Author Topic: Solar BMS a solar charge controller for LiFePO4  (Read 30515 times)

0 Members and 1 Guest are viewing this topic.

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Solar BMS a solar charge controller for LiFePO4
« on: February 22, 2015, 10:37:28 AM »
This will be the second version of the Solar BMS with a lot of improvements over the current SBMS4080.
I had a successful Kickstarter last year for the SBMS4080 (Solar Battery Management System) the 4080 is from 40A charging and 80A discharging so with max configuration 8 cells of LiFePO4 that is 1000W of PV panels possible and 2000W load. I produced about 100 units of SBMS4080 and smaller version SBMS1616 for backers and I also have installed two in my offgrid house (the main reason I designed this Solar BMS).
Here is a photo that I took today of my SBMS4080
 
I made the above photo to show a record of 29.2A from 3x 240W PV panels so 720W total. The reason for this is full sun at 1:16PM snow on the ground and -22C
And below you can see a 3D render of how the new Solar BMS will look like this will be able to handle 3000W of PV and 3000W load at the same time with a TDP of just 22W
   
And here is a comparison of cost for stored energy between the best Lead Acid from Trojan and LiFePO4 from Winston.

So you see why I went for LiFePO4 over the more traditional Lead Acid. There are other problems with Lead Acid especially in a cold climate like mine but the 3x difference in cost for each kWh stored is a good enough reason by his own.
You can see more technical details on my webpage http://electrodacus.com including at the end of the page a nice 7 day graph with my house power production/consumption.   
« Last Edit: April 05, 2015, 10:29:39 AM by electrodacus »
 

Offline Prime73

  • Regular Contributor
  • *
  • Posts: 174
  • Country: ca
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #1 on: February 22, 2015, 11:36:50 AM »
Seems interesting. Where can I take a look at the schematics, etc ?
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2806
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #2 on: February 22, 2015, 12:07:30 PM »
Good stuff but if you want your product to be attractive to the vast majority of serious off grid and grid tie with battery backup installations you'll need to step up the specs:

1) MPPT is a must - I see no mention of this on your website .

2) It needs to be compatible with 48V nominal battery systems.

3) It needs a max PV VOC input of at least 120V to accommodate strings of 3 in series of the commonly used and cheapest per watt PV panels which have VOCs of 35-40V.  At least 3 of these in series are needed for 48V systems.

Right now many are using standard programmable MPPT controllers for use with Lithium batteries in off grid and grid tie with battery back up systems.  BMS if used is separate from the charge controller.

As more people transition from LA to LiFePO4 batteries there is definitely a need for MPPT charge controllers with integrated BMS but you will need to step up your specs if you want to compete with the big guys (Midnite, Outback, Morningstar, Schnieder).

You've got a head start on them - good luck!
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #3 on: February 22, 2015, 12:18:17 PM »
Seems interesting. Where can I take a look at the schematics, etc ?
Links to schematics and firmware are in the first Kickstarter project updates. Search Solar BMS on Kickstarter (only one project) and look at the latest updates. 
« Last Edit: February 22, 2015, 12:37:22 PM by electrodacus »
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #4 on: February 22, 2015, 12:35:03 PM »
Good stuff but if you want your product to be attractive to the vast majority of serious off grid and grid tie with battery backup installations you'll need to step up the specs:

1) MPPT is a must - I see no mention of this on your website .

2) It needs to be compatible with 48V nominal battery systems.

3) It needs a max PV VOC input of at least 120V to accommodate strings of 3 in series of the commonly used and cheapest per watt PV panels which have VOCs of 35-40V.  At least 3 of these in series are needed for 48V systems.

Right now many are using standard programmable MPPT controllers for use with Lithium batteries in off grid and grid tie with battery back up systems.  BMS if used is separate from the charge controller.


As more people transition from LA to LiFePO4 batteries there is definitely a need for MPPT charge controllers with integrated BMS but you will need to step up your specs if you want to compete with the big guys (Midnite, Outback, Morningstar, Schnieder).

You've got a head start on them - good luck!

I have no intention to attract grid tie users this is a product specifically for OffGrid, RV, Boats.

