January 24 Through February 4—TUCSON, ARIZONA: Annual show
Welcome to the GemologyOnline.com Forum
A non-profit Forum for the exchange of gemological ideas
It is currently Thu Mar 28, 2024 10:51 am

All times are UTC - 4 hours




Post new topic Reply to topic  [ 41 posts ]  Go to page 1, 2, 3  Next
Author Message
 Post subject: Spectrometer DIY Kit
PostPosted: Mon Jun 06, 2011 2:56 pm 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
I'm thinking it would be fun to play with this one.
What makes it interesting for me is you need to align and calibrate it yourself.
They seem to have quite good alignment instructions. Having read all the spectrometer threads here in GO one thing I know I don't know how to tell if this one is total waste of money for gemological purposes.

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Mon Jun 06, 2011 8:18 pm 
Offline
Platinum Member
User avatar

Joined: Tue Jan 02, 2007 8:20 pm
Posts: 2659
This unit has a grating that yields a fairly narrow spectral range onto the ccd chip. The auction says that the typical range is 250 nm.

One gemological thing you could probably do effectively would be to center the coverage on the 415nm line and use it for diamond color grading. Provided you have color masters to measure the spectra of.

Or you could get two of them and feed them with a so called bifurcated fiber. have one range from 390 to 640nm and another range from 550 to 800 nm. You would then cover the whole visible range with over lap.

I have no idea how good or bad the software that they provide is. Or how good B&W Tek spectrometers are compared to better known spectrometers such as Ocean Optics or Stellarnet or others that are available.

I would not spend money that you cannot afford to lose. Purchases like this can be big wins or they can be total wastes. Very hard to tell in advance.

http://www.bwtek.com/product/spectrometer/index.html
http://www.science-surplus.com/products/spectrometers
http://www.science-surplus.com/products ... -resources


The manufacturer does not seem to have any models with serial interface. They are all USB now. so these are probably a bit out of date. Also aligning units like this may involve placing or removing order sorting filters (since these were said to be for some specific medical purpose.) The auction warns that "aligning the spectrometer is not trivial" I recommend heeding that warning. You might end up with a $200 plus shipping paperweight.

You will want to have some sources of accurate spectral lines such as a gas laser (diode lasers are not accurate enough in wavelength) a sodium lamp or a swuv tube or other mercury argon source. A neon tube would help too.

Let us know how you do with them if you buy them.


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Thu Jun 09, 2011 6:05 am 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
Thanks for taking time to give good answer. I decided to buy it just for getting my hands on spectroscopy. I have no unrealistic expectations what can be achieved with this unit. How many hours you can have fun with $200 anyway? I asked the seller about possibility to add sorting filter. He promised to include some gel filter material and spacers for making filter to reduce the 2nd order stray.

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Thu Jun 09, 2011 12:13 pm 
Offline
Platinum Member
User avatar

Joined: Tue Jan 02, 2007 8:20 pm
Posts: 2659
You will probably do fine with this unit, and as you say have fun too.


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Wed Jun 22, 2011 4:53 pm 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
Alignment was surprisingly easy task, just taking little bit time. Actually I aligned the optical bench three times with different wavelength ranges just to get experience.

I can squeeze 275 nm range out of the box, ranging barely VIS area (from 405 to 680nm). The detector has 2048 elements and 16 bit A/D. The slit is 50 microns. Typical signal to noise ratio seems to be about 800:1. I calibrated the scale with gemological mercury UV lamp. I spent lot's of time googling the correct wavelength for the Hg peaks on red :oops: . These were actually 2nd order strays from UV/violet area and proved the need of high order filtering.

The software is easy to use but very basic. It has dark signal reduction, single or continuous spectra reading, adjustable integration time, averaging and zooming features. It does not have any tools for correcting the spectra against the light source. Good thing is that you can easily copy the spectra data to the clipboard. I made small .net program for correcting the spectra and importing it to SpekWin32 program.

