[time-nuts] Timebase Replacement - Seiko Epson RTC-72421

Gregory Beat w9gb at icloud.com
Thu Oct 3 13:56:56 UTC 2019


Corey -
These Seiko Epson RTC chips can and do fail (oscillator failure or moves “out of spec”).
These chips have a rough accuracy of ~ 1 second/day.  
That level of accuracy can produce ~ 365 seconds/year (6 minutes)

Your question suggests that you are searching for a Clock with a higher frequency accuracy/stability — than this 1990s RTC solution provides (10-6).  

Seiko Epson RTC-72241 RTC - Full Datasheet
http://www.s100computers.com/My%20System%20Pages/PIC&RTC%20Board/Epson%2072421%20manual.pdf

Page 20 of the full datasheet provides insight & finding the frequency stability (time).

1.) Frequency and temperature characteristics can be approximated 
using the following equations.
** Finding the Frequency Stability **
∆fT= α ( θT- θX)2

∆fT   : Frequency deviation in any temperature
α (1/°C2) : Coefficient of secondary temperature ( −0.035±0.005 ) × 10-6 / °C2
θT ( °C )  : Ultimate temperature (+25±5 °C)
θX ( °C )  : Any temperature

2.) To determine overall clock accuracy, 
add the frequency precision and voltage characteristics.
∆f/f= ∆f/fo+ ∆fT+ ∆fV

∆f/f   : Clock accuracy (stable frequency) in any temperature and voltage.
∆f/fo : Frequency precision
∆fT   : Frequency deviation in any temperature.
∆fV   : Frequency deviation in any voltage.

3.) How to find the date difference
Date Difference = ∆f/f × 86400(s)
** For example: ∆f/f = 11.574 × 10-6 is an error of approximately 1 second/day. **
—
Greg
==

Date: Thu, 3 Oct 2019 00:00:44 -0500
From: Corey Sukalich <zappy at xzapx.com>
To: time-nuts at lists.febo.com
Subject: [time-nuts] Timebase Replacement - Seiko Epson RTC-72421

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?

Corey
N9WIV




More information about the Time-nuts_lists.febo.com mailing list