diy solar

diy solar

Best Sol-Ark 15 Battery?

You really cannot "expand overtime". You cannot mix old and new batteries. You have about a year window to add more capacity, but that's about it.
Are you talking FLA/AGM or LiFePO4? If FLA/AGM that is generally accepted to be true but I'd be interested in seeing links supporting this as it relates to LiFePO4 batteries. Is the window you refer to a time based window or cycle based window? Thanks.
 
You really cannot "expand overtime". You cannot mix old and new batteries. You have about a year window to add more capacity, but that's about it.
I don't think that's the case at all. A BMS is going to control each battery. Each battery is going to perform to it's ability. Batteries connected via coms will work with each other if they are compatible, those not connect via coms will just do their thing.
 
I don't think that's the case at all. A BMS is going to control each battery. Each battery is going to perform to its ability. Batteries connected via coms will work with each other if they are compatible, those not connect via coms will just do their thing.
I was also told by Fortress Power that adding another LFP battery after one year would shorten the new batteries life.
The reason I want as told is because the charge and discharge curves change slowly over time.
So the new battery will be bearing more load than the others and it will discharge faster.
My thinking is that the cells internal resistance gets larger with age.
 
I would get the cheapest ul9540 battery with a bms you can access with an app. Run solark in voltage mode if communications are not supported.
 
I was also told by Fortress Power that adding another LFP battery after one year would shorten the new batteries life.
The reason I want as told is because the charge and discharge curves change slowly over time.
So the new battery will be bearing more load than the others and it will discharge faster.
My thinking is that the cells internal resistance gets larger with age.
With my Redodo mini 12V 100Ah there's a warning not to series batteries that are over 3 months older than each other - presumably for same reaso.ln.
 
I was also told by Fortress Power that adding another LFP battery after one year would shorten the new batteries life.
The reason I want as told is because the charge and discharge curves change slowly over time.
So the new battery will be bearing more load than the others and it will discharge faster.
My thinking is that the cells internal resistance gets larger with age.
That may be, but then, aren't you extending the life of the other batteries? Seems they would balance out at some point.
 
I’m sure fortress and home grid are great but $10,000 for 18kWh?

Home grid is $22,000 for 38.4 kWh.

Pytes is $1750 for 5.1kWh.
  • Safety UL9540A listed, UL9540 with Sol-Ark, and CEC Listed

I could maybe understand if it was an inverter but that much just for cells, BMS and a case?
For the HomeGrid Stack'd it's more than cells, BMS and a case.

- Each 4.8 kWh "slab" has a BMS that manages the cells (just like everybody else)
- Each stack has an overall BMS in the head unit that manages the 3-8 slab stack
-- That overall BMS communicates externally to inverters (charge/discharge limits and the like) and distributes power to/from the slabs
-- Internal to the stack it actively balances the slabs to keep them in lock step (the slab SOC's don't wander) and it monitors SOH for each slab

That overall BMS can also serve as the "master" for multiple stacks and communicate with the inverter system on behalf of the entire battery array. If you had a maxed out array of 15 stacks the master BMS/stack could request as much as 4500A Charge/Discharge (200+ kW) via closed loop.

Additionally, there's centralized fault isolation (w/ look-up codes - voltage/current/temp/comms/internal failure) for each stack and cloud and local (app and software based) monitoring (via WiFi) for the battery array (albeit the phone app was waaaay beta when I tried it). I can't comment on the cloud-based monitoring as mine are kept offline.

I'm not trying to convince anyone on here as there are WAY more cost-conscious options with 90% of the functionality for well-read hobbyist like those of us on this board...BUT, it is a product with unique features and to say it's equivalent to a server rack setup is incorrect. HomeGrid makes the most sense if you were going to build a setup for your rich grandma one state over who knows nothing about batteries/power nor would ever care to. Totally set and forget for the life of the system; bulletproof and safe while not looking like a bowl of spaghetti in grandma's 4-car garage. :p
 
Last edited:
For the HomeGrid Stack'd it's more than cells, BMS and a case.

