Solar BMS (Solar Battery Management System)is a solar charge controller designed to replace the Lead Acid solar
charge controllers most people use today in Offgrid, RV, Boats and
multiple other applications with 12V and 24V systems.
Solar BMS can be used with 3 up to 8 Lithium cells (any type) or supercapacitors.
The new SBMS100 will have multiple improvements over the first generation SBMS4080 see further for details.
Lithium and in particular LiFePO4 is a better long term investment than Lead Acid batteries.
--LiFePO4 has 2000 to 8000 cycles (70% to 100% DOD) vs Lead Acid 250 to 1200 cycles (20% to 50% DOD).
(This means you can get LiFePO4 with half the Lead Acid capacity since LiFePO4 can be discharged deeper and does not have to be fully charged as Lead Acid).
--LiFePO4 has a charge / discharge efficiency of 95 to 98% vs Lead Acid just 50 to 75%.
--LiFePO4 will cost about the same as 2x capacity Lead Acid.
(A half capacity LiFePO4 will perform the same or better do to ability to discharge deeper and stay discharged with no effect on life cycle and do to better charge / discharge efficiency)
--LiFePO4 protected with Solar BMS can last 20 to 30 years where a typical Lead Acid will only last 4 to 6 years.
--LiFePO4 can be 5 to 10x better value than Lead Acid over the life of the battery
--The cost benefit are not the only benefits.
- LiFePO4 can be installed indoors with no need for external venting since it does not produce flammable Hydrogen gas as Lead Acid.
- LiFePO4 even at the same capacity as Lead Acid is much smaller and lighter (in some applications this can be important).
- LiFePO4 is maintenance free (AGM also claims that but in solar applications you probably need an expensive (1 Litre/kWh) gasoline or diesel generator to recharge the battery if there are more than two consecutive cloudy days else the battery life will be drastically affected).
Some Links in support to my claims:
Bosch has a similar grid tied solar storage system see Link using Lithium Iron Phosphate with a 7000cycles and 25years life claim.
For Batteries you can check the Winston specifications for their Lithium Iron Phosphate batteries Link There are other manufacturers of Lithium Iron Phosphate personally I use GBS cells since they where available about 3 years ago locally and where the best option for me at that time. But Winston and others seems to have better specs. My battery has already two years of full time offgrid with daily deep discharge and there is no measurable degradation see my house power consumption graphs below for more details.
1) Double the power
The SBMS100 can handle a total of 6kW (3kW PV array and 3kW Load) where SBMS4080 is able to do 3kW (1kW PV array and 2kW Load).
In order to make this happen a few changes are necessary:
a) Larger power connectors that can take 35mm2 (#2 AWG) cables to handle 120A limit on the SBMS100
b) Thicker copper traces on the metal core PCB.
c) Higher power mosfets and more of them to handle the increased power and dual PV inputs.
d) Ideal diodes for the dual PV input.
(ideal diodes are made of a power mosfet and an ideal diode controller. A normal Schottky diode will dissipate to much heat at this current level that is why ideal diodes are the only option).
2) 24bit ADC for increased current, power and energy measurement accuracy.
The SBMS4080 uses the internal 14bit ADC of the ISL94203 for current measurement. While this was a cost effective method it did not offer great resolution at just 100mA and because this was also used for HW over-current protection there where some aliasing issues that needed SW filters.
The new SBMS will use a 24bit ADC from Linear Technology with multiple differential inputs to measure current on Battery and the two PV inputs with increased resolution (one or two orders of magnitude better).
Some of this 24bit ADC inputs will also be available to user on the 26pin connector for measurement of external sensors and possible automations or future add-ons for the SBMS100.
This will require:
a) The 24bit ADC IC and multiple current sense amplifiers 3 for SBMS100 instead of 1 on the SBMS4080.
b) A 4 layer main PCB do to increase in complexity over 2.5x more parts that on the SBMS4080.
3) Higher resolution and size LCD
As those that have the SBMS4080 know the small 1.4" Nokia LCD is not that great with his limited 84x48 B&W resolution.
