Off Grid Solar

[nwdiver]

You don't know what you talking about. BMS absolutely has to send cell voltages to the host controller. What happens if few cells crap out and now instead of 74 cells in parallel you got 72 cells and now your 230Ah cell group lost 6Ah of capacity? Not monitoring cell voltages at a host level would be crazy.

That might be true... I can see that to keep module level voltage in tolerance but if you're working with a 24v system then you're already regulating voltage on the module level. So for a full pack at ~400v you have a point.... for a single module it's irrelevant.

I haven't read up enough on how the modules communicate with the pack... but I know each module has it's own cell monitoring / balancing. Either way... he's only using 1 module so that 1 module is now the 'pack'...

 

[AntronX]

It does not matter if one module or 100. Each module is it's own power unit and has 6 cells in series. You need to know what those cells are doing to ensure they do not get taken outside their safe voltage range. Right now you are assuming BMS will keep all 6 cells in balance always and there is no way cells can get out of balance. They can, suddenly and unpredictably. Consider this scenario. You top balance the 6 cell groups to 4.2V. All groups are exactly 230Ah each. That means that they all should reach the same voltage at the end of discharge cycle at lets say 5% SOC. Not you repeat this, but with one of cell groups missing 2 - 3 cells due to broken cell bonding wires. Now the good cell group will get down to ~3.1V while that group with missing cells will overshoot down below 2.8V below 0% SOC. Using these modules in unattended automatic charging setup that way is irresponsible and is asking for fireworks.

Also, what happens if a MOSFET in the charge controller fails on and passes PV voltage directly to the batteries? His PV voltage is way above 25.2V limit of the modules. Once the modules get fully charged, they will overcharge and all 6 of them will burn up at once in a very impressive fireball. There has to be separate cutoff relay triggered by cell voltage monitor to make this setup fail safe.

 

[MP3Mike]

@AntronX I think part of the problem is that you keep saying cell(s), but I think you mean string(s) and/or brick(s).

So how does Tesla deal with this since they don't monitor individual cells? (Which you can't do when you have 74 cells hooked up in parallel.) And they have no way to control charging at the individual cell/brick/string/module level. The only thing they can do is detect something is wrong and disconnect the entire pack. (Still leaving all of the cells/bricks/strings/modules hooked up an powered internally.)

 

[S4WRXTTCS]

For me it comes down to not really trusting the grid. While I haven't experienced the black outs that other parts of the country have I think it's only a matter of time.

What I want is two things

20-30 miles per day of Solar charge for my Tesla
Battery backup for critical loads.

If I can get both for under $15-20K I might go for it. If I can get one or the other then maybe.

The tough part is I live in the pacific northwest where we have lots of darkness, but electricity is cheap. I don't really care about the electricity cost though. What I care about is just some independence from the grid if that makes any sense. I also don't want anything obvious so I love the Tesla roof tiles.

 

[AntronX]

So how does Tesla deal with this since they don't monitor individual cells?

They don't need to monitor each of 74 cells in a parallel group since they are guaranteed to all be at the same voltage due to them being connected in parallel via fuse bond wires to the same bus bar. They can only monitor the voltage of that parallel group and compare it to other parallel groups at different state of charge. If you remove one cell, for example, from 74 cell group, that group will have slightly lower voltage than the rest at lower state of charge. That difference can be measured by that BMS circuit board and reported to the host for analysis. But because of that voltage monitoring, the whole pack remains safe as the host ensures that the charge or discharge process will be cut before that now weaker "73 cell" group gets over charged or discharged. There is no need to disconnect whole modules from each other.

 

[nwdiver]

I don't think I posted this on this thread... here's an awesome racking idea that a carpenter we found at the Oceti Sakowin Camp had. They built this in ~2 hours. He simply drilled 3 holes in the side of each frame and fastened it with a word screw. Yeah... it's not grounded... this was a 'battlefield install'...

The array is 12 95w Shell panels (I know... right?) I ran 3 strings of 4 to a phocos 24v MPPT controller. Should be able to light 3 or 4 tents and keep all their phone, radio and flashlight batteries charged. I'm hoping it'll generate at least 1.5kWh/day in the winter... They also have 2 300w panels that should add another ~1kWh/day.

I'm the handsome guy crouching...

 

[Sodamo]

Interesting thread.

Just a thought on cold temp concerns, what about an insulated box with low wattage heat source (lightbulb?) with a thermostat?

