B&W-tech Spectrometer & 473 module: Setup+Mods+Info

[paul1598419]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

There are several ways, but the easiest I know is with MS Excel. If you have the pixel numbers and the corresponding wavelengths, I could do it for you. I don't have Excel, but have good math skills. :yh:

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

There are several ways, but the easiest I know is with MS Excel. If you have the pixel numbers and the corresponding wavelengths, I could do it for you. I don't have Excel, but have good math skills. :yh:

Thats what I was thinking. I have read the calibration instructions which recommend using Excel. It would be nice if I could add the full method to my code though.

I will use my Argon and build the Pixel / Wavelength table. 6 Lines should give good calibration.

 

[hakzaw1]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

Hope none of this will be on the final... :undecided::shhh:


I'm just here to mod my 5&6 mW B&W 473s into 532 and/or 462-465 (07E) lasers for projector usage.
I don't have the spectro.:tinfoil:

but

:thanks:for doing all the maths...

 

[paul1598419]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

The only problem I see is that the spectrum is limited to 400nm to 650nmn. If we had a better program, we could get the full spectrum it was aligned to capture. That is why I really want the BWSpec OEM software. Have you been able to get a pause function in this one yet?

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

The only problem I see is that the spectrum is limited to 400nm to 650nmn. If we had a better program, we could get the full spectrum it was aligned to capture. That is why I really want the BWSpec OEM software. Have you been able to get a pause function in this one yet?

The software wont make any difference to the bandwidth. The 400-650 is due to the 1800 grating installed. You will need to change the grating to change the bandwidth.

My software will allow it to work with any grating. Just use the correct calibration values for the grating you are using.

I will probably leave the pixel numbers on the scale. What I will do is use the Mouse position to show the wavelength were its pointing to. A Box on the screen that shows were the mouse is and the wavelength according to the position. I might need to make a zoom on it if you need better accuracy though.

Pause function fully functional.

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

I found some time to do some more work on my Spectrometer software.

For now we can just use Excel to calculate the 3rd Order Polynomial values. It was pretty easy to use excel to calculate the variables I, C1, C2, C3. Link to the Ocean Optics PDF file explaining how to do it - Instructions.pdf


I have added the calibration to my code. I have calibrated my Spectro using 4 pointers. 473, 505, 532 650.

Wow, After plugging the data into Excel. Calculating the calibration data. Putting the data back into the Spectrometer software. It's bloody accurate. All 4 pointers read exactly what they should read.

I have added two text boxes under the graphed data. One shows Pixel Number and the second shows Wavelength. Just move the mouse to the point on the graph you wish to find the data and its shown under the graph.

One interesting thing I have found after calibrating is the bandwidth on my Spectrometer is actually 457 to 651. It looks like the Grating does need changing. The bandwidth is very small.


I have also made a full install package this time. It should install Mscomm32.ocx correctly. If not you may still need to register it manually.

Download my Spectrometer software

 

[Cyparagon]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

I grabbed the spectrum studio software if people want to try it. I just grabbed these from the program files directory. I also have an installation CD I can grab stuff off of if this doesn't work.

right click; save as:
https://www.dropbox.com/s/zv7h03b0c1ccguz/Spectrum Studio.rar?raw=1

Assuming the formula is correct
d*sin(angle) = wavelength

sin(angle) can be at most 1 when angle = pi/2, so maximum wavelength equals the distance between slits

1800 ln/mm = 0,0018 ln/nm
d = 1/0,0018nm = 555nm

This does agree with experimental results I've had with a 1000ln/mm grating.
What did I miss?

Are you trying to say the distance between the slits is the same as the minimum wavelength that can be transmitted by the slits? That is not the correct relation. What exactly is the relation you're proposing between "d" and the readable span of the system? You're forgetting about the distance to the sensor and the collimating mirror and the focusing mirror. There is a lot more geometry at play than you think there is.

"The amount of energy necessary to refute bullshit is an order of magnitude bigger than to produce it" so I'd rather not, but I will if you keep pressing the issue.

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

I grabbed the spectrum studio software if people want to try it. I just grabbed these from the program files directory. I also have an installation CD I can grab stuff off of if this doesn't work.

right click; save as:
https://www.dropbox.com/s/zv7h03b0c1ccguz/Spectrum Studio.rar?raw=1

Thanks for the software. It does run but it fails to find the Spectrometer.