RS323 port is slow. It takes 350ms to read spectra no matter how much smaller the exposure time is. Anyway, the response feels fast enough when seeking the max signal. Please correct me if i'm wrong but taking averaged spectra can take same time with faster usb spectrometers if exposure time goes up to more than 350ms.

After all, I'm very pleased. This is nice and affordable way to get into spectroscopy :D :D :D



And now couple of rookie questions for our experts:

The seller of this unit has 900l/mm gratings available yielding 365 - 910nm and 2nm resolution. I'm tempted, but dunno if it's good enough for gemological work. What do you think is the limit of acceptable resolution? Currently I can get 1nm FWHM and and about 9 pixels/nm.

Only light source I currently have is 12V halogen fiber optic. It's very difficult to find halogen bulb having enough power at 400-450nm area. The spectra is just too noisy at this area to get any clue about absorption features of the material. I have tried to find seller for SoLux 5000K bulbs in europe with no success. If anyone have easy access to these I would be very pleased to buy or trade it into something.

There seems to be no universally accepted light collection method for gemological purposes. I see integrating spheres, half spheres, pseudo-spheres, microscope setups with and without immersion liquids, cosine correctors with and without collimation and setups with nothing but two ends of the fibers and a stone between.
Obviously most of the setups are for serious scientific use having serious price tag. What would be good and affordable approach for average gemmo who is just collecting spectra library for his own fun and future ID work.

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Wed Jun 22, 2011 8:22 pm 
Offline
Platinum Member
User avatar

Joined: Tue Jan 02, 2007 8:20 pm
Posts: 2659
:twisted: :twisted:

Sounds like SOMEBODY needs to buy hisself either a xenon lightsource or a combo tungsten halogen and Deuterium lightsource.

You COULD get a Solux bulb. That is what Marty Haske uses in his SAS2000. But anyone ought to be able get a little more blue out of a qh bulb by turning the voltage up. Solux , like Ottlight have not rewritten the laws of physics. All qh bulbs transmit blue and even UV just fine. You just need to turn the voltage up. This shortens bulb life. But 100 watt qh lamps have been used to for blue excitation fluorescence microscopy as a cheaper alternative to the arc lamps that used to be used.

Arc lamps are noisy but there is so much signal you just average it out.

You could use a white or blue LED source. Ocean Optics has a bunch of literature on colored LEDs for use in spectroscopy (not whites but they have a huge peak in the blue and its should be pretty stable.

Do they want to sell you a new grating or another spectrometer with a different grating? Have you used the order sorting filters they sent.

This spectrometer ought to be an good for diamond color grading.

Gene


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Thu Jun 23, 2011 12:55 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
wow, mikko, sounds like you are really dug into it. I can add a few comments, but first... congrats on your progress.

Checking my own spectra, you really need less than 1 nm resolution to see that ruby fluorescence at ~ 693 nm is composed from two separate peaks. But resolving the ruby lines is the "gold standard" for gem spectroscopy. Beyond ruby and some other examples of chromium lines, nothing else needs better than 2 nm resolution. So if you don't have the resolution to clearly separate ruby lines (i.e., less than 1 nm)... might as well increase wavelength range.

If your spec software can output tab-delimited text files, I can send you a set of excel templates and a step-by-step guide to using them, to subtract out dark and divide out initial light files to create real transmission spectra averaged over as many scans as your computer can handle. As I've mentioned before, 20 - 30 scans should be enough for any gem purpose. By collecting the spectra separately (if you are inclined) you can check all your spectra, light, dark, light-dark, and gem, for consistency. But if your system works for you, and you are happy, certainly I'm not one to encourage the use of excel.

Oh, did I mention, I never trust in-software suppression of dark signal or averaging of signal? I output all individual spectra, and I do all my own averaging of spectra, dark subtraction, and input light division in subsequent software (much better in mathcad than excel), so I cannot comment on output times. Although it looks like I'd be waiting forever if I was using your RS232.

