====== Receiver tuning ====== The narrow band 345 GHz and 650 GHz receivers in the sidecab and the 850 GHz receiver are tuned manually. //Always wear an antistatic wrist strap// with a grounding wire when you touch receiver components. ===== General procedure ===== * Select the line you wish to observe with the ''UIP'' ''LO'' command. If you are observing a source with a large velocity, slew to your source with the ''OBS'' command before tuning the receiver. For example, \\ ''__UIP__> lo 12co3-2 /side up'' * Go the the receiver and place a ambient load (Eccosorb) in front of the horn. * Tune the local oscillator and the receiver. * Measure the receiver performance. * Measure the IF power on the power meter with an ambient load and with a cold (LN2) load in the beam. * Calculate //Y// = //P//hot / //P//cold. * Calculate //T//rx = ([280 K] - //Y// [80 K])/ (//Y// - 1). * Before observing, remove the ambient load (Eccosorb) placed in front of the horn. Then level the total power on the sky. This ensures the spectrometers are presented with an appropriate IF power level. * After tuning the sidecab receivers, fold the ladder and lean it against the wall. ==== Shutdown ==== * After observing, turn off the receiver. - Turn off the Gunn oscillator: ''__UIP__> lo /nogunn'' - Attenuate the oscillator (ccw) - On the hand paddle switch to the dummy (RX/DL -> Dummy), then ''Exit.'' - Turn off the magnet bias supply. ==== Local oscillator ==== - Close the LO attenuator. Turn it CCW but don't drive it hard into the stop. - Note the Gunn oscillator (F Lock) frequency on the antenna monitor (85 GHz for this example). - Look up the correct micrometer settings on on the tables posted on the sidecab wall. - Set both micrometers on the Gunn oscillator to the chart values. The tuning micrometer sticks out the small end of the oscillator. The backshort is on the long side. * Adjusting the tuner until the PLL locks without also adjusting the power backshort to the correct position might result in locking to an incorrect frequency. The only way to check is to look at a known source, such as Orion. - Then tweak the micrometers while watching the reference tone on the spectrum analyzer to lock the PLL. Usually the tuning adjustment is most sensitive. - When the PLL is locked the green lock LED should stay on! - Tune the multiplier if necessary. ==== Sidecab PLL ==== - With the Gunn oscillator unlocked, the ''Gunn Supply'' on the Gunn section of the PLL box should be 10.0 V. - Monitor the ''Gunn Monitor'' signal with a spectrum analyzer. Center frequency = 20 MHz and bandwidth = 2-5 MHz. - Turn on (up) the ''sweep'' switch on the PLL box. - Tweak the Gunn oscillator tuners to lock the PLL. * If it is difficult to lock the PLL, turn off (down) the ''sweep'' switch. After locking the PLL, turn on (up) the ''sweep'' switch. The PLL sweeps the Gunn oscillator bias to keep it locked as its tuning drifts with time. - Insure the ''Loop Gain'' knob on the PLL box is not too far clockwise (high gain). Once the PLL is locked, adjust the ''Loop Gain'' so the loop corner noise is flat and not peaked. The loop corner noise appears on the spectrum monitor as bumps on either side of the main line. The bumps will move away from or towards the central line at 20 MHz as the loop gain is increased and decreased, respectfully. Try to make the bumps flatten out and disappear. If the loop gain is too high, then the PLL may oscillate and there will appear three lines of similar appearance instead of only one, which is normal. ==== Receiver ==== The general idea is to maximize the total power and obtain a positive slope on the //IV// curve. - On the hand paddle, select the receiver, turn it on (RX/DL -> Rxc on), and place the SIS bias in constant voltage (CnstV) mode. - Set //V//sis = 2.1 mV and set the magnet current to 20-40 mA. The SIS current should be zero. - Open the LO attenuator (CW) while watching the SIS current. Stop when it reaches about 9 µA. * If the receiver is detuned, there may not be any SIS current to start. - Adjust the tuners on the sidecab receivers to peak the total power level. (The 850 GHz receiver is tunerless.) * Turn the mixer backshort (//H// plane tuner) to zero and then turn it the other way until you see a peak in the total power or in the SIS current. This first peak is likely correct for the backshort. Attenuate the LO signal to keep SIS current about 9 µA. Relevel the total power as needed with the (Level) button on the hand paddle. This set the total power to 1500 mV. * Now adjust the //E// plane tuner. Turn it to zero and then turn it the other way. The total power will not vary for a long time--it is on a plateau. Then it will drop a bit, rise sharply to a peak, drop again, and rise to a second peak. It is best to use one of these