Jason, This is amazingly and scarily impressive. Very useful, and in my luddite oriented mind, very worrisome at the same time. Thanks for the demonstration. Howie
On Sun, Mar 8, 2026, 10:11 AM Jason Szumlanski via RE-wrenches < [email protected]> wrote: > I wanted to share a tip for using AI to extract from the amazing knowledge > base we have accumulated over time. I haven't deleted a single wrench email > in over a decade. They are all in my Gmail account as archived emails. If > you are like me and never delete emails, there is a great way to dig up old > information. > > I recently became enamored with Anthropics's Claude Co-Work, which > requires the paid version, but I feel it's well worth it. You can connect > it to your Gmail account (not available in the free version). When you do > that, you can ask it in plain language to search and extract data for you. > > When I get some time I am going to see how I can expand this method to the > online archives rather than using my personal Gmail data. Regardless, > whether using the archives or Gmail, the results are very helpful. I have > used this to find historical discussions with great success. > > I will give you an example of what I asked and what the response was. It > was exactly what I was looking for. > > *Claude Prompt:* > > Go into my email and find the RE wrenches list email list server address > and then search for all emails to or from that list server regarding NEC > section 690.12 and summarize the content of recent discussions at a high > level > > *Response:* > > First, let me find the RE wrenches list server address. > > Got it. The list server address is `[email protected]`. > Now let me search for 690.12 discussions. > > Good, I can see several distinct threads touching 690.12. Let me read the > key threads to get enough detail for a solid summary. > > I now have solid coverage across four distinct threads. Here's your > summary. > > --- > > The list server address is `[email protected]`. > > --- > > ### 690.12 Discussions on RE-wrenches: High-Level Summary > > Four threads surfaced with meaningful 690.12 discussion, covering the > period from roughly August 2025 through January 2026. > > **Thread 1: "Battery RSD" (August 2025)** > > This thread clarified an important terminology distinction: Rapid Shutdown > is a term that belongs specifically to PV systems under 690.12 and should > not be conflated with battery or ESS disconnecting means, which are covered > under NEC 480 and 706 respectively. The discussion resolved a real > inspection dispute where a building inspector was demanding a disconnect > between the battery and the inverter. The installer eventually got the > supervisor to agree that what was actually needed was an AC disconnect > within sight of the ESS, not a DC disconnect between battery and inverter. > > **Thread 2: "Help understanding 690.12/UL 3741 and Ironridge" (October > 2025)** > > A wrench working on a new home with a Sol-Ark/battery system was confused > about whether Ironridge's UL 3741-listed XR Flush Mount racking eliminated > the need for MLPEs. The group clarified that UL 3741 compliance is a > system-level listing that requires specific matched combinations of > racking, inverters, and (in some cases) RSD devices. Ironridge's listing at > the time only covered Tesla inverters with MCIs or Solis with MidNite MNSSR > units, so Sol-Ark was not a covered option. One variation of the UL 3741 > path requires the inverter to be mounted within roughly one to three feet > of the array, making it impractical for most residential installs, while > the other variation uses RSD devices and allows a remotely located inverter. > > **Thread 3: "Tigo TS4-A-F" (November 2025)** > > This was the most technically substantive thread. It started with RSD > device reliability concerns and evolved into a broader interpretation > debate about when 690.12 actually applies. > > On applicability: Zeke Yewdall argued that 690.12 module-level RSD is only > required on inhabited structures and that even unoccupied detached garages > might not require it. He read Exception #2 of the 2023 NEC as clearly > exempting nonenclosed detached structures such as parking shade structures, > carports, solar trellises, and similar, which he argued extends to > ground-mounted arrays. William Miller pushed back, paraphrasing the rule > as: if PV circuits are located on or in any structure, RSS is required, > with the only exception being a structure whose sole purpose is to house > solar equipment. His colorful summary was that if your inverter is in a > shed with one rake in it, RSD is required. Howie Michaelson raised the > additional open question of whether RSD is still required when the array is > on a nonenclosed structure but DC conductors are routed into a dwelling > where the inverter is located, noting this is a particularly important > consideration for cold-climate battery installations. > > On device reliability: William Miller argued that the RSS requirement > outpaced the availability of reliable hardware to implement it, noting that > interrupting high-voltage DC under load requires serious arc suppression > that a small plastic RSD nodule is not well suited to provide. He also > flagged a nationwide RSD device shortage at the time. > > **Thread 4: "MidNite MNSSR-600-SS" (January 2026)** > > MidNite Solar confirmed directly on the list that the MNSSR-600-SS > string-level RSD product has been discontinued and there is no replacement > on their roadmap, citing new UL double-redundancy failsafe requirements > that would have demanded a substantial redesign. The original question was > whether string-level shutdown alone could satisfy 690.12 under NEC 2023 or > 2026 without UL 3741 involvement. William Miller's practical takeaway was > that the most viable design path to avoid 690.12 compliance headaches > altogether is to keep all electronics in a dedicated structure used for > nothing other than the PV system. > > --- > > **Bottom line across threads:** The list has been working through 690.12 > from multiple angles simultaneously -- what structures trigger it, what > exceptions actually hold up with AHJs, whether UL 3741 is a realistic > residential compliance path given equipment compatibility constraints, and > the shrinking market for string-level RSD devices. The ground-mount > applicability question in particular had no clean consensus answer, with > experienced wrenches reading the code exceptions differently. > > > Jason Szumlanski > Florida Solar Design Group > > _______________________________________________ > List sponsored by Redwood Alliance > > Pay optional member dues here: http://re-wrenches.org > > List Address: [email protected] > > 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/[email protected]/ > 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: [email protected] 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/[email protected]/ 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