As for light sources, welcome to the conundrum everyone faces. Most light sources are designed to protect you from the least little bit of stray UV. So normal tungsten outputs can be cancelled by UV-absorbing glass that actually strays into the violet. If you find the right halogen lamps (bulbs), the glass doesn't filter the high frequency, and as Gene suggests, you can overdrive it to get more violet light... the fundamental laws of physics say "more power in, more power out at high frequency." So if you have one of those lamps that don't cut off at 420 nm, just increase integration time a factor of ten. Forget about UV unless you want to get into the expensive lamps... deuterium-halogen, xenon (swoon), or some other fancy options.

Personally I've not ever looked at solux lamps. So I can't comment on their application. But if Marty favors them, I'd be willing to give 'em a spin.

To me, the perfect light source for the visible range would be a halogen lamp supplemented by a violet LED, but I haven't found a standalone unit. Even using a "white" LED probably isn't really better than a halogen, with its big blue spike and cutoff in the violet.

As for light collection, there are lots of fancy systems out there, as you mention. But for normal transmission spectra, where you aren't concerned with determining the number of absorbers in the species, there is nothing particularly wrong with the straight-through method of fiber optic source on one side, gem in the middle (with table turned toward source), and spec fiber on the other side. I've set up a couple little benches that do this, and they do fine by me. Even with this setup if I was hard pressed, and the stone has the right shape, I can come up with a pretty good estimate of absorbers. One can argue about fraction light reflecting from gem surfaces in this arrangement, but if all you care for is the relative scale, and if enough light is getting to the spec, then this method will do. One does have to be a bit careful that all light entering the spec's fiber did traverse the stone, but careful shielding will do the job.


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Thu Jun 23, 2011 1:50 pm 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
Quote:
All qh bulbs transmit blue and even UV just fine. You just need to turn the voltage up. This shortens bulb life.


My halogen source has "boost switch" which helps a bit but not enough. Actually it seems to be "undervoltage switch" giving 10.3V / 12.1V out under the load of 100W bulb. SoLux bulb could be easier than messing with the transducer.

Quote:
You could use a white or blue LED source. Ocean Optics has a bunch of literature on colored LEDs for use in spectroscopy (not whites but they have a huge peak in the blue and its should be pretty stable


I found nice combination of halogen and 3 led sources, but it would be major PIA to install and adjust it all into the box.

Quote:
Do they want to sell you a new grating or another spectrometer with a different grating? Have you used the order sorting filters they sent.


Just an extra grating hopefully pre-installed into mounting bracket.
Nope. Need to find acetone resistant mask tape to make it. First attempt was miserable failure. Haven't tried ethanol or isopropanol yet, they may dissolve the gel material as well.

Quote:
So if you don't have the resolution to clearly separate ruby lines (i.e., less than 1 nm)... might as well increase wavelength range.


Now that was an exact answer to vague question :shock: I'm gonna update the grating just to be able to record little bit more from both ends of the VIS range.

Quote:
Oh, did I mention, I never trust in-software suppression of dark signal or averaging of signal? I output all individual spectra, and I do all my own averaging of spectra, dark subtraction, and input light division in subsequent software


The software does not take avg from the spectrometer because it goes unresponsive for the time making multiple measurements. Instead it asks one spectra at time from the spectrometer and makes averaging by itself. For the correction I'm fine with the formula you gave earlier here in GO. The output has all data needed for each pixel: integration time, nm, averaged output, summed output and dark signal. Spekwin has nice tools like normalization and spike removing. Please, send me your excel templates anyway, I'm sure they will teach new things I wasn't aware of.

Quote:
for light collection, there are lots of fancy systems out there, as you mention. But for normal transmission spectra, where you aren't concerned with determining the number of absorbers in the species, there is nothing particularly wrong with the straight-through method of fiber optic source on one side, gem in the middle (with table turned toward source), and spec fiber on the other side.


