Research Lab Technologist
ALPS3 Laser Power Controller
||I am curious who uses what. Are these webpages a waste of time, or are they any help to others? Are the circuits, software and utilities appearing in other labs? Please send your comments or suggestions or what you have used (or not) or schematics of your version or pictures or anything! Email me, or be creative and send a postcard! I want to hear from the vacuum!||
NOTICE: This webpage and associated files are provided for reference only. This is not a kit site! It is a collection of my work here at the University of Toronto in the Physics department. If you are considering using any schematics, designs, or anything else from here then be warned that you had better know something of what you are about to do. No design is guaranteed in any way, including workable schematic, board layout, HDL code, embedded software, user software, component selection, documentation, webpages, or anything.
All that said, if it says here it works then for me it worked. To make the project work may have involved undocumented additions, changes, deletions, tweaks, tunings, alterations, modifications, adjustments, waving of a wand while wearing a pointy black hat, appeals to electron deities and just plain doing whatever it takes to make the project work.
Started March 2012 for Stefan Trotzky for Joseph's lab, eventually realized in March 2013. This is a remake of the ALPS project. This project stabilizes a laser's output power by monitoring the system's power and attenuating that output beam with an AOM (Acousto-Optical Modulator). The control voltage is logarithmic, giving at least a four decade dynamic range. There are two halves: the main unit and the remote logarithmic PD unit. The PD unit can be mounted on the experiment table, connected via a cable to the main unit.
Note that any reference below to laser amplitude means the 1 st order diffraction of the laser from the AOM.
The Rev:D version of the units have two significant changes. First, an RF AM input has been added to allow MHz amplitude modulation of the RF output. Second, there are two versions made of the unit: a log control version as described elsewhere and a linear version (the optical power is proportional to the control input). This linear version does not use the log PD amp but a regular amplified photodiode such as the Thorlabs PDA36A on the PD Monitor BNC. Athird but minor change with Rev:D is that the LED indicating saturation will remain on as long as the loop is saturated (AKA unlocked) and a few seconds afterwards.
The Rev:E version is very similar to rev:D . The major change is to remove the VCO duaghter board and mount the VCO directly on the main board.
Note that all references to logarithmic control can also be interpreted as linear control.
There are two internal calibrations: VCO frequency and VCO amplitude. Put a scope with 50 Ohm termination to the RF output. Set the unit in Manual and fully turn the manual pot CW (max output).
RF output : This is the low level RF output that is sent to an amplifier then the AOM. Its amplitude and frequency are controlled as described elsewhere. It has no specific frequency or amplitude because those parameters are determined by the VCO (voltage controlled oscillator) used and experimental requirements. That said, it will probably be 75-300MHz and maximum +10dBm. The shield for this output is at the unit's ground potential, it sets the ground level and should be the only ground connected.
RF Monitor output : This output is used to monitor the VCO frequency. Its amplitude is indeterminate and constant, depends on the VCO used. A connected frequency counter must be floating (not electrically connected to ground). If not used, this output should be terminated with 50 Ohms or the loop gain will be changed.
PD Monitor analog output : This monitor gives a logarithmic output of the PD over several decades. Nominal value for 0dB is 0V, offset errors are not removed. Specifically, scaling is 0.375V/decade. A DVM can be used to continiously monitor the average output, alternatively a 'scope will monitor it up to the 50KHz bandwidth. Note that the shield is connected to the unit's ground: be sure to float (no ground connection) whatever is connected so that ground loops are not introduced.
Auto : This is the normal operation. The Amplitude input sets the laser amplitude. This may be overridden by the Manual Mode TTL input to make the Amplitude input directly control the laser amplitude in an open loop fashion.
Manual: The Manual knob beside the switch directly controls the laser amplitude in an open loop fashion.
External: The Amplitude input directly controls the laser amplitude in an open loop fashion.
Pulse: The RF is switched full on for 1mS then held at a low value for 100mS. Effectively, the laser is pulsed at 10Hz with a 1% duty cycle. This is the same as frantically turning the Manual knob full up for 1mS then full down for 100mS.
Manual knob : When the Mode switch is in the Manual position, this knob logarithmically adjusts the laser amplitude over three decades. It is a ten-turn control so that the lower levels stand a chance to be seen.
Amplitude analog input (AM) : In the normal closed loop mode, this analog input logarithmically controls the amplitude of the laser [after the AOM]. Covers four decades with nominal value (0dB) at 0V in. Operating range is approximaltely -5V to +10V; the lower value depends on the noise of the PD (photodiode). Scaling is +3.33V/decade. Absolute safe range is ±10V. This input is galvanically isolated from the rest of the circuit, except that its shield is connected to the Frequency input shield. In Manual (AKA open loop) mode, as selected by Manual Mode TTL Input, this input directly adjusts the laser level.
Frequency analog input (FM) : This analog input adjusts the frequency of the RF sent to the AOM. Although generally positive voltages, the exact range depends on the VCO (voltage controlled oscillator) used. This input is galvanically isolated from the rest of the circuit, except that its shield is connected to the Amplitude input shield
RF AM analog input (on Rev:D only) : An AM can be applied to the RF output with this input. The frequency is generally >1MHz but must be above the frequency of the PID closed loop and above 30KHz. It is transformer coupled at 50 Ohms.
Manual Mode TTL input : A high level TTL will put the unit in open loop mode where the laser level is controlled logarithmically by the Amplitude analog input. It can be left unconnected and by default in the closed loop mode. It is galvanically isolated from everything else.
