Turn to zero both mixer currents with L (=Mixer Bias) and at the E-Box. Turn off both Gunns (C). Switch off both two junctions (E-Box).
	Switch off the Cold Head in case it is still running. Check that the water cooler (Rückkühler and pump) and the ventilators in the Anbau are still running. Switch off the Ionenzerstäuberpumpe.

	Put the Turbopumpstand on the caroussel near to the receiver.
	In the case it's cold in the dome, first warm up the pump with a fan heater for about 15 minutes. Otherwise it won't start.
	Connect the tube of the pump with the connecting piece of the vacuum valve of the receiver.
	Switch on the pre-pump (Switch 1) and evacuate the tube which takes about a minute.
	Open the vacuum valve at the receiver.
	When the pressure is less than 1 mbar (check the analog display of the E-Box) switch on the turbo pump (small buttons ON and START of the red-box). If it is still too cold it won't start. Switch to level 2 at the pre-pump.
	Open the JT-valve of the receiver by 8 turns by turning the knob counter-clockwise.
	Wait until the pressure is less than 0.01 mbar. This may take several hours or one night.

	Before starting the compressors make sure that the drive of the expander (cold head) is running.
	Switch on the compressor RW to the right of the JT compressor. Occasionally it doesn't work and you have to push the RESET button first.
	Now switch on the JT-compressor. The pressure in the JT-Loop (check pressure display at JT compressor) before switching it on should be 6.5 bar. After switching on the JT compressor, the pressure increases to approximately 9 bar.

	When the 20K-stage temperature drops below about 30 K, the JT-valve can be shut by 7 turns.
	Switch on the power supply of the Ionenzerstäuberpumpe, but don't push the START or ON button, since the pump starts automatically when the pressure is less than 10-5 mbar. Don't be bothered by the clicking noise this machine makes for quite a while when trying to start.
	After approximately 10 hours the 20K-stage and the 4K-stage have reached temperatures below 20 K and the JT-valve can be closed step by step until the final temeratures are reached, e.g. 13 K and 4.3 K, - but the valve must not be closed completely. By closing the valve, the temperature of the 4K-stage may increase for one or two minutes but then it will decrease again.
	The temperature of the 4K stage can accurately be read from the red digits - Voltages - at the receiver. For lower temperatures these become higher. 4.3 K corresponds to 1.623. At the end of the green folder there is a conversion table voltages to temperatures.
	If the pressure is between 10-6 and 10-7 mbar the pump can be removed. First close the vacuum valve at the receiver! At the end, the pressure will be about 4 10-7 mbar.

	In the ``Anbau'', check that the ventilators in the window and that of the ``Rückkühler (on the left) are running. The small pump in the background should also work. The big water cooler in the center is not used anymore. The machine to the right is used for the ``Deckenklimaanlage'' in the computer room.
	In case you have closed the JT-valve too far, i.e. the JT-valve was shut accidentally, you should close it completely with a few turns. Then open it again a few turns until you here a ``blob''. After that, only a 1/4-1/2 turn is needed to bring it to the correct position.
	The Penning vacuum measurement instrument should not be switched on until the pressure is below 10-3 mbar (check analog display of the E-Box).
	If the Penning vacuum measurement instrument has failed, the Ionenzerstäuberpumpe cannot start. Remedy: unplug the measurement instrument and plug it again.

 

Warning: If this procedure is not clear, consult the receiver group in Cologne or an expert observer before you start an experiment with a dubious result.

	set the synthesizer frequency to the correct value (A).
	turn on the Gunn (C).
	set the Gunn Bias to approx. 9V (D).
	change the Gunn Tune (E) until the Gunn spike appears in the spectrum analyzer (cf. RX 230 tuning table)
	lock the Gunn spike: use the PLL sweep (K) if necessary. Make sure you lock the correct frequency, by changing the SYN frequency to another harmonic frequency (SYN*HARM/new HARM), where it should remain locked.
	check the I(V) curve in the oscilloscope for the SIS photon step by turning on and opening the mixer sweep on the bottom of the cryostat.
	now switch the BNC cable to see the conversion curve for the SIS junction.
	move the Gunn Backshort (F) and the Multiplier Backshort Harm. (I) and Fund. (H) to find and optimize the conversion curve. Try to set these in such a way, that at first a double-peaked curve (a broad peak near the center and a narrow peak on the right shoulder of it) appears in the spectrum analyzer. Now optimize (F), (I) and (H) to totally suppress the narrow peak on the right shoulder in order to get a symmetric curve with a smooth curvature on top. Be aware that you may find a setting where you can get a lot of LO power, but at the wrong frequency, i.e. the wrong peak in the conversion curve is maximized. Make sure, the left peak is at its maximum - even if that means operating at lower LO power. If moving (F) unlocks the Gunn, move back the Gunn spike with (E). Note, that increasingly higher values of (F), (I) and (H) will eventually collapse the curve you try to optimize, but there are more than one maximum and you should find the highest one for the left peak.
	minimize the sharp Josephson spikes to the right and left of the central peak by changing the magnetic field strength (S).
	the peak amplitude of the curve can be increased by increasing the power for the HEMT amplifier, but be aware that more than 2V may cause interefrences, which will materialize as ripples in your 230 or 345 GHz baselines.
	now turn down the sweep and change the Mixer Bias (L) to set the correct working point to the maximum in the curve.
	check the Y-factor: it should be around 2.