1) MPPT is obsolete (not going to get in to details) but if PV panels are under 5$/Watt it makes no economic sense to implement MPPT and currently at 1$/Watt or under.
Look at the first photo 720W PV array generating 791W that is because I use 8x LiFePO4 with 60 cells PV an MPPT will do worse with this setup.
2) I will stay with 12V and 24V compatibility no 48V variant for safety (close to 100V OC on PV with 48V battery) and a completely different strategy is better.
3) Again max OC input is limited to 50V ideal for 24V setup (60 cells recommended but 72 cells can be used for long wire runs or very hot climate)
BMS will not work good separate for the charging part.
I do not want to compete with anyone this was designed for my on use and since there was some interest I made a small kickstarter. Now I do a new model for those that want larger PV array 3000W possible with the new Solar BMS.
If you look at my Offgrid house you will see my power consumption is between 60 and 90kWh/month mostly for electric cooking and my PV array is just 720W. With the new SBMS 240 to 360kWh/month is possible depending on climate and use pattern.
Many technologies are obsolete because of low cost PV panels not just MPPT  but solar tracking and even solar thermal panels.
I will do the heating for the house with solar PV since is by far the most cost effective option.
But I will need to build a digital MPPT for that since it dose not exist from what I know. 

Here is a 7 day graph with my energy production consumption


Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2806
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #5 on: February 22, 2015, 12:52:44 PM »

1) MPPT is obsolete (not going to get in to details) but if PV panels are under 5$/Watt it makes no economic sense to implement MPPT and currently at 1$/Watt or under.

If you don't understand why cheap PV does not negate the need for MPPT then you have some gaps in your knowledge.

What cheap PV panels has done is make trackers (almost) obsolete and direct solar hot water less necessary as well (PV with a heat pump water heater can work very well)..

Quote
2) I will stay with 12V and 24V compatibility no 48V variant for safety (close to 100V OC on PV with 48V battery) and a completely different strategy is better.
?????

Quote
3) Again max OC input is limited to 50V ideal for 24V setup (60 cells recommended but 72 cells can be used for long wire runs or very hot climate)

The problem with a max input of 50V for a 24 V system is that it would require a  string of 2 12 V panels or a "real" 24V panel with a Vmp of 32+.  The most inexpensive PV panels, due to economy of scale, are the ones produced in large quantities for the large market of "grid-tie" systems and that have VOC's of 34-40V and Vmps of less than 31V which is not high enough as a single panel string to reliably charge a nominal 24V system..  These panels are by far the most commonly used in large off grid installations now because they are more common and less expensive. Using 2 in a string for a 24V system or 3 for a 48V system is one reason why MPPT is needed.  Now as LiFEPO4 battery options expand it may be that  the traditional LA nominal battery bank sizes (12V,24V, 48V) are no longer the norm but that will also require some changes in available inverters.  But for now, as a "plug and play" replacement for current systems that is not the case.

Quote
BMS will not work good separate for the charging part.
  It works just fine - it's just that an integrated charger with BMS would be a more efficient solution.

It's great that your doing this. As I said - I think charge controllers with integrated BMS for Lithium batteries is needed. If you don't plan to compete for the larger off grid systems and grid tie with battery back-up systems then that's cool.

« Last Edit: February 22, 2015, 12:58:07 PM by mtdoc »
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #6 on: February 22, 2015, 01:35:34 PM »

If you don't understand why cheap PV does not negate the need for MPPT then you have some gaps in your knowledge.

What cheap PV panels has done is make trackers (almost) obsolete and direct solar hot water less necessary as well (PV with a heat pump water heater can work very well)..


