Solar electric installed nine years ago is down
kunstler.com
kunstler.com
Hacking high energy DC systems is perhaps a little more responsibility and risk than was desired tho.
Not sure what the permitting / venting issue would be with VRLA (sealed lead acid) as they are used in UPS's everywhere and only vent gas under extreme overcharge conditions. However if I where getting new backup batteries I would definitely go with LFP due to less space used and extremely long cycle life while being very affordable, they also have much more usable capacity than a lead acid and a much flatter voltage curve. I replaced my lead acid in my RV and got 4 times the capacity in the same space at half the weight and I should never have to replace them, they will probably outlive the RV.
The biggest failure mode with both solar systems and battery systems is the electronics components of the solar controller / inverter and battery BMS. The mosfets or capacitors or digital controls seems to fail much sooner than the solar panels or lithium battery cells. Not sure if its just using cheap components to save money or what, but it definitely helps to oversize the controller, inverter and battery BMS's. Would be nice to see GaN or SiC FETs and components in residential solar equipment.
[1] https://www.us.sokbattery.com/product-page/sok-48v-100ah-ser...
Or maybe he just needs to replace the capacitors on the board.
Lead-acid can emit hydrogen during charging which is easy to get rid of with good ventilation. Worst-case scenario it produces a single explosion, hopefully in a utility room with not too much risk to life.
Lithium in contrast will give you very hot, unextinguishable fire.
Lithium Iron batteries which are what's typically used in stationary backup are extremely safe and do not burn as hot compared to more unstable high energy density chemistries like NMC typically used in mobile applications. They are also relatively inexpensive and have extremely long lifespans if not used in extremely hot environments, we are talking 20 years to 80% capacity in a stand-by application.
It is likely that he will need in the next 18 years another charge controller and another set of batteries, so between:
35+2+4+2+4=47 thousands/27=1,740 US$/year
and
35+2+16+2+16=71 thousand/27=2,630 US$/year
I believe that the issue about the lead batteries is due to them being "in a cabinet", most probably they can be relocated or however the cabinet made ventilated for far less than the difference between lead and lithium batteries.
Still, even if lead batteries can be used, it does look like expensive.
[0] and it has to be seen how much it will be the cost for the disposal of the panels