two peaks. * Iteratively tune the LO muliplier micrometers to peak the total power. Then tweak the receiver tuning. * The tuners are hard to reach if the zenith angle is greater than 30° or so. Use the small ladder if necessary. Please take care when tuning because tuners connect to the mixer block by thin, fragile fibre-glass rods. - Once there is some sort of receiver response, check the magnetic field setting. * The aim is to set the magnet current to minimize the total power, which means that the Josephson noise is suppressed. * On the paddle, press the B-field button. This sweeps the total power versus junction bias voltage on the osciloscope. Increase the magnet current from zero while watching the oscilloscope. The curve should become more stable and smooth and the general amplitude can drop significantly. - Continue tuning the receiver for optimum performance. After peaking the total power, press the IVsweep mode softkey and observe the IV sweep on the Oscilloscope or the handheld display. Look to see if the slope is negative, flat or positive. If it is negative, stay in this IV sweep mode, and gently adjust the E-plane (or alternatively the backshort, thought the E-plane may work better) tuner until the slope changes to flat to slightly positive. (Refer also to the section on Tuning Hints or see the IVsweep menu or Receiver Control menu below for more detail on this. Set the LO attenuator so that you get about 8-10 uA of current on the junction. If you draw too much current on the SIS junction, the receiver may be unstable. If you see bad baselines, check the current - if it is much above 12 microamps turn the attenuator up to reduce the SIS current to something less than 12 microamps. - Finally it is best to put the SIS bias voltage on a flat spot of the P/V curve (dP/dV=0). This is critical because in this way the IF-fluctuations with time are minimized. - Press "B Field" on the handheld and there should be a sweep displayed on the o-scope sitting down on the floor. There are two humps and the left one is to be suppressed by adjusting the magnet current. If the sweep doesn't work then manually sweep the SIS voltage from 1.2 to 2.2 mV and check for the unwanted total power bump below the 2.2 mV operating point. - End "B Field" sweep and reconfigure for "receiver" and "constant voltage" ===== 345 GHz ===== * The LO chain consists of Gunn oscillator H121 and a frequency quadrupler, Millitech CDA-02-990548. The harmonic mixer is a Pacific Millimeter WM115. * Although there are harmonic mixers for the different bands, they actually work interchangeably! The high frequency ones with the tiny oval waveguides will not work at lower frequencies than their waveguide band but the larger size waveguides work fine at higher frequencies. * The most common cause of harmonic mixer failure is accidental connection of the Gunn oscillator bias cable to the harmonic mixer. Hence only lesser quality, silver coaxial cables are connected to Gunn oscillators and better quality, blue or purple coaxial cables are connected to harmonic mixers. ==== Frequency Multiplier ==== - After switching the multiplier bias to ''normal'', adjust the voltage to 4.0 V. - Open the LO attenuator (CW) until multiplier current is observed. Do not adjust the multiplier tuners yet. Do not let the multiplier current exceed 0.2 to 0.5 mA. Higher currents might damage the multiplier. - If multiplier current appears, adjust the multiplier tuners to peak the multiplier current. Simultaneously adjust the LO attenuation to keep the current below 0.2 to 0.5 mA. Continue until a peak in multiplier current is found at reduced LO drive and where there is a stable attenuator setting. (The attenuators have noisy spots that will make the SIS current unsteady.) - If multiplier current is absent: - Fully close the LO attenuator (CCW). - Reduce the 4.0 V multiplier bias to 1.0 V. - Slowly open the LO attenuator (CW) until multiplier current is observed. - If multiplier current appears, peak it using the two multiplier tuners. Simultaneously raise the bias voltage and close the LO attenuator to keep the current below 0.2 to 0.5 mA. The final bias voltage is 4.0 V. - If multiplier current is still absent when the LO attenuator is fully open (CW): - Close LO attenuator by two turns CCW as a precaution. - Adjust the two multiplier tuners while watching the multiplier current. When the tuners are close to the optimum, the multiplier current will increase very quickly hence the precaution of attenuating the input power. Once multiplier current is observed, peak it with the tuners. Simultaneously raise the bias voltage and close the LO attenuator to keep the current below 0.2 to 0.5 mA. The final bias voltage is 4.0 V. - At the peak there should be ample LO power. ==== Receiver ==== * Set the magnet current to 34 mA. * Optimum SIS current is 8-9 µA. * The //Y// factor should be 2--2.4. * If the total power goes to zero after releveling, then push the little button next to the red light on the warm IF to gain control over the internal attenuator. Use the push buttons on the mechanical digital readout to obtain a total power level of 1500 mV when looking at the ambient load. On the 345 GHz receivers, use the B-field display or Ocsiloscope (Total Power vs SIS bias) and/or the IV curve to check you are not getting the second photon step. Finally, check the IV curve to make sure the receiver is responding to the correct harmonic. (This is usually only a problem with the 345 GHz receiver). Refer to Tuning Hints. to make sure that there is no second photon step!! Finally it is best to put the SIS bias voltage on a flat spot of the P/V curve (dP/dV=0). This is critical because in this way the IF-fluxuations with time are minimized. ===== 650 GHz ===== //Caveat:// This receiver has respectable performance above 700 GHz. The atmosphere, however, is opaque above 700 GHz. So if the receiver is mistuned, the system temperature will be very high and observing will be very confusing. Use, therefore, the correct columns on the tuning sheet to set the Gunn oscillator tuners. Presently the tuning sheet has red arrows to the correct columns. This has been a recurring error even committed by the local staff! * The LO chain consists of Gunn oscillator H181 and a frequency sextupler, Virginia Diodes WR1.5x3R3 2-13. The multiplier is tunerless and self biased. * If it is difficult to lock the PLL, increase the sythesizer output power to 20 dBm. * Set the magnet current to 34 mA. * Optimum SIS current is 9-11 µA. * The //Y// factor should be 1.6. ===== 850 GHz ===== When the receiver is mounted, connect the the warm IF output to a labeled cable routed to the sidecab. This bundle also has the cable for the 109 MHz PLL reference. The IF cable in the bundle is connected to the old 492 GHz IF line at the dog house behind the side cab in the dome and is then routed through the IF matrix to the IF processor in the spectrometer room. The IF processor must be switched on or else a ''RF level too low'' message will be seen. This receiver has an internal calibration load. On the hand paddle, there is button labeled ''PHot'' or ''PSky'' depending on state of the load. If the button says ''PSky'' then the receiver is looking into the ambient load. If the button says ''PHot'' then the receiver is looking at the sky. \\ //Caveat:// Sometimes the internal load changes by itself. When this happens the receiver will not see the sky the system temperature will be very high system. Press ''TSky'' on the hand paddle to correct. ==== Local oscillator ==== * An SMA PLL is used. * The LO chain includes Gunn oscillator 186. * The Gunn oscillator frequency is about 135 GHz, higher than the frequency of the other Gunn oscillators at the CSO. * This Gunn has a very high pushing figure, above 200 MHz per volt. Therefore it is difficult to lock with the SMA PLL. The SMA PLL which is used on the 850 has been "tuned" to work with this Gunn source. In addition the gain of the SMA PLL should be set to 58. (Login to hau and run ~bin/csopll). There will be no lock indication or green light available, you must determine the lock status by observing the 109 MHz lock IF spectrum. A lock will be evident by seeing the amplitude of the spectrum rise up and there will be two side bands, one on each side of the central 109 MHz spectrum. Use a 5 MHz span to see them. * //Caveat:// This Gunn is locked by "bouncing" the source into the pull in range but not tuning past the hold in range. The pull in range is narrower than the hold in range. One must tune the frequency back short back and forth and be ready to halt when the lock happens. There is no automatic sweep like the default PLL in the side cab. * //Note:// Due to a lack of power available the 850 GHz LO chain has no isolator or attenuator. This results in the multiplier "pulling" the Gunn source during tuning. ==== Receiver ==== * Located on the side of the 850 GHz receiver which has the PLL plate mounted on it there is a chart. There are not many frequency points on the chart. One has to choose the one closest to where he or she is going to observe. Also at the end of this document is the tuning sheet for Gunn H186. An iterative procedure is used to bring the 850 GHz receiver into an active state. This will become intuitive with experience. * Refer to the chart, find a close data point and preset all the micrometers on the multiplier and Gunn to the values. * Lock the Gunn. * Set the SIS voltage to 2.0 mV * Watch for SIS current during tuning--try to attain 20 µA of SIS current. * Adjust multiplier