Cold Temperature Shut-Off

544 views
Skip to first unread message

Grant Tanner

unread,
Mar 23, 2019, 12:10:00 AM3/23/19
to electrodacus
Hi everyone.  Does the Electrodacus BMS have a cold weather shut-off so the solar isn't trying to charge the batteries if they get too cold?

If not, what kind of solutions are available?

EnergyStow

unread,
Mar 23, 2019, 12:12:20 AM3/23/19
to electrodacus
I am sure that it does have both over temp and low temp shut off

EnergyStow

unread,
Mar 23, 2019, 12:14:08 AM3/23/19
to electrodacus
You do have to buy the temperature probe separate

Dacian Todea

unread,
Mar 23, 2019, 12:24:13 AM3/23/19
to electrodacus
Yes there are limits for high and low battery temperature but ideally you will have the battery in a climate control environment (like where you live :) as batteries like that same temperature range as we do).
If for some reason that is not possible then using a thermal isolated battery box and a thermostat control heating element to keep the battery at optimal temperature is a good idea.
If box is well insulated then not much energy (not significant) is needed to maintain battery temperature.
 

Grant Tanner

unread,
Mar 23, 2019, 11:18:17 AM3/23/19
to electrodacus
So could this for example be wired into ElectroDacus BMS?

Victron Temperature Sensor for BMV-702



nebulight

unread,
Mar 23, 2019, 1:30:07 PM3/23/19
to electrodacus
Dacian states that the SBMS is calibrated for a particular 10kΩ thermistor made by Murata NCP21XV103J03RA. You can wire that in for your temp probe.

However I have some questions regarding what can be programmed in advanced settings as I don't have my unit yet so I can't go through the menus and test. It would be nice if you can trigger the EXTiox ports for a low condition. Example would be if temp is below 5 degrees c, then open EXTio4 to disconnect chargers. If it's above 55 degrees c (or below -20 degrees c), open EXTio4 to disconnect chargers and open EXTio3 to disconnect inverter/loads.

Is that how it works? Or could it be implemented to work that way? There are a lot of off grid campers that can't really install them in temp controlled spaces.

Jim A

unread,
Mar 23, 2019, 1:56:06 PM3/23/19
to electrodacus

Dacian Todea

unread,
Mar 23, 2019, 2:13:42 PM3/23/19
to electrodacus
Grant,   They do not say what sort of sensor is that but SBMS needs an 10K NTC 3950 sensor and if you search for that you will find plenty of option for as low as $1 like this  https://www.ebay.ca/itm/223101417089 so just search for "10K NTC 3950" and you will have plenty of options.

nebulight,  That Murata sensor is what is installed inside the SBMS to measure the internal temperature and since is a small 0805 SMD part is not that practical for battery temp measurement see the above answer for that.
My advice for campers in cold climates is to move to a warm climate :) (sorry that was a joke). But the advice is to have an insulated box and then keep the battery at +10 to +15C with a thermostat controlled heater as the amount of energy that will use will be very low (how low will depend on the amount of thermal insulation and how cold it is but can be very low). For example my fridge is a 40 liter isolated box and it requires about 12W to maintain around 18C below ambient. Since my fridge is based on peltier and COP is likely not better than one the same amount of energy will be needed to keep it warm 18C above ambient.  Keep in mind that is average ambient so 18C above ambient will mean that it will be above +5C even in an extreme climate like mine where average temperature in February is -13C
Also battery may be less than 40 liters so less surface for lost energy meaning you can get away with less than 12W and if you double the isolation of my fridge then you reduce the loss to half (probably not needed).
I expect that in most climates in northern Europe you can make an isolated bot that will not require more than a 5W average that is at most 120Wh/day so not significant at all for a 2.5kWh to 5kWh battery typical used in a camper.
Sunny days are usually the coldest and you will not want to not charge because of battery temperature. 
     

Grant Tanner

unread,
Mar 23, 2019, 11:15:39 PM3/23/19
to electrodacus
Thanks Dacian. Insulating the batteries is definitely a great idea. I've heard that rv tank heater pads could work for heating up the battery bank. And thanks for the probe source.

Some things are still unclear. If the probe is installed, 1) can the BMS be programmed to cut charging if the bank gets too cold automatically and 2) can the heating pad charge the bank to temp automatically somehow (then charging can resume as normal).