On the new SBMS100 a 2.2" Color LCD with 320x240 pixels will be used.
The new LCD will offer much more space for displaying data and maybe even some energy graphs (not sure I will have time to implement graphs at release time but will sure be an option in later firmware updates).
4) Bluetooth 4.1 or WiFi
You will decide what wireless connectivity will the new SBMS have, Bluetooth 4.1 or WiFi.
Options are for Bluetooth 4.1, RN4020 with 100m range or for WiFi ESP-03 based on ESP8266 (this seems quite popular at the moment for IoT applications).
If you have any other suggestions for this please let me know the decision is not finalized so you can decide if you prefer Bluetooth or WiFi the one with most votes will be implemented maybe even both (as option not at the same time) if request for both are close.
* Dual PV input is available only on the SBMS40 and SBMS100.
** Cell balancing can be set lower from the SBMS menu if needed.
*** It will be your choice if is Bluetooth or WiFi that will be implemented. Just let me know what you preference is and based on your input one will be selected.
**** Max TDP (Thermal Design Power) is similar to what you may be familiar from CPU's is the total amount of power lost as heat on the device. The SBMS will need to be installed on a small passive heat sink (can even be a piece of aluminium sheet) able to dissipate that TDP.
As you can see from this small TDP the new SBMS is even more efficient with 99.6% transfer efficiency.
STC power and current rating is what you will find on a sticker on the
back of the PV panel and is the power output with 1000W/m2 solar
irradiation and 25C cell temperature.
In a really cold winter day you can expect to get that and even a bit more that is why SBMS is designed to support 20% higher current than rated for example the name SBMS100 comes form the 100A STC rated max current but it can support 120A (20% more) to take in consideration those cold winter days with bright sun.
** Battery capacity size recommendation is just that a recommendation based on max PV size configuration and you can use smaller or larger battery capacities based on your PV array size and type of your battery. For typical LiFePO4 a good value will be 0.2C to 0.5C charge current.
Below table gives an example of max configuration for each Solar BMS and expected energy production and consumption.
* STC (Standard Test Condition) power what the panels are advertised as normally and written on the back of the PV panel.
** Data extrapolated from my own power production/consumption in the worst winter month (December in my case) see below my OffGrid House power consumption for details.
*** Low power mode is the power consumption when there is no sun and normally lower than a sunny day for more details see my OffGrid House power consumption below.
I live in Canada so the rewards levels are listed in CAD (Canadian Dollars). In order to make things easier for you I listed the conversion in other currencies based on google conversion tool the actual amount will depend on your financial institution (bank) so the conversion is just for reference.
A short photo history of the Solar BMS from the very first PCB of a SBMS2040 to the first production of the SBMS4080 for the bakers of my first Kickstarter project.
The SBMS4080 will be the basis on the new SBMS100 and used as prototype since all the important HW in the SBMS4080 will be present in the SBMS100 including the ISL94203.
The photo above describe my solar setup and is made out of a 720W PV
array (3x240W 60 PVcells/panel) a storage battery (made out of 8 cells
from GBS each cell is 100Ah and 3.2V nominal all connected in series for
a total of 24V ) and the SBMS4080 made a reality with the first Kickstarter campaign.
The power/energy graph below was done using the data logging functionality of the SBMS4080 installed in my OffGrid house
Data acquisition was done every minute for 7 days so 1440 datasets /day x 7 days = 10080 data sets
Each dataset looks like the one below:
And the content is Date,Time, individual voltage on the 8 cells, battery current and solar PV current.
The format is this Y,M,D,H,M,S, ,Cell1[mV],,,,,,,Cell8[mV],BattCurrent[A/10],SollarCurrent[A/10]
Total battery voltage and current was used to calculate the power from the PV panels and power getting in or out of the battery.
The orange graph is the power generated by the 3x 240W PV panels and is always positive. Since this is done in winter here the outside temperature was low (-10C to -20C) allowing for good output from the solar PV panels as you can see in the graph getting around the nominal 720W in the afternoon. The total energy from the Solar PV panels is the area in orange including the one not visible because of the blue graph overlap. The visible part of the orange area is actually the Load energy as the difference between the orange area PV energy and blue area above the zero line that represent Battery charging energy.