tuner 1 (left top), 2 (right top), 3 (left bottom), and 4 (right bottom). * Look for any of the following: - SIS current - Gunn spectrum smearing - A Gunn unlock * When any of the above three conditions occur you are getting close to a good place because with case 1 you are getting LO power on the SIS junction and with case 2 and 3 you are "pulling" the Gunn, which means that the 1st resonator is coming into "tune". * Keep peaking current with the four multiplier tuners * You will cause the Gunn source to unlock during tuning! So just re-lock it and continue * Once you are obtaining reasonable current detune tuner #1 and dither the Gunn power back short so as to find a nice 109 MHz lock spectrum. Do be afraid to tune the Gunn power back short from end to end to find the best lock point, but don't touch the frequency back short. * First tuning can take 2 or more hours, be patient and keep starting over if you get lost * It is suggested that one should first observe a known source to make sure that the receiver is indeed tuned to the desired frequency. * If the receiver is tuned during the day, expect to have to relock the Gunn source during the first part of the night when you are open and pointing. Sometimes the receiver can still have a reasonable TSys and not have a Gunn lock. Once the receiver comes to outside ambient temperature then the receiver will work fine all night long with only occasional checks of the lock spectrum. You can set the spectrum analyzer up so you can open the blinds and have a peek at it. There are Two Multipliers for this Receiver: 720-840 GHz, RPG Physics, x12 , Self biased (1998 model, 4 tuners, refer to data sheet for settings) 600-660 GHz, RPG Physics, x6 Self biased (2000 model, 4 tuners, refer to data sheet for settings) Note also: The 850 LO chain have no attenuators! LO Bias: Do not exceed 4 mA on Doubler Do not exceed 2 mA on Tripler ===== LO multipliers ===== Note that all the multipliers are EXTREMELY sensitive to static discharges. These are delicate, expensive and hard to repair items. Please be very careful - wear a grounding strap, keep in contact with a chassis, do not power cycle anything unnecessarily. Also, be very careful when biasing the multipliers. Make sure you have the correct supply, that the current on the multiplier doesn't get too high (less than 2 mA for all), and that you do not ram the various tuners into their stops . Also, please note that we now have just one kind of attenuator (Please note: NO attenuator has been implemented in the 850 LO chain. The Attenuator is a gold colored unit from Custom Microwave, and the attenuator is fully closed with the knob all the way out(anti-clockwise). Multiplier Tuning instructions Make sure the 230-690 GHz attenuators are fully attenuated (knob all the way out). Tuning Procedure: Set the doubler and tripler voltage to Zero, then put the manual shorting switch to normal, so that the multiplier can be biased as follows 280-330 GHz, Millitech, X4: 2 Tuners, Bias at 2-3V 320-380 GHz, Millitech, X4: 2 Tuners, (1990 model, Both 4th, 5th harmonic), Bias at 2-3V 320-380 GHz, Millitech, X4: 2 Tuners, (1999 model, no 5th => more RF Power for 4th harmonic), Bias at 2-3V 600-660 GHz RPG Mulitplier(X6): 4 Tuners, Doubler at 3-5V, Tripler at 2V. 660-720 GHz RPG Mulitplier(X6): 4 Tuners, Self Biased 780-840 GHz RPG Mulitplier(X6): 4 Tuners, Self Biased 840-920 GHz RPG Mulitplier(X12): 4 Tuners, Self Biased Switch in the line with the UIP command and try to lock the Gunn. Set the Gunn frequency Tuner and Backshort. Instructions are laminated on the sidecab wall near the door. General multiplier settings: * Set the 650 and 850 Multiplier tuners (4) to the position suggested on their respective specification sheets. The rebuild tuners specification sheets are quite accurate! * The 345 Quadrupler has two tuners only. Tune the input tuner to maximize the varactor current and tune the output tuner to drop the peak varactor current by about half (couples power to the 4th harmonic). Be aware that the older Millitech X4 Multiplier puts out as much power at the 4th as 5th harmonic. Though it makes to convienient to tune the 345 GHz Rx in the 400 GHz Range, it also adds in noise from the other harmonic frequency!) * Once locked, slowly open the attenuator and carefully monitor the current drawn by the multiplier. Note: that the 650 and 850 multiplier now have an internal 100 kOhm protection. This means that w/o LO applied the multiplier bias will indicate a 10 µA/volt bias positive current. Once LO power is applied and coupled into the multiplier, the current will go negative, though on some multipliers very little effect on the bias (with increasing LO) is noticed. . Do NOT exceed 3 mA on the doubler and 2 mA on the tripler. * Tuning up the