I promise I understand better than you think. An mppt needs a DC-DC converter that is large heavy and expensive (think about a 3000W DC-DC converter for my new Solar BMS able to output 120A) then take an average of 10% gain over a year for MPPT (I'm quite generous with 10% and consider using 72 cells PV panels or larger voltage for long runs) With 60 cells there is no gain at all maybe some loss if it even works.
3000W at 1$/Watt is 3000$ for the array.
A 10% more will cost 300% so the 3000W 120A DC-DC converter should cost less (can you provide me with a source for one less expensive than 300$) then if I ship internationally that monster MPPT charger it will cost at least 100 to 150$ do to large size and especially weight.
The charger should last at least 25 years so no electrolytic allowed in the design or you will need to change that MPPT a few times over the life of the system making that even worse value. My current SBMS is about 240g and the next version will not be larger than 350g and cost about 20$ to 25$ to ship anywhere in the world.
I have no electrolytic in my design so 25 years no problem all is solid state there is also no PWM or so absolutely no noise (electrical noise or interference with other devices)
There is absolutely no place for MPPT in OffGrid solar system at 1$/Watt or under PV panels In grid tie is a different story since you already have the expensive DC-DC and you only need the MPPT algorithm.


Quote

The problem with a max input of 50V for a 24 V system is that it would require a  string of 2 12 V panels or a "real" 24V panel with a Vmp of 32+.  The most inexpensive PV panels, due to economy of scale, are the ones produced in large quantities for the large market of "grid-tie" systems and that have VOC's of 34-40V and Vmps of less than 31V which is not high enough as a single panel string to reliably charge a nominal 24V system..  These panels are by far the most commonly used in large off grid installations now because they are more common and less expensive. Using 2 in a string for a 24V system or 3 for a 48V system is one reason why MPPT is needed.  Now as LiFEPO4 battery options expand it may be that  the traditional LA nominal battery bank sizes (12V,24V, 48V) are no longer the norm but that will also require some changes in available inverters.  But for now, as a "plug and play" replacement for current systems that is not the case.


The most inexpensive PV panels are the one that I use 60 cells and as you see works great with 24V LiFePO4 battery. Max power point for those vary based on cell temperature from no less than 28V to 31V there is no voltage loss on the SBMS (there is so small that it dose not count will be under 2mV/A on the new version). So yes 31V is much more than good enough for LiFePO4 and SBMS.
12V and 24V will probably remain the norm for a long time do to automotive industry and there are to many things that accept this level. 4 or 8 cells LiFePO4 are a perfect replacement for this.
LiFePO4 will have an almost constant voltage from 10% SOC to 90% SOC around 3.2 to 3.3V / cell and will not sag under heavy load.
   

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2806
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #7 on: February 22, 2015, 03:32:43 PM »

I promise I understand better than you think. An mppt needs a DC-DC converter that is large heavy and expensive (think about a 3000W DC-DC converter for my new Solar BMS able to output 120A) then take an average of 10% gain over a year for MPPT (I'm quite generous with 10% and consider using 72 cells PV panels or larger voltage for long runs) With 60 cells there is no gain at all maybe some loss if it even works.

The increased efficiency obtained from an MPPT controller is only part of it's advantage. Much more important in the world of system design is that it allows you to run higher string voltages.  This has several advantages:

1)  Use of less expensive panels as I previously described.
2.) The ability to have an array farther from your battery bank - higher voltage string=lower amps= less voltage drop = smaller wire gauge/less copper needed.  This is a big issue since copper is expensive! Off grid systems in particular often need to locate their arrays far from their battery bank.
3) Higher string voltages gives you more headroom which makes a huge difference in non - full sun conditions- i.e. partial shading and cloud cover.

These are the real world system design issues that need to be addressed for a controller to have widespread appeal.

Quote
A 10% more will cost 300% so the 3000W 120A DC-DC converter should cost less (can you provide me with a source for one less expensive than 300$) then if I ship internationally that monster MPPT charger it will cost at least 100 to 150$ do to large size and especially weight.

No doubt that it costs less to build a low voltage non-mppt controller. But quality high input voltage, high output MPPT controllers are being made and sold for very reasonable prices now.  The market has spoken. The only place for non-mppt controllers now is small 12V systems.

Quote
There is absolutely no place for MPPT in OffGrid solar system at 1$/Watt or under PV panels .

Well that's just nonsense for the reasons I stated and because the market has already spoken on this issue. Spend a couple of hundred dollars more on a MPPT controller or spend many hundreds or thousands more on extra PV, mounting hardware and copper wire?  :palm:

None of the serious charge controller manufacturers are selling non-mppt controllers for anything but very small systems. The vast majority of controllers they sell are mppt - even though PV prices have been below $1 per watt for several years now.