 

[glhs272]

Just thought I would give an update. A couple weeks ago it started getting too cold to continue operating the Lithium pack. So I am currently running a much smaller capacity Lead-Acid based battery. Before switching to lead, I was able to run several charge/discharge cycles on the Tesla modules. Everything worked great. I started off with just one module, when that worked well I added a second module in parallel. Then a third. I did keep an eye on cell voltages. Throughout the many cycles, no issues with cells getting out of balance.

AntronX, I will just say generally speaking you are right in that there should be a BMS on lithuim-ion packs for a long term safe robust product. That is why Tesla and every other OEM manufacturer of lithium packs puts one on their product. No argument there.

However, I will not be using a BMS. At least not initially. I don't want to start an argument or ridiculous BMS vs. No BMS back and fourth on this thread. For now, I am the BMS. I keep an eye on temperatures and cell voltages using a cell log device. I have removed the Tesla BMB boards from the modules and replaced them with a cell tap connector to make it easy to plug in my cell log. I am not the only one to try this. I am using the Micheal Bream (EV West) philosophy of no BMS for Tesla packs. He is putting Tesla modules through a much tougher EV application without any issues so far. Just needs a routine (lets say annually for a low usage EVs) maintenance check on the pack to make sure no cells are getting too out of wack. The goal is to not push the pack to the extremes (ie no 100%SOC charging), so going without a BMS is not an issue. WK057 also didn't utilize BMS, at least not initially (he did or is working on his own custom BMS, but not sure if he finished it) on his huge off grid solar setup using Tesla modules. Keep in mind that these are not packs made up of a motley collection of used laptop cells with various capacities and internal resistances. These are Tesla modules, precisely matched groups of cells, all from the same car. The "creme de la creme" if you will of large 18650 packs. Also, in many instances a DIY BMS is known to damage packs more often than help them. So I will never be putting a BMS (Battery Management System) on my solar pack. However, I may build or utilize a Battery Monitoring System. This would simply be a passive cell voltage monitoring system to interrupt charging if any cell voltage goes beyond a predetermined threshold. No active battery balancing. I will be doing all balancing manually as necessary.

More about the Tesla modules. I removed the BMB boards on the Tesla modules because they are useless to me. For one, I don't have the rest of the BMS system needed in order to reverse engineer and utilize the BMB boards. If someone else can or does this, great. I used the cell tap wires to make up a connector to easily plug in or remove the cell log. Works good. Secondly, even if I had the ability to talk to and command the BMB boards that came with the modules, they would still be useless to me. This is because of the specific way I am utilizing these modules. I have to run multiple modules in parallel. This is necessary to have the capacity I need (30kwh) at the voltage I need (24 volts). If I were running all the modules in series, such as in an EV pack operating at higher voltages, then I might be able to utilize them.

Why does it matter when you connect Tesla modules in parallel? What prevents the BMB boards from working in this situation? Well, I might be bad at explaining it but I will try. Jason (AKA WK057) ran into this same issue with his setup, so you might look at his explanation.

First off, the BMB boards: The BMB boards not only monitor the cell voltages and temperatures, they also have bleed resistors connected to FETs (electronic switches) that turn on to slowly bleed current from specific groups of cells that it determines to have higher voltages than the rest of the cells in series. The computer calculates how long to run these bleed resistors on specific cell groups such that when the battery fully charges, the individual cell group doesn't exceed 4.2 volts sooner than the rest of the pack, such that all cells reach 4.2 volts simultaneously (or as close to it as possible) when the pack is fully charged (balanced pack). If any cell reaches 4.2 volts sooner than the rest, it has to cut off charging early (unbalanced pack) to prevent damaging the cells. So in short the BMB boards are for monitoring and active balancing the module. This works fine when modules are connected in series.

This setup fails when you have modules in parallel. When the BMB turns on the bleed resistors, if there is another module connected to this module, as the BMB bleeds down the cell group, the overall module voltage goes down. This would cause current from the other module connected in parallel to flow to the lower voltage module. Thus raising up voltage of other cell groups and lower voltages in of the cell groups in the other module. This could actually cause further imbalance. I know this is a poor explanation and it would be easier to explain with graphics and such, but don't have time at the moment.

One way around this issue would be to add high current cell taps to the modules, and actually connect all of these cell groups to each other, thus locking them together. In effect creating one large 24 volt module. So instead of a 74 cells in parallel per group, it would be 444 cells in parallel (6 modules x 74 cells per group). Thus in effect creating a 6s444p pack.

