| 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: Recommended practice for cvtRecord chain initialization |
| From: | Emmanuel Mayssat <[email protected]> |
| To: | Benjamin Franksen <[email protected]>, EPICS mailing list <[email protected]> |
| Date: | Thu, 10 Apr 2014 11:15:14 -0700 |
|
Ben,
I tried the inverted table method without any success. My test setup is simple 1/ I use the identity transformation table 0 0 1 1 2/ I use 2 analog out records one as my input X to the cvt record, the other as the output 3/ I test the record chain with the method "1D TABLE" everything works as expected 4/ I change the method to "1D TABLE INVERTED" and reboot the test IOC Because of the identity table, I expected the same behavior. But whenever my cvtRecord is processed, the output is set to 0.0. Am I missing something? -- Emmanuel > From: [email protected] > To: [email protected] > Subject: Re: Recommended practice for cvtRecord chain initialization > Date: Fri, 14 Mar 2014 14:05:45 +0100 > > Hi Emmanuel > > On Thursday, March 13, 2014 15:58:44 Emmanuel Mayssat wrote: > > I have a recordAO (let's say a phase in deg) that is converted through > > a cvtRecord (~ breakpoint table) to a voltageAO. We could assume that > > the voltage on the device doesn't change while the IOC is off and > > autosave everything. > > > > But... in this particular case, where the hardware 'remembers' its > > voltage setting (EEPROM or hardware stays on while IOC is off), the > > chain above should be executed in the opposite direction at iocInit. > > A possible way is to have another cvt record with the same > > "breakpoint" table but with "1D table reversed" method and PINI=1. Is > > this the recommended way? > > I did not have this problem in practice yet, but yes, this is how I > would do it. You might need some additional record logic that disables > the readback and reverse-conversion after the first (PINI) processing. > You might also want to disable processing in the forward direction > during the readback (this might be necessary if setting a value to the > device has other side effects that you want to avoid). > > Cheers > Ben > -- > "Make it so they have to reboot after every typo." ― Scott Adams |