Just because there are some small installations where your controller will make sense does not mean it will make sense for the majority of battery based PV systems.   Not a problem if your plan is to fill a niche market but a big issue if you expect your controller to have a wider appeal.

Oh, and I didn't even get into the advantages an MPPT controller brings to small wind or hydropower systems...


« Last Edit: February 22, 2015, 03:45:22 PM by mtdoc »
 
The following users thanked this post: fourtytwo42

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #8 on: February 22, 2015, 04:27:42 PM »

The increased efficiency obtained from an MPPT controller is only part of it's advantage. Much more important in the world of system design is that it allows you to run higher string voltages.  This has several advantages:

1)  Use of less expensive panels as I previously described.
2.) The ability to have an array farther from your battery bank - higher voltage string=lower amps= less voltage drop = smaller wire gauge/less copper needed.  This is a big issue since copper is expensive! Off grid systems in particular often need to locate their arrays far from their battery bank.
3) Higher string voltages gives you more headroom which makes a huge difference in non - full sun conditions- i.e. partial shading and cloud cover.

These are the real world system design issues that need to be addressed for a controller to have widespread appeal.

1) As I mentioned before less expensive panels 60 cells work the best and is what I'm using for my system and with them MPPT will not be able to work or provide any advantage.
2) You can have the allowed 3% voltage loss that will still be enough. Longer and thicker wires will cost more (not the case for RV and boats you can not have long runs there :) ) but I just calculated the cost of wires for a long run on my future 6kW array (for heating) and the cost of cable is 500$ for a 3kWh array that will be half at 250$ and with higher voltage how much will you save from that 250$ maybe 100$ you still need cables.
3) Not sure you understand how solar PV works voltage is not affected by amount of solar radiation only current is look at some PV panels datasheets.

Again I'm not interested in widespread appeal this is for people like me that see the benefit of LiFePO4 and do not want to spend almost two years developing their own custom version of this Solar BMS if they are even able to do that.   

Quote
No doubt that it costs less to build a low voltage non-mppt controller. But quality high input voltage, high output MPPT controllers are being made and sold for very reasonable prices now.  The market has spoken. The only place for non-mppt controllers now is small 12V systems.

The market is usually bad informed. There was a time when PV where 5$/watt even 10$/Watt and MPPT controllers where considered exotic and expensive (that was the time it made sense to invest in them since 10% with PV prices 10x higher than now made quite a bit of sense) Now after a lot of development in MPPT people seen the advantage but is to late since the advantage is gone with the much lower cost of PV panels.
Another example is all those wind turbine with more than 3 blades because uniformed people thing more blades is better when the opposite is true.  But there are enough small and medium wind turbine manufacturers to build this even if they make absolute no sense just because there is a request form uniformed customers.

Quote
Well that's just nonsense for the reasons I stated and because the market has already spoken on this issue. Spend a couple of hundred dollars more on a MPPT controller or spend many hundreds or thousands more on extra PV, mounting hardware and copper wire?  :palm:

None of the serious charge controller manufacturers are selling non-mppt controllers for anything but very small systems. The vast majority of controllers they sell are mppt - even though PV prices have been below $1 per watt for several years now.

Just because there are some small installations where your controller will make sense does not mean it will make sense for the majority of battery based PV systems.   Not a problem if your plan is to fill a niche market but a big issue if you expect your controller to have a wider appeal.

Oh, and I didn't even get into the advantages an MPPT controller brings to small wind or hydropower systems...

Look as I mentioned before I know MPPT makes no economic sense for OffGrid where you need the expensive DC-DC converter. Still this is open source so you can add an DC-DC converter to the design and try to commercialise that or add an external DC-DC converter that is also a possibility.
As for wind and hydro MPPT "advantages" ... now I know for sure the one that dose not understand how things work is you (sorry If I seem a bit rude but I will not have time to make elaborate technical explanations about all this issues).
I'm an electrical engineer. I love to know what is your qualification?

