Date: Tue, 15 Sep 2009 18:35:13 -0400 Subject: Re: 2.5m obslog 20090914 (MJD 55089) From: Craig Loomis --0016e64719b04cfbae0473a5652f Content-Type: text/plain; charset=ISO-8859-1 Interesting. The designed behavior when hitting the home switch is to come off it by 4000 steps. It seems something similar happens at the other end. It's probably worth reading the specmech code; I haven't done so in years. Thanks for the notes. - craig On Tue, Sep 15, 2009 at 5:30 PM, Daniel Oravetz wrote: > When we actually hit a limit, the value would hit the limit and then > rebound so that the desired move was very clearly not reflected. For > example, if the limit were 20,000 and we were at 19500 and then issued a > move of +1000, the next motorPosition value would be something that was > NOT 19500 + 1000, like 17300 or something. So we just looked at where the > instructed move was not reflected in motorPositions and determined the > limits to within about 1000 steps that way. > > > > In order to test the limits, we moved the collimators by 1000 step > > increments until we found one limit. Then we began issuing 10,000 step > > moves, since we knew that the range for each motor was ~50,000 steps. > Once > > we got close to the other limit, we began issuing smaller steps again > > (1000). > > > > Audrey and DanO > > > >> On Tue, Sep 15, 2009 at 7:52 AM, Audrey Simmons > >> wrote: > >> > >>> > >>> We determined the approximate limits of the BOSS spectrograph > >>> collimators. > >>> The > >>> results are below: > >>> > >>> Sp1: > >>> ---- > >>> a -37000 -> 17000 > >>> b -36000 -> 16000 > >>> c -36000 -> 18000 > >>> > >>> Sp2: > >>> ---- > >>> a 0 -> 53000 > >>> b 0 -> 52000 > >>> c 0 -> 54000 > >>> > >> > >> That shows that sp1 has been reset since it was homed. Too bad. I > >> wonder > >> if it would be worth adding a 'homed' flag to specmech? As far as I know > >> there is no way to tell whether the motors positions have been reset. > >> Take > >> that back: we could declare that motor travel ends at 100 or so (this > >> trick > >> has been used before...); if the ICC sees a motor position of 0 it would > >> require that the motors be properly homed, then moved away from the > >> switches. > >> I understand that SDSS never really tracked the absolute motor > >> positions, partly because the mechs did occasionally get rebooted during > >> the > >> night, and folks either a) didn't want to take the time to home the > >> motors > >> or b) did not want to stress the motors. If there is issue with b, I > >> don't > >> quite understand it: aren't these stepper motors? > >> Anyhow, the current scheme is probably the worst of both worlds. We > >> are > >> paying attention to and recording the absolute positions, but aren't > >> requiring them to be valid. Can we (please) take the hit on requiring a > >> home > >> procedure after a reset? > >> > >> > >>> Note: The 'boss moveColl a=* b=* c=*' command was consistent with the > >>> amount > >>> of ticks we wanted the collimators to move by. ***Do not use the 'boss > >>> pistonCollimator' command because no matter what values you want to > >>> move > >>> the > >>> collimators by, it will always move by +1000 ticks.*** > >>> > >>> For example: > >>> > boss pistonCollimator spec=sp1 ticks=5000 > >>> sp1 went from -2788, -2668, -2672 to -1785, -1668, -1673 > >>> > boss moveColl spec=sp1 a=5000 b=5000 c=5000 > >>> sp1 went from -1785, -1668, -1673 to 3215, 3332, 3326 > >> > >> > >> pistonCollimator was written before the 1000 step motion limit was > >> understood [it will move distances other than 1000. They just have to be > >> >> > >> Oh, how did you tell what the upper limit was? There is no limit > >> switch > >> at > >> that end, so I didn't have the courage to simply drive the motors to > >> 100000 > >> and see where they wound up. Is that what you wound up doing > >> > >> - craig > >> > > > > > > > > DSS DSS DSS DSS DSS Digital Sky Survey DSS DSS DSS DSS DSS DSS > > DSS *** The information posted to SDSS mailing lists is proprietary *** > > DSS This is message 4725 in the sdss-25mlog archive, URL > > DSS http://www.astro.princeton.edu:81/sdss-25mlog/msg.4725.html > > DSS The index is at > > http://www.astro.princeton.edu:81/sdss-25mlog/INDEX.html > > DSS To join/leave the list, go to > > DSS > > > http://www.astro.princeton.edu:81/cgi-bin/mailinglists.pl/show_subscription?list=sdss-25mlog > > DSS To post a message, mail it to sdss-25mlog@astro.princeton.edu > > DSS > > DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS > > > > > --0016e64719b04cfbae0473a5652f Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable =A0=A0Interesting. The designed behavior when hitting the ho= me switch is to come off it by 4000 steps. It seems something similar happe= ns at the other end. It's probably worth reading the specmech code; I h= aven't done so in years. =A0=A0Thanks for the notes.