Looking at the datasheet I'd expect finding a drop-in replacement is likely problematic. A rather involved RTC module providing seconds out to 100 years. (Although it looks a lot like a Seiko RTC chip I used in the late 70s.) Emulation would probably require an FPGA approach, or one using a 2-port memory, either driven from a more stable clock that, ideally, is GPS disciplined. But a full emulation is an involved design, and all you really need to do is sync the seconds, so what's simpler?
One possible approach is to try to do a "jam" correction on the fly. With a little study on how the module interacts with the rest of the carillon there is likely dead time when another controller can take over access to the RTC module. Read the seconds, compare to UTC (or GPS) seconds, adjust if needed by blipping the 30 second adjust bit (D3 in register CD, see full datasheet link below). If GPS is not present, do nothing, assuming GPS will come back before the RTC module skews a full 30 seconds. Depending on how the STD.P pin is used there may be no need for read access, as it may provide pulse per second or minute that could be used for comparison against GPS. In any case register access would require an intercept board that multiplexes all the control lines such that your added controller can get to the necessary registers. Unless I'm wrong (always a possibility) there is no need to access most of the registers in the part, simplifying the design. Or maybe the chip core matches that external crystal RTC I used years ago (sorry, don't remember the Seiko part number offhand). You could then build a module replacement without jam logic but with a GPS disciplined 32.768kHz oscillator. That would be a much simpler approach for minimal drift. Bob LaJeunesse Found a full datasheet at https://www.manualslib.com/download/47987/Epson-Rtc-72421-A.html > Sent: Thursday, October 03, 2019 at 1:00 AM > From: "Corey Sukalich" <[email protected]> > To: [email protected] > Subject: [time-nuts] Timebase Replacement - Seiko Epson RTC-72421 > > Hi all, > > New to the list, but have known of its existence for a while as I know others > that have been subscribers. > > My question pertains to a timebase used in a Schulmerich carillon bell tower > system from the 1990’s. > > The Seiko Epson RTC-72421 Real Time Clock Module (4-bit) is used, but the > clock ends up walking to a noticeable degree (minutes) over a period of > months. Is this expected for the device in question, or could it possibly > have a defect? > > Datasheet available here: > https://support.epson.biz/td/api/doc_check.php?dl=brief_RTC-72421&lang=en > <https://support.epson.biz/td/api/doc_check.php?dl=brief_RTC-72421&lang=en> > > From what I can tell it uses an internal oscillator, so it sounds like I > would need to find a drop-in replacement or build a module that will replace > it with a significantly better time standard. Any advice is appreciated as > to whether could be accomplished without too much trouble. I’ve repaired a > few of these systems over time and there are enough of them still in the wild > that it might be worth me fabricating a module for them. > > Also, on the off chance someone is versed in EPROM data recovery or knows of > a resource for it, I would appreciate that information as well. > > Regards, > Corey > N9WIV > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow the instructions there.