	set the synthesizer frequency to the correct value (A).
	turn on the Gunn (C).
	set the Gunn Bias to approx. 8V (D).
	now check the tuning tables for the corrcect microscrew settings for the Gunn frequency and the Gunn Backshort. The Gunn frequeny given is 1/3 of your actual LO frequency. Set the microscrews <F> and <B> to the values given in the tuning table. These are approximate values and you will have to move <F> until the Gunn spike appears in the spectrum analyzer.
	lock the Gunn spike: use the PLL sweep (K) if necessary. Make sure you lock the correct frequency, by changing the SYN frequency to another harmonic frequency (SYN*HARM/new HARM), where it should remain locked.
	check the I(V) curve in the oscilloscope for the SIS photon step by turning on and opening the mixer sweep on the bottom of the cryostat.
	now switch the BNC cable to see the conversion curve for the SIS junction.
	set the attenuator (see schematic) to maximum LO power.
	iterate between the Gunn and Tripler Backshorts to find and optimize the conversion curve. The conversion curve should be symmetric with a smooth curvature on top.
	minimize the sharp Josephson spikes to the right and left of the central peak by changing the magnetic field strength (S).
	you can now move back the attenuator and try to optimize the shape of the converison curve.
	don't touch the Harmonic Mixer Backshort.
	check the Y-factor: it should be around 2.

to find the correct operation point for the doubler (tripler), first set the corresponding I and V switches at the E-Box to the 660 doubler (tripler) setting (note, that the switches can glitch, so count how often you click) in order to read the current and voltage on the display. Now set the voltage for the doubler (tripler) according to the table using the corresponding labeled V potentiometer (e.g. 4V).

if you now turn the micrometer screw of the doubler (tripler) at the LO, you can see the current change on the LC display at the E-Box. Set doubler and tripler to maximize the doubler(!) current (the tripler current is not sensitive enough and is read only if you want to check the tripler's I(V) curve). This should correspond to the maximum output power: make sure that doubler and tripler are set to optimize the total output power when measured directly with the voltmeter at the CONT output.

First, you need to check the receiver temperature. Make sure, that there is enough liquid nitrogen in the cold load and that the temperature for the hot load is set correctly in MOK under >MENU 3<. This is also a good time to check if the weather parameters have been updated automatically in >MENU 1<. Send a >LOAD< and make sure that the AOSs bandpasses are not saturated during the hot load measurement: if you need to lower the bandpass during the measurement, repeat with >LOAD N< only. A succesful load should now give you an acceptable receiver temperature.

Note:

	big spikes in the bandpass when using the 230 can occur at some frequencies if the 345 Gunn is on
	even/odd noise can occur if the temperature of the AOS water cooler is too high: try lowering it by 0.1-0.3 deg.

Now, you may want to point up the telescope on a planet followed by a focus of the subreflector. For pointing on a planet load the corresponding >PLANET.CAT< catalog. Make sure that >AUPOI< is running in MOK. The pointing is set up in such a way, that >GO< will map a cross around the planet, giving you the pointing offsets in azimuth and elevation, which you may then enter under >MENU 2< (enter >WAIT< after the beginning of each map, in order to enter the pointing corrections before beginning the next map). To do a focus, you need to load >PLANETP.CAT<, which enters on the planet itself. Use e.g. >FOCUS /Z 700 5/CENTER X< to take 5 measurements at X-700 mum, X-350 mum, X, X+350 mum, and X+700 mum. This should give you a reasonable fit for the best z position (use >HELP FOCUS< for more details).

Before you start observing your source, you need to do a calibration on the sky. In principal, there are two ways to do this: the chopper wheel calibration or the sky dip measurement.

Skydip:

	turn off the chopper wheel with >CHOP<.
	type >SKY< to load the skydip catalog, choose the desired azimuth and do a >LOAD<.
	type >GO< to start the measurement.

Chop:

	turn on the chopper wheel with >CHOP<.
	load an object catalog that uses Total Power mode.
	do a >LOAD< and send the >GO<.

