Make Your Own Spectrometer for Under $1

[paul1598419]

Are you aware your device only has 2048 pixels to cover the spectrum? Depending on your grating, That's probably less than 3 pixels per nm. That second decimal is an artifact of the polynomial calibration regression, and should be rounded off. The average beam position would only need to move one pixel over to account for a drift of 0.5nm. You don't think it's possible ANY thermal effects internal to your device could account for a 1-pixel shift?

Of course I'm aware of that. I plan on changing the overall spectrum which at the moment is 340nm to 889nm. That actually gives me about .25nm per pixel, so it is close enough to get the ~.50nm change correct.

 

[Cyparagon]

it is close enough to get the ~.50nm change correct.

I'll ask the question again. Do you believe your device is immune to all internal thermal effects? If so, can you prove it? If not, how did you rule those out?

 

[paul1598419]

I don't believe any device is immune to thermal effects. It was in a room on my bench at the time and had been on for an hour where the temperature was not changing. If it were to be a thermal effect on my device, I would expect it to happen during a change in temperature, not during a period where its temperature was mostly at equilibrium. It would have to move over two pixels to be off. The optical bench on this spectrometer is cut from a solid piece of steel alloy and the optics are made of polished metal. The CCD is close to three inches long and I would be surprised to find this Ocean Optics "commercial" spectrometer would shift over a single pixel in a room at 25 dgrees C.

 

[lazeristasUVISIR]

Source?

Experience. I had many green lasers passed through my hands, and observed a similar multi line behavior with some of them. I would need to open quite old papers to look at spectra.

update:

Looking through papers:
10.2478/s11772−009−0028−5
with a multimode operation at 1064

 

[Cyparagon]

Experience.

I think you've misunderstood the issue I have with your statement. I think everyone knows Nd can be multimode. What you and paul are claiming are that the atomic emission lines change with temperature. I'm reasonably certain this is false.

 

[paul1598419]

I can tell you that my 6328 Angstrom lasers don't change as they heat up on my spectrometer, but the cheap 532nm I have checked twice, now, does.

 

[Cyparagon]

Even if I got the same results you did, I would still be suspicious of the spectrometer or other environmental variables, and I'm rather disappointed that you're NOT. The linearity isn't 100%, the power involved is very different for example. Even if you were to rule that out, 1 pixel isn't statistically significant, and the accuracy of your device isn't nearly as high as you think it is.

 

[dubious speculation]

I think you've misunderstood the issue I have with your statement. I think everyone knows Nd can be multimode. What you and paul are claiming are that the atomic emission lines change with temperature. I'm reasonably certain this is false.

I can tell you that my 6328 Angstrom lasers don't change as they heat up on my spectrometer, but the cheap 532nm I have checked twice, now, does.

A basic Chemistry or Physics course would tell you that atomic emission lines don't change with temperature.
The atomic emission spectrum for each element remains constant regardless of temperature.

If they did change it would render a lot of equations useless. The Rydberg formula which measures energy
differences between electron transitions simply would not work if the emission lines weren't constant.

The wavelength changes you're finding would have to be as a result of either measurement inconsistencies,
or possibly due to the DPSS laser emitting multiple wavelengths other than 532nm.

 

[paul1598419]

Even if I got the same results you did, I would still be suspicious of the spectrometer or other environmental variables, and I'm rather disappointed that you're NOT. The linearity isn't 100%, the power involved is very different for example. Even if you were to rule that out, 1 pixel isn't statistically significant, and the accuracy of your device isn't nearly as high as you think it is.

I know the accuracy of my spectrometer and did the statistical regression myself for the calibration. It isn't rocket surgery. I have 3.73 pixels per nm as it is set up currently. That means it would have to sift over two pixels to be a problem with my spectrometer. I don't claim to have an answer as to why this very cheap pointer is behaving this way, but I am not the only person to note it. CaliKirk found the same thing.

 

[lazeristasUVISIR]

I think you've misunderstood the issue I have with your statement. I think everyone knows Nd can be multimode. What you and paul are claiming are that the atomic emission lines change with temperature. I'm reasonably certain this is false.

I don't think that I wrote about the spectroscopic line (energy levels) changes with temperature. :tsk:

The Nd main laser line is not that narrow as with HeNe, thus it's possible that emitted line moves slightly or becomes double/triple due to a multimode operation.

However, Nd also have other weak laser lines around 1064 what could be generated when conditions are right. Their SH and SF create green lines what other observed here.

I wrote in one message line of Nd:YAG. Though most of the green lasers are Nd:YVO.

 

[paul1598419]

A basic Chemistry or Physics course would tell you that atomic emission lines don't change with temperature.
The atomic emission spectrum for each element remains constant regardless of temperature.

If they did change it would render a lot of equations useless. The Rydberg formula which measures energy
differences between electron transitions simply would not work if the emission lines weren't constant.

The wavelength changes you're finding would have to be as a result of either measurement inconsistencies,
or possibly due to the DPSS laser emitting multiple wavelengths other than 532nm.

I studied Rydberg's equation and the Rydberg constant over 40 years ago. Also, we studied Bamer's series, Lyman, Paschen and Brackett series. I understand these as well as Niels Bohr's contributions as well. I only stated I observed this happening, not that I could explain it. And, I'm not the only person who has now reported this.


Edit: Look, when i first found this cheap 532nm pointer's wavelength changing over time, I was suspicious too. But, when CaliKirk found the same thing using a diffraction grating I sent him and knew to be good, I checked it again. It was only then that I decided that it was likely the pointer and posted about it.

 

[paul1598419]

I have been having to go to Harborvieew in Seattle to get ready for surgery I'm having soon, so I won't be able to make the tests right away, but i now have a six line 532nm and a single line that increase in lines as it warms up to a maximum of five. I plan on testing them all and positing the results here.

 

[CurtisOliver]

I will be interested to see the results when you are able to. Good luck with your surgery Paul.

 

[CaliKirk]

Best of luck on your surgery Paul!

I'll be very interested to see the results on both of the lasers, but especially the one that the lines appear while warming. The 8 line I have, while cold, only has 3 lines and then makes its way to 5 to 6 then to 8 lines while warming. The 8 line is also the only 532 I have observed its lines blue-shifting while warming up and am very curious if either of those do the same?

 

[paul1598419]

I had a little time today and was able to get some quick shots of one of my 532nm which was right on the money. I checked one of the multi-line 532nm pointers that CaliKirk sent me and got four lines. The are 532nm, 538nm, 543nm and 546nm. I didn't have enough time to get more than this for now. I will need to wait until I get back from Seattle to get a long time shot to see if it is shifting with temperature. I wanted to try and get something for right now, though. Thank you for all the good wishes.

 

[CaliKirk]

Very interesting numbers! Too bad we couldn't use a filter and only have 546nm coming through. I'm definitely curious about the temperature shift on those pens as well if there will be a difference of WLs between the two pens.

Thanks for testing and I hope all is well with you Paul!