- Each 4.8 kWh "slab" has a BMS that manages the cells (just like everybody else)
- Each stack has an overall BMS in the head unit that manages the 3-8 slab stack
-- That overall BMS communicates externally to inverters (charge/discharge limits and the like) and distributes power to/from the slabs
-- Internal to the stack it actively balances the slabs to keep them in lock step (the slab SOC's don't wander) and it monitors SOH for each slab

That overall BMS can also serve as the "master" for multiple stacks and communicate with the inverter system on behalf of the entire battery array. If you had a maxed out array of 15 stacks the master BMS/stack could request as much as 4500A Charge/Discharge (200+ kW) via closed loop.

Additionally, there's centralized fault isolation (w/ look-up codes - voltage/current/temp/comms/internal failure) for each stack and cloud and local (app and software based) monitoring (via WiFi) for the battery array (albeit the phone app was waaaay beta when I tried it). I can't comment on the cloud-based monitoring as mine are kept offline.

I'm not trying to convince anyone on here as there are WAY more cost-conscious options with 90% of the functionality for well-read hobbyist like those of us on this board...BUT, it is a product with unique features and to say it's equivalent to a server rack setup is incorrect. HomeGrid makes the most sense if you were going to build a setup for your rich grandma one state over who knows nothing about batteries/power nor would ever care to. Totally set and forget for the life of the system; bulletproof and safe while not looking like a bowl of spaghetti in grandma's 4-car garage. :p
I guess if I had a rich Grandma or installer for some rich puff that can’t mess up his manicure otherwise they are way overpriced for what they are.
IMO

I haven’t looked at my DIY batteries in months.
 
I contacted a local solar company to ask about batteries. He threw me a ball park number of $20,000 for a 19.2 KWH HomeGrid. I think his number was pay back for not hiring him to install my Sol-Ark system last year.
 
The reason I want as told is because the charge and discharge curves change slowly over time.
That is consistant with what Orion BMS said in one of their papers on paralleling batteries. They used the term eddy currents but someone here said it was not exactly the same but the concept was similar. I am guessing that with different internal resistance the voltage sag is not the same between the modules and the phyics of paralleling will cause current to flow from the higher voltage to the lower voltage.

Years ago I had an eight year old Nissan Leaf pack in which I added some modules from a newer Leaf which had a different capacity. I had no way of measuring the resistance or the current between the cells but I did notice the groups of cells in parallel would deviate under load. Therefore my settings were conservative. That was in 2019 before I joined this forum and discovered the great prices on EVE LFP cells.
 
That may be, but then, aren't you extending the life of the other batteries? Seems they would balance out at some point.
No, if you think about it the one pack tied to the five you already have in place gives it no chance of doing anything significant to them. The Curve for the New Battery will quickly be bent into the shape of the older batteries IOW it will age fast, it's IR will degrade and catch up with them and then steady out.

It did not matter to me, I needed another eFlex and a new damaged one was presented to me at a good price, so I fixed it and now it is running with my other five batteries that are roughly 2 years old.

BTW nothing new about this. If you look on every pack of AA batteries it says do not mix old batteries with new ones.
I think we have all done this at one time or the other out of desperation and then found out that the New Battery died fairly quickly. Of course in that example it's like putting a 12 year old LFP battery and a brand new one together.
 
Last edited:
I thought the physics of paralleling the slabs endured they are at the same voltage? Is that what you meant by "lock step"?
That's passive balancing. The HomeGrid actively balances the slabs. The put it another way the head unit BMS distributes/manages/regulates (pick a word) incoming charging current and outgoing load as appropriate to keep the slabs matched. Balancing even occurs across multiple stacks. When first installed my stacks would deviate by as much as 12% over multiple poor production days (I thought something was wrong). Over a week or so that tightened up to 4-5% and now (a year later) my two stacks are no more than 1% off as indicated on the displays. I also took a stack out of the mix for a bit, causing it to be 20% SOC lower than the master when added back to the circuit. The master stack quit charging (or discharging for that matter) and all the power from the charger(s) went to the slave stack until it matched the SOC of the master...then they both proceeded to charge equally as a pair. Trick stuff.