In case you continue to observe in (Dual) Beam Switch mode, make sure with >CHOP< that the chopper is not running before you enter the tau in >MENU 3< after the sky calibration. It's a good idea to add a comment about the tau and weather conditions in the observation log with >REMARK "comment"<. Use >REMARK/URGENT< to write more important comments into the observation log sheet.

Now you're ready to load your observing file with >CAT<, send another >LOAD< and then start the observations with >GO<. Make sure the number of scans per calibration is set correctly in the load cycle with >LOCY<. All spectra will be written in CLASS format and you may want to open a new CLASS file for your spectra with >NEXTFILE<. To stop the observations after the current spectrum, type >FINISH<. For an immediate stop, type >ABORT<, but be aware that this may result in problems, if you abort in the middle of an integration (do not abort a comb/dark/load measurement !).

In a seperate window on Jansky, you can run a >CLASS< session after setting the data directory with >SET DEF DATEN:< (you can set the size of the CLASS file indow with >SET TERM/PAGE = 40<). The name of the current catalog can be seen on the VT510 display. Just use the common CLASS commands to read, reduce or print your spectra (>HA/PL PS_KOELN< will print a spectrum in Köln).

 It's a good idea to frequently observe one of the standard calibration sources listed in the binder you'll find at the observatory. See if you get the correct integrated intensities and if so, put a copy of the spectrum into the binder. This will help the next observer and is a simple way to monitor the overall performance of the system. 
 

	Telescope in Elevation Limit: If you stoptel the telescope without closing the dome, the telescope may hit the low elevation limit. In order to drive it out by hand, deactivate the computer control in the telescope rack, press the "rate" button and then turn the "rate" knob above the buttons to "up". If nothing happens, check if the green fuse behind the metal door on the bottom of the telescope control rack (you can see it centered in the middle without actually opening the door) is in! If it´s out, press it in (again, you can reach it without opening the door), keep it pressed and now turn the "rate" knob. After a few seconds the fuse will pull itself in and stay in. Drive the telescope for about a minute, then set the control back to the computer. Make sure the El is above 10 and Az above 0 for a starttel. Otherwise the telescope will miss the 0,0 position and run into Az or El limit again (from where you will have to drive it out again by hand).
	Hexapod Move Error: If this occurs, you can first try to reinitialize the hexapod motors by resetting x,y,z to their current default values using >SUB< in MOK and following the menu.
	Wobbler Reset: If you need to reset the Wobbler (see manual next to the wobbler control rack) and follow the standard procedure "Reset Wobbler" - "Start Wobbler" - "Start Wobbling" the wobbler may still not move. Try setting the wobbler frequency (F) again last - now it should start. Standard settings are 1800 for amplitude, -60 for position, positive phase and frequency of 1 Hz. If you need to reset the Subref PC you need to initialise the wobbler and then start MOK while the wobbler is Off and not wobbling. Makes sure the values in the >SUBREF< menu in MOK are ok as well. Here, the frequency setting should be -1 Hz, i.e. extern!
	Hexapod Leg stuck: If one of the hexapod legs is in error (usually a value near -200 is being reported), drive it manually "down" for some time (the readout will not update and stay at -200, but the leg will move) and then wait (maybe hours). To drive the leg from the Simatic rack, turn the computer control off (push the button), flip the key and set the knob to the corresponding leg. Try a few stoptels and starttels - eventually, the correct readout will return and you can now drive the leg to its nominal setting. When you´re near these settings use x,y,z and i,j,k in the other menu (F1/F2) to find the true leg settings (they always vary by a few, whereas x,y,z & i,j,k are absolute) - now continue observing.
	Simatic Reset: If you need to reset the Simatic rack, hold the little flip switch down at the OR position. Now flip the Run/Stop switch twice and let go of the other switch again - it will automatically flip back to the center position. You have now reset the internal memory banks of the Simatic unit. If the Stop light is now on, flip "only" the Run/Stop switch again. Now the Simatic control is back (this procedure is also outlined in the Vertex manual next to the Simatic unit). If you need to turn off the PC and it doesn´t come back on, reset the red main power handle at the right bootm of the Simatic rack.
	OTF timing error: Opening the CLASS windows on Jansky to display the OTF calibrated spectra may cause a time error in the data transfer for short integration time spectra. Close these windows with >F4< to avoid time errors, if you have short integration times.
	Jansky Prompt: Don't type any messages following the MOK> prompt without hitting <CR> (like remarks). This will block the dispay for message outputs and thus halt your loop until the messages have been written.    Suggestions, Improvements, Additions are welcome by e-mail [created 1999-10-12 by atieftru@ph1.uni-koeln.de; updated 2000-11-11]

   (back to top)