Solar PV battery options

SD_DRACULA

Interesting, I was looking at the Seplos site, do the kits come without any cases, boards only?

Also can they be hooked up to the Solis so that you get the battery values from the Solis app and also connect them to an rs485 interface at the same time? Or is it one or the other?

championc

Indeed, it seems odd that they are not in a case. The CANBUS and RS485 connectors are on one side while the heatsink and other stuff is on the other.

But yes, the CANBUS and RS485 can be used simultaneously.

I have a HEYO BMS currently, but of course, the Seplos integration with a Solis or Sofar makes a perfect overall solution

graememk

CAN can have numerous devices on it too, and you can have passive devices just listening on it.

(Can you tell that I've spent the weekend working on it? I can now remotely, wirelessly monitor & log all the info from my BMS) using an esp32 and a can interface.

I've also got docs for the SMA (growatt, byd, LG all use this protocol) and Pylontech, use a slightly different one

Thinking of firing up a Home Assistant instance.

SD_DRACULA

https://www.boards.ie/discussion/comment/117740080#Comment_117740080

You can connect an rs485 adapter directly to your Home Assistant server and read from the Solis inverter via modbus integration BUT you lose the Solis stick logging - they trip over each other and then you get crazy readings in your Solis cloud data (been there)

They implemented this in a stupid way it seems, inverter(s) are slaves and the wifi logger is the master so if you daisy chain another rs485 usb adapter to it you get two masters and that's a no-no.

@reklamos found a way around it, a relay on the power pin from the wifi stick and turn it off, read from HA, turn it on but too much hassle.

You can also purchase specialized equipment that allows multiple master on the network but was like over 300 euro, so not worth it.

I just got them to refresh my data to SolisCloud every 1 min instead of 5 and it's close enough for my needs.

graememk

Cutting the power to the logger is a smart idea. So you don't get crashes.

Thankfully don't need to talk to my inverter, I'm measuring everything directly though ct's.

championc

https://www.boards.ie/discussion/comment/117737593#Comment_117737593

Any more takers for orders for Seplos BMS's ? I plan to place the order this weekend

reklamos

https://www.boards.ie/discussion/comment/117741479#Comment_117741479

It has been working like this since June very simple design just esp8266 with relay on top of it. I do not use much Solis page/app for monitoring as I get more data directly via rs485. The beauty of rs485 is that allows me to make changes on inverter remotely and automate it.

SD_DRACULA

https://www.boards.ie/discussion/comment/117746327#Comment_117746327

Curious about that, can you get it to stop discharging the battery if for example want to charge EV directly from the grid without wasting the battery?

corglass

https://www.boards.ie/discussion/comment/117744864#Comment_117744864

For a diy 48v lipo4 battery? I'm very interested. Pm me please

reklamos

https://www.boards.ie/discussion/comment/117746620#Comment_117746620

Yes, you can. The way of doing that is to set Time Charge in Solis to charge the battery at the time when EV is being charged. The crucial part is to set the current to 0(I checked mine and it does allow to set it to 0). While this in place, all the power will come from grind and not the battery. Once EV is charged you can disable Time Charge and this will allow battery to be charged/discharged as normal.

garo

https://www.boards.ie/discussion/comment/117744864#Comment_117744864

Sent you a PM. The new boards is wonky so let me know if you saw it already.

championc

https://www.boards.ie/discussion/comment/117747107#Comment_117747107

Saw nothing so I have sent you one.

championc

https://www.boards.ie/discussion/comment/117746638#Comment_117746638

PM sent

graememk

https://www.boards.ie/discussion/comment/117746731#Comment_117746731

If/when peak evening rates come into action, you can set it up to restrict discharge/keep a minimum % for peak evening.

reklamos

https://www.boards.ie/discussion/comment/117748269#Comment_117748269

I have smart meter for year but so far I have not seen a smart plan that would make financial sense. Hopefully will get something sensible when time comes to renew.

bullit_dodger

Wow after many years lurking, this is my 1st time posting on Boards.ie. Look at me - I’m a big boy now! LOL!

I’d like to give a shout out to all the regulars posting here. I’ve learnt a TON of information about this space from you guys. I’m a software engineer myself, so somewhat technical minded, but the electrical space wouldn’t have been my forte. To namecheck a few, thanks to DrPhil, unkel, championc, Slave1. Forgive me if I don’t list everyone. Very informative and (mostly) well thought out posts folks. Love the willingness to help out fellow forum members!  The occasional rant, but sure we all do that, incl me. haah

I recently ordered a 4.5Kw system, 8.4Kwhr battery, 5Kw inverter along with eddi for the hot water. €8200 for the lot, after grant taken off. Yeah, I might have gotten it for slightly cheaper elsewhere, but I got a good feeling from the guys out of Sligo so happy enough to go with these guys. We’ll see in a week or two how it turns out.