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2806
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #9 on: February 22, 2015, 06:22:13 PM »
Look, I was only trying to give you some constructive criticism in case you were interested in designing and/or  selling a charge controller that would appeal to more than a niche market and kickstarter backers.

As far as my "qualifications" - no I'm not an EE and do not claim to have the skill to design a charge controller. But designing a charge controller is very different from designing a complete PV system.

What I do have is several years experience with PV systems.  I have designed and installed several, including my present 4500 watt battery based system.  I have helped several others design their systems.   I am in regular communication with several people who design and install off grid RE systems for a living and also with some who design charge controllers. I know what people need in a charge controller for Lithium battery based systems. I know the difference between a system that barely works and one that works efficiently and is cost effective.

Good luck to you with your project.  In a few years when several of the big guys are selling MPPT controllers for LiFePO4 battery based systems you may think back to this discussion.



 
The following users thanked this post: fourtytwo42

Online Seekonk

  • Super Contributor
  • ***
  • Posts: 1578
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #10 on: February 23, 2015, 04:25:52 AM »
So the story I got here was BATTERY, your on/off controller is irelevant.  Everyones solar experience is different.Someone tells me 20 years life on anything and I give them a look. Nothing I had ten years ago still works.  I find RE people the biggest energy wasters non the planet.  I designed my system to produce hot water, it charges a battery and runs refrigeration etc on the side.  No MPPT because you don't like capacitors.  I share your dismay, the last MPPT controller I saw had two surface mount electrolytics on the input.  Most controllers aren't much better.  Sounds like a good new product, a stand alone capacitor bank with self test to place before a controller.  MPPT is like your vehicle ECM.  It works great till you get a faulty injector.Then it screws up everything else that is working. Some will just track down to oblivion.  A controller should be smart and tell you when things have gone to crap. Don't believe in those 150A controllers. I have multiple banks and controllers for each.
All are under master control.  I don't have much of a battery.  It is only there for startup current.  Everything is real time power useage.  If fridge needs to turn on, the washer turns off for a period.  It is a totally different way of thinking.  Seven days of power generation, what does that have to do with anything?  It is those four days when the system can't even generate an amp that have to be designed for.

There are a lot of solar products out there and they all meet someones need. My concept of what solar should be has changed many times over the years. I appreciate your challanging my current thoughts. There is a solar product I think will change the developed world in the next ten years, like adding extra insulation to the attic did.  There are millions of electric water heaters in use.  If all their lower heating elements were connected to 300-400W of solar panels through a MPPT controller, dramatic electric savings would result.  Everyone has some place they could mount a couple solar panels.  The cost would be farless than the $1800 many spend for a heart pump water heater.  It would dramatically improve the performance of those too.

 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #11 on: February 23, 2015, 06:32:57 AM »
Look, I was only trying to give you some constructive criticism in case you were interested in designing and/or  selling a charge controller that would appeal to more than a niche market and kickstarter backers.

As far as my "qualifications" - no I'm not an EE and do not claim to have the skill to design a charge controller. But designing a charge controller is very different from designing a complete PV system.

What I do have is several years experience with PV systems.  I have designed and installed several, including my present 4500 watt battery based system.  I have helped several others design their systems.   I am in regular communication with several people who design and install off grid RE systems for a living and also with some who design charge controllers. I know what people need in a charge controller for Lithium battery based systems. I know the difference between a system that barely works and one that works efficiently and is cost effective.

Good luck to you with your project.  In a few years when several of the big guys are selling MPPT controllers for LiFePO4 battery based systems you may think back to this discussion.

Thanks for the comment. Sorry if I was looking a bit less polite in the last comment.
There is not much knowledge in this field of OffGrid energy storage and many things have changed fast in the last few years. 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #12 on: February 23, 2015, 06:40:29 AM »
isnt MPPT mainly used because of trying to match PV conditions to load? if PV incoming power fluctuations can be made irrelevant, would a MPPT not be required? ie : a PV power conversion that will just keep on supplying power to charge something? assuming battery bank is infinite or large enough to enable the PV conversion side to be always on.

btw, which kind of lifepo4 cell did you buy? the huge blocky ones? i bought some to try ... but im stuck at making AC mains PSU as step 1 first lol

a very interesting review of battery


lifepo4 vs AGM, im unfamiliar with AGM. but i think this practical use review was very very informative

