Subject: Re: 2.5m obslog 20100529 (MJD 55346) From: French Leger Date: Mon, 31 May 2010 09:22:54 -0600 Jim et. al.: It seems that the cause of the Marvels long fiber bundle snag reported = by Howard B., a couple of days ago, was due to the rotator in motion = while at low altitude, below 15 degrees.=20 The good news seems to be that no damage was done. Although I can not be = absolutely sure of this until Tuesday morning once I have completed a = thorough investigate. Neither Howard nor Robert have reported any = obvious damage to the the fiber link. General information: This section of the fiber link which, runs from inside the cone to the = telescope mounted cartridge has been designed and built with the = potential of this type of failure in mind. We therefore ran some tests = on the rotator to determine what force it would take to stop it by = causing an over current condition. We found that 50 to 60 lbs of force = applied to the outer surface of the rotator drive ring was sufficient to = produce the abort current. We then designed an outer sheathing for the last 18 feet of the link = that would protect the fibers from this type of "snag". The outer = webbing type sheathing provides the axial protection. This would be = protection against tensional type pulling conditions. Inside of this = webbing is a spiral wrapping that provides the radial protection against = sharp bending conditions. Both sheathings are anchored at their ends to = a custom designed and modified Swagelok compression fitting. Pull tests = were conducted to determine the force needed to separate the sheathings = from their fittings. We were able to apply a 100 lb force to this joint = without separation failure. The gang connectors are heavy enough to = accept these type of loads, both axially and tangentially, without = damage.=20 Lastly we provided excessive spacing inside the link sheathing. The ID = of the sheathing is .750" while the fibers take up no more then 20% of = this cross-sectional area. The fibers are anchored at the gang connector = end just behind the USCon connectors to prevent tensional forces being = applied directly to the ferrules. The fibers are not anchored at any = other position along the sheathing. This allows the fibers to "float" = inside the sheathing which means that forces applied to the sheathing do = not transfer directly to the fibers. Although this design and configuration does not guarantee complete = protection against mechanical failure due to the episode experienced = this week it does provide a degree of robustness that an unprotect fiber = would lack. One of the possible solutions to this "Snag" problem would be to prevent = the rotator from moving whenever the telescope finds itself below 18.5 = deg altitude. This can easily be done in the interlock software or in = combination with a feedback signal from the gang connector identifying = its location at stow. =20 G'Day, French =20 On May 30, 2010, at 2:23 PM, jegunn wrote: >=20 > It would be well if the `cold' observer could go down before the > slew to zenith started and stay down until the slew to field is = completed > and watch; I do not think this would even be an efficiency hit. >=20 > --jim >=20 > On Sun, 30 May 2010, Daniel Eisenstein wrote: >=20 >> Until the cause is diagnosed and suitable mitigations in place, >> do we need to maintain visual contact with the region during >> any motion that could repeat the problem? I don't have a clear >> picture of the geometry, so this is just a question. >>=20 >> Also, I don't know if we have a way to test *throughput* of >> the fibers without getting on-sky, but that would be another >> check of the state of the long fibers. >>=20 >> Thanks, >> Daniel >>=20 >>=20 >> On Sun, May 30, 2010 at 03:43:13PM -0400, jegunn wrote: >>>=20 >>> This MUST not happen, EVER. We must understand how it did happen, = and >>> do whatever is necessary to prevent its happening again. If nothing >>> was damaged, it was at least a minor miracle. There should never be >>> enough slack in the umbilical to allow it to even get close to the >>> autofill connectors. If we have to install rails around all the >>> protuberances on the back of the telescope to prevent this, that is >>> what we have to do, but I suspect that what failed was the slack >>> taker-upper. In any case, this is not something that we can let >>> slide. It HAS to be fixed, ASAP, and whatever mods need to be made >>> to the new fiber installations have to be understood and = implemented. >>>=20 >>> --jim >>>=20 >>> On Sun, 30 May 2010, Howard Brewington wrote: >>>=20 >>> 05:30Z >>> Closed because of clouds and nearby radar returns. During the = process >>> of closing, we noticed that MARVELS's fiber bundle was snagged on = the >>> LN2 auto-fill connector for sp2. This wrap pulled all the slack >>> fiber from the lower enclosure. Not sure how this happened; all was >>> normal when the fiber was plugged into cart 06. We did marvels = calibs >>> after closing. The TIO and ThAr images looked fine, so we're = guessing no >>> damage was done. >>>=20 >>>=20 >>>=20 >>>=20 >>> 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 5215 in the sdss-25mlog archive, URL >>> DSS = http://www.astro.princeton.edu:81/sdss-25mlog/msg.5215.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_subscriptio= n?list=3Dsdss-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 >>=20 >=20 > 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 5218 in the sdss-25mlog archive, URL > DSS = http://www.astro.princeton.edu:81/sdss-25mlog/msg.5218.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_subscriptio= n?list=3Dsdss-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