I was interested in the fiscal viability/return on the battery. What the analysis threw up wouldn’t have really changed my mind one way or the other, such as, if I never would recoup my “investment”, but I wanted to know. So, the problem when I started to think about it, is that it’s actually kind of hard to model the usage/savings as there’s a lot of variables in play. What would you do if you didn’t have a battery, load moving to night rate etc but after doing the work, I thought others may like to see the numbers I have. Maybe you find them useful, maybe not. Everyone’s parameters will be different, but this may serve as a useful (rough) guid.

First some background numbers. I’m with Airtricity so I used their figures. 1YR-ELEC-10.pdf (sseairtricity.com) (21.07 cent day / 10.43 cent night) – Aside: wow didn’t think they’d gone up so much! Note to self to look at other providers!

The “base load” for my place is about 500-600 watts. Lots of computers, smart plugs, automated lights etc.

The way I approached this is that I have two models.

• Model 1 is where I ignore the PV side of things. I use the battery as just a way to charge the thing fully at night rate and then discharge during the day (365 days a year)
• Model 2 is where I do that for 4 months of the year. Basically 120 days, where I can’t charge the battery due to insufficient solar and then for the other 200+ days I utilize the battery to generate savings.

In reality for model 2, it’s no quite so binary. There may be days where I can’t fill the battery in summer, likewise there maybe days in winter that I could fill it but it’s too hard to model that, but I simplified it for the sake of getting “some” number.

Next post will have “model 1”…

bullit_dodger

Model 1 – the easy one to compute!

  Battery Capacity (Kwhr)             8.4 Kw

  Battery % usable                       85% (I’ve a Givenery 8.2Kwhr ordered, it states 100% depth of discharge but let’s go with 85% usable)

  Usable capacity (Kwhr)              7.14 Kwhr

  Day cent/unit                           €0.211

  Night rate cent/unit                  €0.104

  Night rate (with losses)             €0.12    (Upped the rate to €0.12 due to losses putting into/taking out of the battery, I assumed 15% losses hence €0.12 cents)

  

Savings per day in using battery night rate during day  = €0.65/day       (day rate – night rate) x capacity

Number of days                                                         = 365

Saving per year                                                         = €236.37

Battery cost (roughly)   = €3200

Break even point          = 13.5 years.

I’m sort of taking Model 1 as the “lower end” in terms of ROI. 

bullit_dodger

Model 2 – Using PV to fill the battery

  Battery Capacity (Kwhr)             8.4 Kw

  Battery % usable                         85% (I’ve a Givenery 8.2Kwhr ordered, it states 100% depth of discharge but let’s go with 85% usable)

  Usable capacity (Kwhr)             7.14 Kwhr

  Day cent/unit                               €0.211

  Night rate cent/unit                   €0.104

  Average rate                                €0.16    (As opposed to model 1, I’ll be using my battery I filled via solar in the evening (day rate) and also during the night instead of importing from grid – hence an average rate)

  

  Savings per day in using battery    = €1.07/day       “average rate” x capacity

  Number of days                            = 245

  Saving per year                            = €262.40

  Savings per day in using battery night rate during day  = €0.65/day       (day rate – night rate) x capacity

  Number of days                            = 120

  Saving per year                            = €77.71

  Total savings for the year              = €262.40 + €77.71 = €340.11

  Battery cost (roughly)                  = €3200

  Break even point                         = 9.4 years.

This is probably closer to the mark. Before people say it, I know there’s lots of over simplifications here, such as battery degradation over time. Vacation time where no one is using anything in the house, forthcoming FIT benefits (if any)  etc etc etc. I get it, but electricity prices are at best staying flat, but most likely going up, so I’m sort of just generally speaking here.

If you go the route of DIY batteries such as unkel, DrPhil etc have so excellently done, your payback time is probably 6-7 years. Overall, my own case aside as I wanted one anyway, but I’d slightly come down on recommending “for” rather than “against” a battery.

Only works though if you have a decent base load of 500w or more to exhaust the storage overnight/during the day. If you have a load of 200-300 watts it starts to become more uneconomical 

SD_DRACULA

I'd say that the battery is actually 10kwh and 8.5kwh is the usable 85% hence the fake 100% discharge i.e. it only works 85% of the time, all the time 😁

bullit_dodger

Assuming that was the case Dracula, that the full 8.4Kwhr was available to take advantage of, updating that in my spreadsheet yields

Model 1 - Savings per year = €278 (11.5 years)

Model 2 - Savings per year = €415 (7.6 years)

So if someone asked me if they should get one then I'd be more inclined to recommend "yes" over "no". Naturally you'd have to match the capacity to the load and not be stupid getting a large battery when you don't have the usage....and again.....load of simplifications there in the modelling, but generally like the numbers. I thought it would be worse to be honest. For me, it's sort of I have the money now and I want to future proof myself a little, hence why I was doing it even though it was potentially 50-50 if there was an ROI on it.

