not for long.) We have an echelle spectrograph on the telescope that works. We have observed 3.5 nights with it (Dec. 14, 1998; Jan. 6,7,8(second 1/2). The instrument is operationally stable. It has now been cold on the telescope for about 20 nights with no problems. APO staff routinely fill the dewars and have had no problems other than a sticky valve, which all know how to un stick now. We have obtained data on drift in cross dispersion and in wavelength (echelle dispersion) over long sequences with systematic moves of the telescope; on repeatabiltiy night to night (same stars, different nights);on all aspects of spectrograph thruput (except of course the critical three mirrors of the telescope) including slit function (amount of light actually through the slit for given seeing; on instrument operability in the coldest conditions; on dark current; on stray light signals; and on stars used for various types of science (interstellar lines, halo star abundances, stellar line profiles, nebular emission lines) that is already done and published (this, to look for subtle errors). We have not yet done a stable series for radial velocity stability (end to end, with reductions) or on broad line/continuum separation at high signal to noise. These data were obtained in essentially three instrument setups, in the subtle sense that various foci were changed, widgets twiddled etc. so the detailed state was changed three times, the longest stable run being 48 hours. These data are in various stages of reduction and I shall not here quote values for the various parameters derivable from above. Everything looks very good. Drifts of 1/2 pixel have been observed over many hours. The design goal is 0.1 pixel in four hours. Whether the difference is due to unmonitored observing conditions or represents an instrument limitation will only be known after another month of observing. To the eye, the spectra are beautiful. The full format is well included with in the big CCD. The echelle orders are clearly and cleanly separated from each other (2 pix, FWHM). The resolution (emission lines) is 2.2 to 2.5 pixels, and we expect the order FWHM to resolution (Nyquist) to be 3/2. We may have an aberration that is forcing this ratio and Shu-i is working on this. We know the optics delivers the proper ratios (measured with the 800x800 15 micron chip in 1989), but that was over a small field. Clearly, much more work has to be done on this. The spectra can be reduced, to first order (all we have checked) with standard IRAF packages without evident failure. We have not carried the reductions through to completion to learn of subtle effects in photometry. All of that is underway. The echelle software works without failure (or at least, only with understood failures). The remote software works as well. The system has been completely controlled from Chicago (slew, acquisition on the slit, guiding for up to 30 minutes) all without site intervention. We have various things on the to do list (such as remote focus within the remark echelle window, etc.) that should be done by the end of February. The main uncertainties at this point are throughput and use of the guider on faint objects. The former is low (1% overall) including top of the atmosphere down to the computer. We are documenting all of the surfaces in between and will publish the error budget and any out of spec performance issues once we are done with the reductions. The low number does not include precise gain, precise slit/seeing/tracking corrections, degraded telescope optics (73% overall assumed), etc. We have also not verified that we are photon noise limited at over 100 counts per pixel (i.e., that the flat fielding is perfect). However, there is no reason to suspect a problem. The chip has no defects of note. As noted above, we have all the data for the throughput calculation, which I will publish when all the reductions are done and I am satisfied we have thought of everything. At the moment we are using only one slit (1.6secx1.6sec, square). We have other slits, and will make a few more (especially if the slit function is ambiguous and we need a straight large slilt measurement for calibration. We have a number of nights in February to finish the undone work (radial velocities, wide lines, tests on QSOs), to verify any realignments necessary and to verify the comparisons between our data and existing data for the stars and gas noted above. The teritary is not automated, though the work is underway. I hope we can test it completely in March 1999. If I try to list all the folks who made this possible, I would certainly miss someone without lengthy research into years of notes. All who contributed know who you are. My personal thanks. This message is being sent to the echelle commissioning team, the group currently defining tests and working daily for the next nights work. Their hard work is evident. Questions may be sent to me directly. Everyone on the echelle commissioning team mailing list has a copy of our current working task list if you wish to see it. DON YORK APO APO APO APO APO Apache Point Observatory 3.5m APO APO APO APO APO This is message 331 in the apo35-general archive. You can find APO the archive on http://www.astro.princeton.edu/APO/apo35-general/INDEX.html APO To join/leave the list, send mail to apo35-request@astro.princeton.edu APO To post a message, mail it to apo35-general@astro.princeton.edu APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO APO