Model label: GalvoScanAssby
Check out the documentation for the Galvo driver box first.
Assembly instructions
Overview:
Y1 - Mirror with 30 degree stroke. This is the first mirror in the path of the laser. This moves in the same direction as the resonant scanner.
Y2 - Mirror with 40 degree stroke. This is the second mirror in the path of the laser.
X - This is the third, smaller galvo mirror.
From left to right: Galvo driver box; in the back driver boards for X, Y1, Y2. Galvo holder with galvos and alignment jig (note pentaprism). Galvo holder with galvos and alignment jig without Y2 mirror (note nacked shaft).
General:
Set up galvo driver box. (Note: be sure to set the input configuration for each galvo driver to full differential. A jumper will have to be removed, as indicated in the Cambridge 671 manual. The galvo drivers are being shipped in single-ended mode.)
Build alignment jig. Glue pentaprism on alignment rig
Mount galvo mirrors on shafts - this should be done with torque wrenches to specs provided from Cambridge Tech
Recommended torques for galvo mirrors and galvos
14 mm x mirror .05” screw socket, 2 inch-pounds torque
20 mm x mirror 1/16” screw socket, 3.8 inch-pounds torque
y mirrors 140 inch-pounds torque
x mirror 110 inch-pounds torque
For this entire procedure: zero the X, Y inputs to the galvos – this can be done with a 50 Ohm terminator or short circuit
Setup alignment laser
Wire up and turn on galvos
Do alignment
1) Slide in 30 deg stroke (written on back) and 40 deg stroke (written on back) and 3rd galvo into mounting block. Push galvos all the way into mount.
2) Attach first mirror to galvo. Remove the shafts the galvos come on for shipping. Push mirror all the way along shaft.
3) Couple a test laser beam into the pentaprism, level and parallel to hole pattern on table. Beam should be roughly horizontal (does not matter too much with pentaprism). Use target card with target in position of pentaprism. Adjust position of entire apparatus such that the back reflection goes along the same vertical plane.
4) Zero the inputs to the galvos – this can be done with a 50 Ohm terminator or short circuit
5) Hook up Y1 first. Power up galvo driver (make sure no faults - LED turns green). Note that X cable / Y cables have different endings. Make sure to put mirrors on before turning galvos on.
6) Set Y 1 voltage offset to 1.5V (position sensor for Y1; BNC on the galvo driver box) (which is approximately 6 V on the input).
7) Rotate angle of Y 1 mirror such that the reflected beam is coming out at an 8° angle up with respect to the table.
8) Do fine adjustment of voltage offset to hit 8° exactly (voltage in this instance was 1.511 V; 2nd time 1.587 V). GETTING THE Y MIRROR RIGHT IS IMPORTANT
9) Mount the other Y 2 mirror, adjust position so beam perpendicular; position singal is at -4.8mV
10) adjust X angle so that beam emerges level with table; note position signal is at 0.6 mV
22 Comments
Rob Campbell
Adjusting the offset pot allows me to go from +0.5 V to +1.1V I can't reach +1.5V. Suggestions please?
Daniel Flickinger
Natalia had this problem, and I think that the solution was that the input jumpers for the galvo drivers need to be reconfigured. They should be set for differential input. I'll find a manual and post the specifics of what need to be done in a minute, if you don't have a manual handy.
Rob Campbell
OK. I see. I have jumper 4-6 connected and that needs disconnecting. This is what the shipped unit looks like:
Daniel Flickinger
Can't find a manual right now... Yes, be sure that you set the input configuration of all of the galvo drivers to full differential, by removing the jumper indicated in the manual.
Rob Campbell
Yep. The jumper to remove is the one nearest the edge of the board.
Rob Campbell
If you need a torque wrench, I suggest the one from SwissTools (http://www.pbswisstools.com/en/tools/quality-hand-tools/torque-tools/product/pb-8320a-10-80/) over the ThorLabs one (https://www.thorlabs.com/newgrouppage9.cfm?objectgroup_id=6724).
Anonymous
The minimum offset was +2V with differential mode in our driver. However, the offset reached +1.5V with original common mode. Can we use common mode in this case?
Daniel Flickinger
Hmm...seems odd. The Janelia designed voltage divider circuit, which generates the y1 mirror control signal, should only work properly when the galvo driver is set to differential mode. The output of the voltage divider circuit is actually a "full differential" signal, with the negative and positive conductors swinging by the same magnitude, in opposite directions (in other words, the common mode voltage between them stays at 0 V). If the 671 galvo driver is set to "single-ended" input mode (with 2 jumpers, as shipped), it shorts the negative signal conductor to ground on the 671 board, which should cause all sorts of trouble. Please check the output of your voltage divider circuit one more time. If you have 0V for the Y command signal going into it from the front panel, then can you get the y1 signal going out, between the + and - signal conductors, to be 6V, by adjusting the offset knob?