You are talking about MPPT in grid tie and there it makes sense since the HW is already there is needed to push energy in to the grid.
Battery bank is not infinite not even close you always have excess energy in offgrid applications.
I got mine GBS 100Ah a few years a go there are better ones now I will suggest looking for Winston they have cells up to 1000Ah.
I'm sure you are familiar with AGM (Absorbed glass mat) is a sealed Lead Acid battery.
If you look in the comment section of that video you will see my comment about his video. I do have youtube videos about this subject.
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #13 on: February 23, 2015, 06:57:30 AM »
So the story I got here was BATTERY, your on/off controller is irelevant.  Everyones solar experience is different.Someone tells me 20 years life on anything and I give them a look. Nothing I had ten years ago still works.  I find RE people the biggest energy wasters non the planet.  I designed my system to produce hot water, it charges a battery and runs refrigeration etc on the side.  No MPPT because you don't like capacitors.  I share your dismay, the last MPPT controller I saw had two surface mount electrolytics on the input.  Most controllers aren't much better.  Sounds like a good new product, a stand alone capacitor bank with self test to place before a controller.  MPPT is like your vehicle ECM.  It works great till you get a faulty injector.Then it screws up everything else that is working. Some will just track down to oblivion.  A controller should be smart and tell you when things have gone to crap. Don't believe in those 150A controllers. I have multiple banks and controllers for each.
All are under master control.  I don't have much of a battery.  It is only there for startup current.  Everything is real time power useage.  If fridge needs to turn on, the washer turns off for a period.  It is a totally different way of thinking.  Seven days of power generation, what does that have to do with anything?  It is those four days when the system can't even generate an amp that have to be designed for.

There are a lot of solar products out there and they all meet someones need. My concept of what solar should be has changed many times over the years. I appreciate your challanging my current thoughts. There is a solar product I think will change the developed world in the next ten years, like adding extra insulation to the attic did.  There are millions of electric water heaters in use.  If all their lower heating elements were connected to 300-400W of solar panels through a MPPT controller, dramatic electric savings would result.  Everyone has some place they could mount a couple solar panels.  The cost would be farless than the $1800 many spend for a heart pump water heater.  It would dramatically improve the performance of those too.

Sony backs this type of LiFePO4 with 20 years warranty in RE condition 100% DOD every day over 7000 cycles with 100% DOD this is a totally different type of battery than what was traditionally use by small DIY offgrid people in the past (actually still present).
You can make MPPT to last 25 years (no electrolytic) but it will be way more expensive and at the current price of PV it makes absolute no economic sense.
Those 7 days where a tipical 7 days at the end of January beginning of February with the 3x 240W PV panels the worst day in December will still produce 0.3kWh (and that is significant part of my 0.8kWh/day in low power mode)
I will actually heat my house with PV in the near future since is the most cost effective heating. But it requites a specially designed house  (I designed and build my own house) that has a large thermal mass storage incorporated in to the design.
I will also need to design and build a digital MPPT (this is what I call it but is totally different no DC-DC converter much more simple) this is really needed for heating to maximize power else loss is in excess of 40 to 50% but this only works for heating not battery charging.
I have a video on youtube talking a bit about this digital MPPT will be open source so freely available and maybe my next Kickstarter if people understand this and need it.
I will have just PV panels and resistive heat elements embedded in the concrete floor that acts as thermal storage (100kWh at 10C delta) There will be 20 to 30 resistive loops and more or less of them will be connected to keep the power from the PV array at the max power point that is a short definition of what I call a digital MPPT. 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 2806
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #14 on: February 23, 2015, 08:44:59 AM »

Thanks for the comment. Sorry if I was looking a bit less polite in the last comment.
There is not much knowledge in this field of OffGrid energy storage and many things have changed fast in the last few years.

No worry. It's all good. I hope you succeed in building a better mousetrap.

BTW - what will the price be for your new CC?
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #15 on: February 23, 2015, 08:51:52 AM »

No worry. It's all good. I hope you succeed in building a better mousetrap.