Click on the graph to download the raw .csv file generated by SBMS4080.
As you can see from the animated graph above the daily PV array
production was between 4.285kWh in a full sunny day (Day 5) and 1.082kWh
in a cloudy day with no sun breaks (Day 7). As I mentioned in the above
tables in December I can get an even worse day with heavy snow clouds
and short day that will only generate as low as 0.3kWh for the entire
day with the same PV array.
The energy consumption for the 7 days has big variations from as little as 1.132kWh to as much as 4.822kWh with an daily average of around 2.7kWh and around 80kWh/month energy use if I where to extrapolate this 7 days. This is normal for February since days are already longer and less cloudy than December in my geographic area (Canada, Saskatchewan closest large city is Regina). In November and December my power consumption can drop to 60kWh/month and in spring summer it can get as high as 90 to even 100kWh/month it will depend on how much cooking we do.
Before getting in to details about my loads I will do a short explanations to the power graph by day.
Day 1 (at about 10:00 there is a sharp increase in power that is do to the fact that I cleaned the panels of snow, at about 12:40 or so there is a short 2.5minute increase in power consumption that was form the microwave reheating some food and you see that pulse at about -900W but the total power consumption is that + the solar PV power at that moment about 500W so the total load was 1400W from the DC side. Then later that day 15:30 we made a bread with the small convection oven and since that has a thermostat you see many on / off cycles to keep the set temperature. Then even later after the sunset around 20:30 I used the propane heater that has 3 pumps that work at different stages and all combined get to about 130W + the base load that was around 50W you see a peak of around 180W).
Day 2 (About the same as first day some cooking with the convection oven and the water pumps from the heater during the day and a bit after sunset).
Day 3 (No cooking so just DC loads mostly the water pumps for heating and you notice around 14:30 the battery was full but you will see this better in day 6)
Day 4 (There was again a bit of snow on the panels in the morning and again the battery was full at around 14:30 even if we done some cooking).
Day 5 (This was a full sunny day and I done quite a bit off cooking and the water pumps where running all day plus I used my larger computer not the laptop that is why you see higher power consumption after sunset).
Day 6 (There where some clouds in the morning then I run the water pumps from the heater. The battery was fully charged by about 12:00 and the water pumps where running for another hour you see more often PV connected and disconnected do to larger load then when went shopping all day so only the small base load was on the refrigerator and one of the pumps at low setting. So all that is a good example of unused solar energy because it was not needed and this is normal in offgrid situations since you can not sell the excess power. Once I will do my electric house heating with solar PV panels all that extra energy will be used to heat the house in winter but I will still need to find a use for extra energy in summer).
Day 7 (This is the lowest solar PV production but it could have been even lower there where portion of the day where you where able to see where the sun is. If it was one of those days the power will have been always below 100W mark and the production will have been as low as 0.3kWh).
Even if this is Canada and you may expect low solar output it is not. As a comparison I get about the same amount of solar energy average over a year as in Sydney Australia. If you want to know how that compares to your location check out this great on-line calculator PVWatts. For direct comparison with my location select Canada (SA Regina).
December 2012 When Solar BMS idea was born.
April 2014 SBMS4080 was successfully funded on Kickstarter.
October 2014 SBMS4080 was delivered to backers.
March 2015 Kickstarter campaign for the new SBMS100.
April 2015 Hopefully a successfully funded SBMS100 campaign.
May 2015 PCB design for the new SBMS100.
June 2015 Build the first SBMS100 prototype.
July 2015 Software development implementing all the new changes.
August 2015 Continue Software development, testing and PCB ordering.
September 2015 Ordering parts for production and start production.
October 2015 Finalize production and testing and start shipping to bakers.
Risks and challenges
I think the SBMS4080 Kickstarter project is a good indication that I have the knowledge and discipline to take a project from design to delivery the way I promised.
Unless something absolutely unexpected gets in the way I expect to deliver in the same way I did on the last project.
- (40 days)