As far as creating a Battery Monitor System, I was thinking maybe just using a simple cell log and tying the alarm output (alarm goes on when a cell exceeds 4.2 volts) to a relay that trips the solar panel input GFCI breaker, thus interrupting charging. You would have to manually reset it before you could resume charging, which for me would cause me to investigate why it tripped. I would see that the pack is getting out of balance and then manually re-balance the pack. Not sure yet. I am looking into how I want to finalize my pack design over the winter. Looking for good feedback, not just negative feedback.

 

[AntronX]

Sounds like you got it. Just setup automatic charge/discharge termination based on voltage of each 72cell block in each module, even paralleled ones. I get what you mean by balancing issues with modules connected in parallel. That issue could be remedied by activating cell bleed resistors of all other modules at same time to prevent current flow into unbalanced module during it's balancing. But that would require custom software in BMB boards and the host. You can go without balancing as long as you have auto charge/discharge termination. So have fun, but try to thermally isolate each module so that one catching fire will not propagate to all of them.

 

[glhs272]

Taking a look at the weather, I see above freezing temps for the foreseeable future (10 days). I am going to disconnect the lead acid batteries and reconnect the Tesla battery modules.

About those lead acid batteries. As I was expecting, they are too low of capacity to be of much practical use for directly charging the car. I get about 3-4 miles of range out of them, directly charging the car. The other issue is the reduced capacity of the array during winter. This winter I had several weeks of complete shading due to snow. So not a whole lot of energy available from mid December into mid February. We did get some warm days and eventually melted the snow from the panels. But I don't see much value in trying to up-size my lead acid battery bank at this time. Although on a sunny day, even in the winter, I can charge the 12 volt bank from dead empty to full charge cut off by noon.

On the other hand I am exploring using this arrangement during the winter to provide constant 12 volt supply/charging for the Tesla. This would eliminate vampire drain at least, which would be like adding 4 miles of range a day. Another side benefit of that would be is it will extend the life of the 12 volt battery on the car, as it would reduce the number cycles it is required to perform daily. I would use a 24 to 12VDC DC-DC converter regulate the voltage going into the car and limit the maximum drain from the battery bank. I would then connect the 12 volt side to a 12 volt plug adapter to plug into the 12 volt supply port inside the car or add an external one to the front of the car. Similar arrangement that @Ingineer uses to store his Tesla's only this would be completely solar powered. I think there is enough power produced by this setup to work.

 

[zag2me]

Wow that's a fair bit of snow!

Thinking about this a little though, my daily commute is only 4 miles so this is something that's quite interesting to me.

I've been looking myself at simple plugin solar without any storage or grid connectivity. There are some really cheap deals in the UK coming up for large solar systems that you can install yourself. I'd be interested in any thoughts.

2kW 2000W Plug-In DIY Solar Panel PV Kit SystemTile/Slate Roof Mount House | eBay

I also found a huge battery storage solutionf or the home here:

NEW 2.4Kw 2400w Lithium Ion Battery Storage System with AC Inverter - Off Grid | eBay

No idea how good it is though.

 

[glhs272]

With a 4 mile commute, you wouldn't need too much solar to make that work. The solar kit you are referencing is a grid-tied version, as is the storage system (AC Coupled). Which is fine if that's what your going for. In my case I went off-grid for a number of reasons, which I won't get into. The grid tied systems mounted on your house need to be designed, setup and installed more carefully than a simple solar shed system would.

I tried to keep my off grid system as simple as possible. Just (6) 275 watt solar panels connected to a battery charge controller, which is connected to 4 lead acid deep cycle batteries (or Tesla Model S battery modules in the summer). Then a simple 24volt pure sine wave inverter (important that it's pure sinewave for car charging) connected to those batteries. Connect your EVSE to your inverter to charge your car. Simple as that.

For me, 4 miles is not enough. That limit is mostly to do with the lead acid batteries, as there is more solar capacity than the lead acid batteries are capable of storing. Simply using the Tesla modules to increase storage will get me closer to 8 miles per day. From there, I will need more panels.

If you have a few quid burning a hole in your pocket, the interest to do it, then go for it. I have fun tinkering with this kind of stuff. Once it's there, the energy is free.

 

[mspohr]

Taking a look at the weather, I see above freezing temps for the foreseeable future (10 days). I am going to disconnect the lead acid batteries and reconnect the Tesla battery modules.