Anonymous
Do you mean the voltage between 2 pins of Y1 output on the divider? It was 0 and did not change by adjusting offset knob.
Daniel Flickinger
There should be 4 pins on Y1 output from the divider. One is not connected. One is ground (this is probably in the middle of the three pins that are connected). The other two should definitely change relative to ground as the offset knob is adjusted. You should be able to get 6V between them. Something has to change on the divider output if the Y1 position sensor output changes when you move the offset knob. Does any of this not match up with what you're seeing?
Anonymous
The voltage between the pins shifted between +4V and +9V as I adjusted the offset knob.
Daniel Flickinger
Sounds perfect. If the jumper config of the Y1 galvo driver is set properly (see manual, should be differential mode, or see Rob's posts above), then everything should work right. If 6V command signal goes into the 671 board, then the position signal out should read 1.5V.
Anonymous
I double checked the manual and Rob's post and confirmed that jumper config is good. Somehow I get between 2V and 4.5V from the Y1 output.
Anonymous
We noticed that we were using Y1 driver as Y2 driver and Y2 driver as Y1 driver. After swapping them, we got 1.5V output. Thanks a lot!
Anonymous
We might have mixed up galvo drivers during assembling. Is there a way to tell which driver is for Y1, Y2 and X? The white adapter for cable coming from a lager blue capacitor is labeled with either '~~~X', '~~~Y', or '~~~~Z'. Is each of them for X, Y1, and Y2?
Daniel Flickinger
The power supply cables (from blue capacitor) can go to any driver. To tell which driver is for Y1, Y2, or X, match up the serial numbers on the drivers with the galvos themselves. The small galvo is X. The large galvo with 40 deg stroke (bumper to bumper) is Y2. The large galvo with 30 deg stroke is for Y1. It's easy to tell which large galvo has the larger stroke if you grab the rotors, spin them by hand (while the driver/servo is off, of course), and see which one turns farther.
Anonymous
Thanks, the serial numbers matched between drivers and galvos. Now, the position out signals from Y1, Y2 and X are +1.5V, +1mV and -2V, respectively. Should Y2 and X be -4.8mV and 0.6mV at this point?
Daniel Flickinger
If you are getting ready to do the alignment with the laser, then the position outputs from Y2 and X should both be near 0 V. I'm not sure why you're getting -2V from the X mirror. This shouldn't happen if you have 0V on the X command input. Check that. The exact voltage coming from the position output for all galvos before you do the alignment doesn't really matter. You get them approx right (~1.5, ~0, ~0), do coarse alignment by hand in their fixtures, tighten down the fixtures, and then do fine alignment by adjusting the PO pots on the drivers. Note that you have to do coarse and fine alignment for Y1 before you can mount Y2 in the block. Then after the fine alignment is done for all galvos, you see what your position output voltages are and write these numbers down. They are unique to your rig and the numbers -4.8mV and 0.6mV were for our rig and irrelevant to anyone else. You'll only use your numbers later is you notice some odd alignment and want to check if the galvos are still in the right places...
Anonymous
I noticed that X mirror is not fixed at certain position unlike Y1/Y2 even during power is on, and the position out signal can be changed between -8 and +8V by moving X mirror by hand. Is this normal? Also, where is PO pot on driver?
Daniel Flickinger
It's definitely not normal. Your X servo is not engaged. You probably see a red light on the driver, and you should see a green light. If the X galvo is plugged in right (sounds like it is if you're reading the position out), then I don't know what's wrong. The mirror is attached, right? If the mirror isn't attached, then the servo might fault out due to instabilities during start up, as the driver is tuned for the rotor inertia with the mirror attached. If the mirror is attached, the wires and connection between the galvo and driver are secure, all the other connections to the driver are the same as they are with your Y1 and Y2 mirrors, and you still can't get the X servo to engage, then it might be best to contact Cambridge. There is some troubleshooting info in the manual, too--there are some pins on the driver which report which fault condition you have.
The PO pot is in the line of pots that are next to the command signal connection. It should have a little piece of paper glued to it with the "PO" label. Some of these might be a bit tricky to get to because they're next to the power supplies in the enclosure.
Anonymous
I see constant red light on voltage divider. The color of LED on X driver changes as I rotate X mirror. The color is mostly green, but it changes to red when the position out signal is below -5V. I will contact Cambridge technology.
Daniel Flickinger
Hmmm...that's very weird and something that I've never seen. If the light on the galvo driver is green, then the servo should be engaged, and the mirror should resist rotation with a good amount of strength. Maybe there is a wrong setting on this somehow, such that it needs some additional signal to engage (perhaps it can be set to use a logic signal on one of its normally unused pins to engage or not).