Dacian,
Thanks for your quick response on this. I am really enjoying working with your SBMS0.
A little more background:
I am building this system for a travel trailer that I sleep in when camping. This affects my perception of what's safe enough versus batteries in a detached shack/garage that is not a living space.
Also, I don't have solar. I always camp in the shadiest place I can find. Folks who put solar on their RVs start camping in sunny spots. Then they get hot and want to run the AC. Then they need more solar, and more batteries to run their AC. I'll stay in the shade. I charge the battery when I have access to mains power, off the alternator when I'm driving, or with an AC generator on long trips.
If shunts where on the negative side, I would connect the fuse to the positive terminal. This is compliant with the ABYC in that any positive conductor connected to a battery bank requires overcurrent protection within 7 inches of the batteries positive terminal. It's pretty hard to follow this requirement with a shunt between the battery positive and the fuse.
There are no relevant IEEE or IEC standards for RVs, so I look to ABYC (marine) standards as "best known practice".
Not sure what you want to use the contractors for ?
1) Battery disconnect means - One of the EV200 contactors isolates the battery from everything. It is operated by the SBMS Type 5 output. This circuit includes a "seal-in relay". The SBMS can disconnect the battery on a Type 5 alarm, but the seal-in relay makes is so I have to manually reconnect the battery (by pressing a pre-charge button followed by a seal-in relay close). I do NOT want any piece of automation reconnecting the battery after it has disconnected. (Protection engineer attitudes here!). I will also have an E-Stop switch in this circuit. This is a big red button that when pressed opens the contactors and shuts down the Victron. You press it when awoken in the middle of the night by the smoke alarm, only to find smoke pouring out of the battery compartment. I don't expect to ever use this feature.
Here is a reference on seal-in relays as they apply to power system protection.
The battery disconnect is also handy when I'm on an extended stay at some place with shore power. I can run all my DC loads off the Victron charger, without have the battery constantly on float charge. I'd rather disconnect the battery in this situation.
2) Load disconnect means - Most of my loads are 12VDC. Turning off the Victron inverter doesn't do much because I only run the microwave, coffee maker and wife's hairdryer off AC. Everything else is DC.
I looked at shunt-trip circuit breakers. But I could not find any that I liked. Firstly, I try to only buy safety products from North American or European vendors (Eaton, ABB, Siemens, Schneider, Victron, Tyco Electronics, etc.) I do not buy safety products, such as fuses or circuit breakers, from companies located elsewhere.
Then I found these used TE contactors for EVs for $12.50 each, normally $160+ from Mouser. They are purpose built for switching large battery banks. I bought three (two plus a spare). They even have a built-in economizer circuit and flyback diode.
The Class-T fuse will be for overcurrent protection. The EV200s for load disconnect. That having been said, these EV200s can interrupt 5000A (twice) on a 24 VDC system. That's pretty impressive for a switch (not a breaker/fuse).
Lastly on the subject of electrical protection, I would like to point out this statement by Victron (bold emphasis mine):
"Worded simply: a battery must be intrinsically safe, and include its own large disconnect mechanism such as a contactor. Only relying on digital signals telling our inverters and chargers to stop charging is not sufficient."
I am in agreement with Victron on this point. Are you disagreeing with Victron on this matter?
3000W is a bit much for just 4x EVE cells
I agree. The MPII 3000 is rated for 3000 VA and 2500 W continuous. But your point is still valid. The 2000 would be sufficient for my needs today. The largest load I have presently is an 1800W hairdryer that my wife runs for 10-15 minutes every couple of days.
I was watching a Will Prouse video where he lamented all the money he wasted over the years buying successively larger inverters. He would always buy the smallest inverter that met his immediate needs, only to buy a larger model next time. Given the small price differential between the two, I decided to buy the 3000. Buy once, cry once.
Bob
Bob,
If shunts where on the negative side you could still not have a fuse between battery and shunts.
In case fuse melts due to high current trough the shunts the energy stored in all the cables (inductors) will either increase or decrease the voltage at the shunts depending on the current direction before the fault and since the SBMS0 is only powered by the cell sense wires it can not protect the internal electronics from this energy pulse and that protection will need to be outside next to the shunt's adding complexity.
Also the shunts either on negative or positive will need to be close to the battery and so the short 10 to 20cm of wire between the battery positive and shunt will need no special protection as it is not much different from the connections between cells.
You can design the shunt so that it can be enclosed in a box that means you select the shunt so that max power dissipation is below 3 to 5W depending on box.
Not sure what you want to use the contractors for ?
The Victron inverter charger has separate ON/OFF controls for charge and discharge and that can be handled by SBMS0 and in the unlikely case something fails unsafe a simple circuit breaker with a trip coil can be controlled with EXT IO6 set as type 5 and also acts as over current protection at the same time and if it is a dual circuit breaker you have the separate pats for loads and charge sources.
I will soon have a new DEX designed that will multiply the available EXT IOx and will have a n output dedicated to EXT IO6 set as type 5 to trip a circuit breaker in case of fault as secondary protection.
3000W is a bit much for just 4x EVE cells. I have a 3500W inverter than normally will not see more than 2500W but is on 24V 8s2p so a total of 16 EVE cells (the 280Ah version).
3000W from 12V is 250A so normally a 500A shunt will be fine if you are not fully enclosing that in a sealed box as with 50mV / 500A = 0.1mOhm you will have 6.25W a bit much in a sealed box.
But if say your setup was limited at 2000W max then is 166A * 166A * 0.0001Ohms = 2.76W and that is not a problem to dissipate in a sealed box.
I am a recently retired Power System Protection engineer. This means, for good or bad, I'm highly opinionated on anything to do with overcurrent protection. I ultimately chose the SBMS0 for my 4S, 12V RV system. But I was initially appalled when I saw the shunt on the positive battery terminal with the requirement to NOT put a fuse between the battery and the shunt. As an opinionated power system protection engineer, I want to see solid dielectric insulation (not air insulation) between the battery and the fuse. Unfortunately, a shunt is power resistor that needs ventilation for cooling. This effectively precludes sealing up the shunt so that it has no exposed metal.
After seeing the cost of the REC BMS (and other competitor BMSs that can control a contactor), I got over it and figured out another way to address these concerns. The system I'm building is comprised of the SMBS0, Victron MPII 3000, two EV200 contactors, four EVE LF304, and a Class T fuse. I'm bench testing this now. More on this later in another post.