BTW - what will the price be for your new CC?

Thanks, There is a table with prices on my website see signature and there are a few variants from SBMS25 to SBMS100. There will be some discount for the early bids on Kickstarter.

Online Seekonk

  • Super Contributor
  • ***
  • Posts: 1578
  • Country: us
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #16 on: February 23, 2015, 09:30:42 AM »
There are markets for everything.  I think the real growth will be for diversion controllers that can utilize excess panel energy in a smart way and are able to be directed by a central control. 
 

Offline HackedFridgeMagnet

  • Super Contributor
  • ***
  • Posts: 1853
  • Country: au
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #17 on: February 23, 2015, 10:16:51 AM »
I like what you have done electodacus and good luck with your next project. I think you have found a nice little niche.
Really nice to see some backers from kick starter actually getting what was proposed.
I might start using LiFePO4 in smaller remote monitoring/control type jobs. (If I ever get another)

Just a few queries.
I know that for you weight and space are a premium. But for me I would prefer having the common power connections coming into the controller too.
Is 24 bit necessary? wouldn't 16 bit delta-sigma be more suitable? I would imagine there is a fair bit noise on your power lines.
Have you decided what processor you are using? Will it have floating point capability?

I think you make a reasonable argument against mppt, but when I look at your graphs it seems you get a lot of periods even in the middle of the day where you seem to get absolute zero solar energy produced. Why is this so? I thought it would produce about 10-20% even if cloudy. To me it seems that energy on cloudy days is most important of all. Possibly this would be the advantage of an mppt device.
 
The following users thanked this post: fourtytwo42

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #18 on: February 23, 2015, 10:19:19 AM »
There are markets for everything.  I think the real growth will be for diversion controllers that can utilize excess panel energy in a smart way and are able to be directed by a central control.

Yes this will be my next project an addon to the new Solar BMS that will be able to heat my house it will do a sort of what I call (digital MPPT) is a new concept I thing nothing to do with DC-DC converters used in battery charging.

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #19 on: February 23, 2015, 10:27:53 AM »
I like what you have done electodacus and good luck with your next project. I think you have found a nice little niche.
Really nice to see some backers from kick starter actually getting what was proposed.
I might start using LiFePO4 in smaller remote monitoring/control type jobs. (If I ever get another)

Just a few queries.
I know that for you weight and space are a premium. But for me I would prefer having the common power connections coming into the controller too.
Is 24 bit necessary? wouldn't 16 bit delta-sigma be more suitable? I would imagine there is a fair bit noise on your power lines.
Have you decided what processor you are using? Will it have floating point capability?

I think you make a reasonable argument against mppt, but when I look at your graphs it seems you get a lot of periods even in the middle of the day where you seem to get absolute zero solar energy produced. Why is this so? I thought it would produce about 10-20% even if cloudy. To me it seems that energy on cloudy days is most important of all. Possibly this would be the advantage of an mppt device.
Thanks.
Weight is important for the SBMS do to expensive shipping and why will I have something heavier than it needs to be? :)
24bit is probably not necessary for larger systems but the integrated 14bit ADC (rounded to 12bit) in the ISL94203 that I used in the old version has anti alias problems and even at 100mA resolution results with small loads are not great and with noisy loads as inverter things where even worse I needed a strong SW filter since HW filter will interfere with Short-circuit and overcurrent protection.
I will need to switch ranges with a 16bit and it is not less costly so I preferred the 24bit. I may use that 24bit for some other applications like a 6.5 digit multimeter.
I will be using the same STM32F072 I used in the last SBMS more than capable to handle all the SBMS functions.
Is an OffGrid system so when battery is full there is no place for solar to go you just have excess unused power.

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #20 on: February 24, 2015, 11:04:27 AM »
The Solar BMS is now on Kickstarter here is a link
http://electrodacus.com/k

Offline eneuro

  • Super Contributor
  • ***
  • Posts: 1411
  • Country: 00
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #21 on: March 01, 2015, 07:57:19 PM »
Those connectors on top of control panel doesn't look great and its mounting high power wire just by inserting it into such connector with screw driver can lead to erosion as well as such connections are exposed to vibrations.


