1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 <2022> 2023 2024 2025 | Index | 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 <2022> 2023 2024 2025 |
<== Date ==> | <== Thread ==> |
---|
Subject: | Re: Timestamp test questions on Linux |
From: | Michael Davidsaver via Tech-talk <tech-talk at aps.anl.gov> |
To: | gjzhai at mail.ustc.edu.cn |
Cc: | tech-talk at aps.anl.gov |
Date: | Fri, 18 Mar 2022 20:27:46 -0500 |
On 3/17/22 19:01, gjzhai at mail.ustc.edu.cn wrote:
The R&S®SMB100A microwave signal generator outputs 1Hz pulses to the EVG. I select the source of event 125 from front panel input 0.
What you describe is exactly what I would expect if the ~1Hz signal originates from a free running oscillator which is different to that used for the system clock. Unless your signal generator and system clock are ultimately derived from, or synchronized to, the same reference oscillator (eg. GPS or similar) one will always drift with respect to the other. What is usually done is to have a GPS receiver with both an electrical 1Hz output (often called Pulse Per Second), and a logical connection to an NTP server. Devices like this are available from multiple vendors. Some examples (which I don't specifically endorse): https://www.meinbergglobal.com/english/productinfo/gps-time-server.htm https://www.microsemi.com/product-directory/synchronization-systems/3818-enterprise-network-time-servers https://endruntechnologies.com/products/ntp-time-servers/ntp-server
epics> dbpf("TST{evg:1-TrigEvt:0}TrigSrc-Sel","Front0") DBR_STRING: "Front0" epics> dbpf("TST{evg:1}1ppsInp-Sel","1") DBR_STRING: "Front0" epics> dbpf("TST{evg:1}1ppsInp-MbbiDir_.TPRO","1") DBR_UCHAR: 1 0x1-----原始邮件----- 发件人: "Michael Davidsaver" <mdavidsaver at gmail.com> 发送时间: 2022-03-09 15:05:05 (星期三) 收件人: gjzhai at mail.ustc.edu.cn 抄送: tech-talk at aps.anl.gov 主题: Re: Timestamp test questions on Linux On 3/7/22 00:47, gjzhai--- via Tech-talk wrote:Hello! I installed mrfioc2 2.3.0 on MTCA platform. According to EVR Usage Guide, when I do the timestamp test, there are two problems:Could you describe your setup in more detail? Where does the 1Hz (aka. 1 pulse per second) signal to the EVG originate? The same GPS receiver as the NTP server in question? https://github.com/epics-modules/mrfioc2/blob/a8cb48d549b7aae13206cf2d168c6f1cfec2e1c4/evgMrmApp/Db/evgMrm.db#L93a. The system is configured with NTP, but the current time provider does not have NTP [opi@localhost ~]$ ntpstat synchronised to NTP server (202.38.64.7) at stratum 3 time correct to within 11 ms polling server every 512 s b. The time interval between the system clock and the EVR time is large. epics> generalTimeReport(2) Backwards time errors prevented 446106 times. Current Time Providers: "EVR", priority = 50 Current Time is 2022-03-07 16:34:25.575015. "OS Clock", priority = 999 Current Time is 2022-03-07 16:36:55.978638. Event Time Providers: "EVR", priority = 50 Thank you for your time. I am looking forward to your reply.