SBMS0 on Sailboat

[Sea Rover]

Hi everyone

Ive got a 10 year old 12 V 400Ah pack wired in 4s4p configuration.  

All charge sources have been set up to work with LiFePO4.  In 10 years I've only had 1 HV shutdown (when the voltage sense wire on the alternator reg fell off and shorted to the engine block).   Still, I need to protect against unforeseen events like this so I need a HV/LV shutdown relay to handle my main contactor which has all loads and charging sources connected after it.

I've used various BMS's over the years so I have a pretty good idea of what Im looking to have the SBMS0 do.  On paper I think the SBMS0 will work.  Just to be sure though, can someone answer if it can accomplish the following:

Q1:  Can the SBMS0 be setup with a battery protect LV/HV shutdown event for cell voltages?  From the manual I see there are HV lock and LV lock settings.  I assume these are the settings I'd work with to kick off my contactor.

Q2:  From the manual I also see more conservative HV and LV settings with a time delay.  I assume these could be used to trigger shutdown of my Inverter/Charger, MPPT's and Alternator regulator in addition to a buzzer...  Thus giving me time to sort out why something is out of sorts before kicking off my contractor and putting me in the dark.  None of my devices work are setup to run on canbus but they all have simple on/off contacts like the Victron Smart MPPT's

Thanks

Gary

  

[Dacian Todea (electrodacus)]

Gary,

Q1: The SBMS0 has two HV and LV levels. The normal operation HV is set by default at 3.55V wile normal LV at 2.8V this will stop all chargers probably every day and in case of LV all loads if it ever happen to have an empty battery.

Then there is a HV lock and LV lock limits set by default at 3.75V and 2.5V respectively and this will trigger an EXT IOx that is et as type 5 thus you can use this signal to trigger the contractor but it can not do that directly as it is just a 50mA capable optoisolated signal.

I currently work on a so called DEXT that will take this type 5 signal and be able to trip any circuit breaker equipped with a trip coil. Should have this done in maybe 2 or three moths so this summer.

Q2: Yes as mentioned already at Q1 the normal HV and LV are to control charge sources and then separately load sources so there are two completely different signals same optoisolated max 50mA capable type.

Those will be normal operation since as soon as say MPPT fully charges the battery (any cell above 3.55V for more than say 3 seconds) the charging should stop completely and so type 5 that triggers the trip coil of the breaker should never be enabled unless something failed to respond to the normal remote ON/OFF so some equipment failure.

The EXT IOx can directly control the ON/OFF signal on the Victron equipment.

You can see last photo in post I made recently how the DEXT looks like and in that photo it controls ON/OFF a DSSR50 and a Victron MPPT 100/50  https://groups.google.com/g/electrodacus/c/cmLykQadmUk/m/YINEUIxCBQAJ using the HV signal from the SBMS0 (the brown and white twisted wire).

[Sea Rover]

I just found the "beginners guide to Electrodacus" v0.61.  This I think has gone a long way to clearing up the different extio types for me and I have a better idea of how to wire things up.  Post a link if there is a newer version I should be reading.

I'll be using a BlueSea 7700 contactor for my type5 shut off.  Its a bi stable latching relay that can be triggered with less than 10ma but requires a pulse.  Will the new DEXT support pulse signals? 

I was thinking of taking a stab at building Ross's or Tim's momentary switch circuit referenced in the beginners guide but Im far from competent in electrical engineering...  I know REC bms packages up a latching relay driver which I'll add to my shopping list if required.  

Due to the various locations of loads and charging sources scattered all over my boat (and the fact that I have an inverter/charger that is one unit) I wont be using the Load disconnect function but here is how I was thinking of hooking up the EXTIO's

EXTIO3: Type2 | Buzzer relay for low voltage notification

EXTIO4: Type 1 | paralleled shutdown relay for MPPT's, Inverter/Charger and Alternator Regulator

EXTIO5: Type 1 | Buzzer relay for high voltage notification

EXTIO6: Type 5 | Bi stable latching relay driver for Fault condition shutoff

Does this make sense?  Has anyone on this forum managed to get a BS7700 contactor working with the SBMS0?

Thanks

Gary

[Dacian Todea (electrodacus)]

Gary,

Not sure you will want a buzzer for type 1 as the battery can get to fully charge many times per day and will likely get annoying.

If the pulse can turn the contractor ON and OFF then that is just not a safe ON/OFF as you can not be sure if the pulse will turn it ON or OFF.

If I remember correctly Tim's momentary switch circuit uses the power ON LED signal in order to know if the inverter is ON in order to only turn it OFF.

The new DEXT will have the option to turn OFF a circuit breaker using trip coil with the type 5 signal. Meaning there will be a circuit capable of handling the 6A or so pulse current required for the trip coil.