Dacian Todea

unread,
Mar 23, 2019, 11:32:01 PM3/23/19
to electrodacus
If you install a battery temp sensor on SBMS you can set a low and high limit but in both cases the charging and discharging will stop as this is not done in software is done in hardware by the ISL94203 BMS IC and that is how that will react to this limits.
It only makes sense to use a heating pad if battery is in an insulated box and if that is the case the battery temperature will never go below whatever you set. There are inexpensive thermostat controllers that you can use for that and you do not need a high power thermal pad 10 to 20W should be plenty and should work at less than 50% duty cycle to keep battery at adequate level even in extreme weather.
Here is a link to a very small and inexpensive digital thermostat https://www.ebay.ca/itm/273135757260 it uses the same type of NTC sensor and you can set a desired temperature connect a low power thermal pad and it will maintain the battery temperature. 

 

nebulight

unread,
Mar 24, 2019, 12:49:26 AM3/24/19
to electrodacus
How is that handled with the SBMS0 since charge and discharge are controlled with extio3 and extio4 since it has no internal mosfets? Could it not be programmed to read temp and open extio ports based on the data provided? I'm not a programmer or engineer so I may not fully understand out the ISL94203 works.

Dacian Todea

unread,
Mar 24, 2019, 1:26:00 AM3/24/19
to electrodacus
The EXT IO3 and EXT IO4 on all SBMS models including SBMS0 is a mirror of the CFET and DFET flags read from the ISL94203. I try to keep the software involvement as low as possible but since EXT IO3 and IO4 are controlled by the microcontroller there is a possibility to do anything I want but that will come at a cost in therms of safety. I like the fact that important decisions are done by hardware as that was designed and tested significantly.
ISL94203 is a stand alone BMS IC meaning it has a build in logic and memory so all those parameter settings you see in advanced parameter setting on the SBMS are saved in to a EEPROM type memory inside the ISL94203 and when the ISL94203 is powered ON that EEPROM memory will be copied to a working RAM memory and based on what is in there (all limits) the ISL94203 will react to protect the battery (no microcontroller is needed it can just work stand alone). Now the microcontroller I added is there to offer the user an interface so that user can program the ISL94203 initially based the specific battery and battery configuration the user has after that is done the ISL94203 will take care of the battery. The other part that the microcontroller is doing is read the data from ISL94203 and display for the user the individual cell voltages and also the microcontroller will measure the current and calculate the battery SOC plus calculate and store all the energy data.
So while the microcontroller and thus the software I wrote has important function all the safety critical stuff is done by the ISL94203.   

Battery should never be below a certain temperature if implemented correctly. Batteries have large thermal mass thus it will take hours to increase or decrease battery temperature thus the only good solution is to find a way to keep the battery in acceptable temperature range at all times.  Easiest way is to have the battery inside the house where you never have extreme ambient temperatures and is ideal for long battery life but if that is not possible then a isolated battery box with thermostat controlled heater is the best option.
Allowing the battery to go below freezing over night will make it impossible to heat the battery back during the day as it will take to long so the day will be done by the time battery is ready to take a charge. There is no way around a thermally isolated and heated battery box if battery needs to be outside.

nebulight

unread,
Mar 24, 2019, 2:27:22 AM3/24/19
to electrodacus
Thanks for the explication. Maybe you could do a combination of things like voltage disconnect and temp disconnect on extio5-6 since the sbms0 will have the options for optoisolators. If I’m understanding it correctly.

Dacian Todea

unread,
Mar 24, 2019, 5:11:30 AM3/24/19
to electrodacus
As mentioned temperature disconnect is already possible and done by the ISL94203 is just that both charging and discharging is stopped when you get to the set limits. I can of course add a separate temperature limit controlled by the microcontroller and then I can disconnect just the charging on SBMS0 but there is no much point in doing so as battery should not be let to cool down below the safe limit for charging as bringing that back by heating the battery will take hours and so it makes no practical sense.
Voltage disconnect is already done of course and based on individual cell voltage that is the main role of a BMS.

nebulight

unread,
Mar 24, 2019, 10:31:14 AM3/24/19
to electrodacus
I agree that temp regulation should always try to be managed, but sometimes things happen out of your control.

The issue with having the discharge cut off in low temps would be if you are on a boat or RV, if there is a low temp condition and the battery shuts off, then you have no power to heat the battery (either with heat pads or furnace controls) and your stuck sleeping in a van with no heat.