Course the real winners are the DIY folks building their own banks from individual cells. Payback in the 4-6 year range I would think depending on how much they paid for the BMS, cells etc. I might go there in a year or two, but I think 8-10Kwhr is probably the upper limit on what most normal households would do.

Hmmm, maybe I should dust off the old cryptocurrency rigs - LOL !

SD_DRACULA

What you can't account for is the garbage months when it's supposed to be good (i.e. this August) where you can't even top up the battery on many days so those numbers are highly optimistic I'd say.

I have a 7kw South system with a 5kwh battery for comparison.

bullit_dodger

yeah, very true - although I did mention that there would be days in the summer where you wouldn't be able to charge the battery, but similarly, there would be the odd day in the winter where you might get it charged or partially. Not to mention the fact that the battery isn't either empty/full, but can get to like 60% full. But I took the decision for simplicity that one would cancel the other....mostly. This was also the reason why I decided that I'd model 2 different scenarios, "model 1" where I ignore the PV generation entirely.

The logic there being that one would assume that if you do introduce PV generation in the mix, then you should be better than that "model 1" figure. So we have a range of 7.6 years -> 11.5 Years. There are also a whole host of simplifications there that I haven't mentioned, different rates from different suppliers, different utilization of summer/winter (more lights for longer in winter, hot water has more work to do etc), but I think 8-9 years is probably an realistic figure. (for me)

Before I did the math my intuition was about 20 years, which is probably outside of the useful life span of most battery packs. I still wanted a battery in the mix for other reasons than simply ROI, but I'm pleasantly surprised that it's viable as it is.

SD_DRACULA

It could also be way way quicker than 7.6 years with the way the clowns are bumping up the prices...

bullit_dodger

100% agreed, rates are mental and not going to improve at least in the short term. The problem is that just so hard to model everything, so many parameters. That's why I tried at least to explain some "shortcuts" I made. Necessary evil I'm afraid.

I was kinda shocked when I saw 21c/day rate Dracula! I've not really paid much attention to it over the last couple of years. Solar was always one of those things that "Ohh yeah, I must get around to doing that at some stage". Educating myself in the space, I was watching a youtube video there recently with a guy taking about his experiences for 5 years with solar in San Diego, and he mentioned peak rate of 53 cent (USD of course, but yeousir) 53 cents!!? if that wasn't an incentive to get solar and a battery, dunno what would be?!

Slightly off topic of this thread (batteries), but has anyone gotta some low capacity (200-400Watt) wind online? Probably not a runner here for me in south dublin city, but if I didn't feel like a spanner with a windmill on my roof, I'd love a small little guy there giving me a trickle charge whenever the wind is blowing. (yeah, I know they are AC and introduce some complexities)

garo

You won't be filling your battery 245 days of the year from solar. It will be closer to 150. You have a 4.5kWh system. With my 4.8kWh system in Dublin for the past week I haven't generated over 15kWh on anyday. Most days have been closer to 10. So if you generate 15kWh and your base load is 600W over the 10 hours from 9am to 7pm you are left with 9kWh. Subtract the dishwasher and a load of laundry and you are left with less than your battery capacity.

SD_DRACULA

I'd be interested in this https://icewind.is/ if you can just put them on the house.

The CW100 looks ok-ish for my night usage or if anyone is aware of any CloudPV instead of SolarPV that would be perfect for Dublin 😅

garo

Also when I did the sums - 2.4kWh battery and 4.8kW PV - I calculated 420 eur saved p.a. Admittedly that was with 14c/7c unit prices. My base load is similar to yours. Now I ended up saving > 450 last year but that is due to WFH and being able to put onb several loads of washing when it was sunny. etc. etc.

bullit_dodger

Ahh super - great to have some "real life experience" data! Nice one.

So chucking 150 days then into my spreadsheet, and using the battery as a "night rate reservoir" to be used during the day for other 215 days.

Approx cost of battery €3200 , so at €392/year savings gives a payback in 8.9 years - and within my "general expectation" of 8-9 years that I mentioned earlier. Course different battery capacities will have different costs, and yields, so....mileage will vary as they say for different individuals.

It's ballpark, for sure. The reality is that you'd never absolutely pick every single right day/night to have the battery charging or left to deplete depending on what the sun the next day is going to do, but it's a reasonable approximation I think.

Course my "savings" above are only for the battery, doesn't include direct solar used during the day that's not gone into the battery. I was only trying to work out the ROI of the battery itself.

slave1

Battery losses, you should reduce and saving by 20% to cover in/out losses.

Personally I'd model zero PV to battery Nov to Feb inclusive to be prudent

slave1

Good stuff though