[sailingharry]

I am also considering the Blue Seas relay on my boat.  I'm also considering the 500A Tyco relay, if I can convince myself that the current draw (times TWO!) is worth cost savings (times TWO) over the Blue Seas.  But, for the Blue Seas, the 7713 is the 7700 with the electronics integrated to make "look like" a normal relay -- you provide 12V to the control wire, and it stays "ON" until you remove the 12V from the control wire.  I believe (but am not certain) that just like the 7700, the actuation power comes from the battery side terminal, and so the control from the SBMSO is control level power.  The big benefit of the Blue Seas is the manual override for both close and open -- you can defeat 100% of the 'Dacus black magic if the magic smoke leaks out.  The downside is $$$ and space (they are big switches).

Note, as I alluded above, you could be in for double the size/space/power/cost issues.  You can very easily implement the 'Dacus with a single shunt/relay/fuse, and just pass on the information that the 'Dacus can give regarding power in vs power out.  But if you want the full benefit of Dacian's efforts, you need two shunts, two "switches," and two fuses.

Dacian is a big fan of using a shunt-trip breaker to address both the switch function, as well as the overload function.  It is a simple, low power, lost cost solution -- but I have not been able to find a shunt breaker that works at 12V DC.  If you running a 24V (or higher) system, it looks like a great solution.

[Sea Rover]

Thanks for the reply.  I can see your logic and understand that the SBMS0 is usually used to control charging sources and loads in addition to keep the battery safe.  In my case I only need it to keep the battery safe so Im hoping that the Type 1 setup relay never triggers.  I'd have it set to trigger above all my charging source voltage settings. 

As for the fault trigger, I need this to only shut down the contactor.  Once it is off I want to have to manually turn it back on.  I watched a brand new 1.5 million $ catamaran burn before it hit the water because the Victron charge controller glitched and reset to 48V on a 24V system.  The BMS while protecting the battery and shutting down would see the voltage drop to safe levels and start up again.  24 hours of repeating this and the pack burst in to flames.

If anyone has a circuit that they are currently using with the Blue Sea 7700 contactor or has experience with the REC BMS latching relay driver please post here or contact me directly.

Thanks

Gary

[Sea Rover]

Hey Sailingharry.

Like you, I've been playing with LiFePO4 on my boat for many years.  I just looked back on cruisers forum and was struggling with the BlueSea 7700 in 2015.  Back then I had a Chargery BMS and before that a House Power BMS.  I managed to melt my Chargery by flooding my battery compartment with water so on this go around Im looking to finally do it right and remove the manual requirement for me to shut off the battery on low or high fault conditions.  In actuality this is more of a want than a need.  I've run my pack for the past 12 years with no fault protection.  In general if Im leaving the boat for any period of time I switch over to my start bank and shut the house bank down.

Im at the stage now where Im considering moving the BS 7700 up to the thruster/windlass circuit and just spending the money on a BS 7713.  Im getting to that age where playing with small electrical components and a soldering iron is no longer fun and I just want to enjoy cruising.  I think the Rec ML relay driver would also work but finding any one who's used it seems to be challenging.

With the 7713 I think all we need is an opto relay to reverse the signal.  When the contactor switches off there is still a small draw but Im ok with that as Im more concerned about the high voltage fault and the possible fire than the low voltage fault requiring me to replace my 12 year old (and very abused) pack.  

[sailingharry]

Do you have more details on this accident?  The general consensus is that it's virtually impossible to get LFP to catch on fire -- and charging a 24V bank at 48V is, from what I understand, not one of the ways to make it happen.  In fact, I don't think there is a generally accepted way that you CAN make it happen.  I'd love to hear more specific details -- was the pack LFP,, was it some auxiliary system, etc. 

However, I do agree with your thought that a Type 5 cutoff (HVC/LVC) should require manual intervention -- it should be so incredibly rare that I would want to know (A) that it happened and (B) that the problem was solved.  It makes the shunt-trip breaker really appealing, as it is the most likely to not self-reset.

[Sea Rover]

The incident was on a high end cat just about to launch, built in Chile.  I was there and saw the aftermath first hand.  Lithionics had just bought out HousePower/CleanPower BMS and their very first production run had a firmware issue with the fault shutdown hysteresis and the on/off cycling was just too quick for the battery to cool down.  I was keen to find out what exactly happened because I also had a HousePower BMS at the time.  Lithionics was blaming Victron and visa versa... You'll notice all Victron products now default to 12V...  Lithionics has long since taken their products well past those first batteries.  It didn't burn the boat down but the cabin where the battery storage was stained black and it made a hell of a mess.  Lots of melted plastic parts around the batteries themselves.  Delayed the launch by almost 3 months.

[Dacian Todea (electrodacus)]

If the BS 7713 requires a continues 12V signal to be ON and max 7A for 40ms when changing state then the DEXT will be able to handle that just requires a small modification to reverse the control logic. I will see how I implement this in the next prototype probably will require removing two resistors or so and installing another two in a different place.

The DEXT will be able to control two BS7713 but both will be controlled by type 5 so they will stay ON as long as there is no error and will turn OFF if any cell exceed secondary levels default 3.75V or 2.5V

The problem with this may be that of the cells get back down to normal the BS7713 will be re-enabled while with a circuit breaker that is tripped that will require manual intervention.