I would like to know more details about SSEF spectrometer sampling. The photos at their website doesn't tell the whole story. Not that I even dream about being able to copy it but it could be a good exemplar.

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Fri Jun 24, 2011 8:51 am 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
mikko wrote:
For the correction I'm fine with the formula you gave earlier here in GO.

Glad someone found it useful!

mikko wrote:
Please, send me your excel templates anyway, I'm sure they will teach new things I wasn't aware of.


Send me a pm with your email addy. There are several files, so if I can't get a good compression, I may have to send them in multiple emails.


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Tue Jun 28, 2011 4:59 pm 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
Many thanks Brian. I found small error at my formulas. The best thing was to learn efficient order for taking spectras.

Please have a look at this SSEF UV-VIS spectrometer brochure: http://www.ssef.ch/fileadmin/Documents/ ... _small.pdf

I have plan to make affordable imitation of this setup using Thorlabs optomechanincs "big boys legos". Do you think I'm interpreting the photos right?

:-k :-k :-k THIS IS GUESSWORK!!! The light source is on the upper part of the tube having probably cooling fan. The power source for halogen is in the separate box. On the front of the halogen source there is hexagonal blue filter, probably halogen to daylight filter and/or IR filter for reducing heat. There is 5 or 6 varying wavelength UV LEDs around the halogen, inclined so that their intensity is concentrated on the center area of the white UV- diffusor (quartz ?). The sample aperture is controlled by lever activated iris. There is some UV-VIS transparent plate right behind the iris. There is no way to tell from the photos about signal pre-processing before it is coupled in to the fiber. Somebody already told me it is something one would call "pseudo integrating sphere". Is that term used for half spheres? There may also be a beam shutter (for taking dark spectra without turning light source off). :-k :-k :-k

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Tue Jun 28, 2011 6:00 pm 
Offline
Gold Member
User avatar

Joined: Mon Apr 16, 2007 12:05 am
Posts: 1499
For the most part, your interpretation of the photos looks correct.

Yes, the halogen source is buried somewhere at the top of the upper black tube... a bare lamp not connecting to a fiber to collimate the light. At the bottom of the document, they show the UV LEDs in a ring around the bottom edge of the tube, tilted to illuminate a region on the central axis of the tube. That blue plate in the middle probably is a "daylight" filter for the halogen source, but I don't see any real need for it.

The metal ring marked "UV Diffuser" probably holds a diffusion plate and it may have an iris aperture.

The point of the light source seems to use several sources whose light is mixed together by the diffusion plate to provide a spread of light across the entire stone, rather than using a collimated source entering a specific spot on the stone. To each their own, but this essentially rules out any kind of "path-length" estimate for use with absorption spectra.

It seems to me that the sample table has a fixed opening where you slot in those little black rings to hold the stone. But it may have an iris aperture as well.

I don't know what is below the sample table. Maybe there is a half hemisphere opening upward, and then the spec's fiber at the sphere center pointing downward to collect light reflecting from the hemisphere. This makes some sense when it is being used with the diffuse light source. If a collimated source was being used, then you'd just put the spec's fiber directly beneath the stone pointing upward to collect the light directly.

As for Thorlabs "optics legos," I use them for everything.


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Wed Jun 29, 2011 1:49 pm 
Offline
Site Admin
User avatar

Joined: Fri May 12, 2006 11:24 am
Posts: 7523
Location: Rome, Italy
Brian wrote:
I don't know what is below the sample table. Maybe there is a half hemisphere opening upward, and then the spec's fiber at the sphere center pointing downward to collect light reflecting from the hemisphere.


uhm :-k i don't think there's an hemisphere there, more likely a right-angle mirror like this one connected to a collimator just in order to avoid a too tall structure (fiber bending..), but this is a guess only.....i'm not sure that will be their deifinitive version of the chamber setup........

ciao
albé

_________________
GemmoRaman-532 - GemmoFTIR - GemmoSphere - EXA


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Sun Jul 03, 2011 4:44 pm 
Offline
Valued Contributor

Joined: Wed Oct 22, 2008 5:00 am
Posts: 344
Location: Finland
I made cosine corrector from teflon. It is working as expected. Of course I have no clue how much better would an integration sphere be.