It's a legit system...not simply a bank of parallel batteries. Now is it necessary? Meh...probably not. But damn if it ain't cool. ?
 
Last edited:
That's passive balancing.
You can call it that but fundemental physics states that batteries in parallel will all reach the same voltage. I do not know what else would be accomplished by "active balancing"? I certainly understand the benefit of active balancing cells or batteries in series. Perhaps I do not understand the distinction between slabs and stacks and whether some of those are in series and that is what is being balanced?

EDIT: Since posting, I have reread your explanation and now better understand the distinction. I agree with your statement that it is probably not necessary but it certainly sounds nice, and it has a nice marketing spin.
 
Last edited:
I am considering the Pytes as an upgrade to my SolArk 12K
I have learned more about the Pytes since the above post and realize there are now two versions of their 48 volt rack mount battery. The original is the E-Box 48100R and the newer version is just known as the Pytes V5. They have different form factors and therefore would use different racks and enclosures. The price seems to be the same and the enclosures vary. The the V5 has internal heating pads and some other features. If there are not significant price differences I tend to favore newer models. I also understand the V5 uses CALT cells and do not know the heritage of the cells in the E-Box. I have not applied for the building permit and so far only have documentation for the E-Box cells but will update the application with the new UL 9450 documents.
Anyone have experience with the Pytes V5 batteries yet?
 
I have learned more about the Pytes since the above post and realize there are now two versions of their 48 volt rack mount battery. The original is the E-Box 48100R and the newer version is just known as the Pytes V5. They have different form factors and therefore would use different racks and enclosures. The price seems to be the same and the enclosures vary. The the V5 has internal heating pads and some other features. If there are not significant price differences I tend to favore newer models. I also understand the V5 uses CALT cells and do not know the heritage of the cells in the E-Box. I have not applied for the building permit and so far only have documentation for the E-Box cells but will update the application with the new UL 9450 documents.
Anyone have experience with the Pytes V5 batteries yet?
I have four of the V5 batteries in closed loop with my Solark 15k. Victron Power In is being used as the bus bar. So far, all is good! It is indoors, and so I am not taking advantage of heater. I also felt that newer was better for long term.
 

Attachments

  • 20231223_141402.jpg
    20231223_141402.jpg
    134.4 KB · Views: 16
You can call it that but fundemental physics states that batteries in parallel will all reach the same voltage. I do not know what else would be accomplished by "active balancing"? I certainly understand the benefit of active balancing cells or batteries in series. Perhaps I do not understand the distinction between slabs and stacks and whether some of those are in series and that is what is being balanced?

EDIT: Since posting, I have reread your explanation and now better understand the distinction. I agree with your statement that it is probably not necessary but it certainly sounds nice, and it has a nice marketing spin.
Given how much people talk about splitting their packs, balancing, patching until stable, etc. I think a large system that does all this without on site staff of DIYers with unlimited time and patience is worth a healthy chunk of value.

IF it works.

(I think a bunch of AC coupled self contained powerwalls or IQBATs does that fine)
 
I wonder if the Pytes V5 uses the CATL QLIN cell which seems more energy dense than the original? Perhaps this explains the change in weight when comparing V5 to the E48100R?

 
I also felt that newer was better for long term.
I agree. I have waited a year to pull the building permit with my SolArk so I can wait a little longer. I like the features and heard that the V5 uses CATL cells. I have UL docs for the E-Box batteries but have not been able to find ones for the V5 batteries. I recently watched a Pytes Webinar that said the V5 has UL9450 approval.
If you pulled a permit, did you use updated UL docs for the V5 batteries?
 
Last edited:
Back
Top