Maybe it only works with the modified Spectrometers sold by Science Surplus store. It is communicating with the Spectrometer as I can see the LED blink on the Spectro card. But it says no Spectrometers found.

There is some very useful information in the help file. It has the compression algo for binary mode data. I should be able to mod my code to use binary mode now. It will speed up data transfer by about 3 times.

 

[paul1598419]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

@Mosc007

I wouldn't leave the pixel numbers at the bottom of the page. After calibration, it is normal to see the wavelength there. The pixel numbers at that point have no meaning. One thing i would caution is if you don't have a good calibration source for the lower and upper ends of the spectrum the formula doesn't work well to fill in extrapolated data. So, your outside lines, below 473nm and above 650nm won't be accurate. BTW, where did you get a 650nm source that's spot on 650nm? If it is a direct diode, it won't be. A neon source will work better for this. Once you are done, I'd love to have a copy of your new software, if possible. :thanks:

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

@Mosc007

I wouldn't leave the pixel numbers at the bottom of the page. After calibration, it is normal to see the wavelength there. The pixel numbers at that point have no meaning. One thing i would caution is if you don't have a good calibration source for the lower and upper ends of the spectrum the formula doesn't work well to fill in extrapolated data. So, your outside lines, below 473nm and above 650nm won't be accurate. BTW, where did you get a 650nm source that's spot on 650nm? If it is a direct diode, it won't be. A neon source will work better for this. Once you are done, I'd love to have a copy of your new software, if possible. :thanks:

I can change the scaling to wavelength. To me though you don't get the accuracy you will by using the mouse pointer to read it.

Hehe. The 650 is probably very iffy. It's a 0.5 mw commercial pointer. Who knows what it really is. It was just a quick test to prove the calibration formulas worked. The 473 and 532 should be accurate. The 505 could be off as well.

What I was impressed with is after calibration all 4 pointers now show exactly what the spectro was calibrated for. Ie, 473 now reads 473, 505 now reads 505, 532 now reads 532 and 650 reads 650. Which proves the calibration method works. I should have used a Neon for 633 but would have taken time. This was just a quick calibration to confirm it works.

The software is posted in my message about the calibration. Click to download it.

There will be another update soon with binary data mode to increase read speed substantially.

 

[paul1598419]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

None of the 505nm diodes I measured turned out to be 505nm. :crackup: If it is one of the ones I measured, you can probably use it to calibrate as many were confirmed by Singlemode in Berlin on his spectrometer at the University of Berlin.

 

[Mosc007]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

When I get time to do a serious calibration I will use my ML argon and a HeNe.

It does look like the grating will need changing. I am not happy with only a 200 nm bandwidth.

The 505 was just one of DTR's.

 

[paul1598419]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

Did you check for baffles in the optical bench on yours? They severely limited the bandwidth on mine, but after removal, seemed to be at least 300nm on mine. The only problems with changing the grating is the cost and it will have to be aligned. If you have never done it before don't think it will be as easy as the directions make it out to be. It could turn into the biggest nightmare of your life. I have done one myself and it was no picnic to say the least.

BTW, I tried downloading your software, but I haven't been able to get it yet. I may be missing something.

 

[diachi]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

Software downloaded and installed no problem on my end Mosc. Haven't tested it with the spectro yet but it does run just fine. Will update once I get a chance to test it.

Cyp, thanks for sharing that! Will take a look.

 

[CynicalBrad]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

I'd be cautious using 532nm lasers as calibration sources unless they are high quality units.. I recall a thread somewhere about a lot of the eBay ones being multi-line or not even 532 but off by a few nm such as I believe 534 due to cheap coatings on the crystals.

 

[diachi]

Re: B&Wtech 473nm unit / Spectrometer Mods & Info

I'd be cautious using 532nm lasers as calibration sources unless they are high quality units.. I recall a thread somewhere about a lot of the eBay ones being multi-line or not even 532 but off by a few nm such as I believe 534 due to cheap coatings on the crystals.

Yup, we've seen that with several of the cheaper pointers.