There has been lot's of discussion about using 405nm blue ray laser to reveal visible range luminesense of many gem species. When playing with the PL spectrums I noticed teflon itself has a remarkable PL spectrum when illuminated with the 405nm laser. That may interfere with the gem spectra.

What would be the PL properties of barium sulphate or spectralon? How about teflon? I'm not sure about purity of the material I used.

_________________
GemmoRaman GemmoFtir GemmoSphere


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Mon Jul 04, 2011 5:22 am 
Offline
Site Admin
User avatar

Joined: Fri May 12, 2006 11:24 am
Posts: 7523
Location: Rome, Italy
Hi Mikko,

have a look here, it seems the material used for OO cosine correctors is opaline glass or spectralon, depending about the uv region effectiveness...
ciao
alberto

_________________
GemmoRaman-532 - GemmoFTIR - GemmoSphere - EXA


Top
 Profile  
Reply with quote  
 Post subject: Re: Spectrometer DIY Kit
PostPosted: Mon Jul 04, 2011 3:39 pm 
Offline
Platinum Member
User avatar

Joined: Tue Jan 02, 2007 8:20 pm
Posts: 2659
Spectralon is a trademark for a sintered teflon used to make integrating spheres. By intention reflective rather than transmissive.
I wonder if Ocean Optics isn't fibbing a little bit or maybe even the owner of the trademark is fibbing too. A cosine corrector is translucent teflon not sintered at least not on any of the ones I have seen nor do any of the pictures look like the sintered material. But if you own a trademark you can do whatever you want with it or license someone else to do whatever you want to let them do.

Some teflon I have seen is quite blue fluorescent. You may have an example like that. My impression is that the slipperier the teflon , the more fluorescent. For spectroscopic purposes , obviously , you don't want fluorescence or photoluminescence. ( I wonder if the fluorescence isn't either lubricant from machining , or mold release agent from molding. Most of these kind of hydrocarbons DO fluoresce. Perhaps you should try cleaning the piece a few times with paint remover and solvents)

Good opaline glass used to be easy to buy at camera stores but like water white glass , is not that easy to find anymore. However for diffusing light white acrylic plastic is actually more efficient at scattering light than opaline glass is. The only thing you need to watch out for is that it has a pretty flat spectrum to begin with AND that you keep an eye on it over the years because unlike opaline glass, plastic (except teflons) can yellow with age , due mainly to attack by atmospheric oxygen and ozone. But you can get an equal diffusing effect with a thinner layer of plastic than with glass. (the glass can also resist far infra red better but if you are after a fiber optic most of that will be filtered out anyway) The thinner plastic results in more diffuse light actually being transmitted which could make a difference if you are scratching for signal.

If you are only working in the visible you could make immersion contact with your stone for either or perhaps both of your fibers using non fluorescent microscope immersion oil.
You would need to take a reference spectrum without the stone.

The black interior of the overpriced SSEF goodie can be duplicated with flat black paint or with black velvet or black flocking which is available from many sources. On my other favorite forum http://www.photomacrography.net/forum there is discussion of this and sources.http://www.photomacrography.net/forum/v ... hp?t=13788

Barium sulfate does not fluoresce nor luminesce in civilized spectral ranges.

Also I don't think you really need a cosine corrector since you are doing mainly transmission/absorbance measurements with the possible reflectance measurement of an opaque specimen


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 41 posts ]  Go to page 1, 2, 3  Next

All times are UTC - 4 hours


Who is online

Users browsing this forum: No registered users and 13 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Gemology Style ported to phpBB3 by Christian Bullock