doing any significant harm topping the batteries off every day, but it would be nice if there was a "storage" mode that said wait until it hits x (Maybe 90 or 95%?) before topping off. That would mean topping off the battery much less frequently instead of daily. Everything I've read is that these batteries will last longer if they are are held a bit below 100%, so it bugs me that it's basically pushing them to 100% every day when it's not really needed. You can kind of adjust the recharge point by changing float voltage to raise/lower the point that bulk charging kicks in, but that's it's not going to work as well as SOC and it's a pain because you have to do it on each device and the multiplus can't do it from the app. It's also not a great approach because a momentary voltage drop can kick off bulk charging when it's not really needed yet based on SOC. I have my bulk/absorption set to 14.1v, so it's pretty conservative and not really pushing the batteries hard. In my testing, the difference between pushing the cells to 14.1 vs. 14.4 is way less than than 1% of capacity, so it's kind of pointless if my mind. Again, all of this is probably overthinking things and the system is working fine. I'm happy whether my battery cells last 5000 cycles or 2500 cycles, so hopefully it's all good.
I still think Victron could/should give the option to incorporate SOC in it's charging logic to make it even better. The system works perfect when it's in actual use. When it's just sitting in my driveway (fridge and toilet fan running), It basically tops off the battery every day with solar even if the battery is almost full (98-99%). I guess it's not doing any significant harm topping the batteries off every day, but it would be nice if there was a "storage" mode that said wait until it hits x (Maybe 90 or 95%?) before topping off. That would mean topping off the battery much less frequently instead of daily. Everything I've read is that these batteries will last longer if they are are held a bit below 100%, so it bugs me that it's basically pushing them to 100% every day when it's not really needed. You can kind of adjust the recharge point by changing float voltage to raise/lower the point that bulk charging kicks in, but that's it's not going to work as well as SOC and it's a pain because you have to do it on each device and the multiplus can't do it from the app. It's also not a great approach because a momentary voltage drop can kick off bulk charging when it's not really needed yet based on SOC. I have my bulk/absorption set to 14.1v, so it's pretty conservative and not really pushing the batteries hard. In my testing, the difference between pushing the cells to 14.1 vs. 14.4 is way less than than 1% of capacity, so it's kind of pointless if my mind. Again, all of this is probably overthinking things and the system is working fine. I'm happy whether my battery cells last 5000 cycles or 2500 cycles, so hopefully it's all good.
I agree; I believe it takes at least voltage, current, time measurements and a lookup Reference Table/Algorithm of sorts. I think a coulomb counter 'automates' these things with its own built-in assumptions. Some systems may even incorporate some sort of battery Health factor and provide the information to an active control system. A simple 'poor man's' implementation using manual V and I measurements that are known or assumed to be constant over a period of time may not be not be as precise, convenient or automated as a sophisticated system. But, I believe, perhaps mistakenly, that manually measuring V, I and t and then looking up values in a table could be useful as an occasional, low-cost 'sanity check' or monitor. I found a Battle Born Table for Remaining Capacity vs. Voltage w/ 0.0C/ No Load, at 'Rest'. That is useful when the battery is disconnected/in storage, but in my application the battery is usually connected and has some Load.
Voltage is really only informative when the battery is very low or very high. The rest of the capacity/voltage curve is so flat that voltage isn't very helpful, particularly if you are pulling a load or charging when looking at it.
That's OK if there is no drain on the system and it's truly idle, but not ideal if the batteries are cycling periodically. If you never "top off" the cells, they don't spend much time in the range where cell balancing happens. Again, it's probably not going to create any big issues repeatedly charging to 13.8, but I'd rather have a setting where the chargers are smart enough to wait until the get down to "X" soc before charging, but still charge to a point where you get some balancing time. I think my BMS's are configured to start balancing once cells hit 3.4v. You'd still get some balancing in at a 13.8V charging target with the 3.4v setting, but I find that most of the balance swings don't happen until cells approach 3.5v.
Exactly, V alone is not enough. That is where other measurements and a Table come into play. I guess haven't been understanding the significance of V and I measurement precision vs. other factors that influence those values. For example, other factors like time/duration of measurements rather than precise, instantaneous E and I measurements could throw meaningful Table Lookup of SOC out the window. So, I guess Tables or Graphs of SOC vs. E, I and t just would not be possible or useful.
Have a read here:
State of charge estimation of LiFePO4 batteries based on online parameter identification
With the research object of LiFePO4 battery, this paper aims to correctly estimate the battery state of charge (SOC) by constructing a comprehensive S…
www.sciencedirect.com
It's pretty much still a research topic - it won't work with a few lookup tables...
Yes so I have mine set a bit low and have a periodic equalize set to 14.2 to have some time in the balancing zone.