=A0- craigOn Tue, Sep 15, 2009 at 5:30 PM, Daniel Oravetz wrote: When we actually hit a limit, the value wou= ld hit the limit and then rebound so that the desired move was very clearly not reflected. =A0For example, if the limit were 20,000 and we were at 19500 and then issued a move of +1000, the next motorPosition value would be something that was NOT 19500 + 1000, like 17300 or something. =A0So we just looked at where th= e instructed move was not reflected in motorPositions and determined the limits to within about 1000 steps that way. > In order to test the limits, we moved the collimators by 1000 step > increments until we found one limit. Then we began issuing 10,000 step= > moves, since we knew that the range for each motor was ~50,000 steps. = Once > we got close to the other limit, we began issuing smaller steps again > (1000). > > Audrey and DanO > >> On Tue, Sep 15, 2009 at 7:52 AM, Audrey Simmons >> wrote: >> >>> >>> We determined the approximate limits of the BOSS spectrograph >>> collimators. >>> The >>> results are below: >>> >>> Sp1: >>> ---- >>> a -37000 -> 17000 >>> b -36000 -> 16000 >>> c -36000 -> 18000 >>> >>> Sp2: >>> ---- >>> a 0 -> 53000 >>> b 0 -> 52000 >>> c 0 -> 54000 >>> >> >> =A0 =A0 That shows that sp1 has been reset since it was homed. Too= bad. I >> wonder >> if it would be worth adding a 'homed' flag to specmech? As= far as I know >> there is no way to tell whether the motors positions have been res= et. >> Take >> that back: we could declare that motor travel ends at 100 or so (t= his >> trick >> has been used before...); if the ICC sees a motor position of 0 it= would >> require that the motors be properly homed, then moved away from th= e >> switches. >> =A0 =A0 I understand that SDSS never really tracked the absolute m= otor >> positions, partly because the mechs did occasionally get rebooted = during >> the >> night, and folks either a) didn't want to take the time to hom= e the >> motors >> or b) did not want to stress the motors. If there is issue with b,= I >> don't >> quite understand it: aren't these stepper motors? >> =A0 =A0 Anyhow, the current scheme is probably the worst of both w= orlds. We >> are >> paying attention to and recording the absolute positions, but aren= 't >> requiring them to be valid. Can we (please) take the hit on requir= ing a >> home >> procedure after a reset? >> >> >>> Note: The 'boss moveColl a=3D* b=3D* c=3D*' command wa= s consistent with the >>> amount >>> of ticks we wanted the collimators to move by. ***Do not use t= he 'boss >>> pistonCollimator' command because no matter what values yo= u want to >>> move >>> the >>> collimators by, it will always move by +1000 ticks.*** >>> >>> For example: >>> > boss pistonCollimator spec=3Dsp1 ticks=3D5000 >>> sp1 went from -2788, -2668, -2672 to -1785, -1668, -1673 >>> > boss moveColl spec=3Dsp1 a=3D5000 b=3D5000 c=3D5000 >>> sp1 went from -1785, -1668, -1673 to 3215, 3332, 3326 >> >> >> =A0 =A0pistonCollimator was written before the 1000 step motion li= mit was >> understood [it will move distances other than 1000. They just have= to be >> <1000] I'll add a piston=3D option to moveColl. >> >> =A0 Oh, how did you tell what the upper limit was? There is no lim= it >> switch >> at >> that end, so I didn't have the courage to simply drive the mot= ors to >> 100000 >> and see where they wound up. Is that what you wound up doing >> >> =A0- craig >> > > > > DSS DSS DSS DSS DSS =A0Digital Sky Surve= y DSS DSS DSS DSS DSS DSS > DSS *** The information posted to SDSS mailing lists is proprietary **= * > DSS =A0This is message 4725 in the sdss-25mlog archive, URL > DSS =A0 =A0 =A0 =A0 http://www.astro.princeton.edu:81/s= dss-25mlog/msg.4725.html > DSS =A0The index is at > http://www.astro.princeton.edu:81/sdss-25mlog/INDEX.html > DSS =A0To join/leave the list, go to > DSS > http://www.astro.pri= nceton.edu:81/cgi-bin/mailinglists.pl/show_subscription?list=3Dsdss-25mlog > DSS =A0To post a message, mail it to sdss-25mlog@astro.princeton.edu > DSS > DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS DSS > --0016e64719b04cfbae0473a5652f--