By default DEXT will be provided set to work with a trip coli circuit breaker but at request I can make the small modification to work with relays/ contractors or the BS7713 witch seems to work the same way just with lower power at steady state.

[Captain Borchik]

hi I have SBMS0. and dmv 712, soc on sbms0 is 89%. and bmv712 100% who is right????

[Dacian Todea (electrodacus)]

Boris,

I can not be sure any of them is right or wrong based on the limited info.

SBMS0 if installed correctly will be the most accurate as it has the individual cell info while BVM712 can not do that.

SBMS0 will calculate the SOC based on current in and out of the battery and correct to 100% SOC each time battery is fully charged so when any cell is fully charged meaning exceeding 3.55V

All I can say based on your info is that battery is not 100% SOC but I do not know if 89% is correct as that will depend on you installing and setting the current shunt / shunts correctly.

The BMV712 can only see the total battery voltage so it will not be able to properly correct for battery fully charged as for accuracy if set correctly the SBMS0 will be more accurate due to higher resolution measurement.

Do both SBMS0 and BVM712 measure the same battery current ?

If you charge the battery to 100% (so both SBMS0 and BVM712 show 100%) then only discharge for a few hours (disable charging) and see what SOC they both show. If they show about the same SOC withing plus minus 1% then current shunts settings are correct and the BVM712 just has no idea when battery is fully charged and guesses wrong as BMV712 can only see the total battery voltage thus it incorrectly corrects to 100% SOC.

[Captain Borchik]

Thank You!

[Captain Borchik]

Dacian  , I solved the problem with SOC reading on BMV72, I simply adjusted charging voltage to 14.2 ON BMV72, and now I have same reading SOC with electroducus , 

Best regards Boris

[Garth Wilcox]

Another sailboat system.

8 cells in series, 24V 163AH

I am seeing a wide disparity between SOC displayed and that indicated by the resting voltage.  The boat is currently in a shed so no solar power and since I have lots of AC power the batteries are just sitting there.  I believe my only load is the BMS itself.  The wiring diagram in the manual shows wires 10, 11, and 12 all connected to the positive post (last cell in the string.  I believe wire 10 is voltage sensing and 11&12 are providing power to the BMS. Is this true?

If I were using the boat instead of working on it this would not be much of an issue but it I noticed dangerously low voltage the other day while the BMS said ~50%.

Is there any reason I cant move 11&12 to the shunt so the BMS will track the current draw?  

Can I add a switch to force the BMS into a dormant state?

Is there a better way to deal with this?

[Dacian Todea (electrodacus)]

Yes that is correct wire 10 is sensing and 11+12 power the SBMS0.

The SBMS0 self consumption is calculated but since that is so small it will take over 200 days for the battery to be fully discharge if it starts from full with no charge source.

Battery will also have a significant self discharge in this long period and SBMS0 will have no way to account for that.

If you are not using the battery best will be to just remove the 12pin connector so no Load on the battery and have the battery fully charged before you do this then you will still need to check the battery every 12 months and charge if needed as there is still some self discharge depending on battery quality ambient temperature and battery age.

Removing/disconnecting the pin 11+12 will reduce the SBMS0 self consumption but not to zero as the ISL94203 will still be powered.

SBMS0 with WiFi ON about 0.8W

SBMS0 with WiFi OFF about 0.35W

SBMS0 with pin 11+12 disconnected below 0.1W

So in worst case starting with a full battery 4kWh and SBMS0 with WiFi ON the battery will be fully discharged in  4000Wh / 0.8W = 5000h / 24h = 208 days but this ignores the battery self discharge so it could be quite a bit less than 200 days before battery is over discharged. 

[Garth Wilcox]

Thanks Dacian, that's very helpful.

In my case with a relatively small battery, the SBMS draw is non-trivial.

3% per month self discharge would be about 0.18W compared to the SBMS0 ranging between 0.8-0.1.

When I had this issue I was running wifi and probably started at 70% SOC.

I will think about switching wires 11/12, it may or may not be worth the effort. In any case I need to be focused on voltage not SOC during storage.

[Dacian Todea (electrodacus)]

The SBMS0 already calculates the self consumption current.  It knows the SBMS0 consumes 0.8W when WiFi is ON and 0.35W when WiFi is OFF it knows the voltage and it can calculate what the self consumption current is and it take that in to account when calculating the SOC.

But since this are very small values compared to max current and resolution used to measure the current the accuracy of SBMS calculation will not be good if SBMS0 is the only load.

Long therm storage multiple weeks or months with no charge source present requires that SBMS0 is disconnected.

If you have a small charge source even a 10W PV panel installed in the sun can compensate for the SBMS0 self consumption and even for battery self discharge then it can stay in storage indefinitely.

But if there is no charge source then SBMS0 should be disconnected. 

[Garth Wilcox]

I had no idea SBMS0 power consumption was already accounted for.  I cant do better than that.

Your entire last post would be nice to include in the manual.

Thanks again.