Maybe it's easier and cheaper ship products at lower weight and size, but automotive connectors seams to be much better for high power applications if well terminated, so sometimes it is better save shipment costs and DIY something yourself which will last for years and easy to repair.


There is huge advantage of using such copper wire terminators, while everybody who has a hammer skils can make it at the cost of copper pipe, cut it into small pieces, press in the vise, make hole and we have nice wire terminator capable to survive many amps  8)
« Last Edit: March 01, 2015, 08:13:08 PM by eneuro »
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
“Let the future tell the truth, and evaluate each one according to his work and accomplishments. The present is theirs; the future, for which I have really worked, is mine”  - Nikola Tesla
-||-|-
 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #22 on: March 02, 2015, 04:21:19 AM »
Grats on the kickstart ! im sure it will hit target soon.

(i might have spotted an error "b) A 4 layer main PCB do to increase in complexity over 2.5x more parts that on the SBMS4080." ... do to increase? im not sure how to rephrase it :P )

Thanks for the comment. What I want to say with that phrase is that the new SBMS has 2.5x more components on that main board and will be almost impossible to route all that on a two layer PCB as I had non the SBMS4080.
If someone can find a better way to express that let me know. English is not my strong point :) 

Offline electrodacus

  • Regular Contributor
  • *
  • Posts: 229
  • Country: ca
    • electrodacus
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #23 on: March 02, 2015, 04:36:39 AM »
Those connectors on top of control panel doesn't look great and its mounting high power wire just by inserting it into such connector with screw driver can lead to erosion as well as such connections are exposed to vibrations.


Maybe it's easier and cheaper ship products at lower weight and size, but automotive connectors seams to be much better for high power applications if well terminated, so sometimes it is better save shipment costs and DIY something yourself which will last for years and easy to repair.


There is huge advantage of using such copper wire terminators, while everybody who has a hammer skils can make it at the cost of copper pipe, cut it into small pieces, press in the vise, make hole and we have nice wire terminator capable to survive many amps  8)

Thanks for the comment. Are you referring at the SBMS4080 or at the new SBMS100 that is now on Kickstarter. They have different connectors but both use screws to keep the wires thigh. I do not think vibrations will get those connection louse. Unless is installed on a tractor without any suspension but even then methods to secure that connection with the connectors I use exist.
I do want to keep the SBMS compact and light and the connectors need to be SMD type to work with the metal core PCB. The ones on the new SBMS where the only ones that got my requirements and where able to take #2 AWG (35mm^2) copper wire.
If you have a better suggestion for connectors with this restrictions I will like to see them. There is still time to improve the new SBMS if that is the case. Same for any other part of the SBMS like better power mosfets or ...
It is open source and I like to take the final decision but I love to here suggestions. The dual PV input was suggested by a few of the first bakers nd since was not that hard to implement I did that. It seems that out of those 4 models (two with dual input and two without)  >90% of the current bakers went for the two dual input models. 

Offline eneuro

  • Super Contributor
  • ***
  • Posts: 1411
  • Country: 00
Re: Solar BMS a solar charge controller for LiFePO4
« Reply #24 on: March 02, 2015, 06:55:58 AM »
The ones on the new SBMS where the only ones that got my requirements and where able to take #2 AWG (35mm^2) copper wire.
If we assume solid coper cross section than 35mm2 leads to close to 7mm diameter which is quite eaasy to fit inside copper pipe I've shown above, but it looks quite dificult to screw, but probably you have some idea how to do it, so it could be interesting to see how this wire fits into your upgraded new connectors  ;)

In spot welder I've used 2 x 35mm2 in parallel into 15mm diameter copper pipe to help disipate welding heat at bulky transformer secondary in spot welder, so in many my projects those copper pipe based wire terminators works for me and are easy to make, but for commercial product maybe something different will be better, so its up to you to choose what better fits your requirements if you want to have happy customers.
12oV4dWZCAia7vXBzQzBF9wAt1U3JWZkpk
“Let the future tell the truth, and evaluate each one according to his work and accomplishments. The present is theirs; the future, for which I have really worked, is mine”  - Nikola Tesla
-||-|-
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf