[RE-wrenches] EG4 v HomeGrid battery systems update?

Christopher Warfel cwarfel at entech-engineering.com
Wed Sep 3 12:03:41 PDT 2025


Hi Jason, I spoke with Solark re Homegrid and EG4. Did not mention you 
name but your experience.  You have been really helpful.  I am centering 
on open loop w EG4. My longest work commute is 15 minutes on this 
island, so it's just having the time to trouble shoot. Plus, one mess 
up, and everyone knows about it. Chris

On 9/3/2025 1:18 PM, Jason Szumlanski via RE-wrenches wrote:
> Nothing like a coincidence...
>
> A client on an island with three stacks of 8 HomeGrid batteries has 
> just called me to report that the entire system is down. All of the 
> battery BMS module displays are dark. I have mySolArk data from three 
> inverters there that show the batteries were topped off and just 
> idling along with solar covering the load at 5:00 pm last night when 
> the power suddenly went out, killing the Starlink connection. I'm 
> going for a boat ride tomorrow to check it out. Ugh.
>
> I will report my findings. Not sure if it's a battery or inverter 
> issue, but the customer's handyman reported nothing on the battery 
> screens, which seems strange to me. Even an error should not turn the 
> display off, as far as I recall.
>
> P.S. No power loss notifications from Sol-Ark is a frustrating loss of 
> functionality. I heard a rumor on Facebook that they are fixing this 
> in mySolArk this month, but I'm not holding my breath.
>
>
> Jason Szumlanski
> Principal Solar Designer | Florida Solar Design Group
> NABCEP Certified Solar Professional (PVIP)
> Florida State Certified Solar Contractor CVC56956
> Florida Certified Electrical Contractor EC13013208
>
>
> On Tue, Sep 2, 2025 at 4:31 PM Jason Szumlanski 
> <jason at floridasolardesigngroup.com> wrote:
>
>     We have some stacks of HomeGrid out there in various capacities,
>     from the smallest with five in a stack, up to four fully populated
>     stacks of eight. In total, I oversee around 150 Stack'd batteries,
>     including many that I installed and some that were installed by
>     others. I will say that they are easy to install, have a nice bold
>     visual interface, look good, and perform up to expectations. They
>     communicate flawlessly with Sol-Ark 15k. However...
>
>     I believe there is a fundamental design flaw in this stackable
>     battery architecture. Here is why I am no longer offering HomeGrid
>     in a nutshell:
>
>       * When one battery in a stack has a fault, the entire stack
>         faults out, which renders the stack non-functional until you
>         either:
>          1. Remove the battery from the stack or turn the circuit
>             breaker off, and:
>          2. Reconfigure all of the dip switches to remove the
>             battery from the communication loop, then:
>          3. Reprogram the master BMS to recognize the new stack
>             members and their positions.
>       * If you have multiple stacks, you have to do all of the above,
>         and in addition:
>          1. Remove a battery from each additional stack to balance
>             them, then perform all of the above steps on each stack.
>          2. But before you reprogram each master BMS you have to take
>             the stacks out of parallel communication, then reprogram
>             the parallel stacks before operation again.
>          3. Making each stack equal is per HomeGrid support, but in
>             practice, I don't know if it is necessary, especially if
>             you are losing one of eight (12.5%). If you have smaller
>             stacks, this might be a bigger issue.
>       * To diagnose a battery issue with a laptop and cable and get
>         warranty support, you can only do that with the master BMS
>         because each battery does not have a comm port. That means you
>         have to have the entire stack non-functional while you perform
>         diagnostics, which is not ideal for off-grid settings,
>         especially if there is only one stack.
>       * If a battery fails to balance and becomes depleted, causing a
>         fault, there are no simple terminal bolts to connect an
>         external charger. I'm not sure how you would even accomplish a
>         manual charge without opening up the case.
>
>     The EG4 server rack batteries work in a fundamentally different
>     way. Each battery has an independent BMS. When there is an alarm
>     in a stack or stacks of batteries, the entire battery bank does
>     not fault out. You can physically take a battery out of the stack
>     without changing any dip switches on the other batteries. You can
>     shut one down or experience a fault on one battery without any
>     others shutting down. I have tried this with the LifePower4
>     batteries, even when there are multiple communication strings of
>     16 batteries connected to a communication hub. The rest of the
>     batteries just keep on working, which is the way it should be! The
>     communication hub will just show zero values for the battery that
>     is missing from the stack. I cannot confirm if this is the case
>     with the LL batteries, but I suspect it would be. In a way, this
>     is like having the batteries in an open loop in terms of
>     resilience, with all of the benefits that closed-loop battery
>     communications offers. I have had a small variety of battery
>     issues with EG4, and not once has the whole bank of batteries been
>     affected by one battery's issue.
>
>     Side note about another server rack option: I can confirm that
>     Pytes Ebox V1 batteries in a communication stack will shut down
>     all batteries if one has a fault, at least confirmed by one
>     situation I had. This is despite each battery having it's own BMS
>     and console port to communicate with the batteries. The situation
>     in my case was a battery that had no "Barcode" programmed into it,
>     which was causing a parallel communication fault and shutting down
>     the whole stack. In this case, physically bypassing the battery
>     with the issue with a Cat5 coupling worked fine. There are no dip
>     switches to set, and the master battery reconfigures the
>     communication stack automatically. With Pytes' support, I was able
>     to manually code in the Barcode to the BMS with a console cable,
>     and the problem went away. I am not sure if all varieties of
>     faults would have the same effect with Pytes EBoxes, but this
>     communication issue definitely caused the whole stack to fault out.
>
>
>     The phenomenon of the new breed of LFP batteries lacking
>     reliability/redundancy inspired a blog post that I did just a
>     couple of weeks ago:
>     https://floridasolardesigngroup.com/homegrid-stackd-batteries-the-redundancy-fallacy
>
>
>     A couple of other notes on HomeGrid:
>
>       * They do not have any way to connect conduit to the BMS. You
>         wouldn't want to anyway, especially with rigid conduit, since
>         you might need to remove the BMS for service. The BMS should
>         be at the bottom, in my opinion, for this reason. You can only
>         run positive and negative battery cables out of the provided
>         strain relief glands in free air, and it requires that the
>         batteries be about 4 inches away from the wall. There is no
>         suitable way to protect 100% of the battery cables.
>       * Along the same lines, if you ever plan to expand the system,
>         make sure you leave enough battery cable length to reach a
>         higher level.
>       * The lack of busbars is a really nice feature (until you get
>         into larger systems).
>       * The discharge rate supports the maximum input for a Sol-Ark
>         15K with, I believe, just three batteries.
>       * I love their "busbar pair" designed specifically for the
>         Sol-Ark 15K. I order a pair with every inverter, regardless of
>         what battery I am using (although I am not actively selling
>         Sol-Ark right now).
>       * You can't monitor the condition of individual batteries with
>         Solar Assistant, or any other tool remotely to my knowledge.
>         You can't even monitor the condition of paralleled stacks.
>       * The "app" for the batteries is mind-bendingly useless – unless
>         I'm really missing something.
>       * For some firmware and hardware versions, over-the-air updates
>         are not possible, and HomeGrid will need to send you an update
>         tool. To be fair, I think this is also the case with EG4 and
>         some other manufacturers.
>       * Once we received a shipment where three of eight batteries
>         were in the right boxes, but there were no guts in the
>         batteries at all! It was just an empty steel battery shell. We
>         had to send them back to our supplier. The boxes actually said
>         13 Kg on the labels rather than 52 Kg, but nobody at the
>         factory caught it. Strange.
>       * HomeGrid Support is very competent, I would say among the best
>         in terms of knowing their products inside and out, but it's
>         50/50 whether I get someone on the phone or get a call back in
>         a timely manner that allows me to complete a service call.
>         They are willing to schedule assistance if you have an
>         off-grid situation that requires help.
>
>
>     Is this a vote for EG4? Not necessarily, but it's hard to argue
>     with the price and the superior reliability/redundancy aspects of
>     the LifePower4/LL batteries. One battery fault should not shut
>     down an entire system unless there is a legitimate safety hazard.
>     It's possible that these are UL issues that require
>     system shutdowns, but EG4 appears to have overcome the problems
>     I've seen with other manufacturers' products.
>
>     Like others mentioned, I prefer to go with the 14.3 kWh / 16 kWh
>     sealed batteries. I feel they have better build quality (other
>     than my recent rant about rust on the MNP PowerFlo16), and keeping
>     components sealed up better just makes sense to me, especially in
>     challenging environments. Of course, if you want more modularity
>     in terms of expansion options and less impact if a single unit
>     goes down, 5 kWh units might be a better option. There is no right
>     or wrong option, I guess – sometimes it just comes down to
>     priorities, space, mounting options, and price.
>
>
>     Jason Szumlanski
>     Principal Solar Designer | Florida Solar Design Group
>     NABCEP Certified Solar Professional (PVIP)
>     Florida State Certified Solar Contractor CVC56956
>     Florida Certified Electrical Contractor EC13013208
>
>
>     On Tue, Sep 2, 2025 at 10:41 AM Christopher Warfel via RE-wrenches
>     <re-wrenches at lists.re-wrenches.org> wrote:
>
>         Hi Wrenches, I went through the archives and read the
>         comparison between
>         these two.  The HomeGrid manual was in such a mess (or my pdf
>         reader was
>         defunct), that I started looking at other options. If anyone
>         has a
>         strong opinion of either of these two, or something better, I
>         would
>         appreciate.  We traditionally install small systems, and this
>         would be
>         one (@15kWh).  I would prefer to use a racking system with the
>         BMS as
>         part of the packaging. Solark 12kPV multimode.  Thank you, Chris
>
>         -- 
>         Christopher Warfel, PE
>         ENTECH Engineering, Inc.
>         PO Box 871, Block Island, RI 02807
>         (401) 447-5773
>
>         _______________________________________________
>         List sponsored by Redwood Alliance
>
>         Pay optional member dues here: http://re-wrenches.org
>
>         List Address: RE-wrenches at lists.re-wrenches.org
>
>         Change listserver email address & settings:
>         http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
>         There are two list archives for searching. When one doesn't
>         work, try the other:
>         https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
>         http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>
>         List rules & etiquette:
>         http://www.re-wrenches.org/etiquette.htm
>
>         Check out or update participant bios:
>         http://www.members.re-wrenches.org
>
>
> _______________________________________________
> List sponsored by Redwood Alliance
>
> Pay optional member dues here:http://re-wrenches.org
>
> List Address:RE-wrenches at lists.re-wrenches.org
>
> Change listserver email address & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>
> There are two list archives for searching. When one doesn't work, try the other:
> https://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/
> http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>
> List rules & etiquette:
> http://www.re-wrenches.org/etiquette.htm
>
> Check out or update participant bios:
> http://www.members.re-wrenches.org
>
-- 
Christopher Warfel, PE
ENTECH Engineering, Inc.
PO Box 871, Block Island, RI 02807
(401) 447-5773
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20250903/fd64d2e3/attachment.htm>


More information about the RE-wrenches mailing list