RF Disable TTL input : A high level TTL input will shut off the RF output under all conditions in all modes. It can be left unconnected and by default the RF output will be on. Note than the RF Monitor output is not affected. It is galvanically isolated from everything else. Measured delay from TTL in to turn off is 2.25uS. Measured delay from TTL in to RF back on is 65uS. TTL high must be >2.2V, low <0.8V.
RF Input : This input is an alternate RF souce to replace the VCO. This input must be manually selected by the RF Source switch (see below) inside the unit, otherwise the VCO is used and this input is ignored. When used, the RF Monitor output will have this signal, approximately -3dB. Be careful when using this input because it is directly connected to the circuit. Amplitude should be limited to +10dBm max. The shield of this input is directly connected to the unit's ground. Note than a ground loop may be created if the RF source connected to this input is grounded, preferably a floating source should be used.
Inside the Main Unit
RF Source switch : This switch selects the RF source between the internal VCO and the external RF Input connector. See "RF Input" above.
RF Level trimpot : This determines the closed loop gain by adjusting the nominal RF amplitude. Turning it CW will increase the RF level and hence the gain, conversely CCW decreases level and gain.
VCO Frequency trimpot : This adjusts the VCO frequency.
VCO Daughter Board connector : The internal RF comes from this daughter board. It plugs into the main board on 1mm pins. Several voltages are available. VCO supplies are regulated and/or filtered on this board, as is tuning voltage scaling and filtering. The RF output amplitude is not important ans it can be adjusted by the RF Level trimpot. Do not hotplug this board.
Feedback Network : These are four sockets to accept 1mm pins. They are on the error amplifier output and inverting input, and a ground. An optional feedback network can be plugged into these sockets to give custom PID response and/or nonlinear response.
Inside the PD Logarithmic Unit
I-ref trimpot and test point : Measure the testpoint voltage and adjust the trimpot for +1V. This sets the reference current to determine the 0dB reference level for the photodiode. Setting it lower will increase the sensitivity of the PD (by setting a lower reference level) and increase noise, the converse applies. Scaling is 1µA/V.
0dB Calibration switch : Flipping this switch disconnects the PD and applies a 0dB reference level signal in its place. With the switch flipped, set the Amplitude input to zero (short or terminate it), then adjust the Offset trimpot to the point where the laser amplitude (or RF output) toggles full on and full off. Set the switch back to normal when done.
VCO Daughter Boards
The VCO duaghter boards allow the use of various VCOs on the one unit, contigiously. Simply power down then unplug one and plug in another to change the frequency, although the RF level may have to be adjusted. Several supply voltages are provided as is the raw (but galvanically isolated) tuning voltage. Two boards have been made, one for the Minicircuits A06 package for VCOs such as POS-150. The other is the for the Minicircuits BK377 and BK276 packages for VCOs such as JTOS-150. If using a TCXO or OCXO, external power may be required for the heater or TEC.
|All RF Impedance||50 Ohm|
|Ext RF Input||+10dBm|
|AM/FM Common Mode||±5V|
|TTL Inputs 0||0.8V|
|TTL Inputs 1||3.0V|
Outside of this project, a laser is locked to a Rb absorption line. The laser output passes through an AOM driven by an RF signal, the 1 st order beam amplitude (roughly proportional to the amplitude of the RF) is sent to the experiment. Part of that beam is split off to a photodiode (PD). The PD output is compared to an analog command signal and the RF amplitude is adjusted to minimize the difference. However, the PD goes through a log amp and the RF amplifier has exponential gain control. At least in theory, this produces a linear amplifier with a logarithmic control input.
Refer to the above block diagram. Note that this project is a rat's nest due to the extensive succesive analog requirements. Good luck navigating the schematic. A VCO (voltage controlled oscillator) has its frequency set by a remote tuning voltage, the FM input. Part of the RF output from the VCO is sent to be monitored externally while the rest is sent to a logarithmic amplifier. The gain of this amplifier is set by the error amplifier, described below. After passing through a switch which can disconnect the RF, the RF goes to an external RF amplifier. This amplified drives the AOM, which causes some of the laser to appear at a PD. The PD current is compared to the PD reference current and logarithmically amplified. Meanwhile, a control voltage (the conditioned AM input) determines the desired PD level. A PID (proportional, integral, deriviative) error amplifier compares this control voltage to the log PD signal and adjusts the gain of the first (RF) exponential gain amp. This log-exponent loop has a net linear gain but allows the control signal to be logarithmic and give a very wide dynamic range. A minor refinement is the nonlinear feedback of the error amplifier which matches the nonlinear response of the AOM as it approaches saturation.
Note that the analog and digital control inputs are galvanically isolated. The PD is floating, in fact the PD bias voltage is on its shield. The circuit gets a ground reference only from the RF output and the RF monitor.
The VCO is on a daughter card with its own LDO (regulator) and with conditioning of the FM input and attenuation of its RF output. This allows a wide range of VCOs to be used just by unplugging the current VCO daughter card and pluggin in another.
On Rev:B, the front panel rotary switch from ElectroSwitch has to be modified because it is shipped as two positions but four are required. Remove the two nuts on the back of the switch and the cover. Turn the knob CW. The metal plate will be touching a stop pin. Move that pin two stops. Close cover. See this picture.
|Sorry, no more chance for asking direct questions, queries, broken links, problems, flak, slings, arrows, kudos, criticism, comments, brickbats, corrections or suggestions.|