My current setup has an external thermostat with my solar panels hooked up to a relay. It’s bulky but works. However moving forward I would like to remove this setup. I’ve learned a lot since then and realize I could use a battery protect with lower power, but this was what I had on hand.

https://youtu.be/YlrP-CwFnlg

Dacian Todea

unread,
Mar 24, 2019, 2:12:36 PM3/24/19
to electrodacus
looking at your video the battery is almost insulated (I see something that looks like thermal insulation but just on the sides). All that is in that box has significant thermal mass so I doubt that even as it is the battery can get down from 20C (68F) to 5C (41F) in 24h that assumes you heat that space every day.
If battery drops below freezing it will take many hours to bring that above 5C (41F) and allow charging and by that time the day is done so you can not take advantage of the solar panel.
That is why in my opinion the only good solution is to have the battery thermally isolated and heated so that it is always at adequate temperature.
I will make sure to add the possibility for the EXT IO to be controlled separate based on temperature limits and disconnect charging only when battery temperature is to low but this should be a separate option not normal type 1 as many people do not use a thermal sensor and they should be able to use type 1 (charge enable if all cells are above 2.8V).
I try to keep the amount of options limited as SBMS is already to complex for most people and that is the reason I try not to add to many options. 
   

Carl Phoenix

unread,
Mar 26, 2019, 5:00:40 AM3/26/19
to electrodacus
 My battery is in my living space (motorhome) so is fine while i'm living there!
 My potential problems arise when i'm not there about 4 to 5 months each year over the winter period in the UK.
Over this rest period i only have minimal draws for security, I also use lower charge cut offs to maintain battery at about 40%  
For Boats and RV's it would be great to be able to save complete profile, one for when we are home and one for when we are not, do you thing that might be possible? 

Dacian Todea

unread,
Mar 26, 2019, 5:15:03 AM3/26/19
to electrodacus
Carl,  Changing the battery settings (including charge limits) will require a power cycle as for security reasons that is how Intersil designed the ISL94203.
I do not think a different profile is needed  if you are using LiFePO4 as the default profile is excellent for any conditions and it will make no difference if you where to use a different profile for the winter will still last the same 15 to 20 years.

Carl Phoenix

unread,
Mar 26, 2019, 8:13:11 AM3/26/19
to electrodacus
Thanks, I was just thinking that as the system was not in normal use when im away it would be held full all the time.

Doug Fentiman

unread,
Mar 30, 2019, 7:05:08 PM3/30/19
to electrodacus
Here is an example of a 12vdc battery box heater for an RV or boat. Runs from PV when energy available so won't deplete batteries.

nebulight

unread,
Nov 28, 2019, 6:20:25 PM11/28/19
to electrodacus
I'd like to revisit this topic as I've now that I've received my SBMS0. I tested with a cup of ice and as expected it shuts of the charge and discharge fet which in a camper van/RV would have undesirable results. I'm just hesitant to install this in my camper van with the possibility of shutting off the entire battery bank and then giving me no way to turn on my heater. Yes, I understand that I could build a box to heat, but with 400ah at 12v, I don't have a lot of space left plus it may not always be enough. You mentioned you may be able to add something in software. Would it be possible just like the over and under voltage conditions to have a recovery temp? This way you could shut off charge at lets say 3 degrees c but not allow charging until the temp recovers to say 6-7c?

Dacian Todea

unread,
Nov 28, 2019, 9:58:47 PM11/28/19
to electrodacus
There is no other good solution in my opinion other than to have a isolated and heated box in cold climates as if the battery gets to cold it will take hours for the temperature to recover due to large thermal mass of the battery.  You can only measure the battery outside temperature and if you try to heat the battery after it went to low in temperature you may think battery is heated but internally battery is still cold and it will take a long time (hours to heat it back up) thus the thermally insulated box and keeping the temperature at around +10C to +15C With a properly insulated box there will be very little energy needed to maintain +10C even with -20C outside temperature.
If there is no isolated box you will just waste huge amounts of energy to try and heat the battery and likely it will not be effective even with the huge energy wasted the battery internally may still be to cold.

Sean Brendan Garrette

unread,
Aug 22, 2020, 12:40:26 AM8/22/20
to electrodacus
Hie everyone
I am trying to buy a thermistor on mouser but I don't know how the thermistor is connected to the SBMS0,all I know is that it must be 10kOhm,3900,tolerance 1%,but I see the thermistors are in the form of blocks,others have a battery terminal, others are probes so am confused which one to choose,forgive my ignorance.

Regards

Dacian Todea

unread,
Aug 22, 2020, 2:36:34 AM8/22/20
to electrodacus
Sean,

That NTC thermistor is like a variable resistor (resistance will vary based on temperature) and so it is a device with two terminals same way a resistor will be. Send me a link to the one you prefer and I will take a look at.
 


