I am having trouble calibrating the Raman. For reference, I'm on step 9 (https://www.open-raman.org/build/starte ... /assembly/), and I'm not getting a similar quality of image as shown in the Youtube tutorial. This is the image I got:
Screenshot 2024-06-23 151105.png (777.49 KiB) Viewed 1250 times
. This is the setup I currently have:
Screenshot 2024-06-23 151105.png (777.49 KiB) Viewed 1250 times
The image you get looks like the spectrum you'd get from a coherent source (which seems to be the case since you're using a fiber locator on the photo). I'm just surprised by how broadband the spectrum is for such source. For step9 you should use a neon spectral lamp instead.
The quality of the image you observe is directly linked to speckle noise which arise from coherent source. If the fiber is a multimode one, you can try shaking the fiber rapidly. If the integration time is not too short you should observe a much cleaner image (averaging of many speckle grains). If the fiber is single mode (which is a possibility when I see the thickness of the line), you need to couple to a multimode first.
The fiber optics cable I'm using is single mode. Could you clarify what you mean by coupling to a multimode first? Do you mean switch out the single mode for a multimode cable?
I didn't check if the slit was horizontal or vertical - I was rotating it until I saw that the laser light passed through. It might be positioned horizontally, so do you know how I can get the laser light to pass through if I turn the slit to be vertical?
Lastly, when you say the image is broad, do you mean that the line is super long (not based on its thickness)?
Yes, the neon light bulbs like these work great. You will have to solder a ~300kohm resistor to it and plug it to the main (110V/230V) -- take care not to hurt yourself.
What I'm surprised is the width of the light and the fact that you don't see it as you rotate the slit.
Do you use a collimator with your fiber?
The idea for this step of the alignment is to illuminate the slit to view it on the camera by reimaging the fiber on it. Since the slit is not perfectly centered (same for the optical axis), you need a very large fiber core (typ 1mm) to be sure to illuminate the slit
I don't think I used a collimator with the fiber optics cable. Is there a reason to add it because I see that it narrows the beam, which could make it harder to go through the slit right?
Yes indeed, but then I don't understand why you don't see the slit in the vertical position :-/
Nonetheless, you can't align with just one ray (laser) and need at the very least two -- the more the better, provided you can cleanly discriminate them. Hence the neon lamp. Other lamps could work but based on my experience neon is the best.
I presume you're simply imaging the slit out of the camera active area. Try rotating the grating when the slit is vertical. You should find the signal then.
Could you clarify more about why I need a multimode cable instead of single mode? In the tutorial video, I only saw you use a single mode.
Also, is there a reason where you switched from a neon lamp to a neon bulb connected with resistors in the video? I can probably just use the neon bulb with resistors for both the alignment and the calibration, right?
It's always a multimode fiber (I don't own single mode ones). Just I don't have one with a large core so I compensate with a XY stage but it's an inferior solution. It's really a question of getting light on the slit and not 100um next to it.
Don't forget to rotate the grating, it's possible that you are at the edge of the available span. Design is 540-650 but it's very sensitive to the actual angle.
You can use the neon source only. In fact, that's what I would do today. The other lamp is not a neon but a fluorescent one which has a different spectrum. One of the advantages is that it has two peaks near 540 nm to tune the orientation of the edge pass filter. The disadvantage is that I since learned there are dozens of variants of fluorescent lamp and it's very difficult to give reliable sourcing for the exact spectrum I have. People in Germany (nearby country to mine) were unable to find one similar to mine for instance!