About those lead acid batteries. As I was expecting, they are too low of capacity to be of much practical use for directly charging the car. I get about 3-4 miles of range out of them, directly charging the car. The other issue is the reduced capacity of the array during winter. This winter I had several weeks of complete shading due to snow. So not a whole lot of energy available from mid December into mid February. We did get some warm days and eventually melted the snow from the panels. But I don't see much value in trying to up-size my lead acid battery bank at this time. Although on a sunny day, even in the winter, I can charge the 12 volt bank from dead empty to full charge cut off by noon.

On the other hand I am exploring using this arrangement during the winter to provide constant 12 volt supply/charging for the Tesla. This would eliminate vampire drain at least, which would be like adding 4 miles of range a day. Another side benefit of that would be is it will extend the life of the 12 volt battery on the car, as it would reduce the number cycles it is required to perform daily. I would use a 24 to 12VDC DC-DC converter regulate the voltage going into the car and limit the maximum drain from the battery bank. I would then connect the 12 volt side to a 12 volt plug adapter to plug into the 12 volt supply port inside the car or add an external one to the front of the car. Similar arrangement that @Ingineer uses to store his Tesla's only this would be completely solar powered. I think there is enough power produced by this setup to work.

Here's the reason I can't rely on winter solar production. There are solar panels under 5 feet of snow on this roof.

And these pole mounted panels usually take a few days to shed (the panels are about 10 feet above ground level)

I don't get much winter production but then I don't expect to get much. I do run a healthy surplus from Spring through Fall.

 

[glhs272]

Here's the reason I can't rely on winter solar production. There are solar panels under 5 feet of snow on this roof.
View attachment 215956

And these pole mounted panels usually take a few days to shed (the panels are about 10 feet above ground level)

View attachment 215957

I don't get much winter production but then I don't expect to get much. I do run a healthy surplus from Spring through Fall.

Wow, yeah lots of snow there.

In my case, I have a large metal building (garage) with a nicely exposed south facing wall. I notice that it gets lots of sun and the snow never accumulates. If we get a big wind driven snow and it attaches to the wall, it won't take long to shed off. Also, it faces my back yard which when covered with snow adds to the additional sun via reflection from the snow. So, I was thinking of having a winter specific solar array made up of vertically mounted panels. Obviously not efficient compared to properly angled panels in the summer, but in the winter this might be the only way to get reliable sun on these panels. So this is something I might experiment with.

In your case, you would have to mount panels very high up (above the max snow line). Perhaps you could do that to your pole mounted ones, however if they are already 10 feet high... geez it would hard to set them up higher. Maybe you could mount to the side of your cabin, if the aesthetics didn't bother you. Beautiful property btw.

 

[mspohr]

Wow, yeah lots of snow there.

In my case, I have a large metal building (garage) with a nicely exposed south facing wall. I notice that it gets lots of sun and the snow never accumulates. If we get a big wind driven snow and it attaches to the wall, it won't take long to shed off. Also, it faces my back yard which when covered with snow adds to the additional sun via reflection from the snow. So, I was thinking of having a winter specific solar array made up of vertically mounted panels. Obviously not efficient compared to properly angled panels in the summer, but in the winter this might be the only way to get reliable sun on these panels. So this is something I might experiment with.

In your case, you would have to mount panels very high up (above the max snow line). Perhaps you could do that to your pole mounted ones, however if they are already 10 feet high... geez it would hard to set them up higher. Maybe you could mount to the side of your cabin, if the aesthetics didn't bother you. Beautiful property btw.

I was thinking of adding some panels to the front of the garage which faces South to get better winter production.
Have to get permission from the chief planner, though

 

[dgpcolorado]

One reason I like my pole mount panels is that it is easy to pull the snow off. I also change the pitch with the seasons and around the winter solstice I have them at 55º, which makes pulling the snow off easier (they are currently at 45º and I will adjust them to 35º in another week or so as the spring equinox approaches). I use a window squeegee on a long 8-16 foot painter's pole to pull the snow off.

However, I don't have to deal with five feet of heavy "Sierra Cement" in this near desert climate.

^ Panels at 55º in early morning in January, just before I pulled the snow off.

 

[Ampster]

Have to get permission from the chief planner, though

You are in California and the Solar Rights Act means the chief planner cannot turn you down.

 

[mspohr]

You are in California and the Solar Rights Act means the chief planner cannot turn you down.

Yes, that's true. However I was referring to my wife who has final veto on everything.
Interesting that the Tahoe Regional Planning Agency objected to my pole mount panels because of the scenic corridor. I was able to move them to reduce their concerns but I was thinking of playing the Solar Rights card.