Sean Brendan Garrette

unread,
Aug 22, 2020, 5:58:56 PM8/22/20
to electrodacus
Dacian and all
These are the thermistors I am seeing,which one should I buy which is suitable for SBMS0/easiest to connect.The specifications are the same.

Regards


IMG_1612.jpg
IMG_1609.PNG
IMG_1611.jpg
IMG_1610.jpg
IMG_1613.jpg

Barry Timm

unread,
Aug 23, 2020, 1:30:49 AM8/23/20
to electrodacus
I have the opposite but similar problem to the cold - I stay in very hot climates in our RV, and when dry-camping (with no external power) the RV A/C running on solar only may not be enough to keep the inside temperatures much below about 90F.
As I understand that high temps are really bad for longevity of LFP, I was wondering about a well insulated battery box solution, but then am unsure how much self-heating from the battery working at an average of perhaps 1000w during the middle of the day, may do to actually create MORE heat inside the insulated box, which defeats the purpose of the insulation.
Would this "self heating" become an issue, or is it a non-issue and thus is it a good recommendation to insulate the battery to help reduce the potential high heat periods?
Pls understand that it's unrealistic to insulate the RV itself much better than it already is, so we are sometimes subject to these high heat conditions. I'm just trying to figure out a way to keep the battery cool enough.
thanks!

Dacian Todea

unread,
Aug 23, 2020, 3:32:42 AM8/23/20
to electrodacus
Sean,

All of them will work but the lug type should be the most easy to use.

Dacian Todea

unread,
Aug 23, 2020, 3:47:22 AM8/23/20
to electrodacus
Barry,

The battery itself will not produce almost any heat and will be almost at ambient temperature at all time. I do not see more than 1 or 2C above ambient on my battery.
Battery will degrade faster at higher heat. The calendar aging is around 1% per year for my battery that is at an average of about 23C (18C to 26C variation over a full year lower temperatures in winter higher in summer).
About two days ago we had a peak of 37.5C (99.5F) outside so inside the house it got to 27C (80.6F) fairly hot and dry here also this year.
Having the battery in average 10C higher temperature will mean about a doubling in battery calendar aging degradation so instead of say typical 1% per year it may be 2% per year.
Not sure if the 1% per year extra calendar degradation justifies having an insulated battery box and cooling maybe with a peltier module.
Say your battery internal resistance is 8mOhm (I doubt it will be higher) then 1000W / 26V say about 40A the voltage drop on the internal resistance will be around 40A x 0.008Ohm = 0.32V x 40A = 12.8W worth of heat that will need to be extracted with peltier from the box to maintain the battery temperature but to that you need to add the box thermal loss (if well insulated and 10C delta maybe another 10W).
So two small peltier set at 6V using about 12 to 15W including the fans will be capable to keep the battery 10C below ambient and save you that 1% or so extra degradation per year.
    



Barry Timm

unread,
Aug 23, 2020, 12:16:55 PM8/23/20
to electrodacus
Wow, thanks Dacien. Very helpful. I really appreciate how you lay out the calculations so that I can learn as well a simply getting an answer. 
Thanks again! 
Now, to find a couple of suitable Peltier modules to check feasibility given my space constraints.... 

Dacian Todea

unread,
Aug 23, 2020, 2:18:13 PM8/23/20
to electrodacus
Barry,

I think that if you are space constrained the larger problem is building a properly insulated battery box as that will take much more volume than the peltier modules.
The calculation is based on the typical 12706 peltier module that my small fridge also uses and based on undervolting that to 6V so you can have those two in series at 12V and install them maybe on same heatsinks with a single large 12V fan.
You can find kits like this  https://protosupplies.com/product/thermoelectric-peltier-tec1-12706-cooler-kit/ that incude the heatsinks and fans and this one looks very similar to what my small fridge has.


Barry Timm

unread,
Aug 23, 2020, 3:14:26 PM8/23/20
to electrodacus
Perfect! Many thanks, Dacien!

Allen Tindall

unread,
Aug 30, 2020, 12:39:05 AM8/30/20
to electrodacus
I purchased this thermistor: Mouser part number 81-NXFT15XV103FA2B14 Mfg part number NXFT15XV103FA2B140.  It is very small and has 2 tiny wires about 3 inches long varnished together.  I used 2 crimp ferrules (smallest size available) and a twisted pair of 24 AWG wire.  The ferrules held the ends together so I could solder them.  I used shrink wrap to insulate the connections.  This allowed me to dangle the thermistor between the batteries in the middle of the pack.  Works well to allow the SBMS0 to sense battery temperature.
Reply all
Reply to author
Forward
0 new messages