high frequency multipliers T1 is the doubler input tuner. Tune to maximize the negative current. If there is no current, lower the varacter bias voltage until rectified current starts to show. Be careful to Moniter the multiplier current at this stage. Since the 850 Multiplier chain has no attenuator, adjusting T1 is very likely to pull the frequency of the Gunn beyond the lock range of the PLL. Be aware of this!! You can still tune with the PLL unlocked because the Gunn frequency is only off by a few hundred MHz at most. Just lock the PLL when you are done tuning the receiver. T2 tunes is the doubler output power. Tune T2 to drop the doubler current by about 20-40%. This should increase the current on the tripler stage. T3 is the input tuner of the tripler, tune to maximize the current on the tripler. Note that T2 & T3 interplay. Moniter the current! T4 is the tripler output tuner. Tune T4 to drop the tripler current by 20-40%. This means that power is coupled out of the multiplier. However completely dropping the current to a local minima is most often NOT the best operation position. Now tune the receiver backshort to get some SIS current reading. From now on the process is iterative. That is, optize the SIS current by adusting the: Multiplier tuners Gunn backshort tuners Variable attenuator on the 230-690 Rx's Receiver tuners Bias of the multiplier The 850 has one extra adjustment, the focus of the multiplier. This micrometer moves the 850 LO chain normal to the dewar, and maximizes the coupling to the SIS mixer. == old text == [[http://www.submm.caltech.edu/cso/receivers/tuning.html|Overall]] and [[http://www.submm.caltech.edu/cso/receivers/instructions.html|specific]] tuning instructions. * Before tuning, check: * LO attenuator: fully attenuated * Multiplier bias: normal, set to 4 V * Magnet bias: off * Turn on one of the sidecab receivers. - UIP> za 44 (to allow the mirror to change the position for your receiver) - UIP> lo yourline /force /lock=default /if=1.5 (For side cab receivers, to set the line at the middle of the AOS) - check the tertiary mirror position. - Don't go beyond Vop of greater than 10.5 V. If you go beyond that, you'll be in the over voltage protection mode. To get out of the mode, do following. The steps below will be done in the phase lock control panel in the sidecab. i. Switch to "local" mode from "remote" mode. ii. Turn the Gunn off. iii. Turn Vop down for 2-3 turns. At the same time, watch the Vop reading on the reader above the HP synthesizer. iv. Turn the Gunn on. v. Switch to "remote" mode from "local" mode. And follow tuning instruction for tuning. If you get a message SYNTH FAIL(06) on the antenna status monitor, powercycle the HP synthesizer (turn off and turn on) located just above the oscilloscope that allows us to check the Gunn lock status. You turn off, wait for 10 sec or so, and turn on the power of the HP synthesizer, and go back to UIP, reload the line (UIP> lo yourline/force) and see if the error message disappears or not. If this doesn't work, wait for 5 min. If 5min doesn't work either, wait for 20 min to let the synthesizer cool down completely. After 20 min, turn the power on and load some other line which is lower than your line frequency first, then load your line frequency next. ===== Troubleshooting ===== * If there's no response when you are tweaking the tuner/power micrometers when locking the Gunn * Turn off the Gunn oscillator. \\ ''__UIP__> **lo /nogunn**'' * Set a slightly lower //V//op value, maybe about 13. * Reload the line and tune again. * If there isn't enough total power, about 1000 or more, even though there is a response from the receiver: * Press the //local// button on the golden total power box (external IF amplifier) attached to the receiver to put it in local control instead of remote control, * Set the attenuation by adjusting the toggle on the total power box. Watch for the total power response on the handset. * Once there is enough total power, press "remote" button and return to the regular tuning process using the handset. * 345 GHz sidecab receiver tuned to fifth harmonic * If the //I//-//V// curve on the handset indicates the frequency is above 400 GHz, the receiver is most likely tuned to the fifth harmonic. * To escape this mode, try: * Look for a log sheet indicating the receiver settings for the line frequency. Preset the backshort and E-plane tuner values. Then maximize the total power by optimizing the LO. * Select the //I//-//V// curve mode on the handset. Change the backshort and E-plane tuner to find the position where the first photon step shrinks. If it shrinks, the receiver is tuned to the fourth harmonic, which is correct. Leave the backshort and E-plane tuner there, then maximize the power by changing the LO multiplier.