Bought my first new spectrometer..... but..

Hoping someone here might have experience with spectrometers.

I purchased a new spectrometer from ASEQ Instruments in Canada, trying to understand how to get accurate readings, unit is rated to a resolution of .35 nm. I've attached a screen shots of a spectrum from a mercury bulb, the lines aren't matching, over 1 nm different than the actual line wavelengths. Are my expectations wrong, or am I doing something wrong? Will using the wrong core fiber cause that amount of difference? I don't know what the core size of the Ocean Optics fiber cable I'm using is.

Chris

Update: ASEQ recommended shifting the Spectrum+ software scale (in this case, to the right as it's reading too high consistently on all lines) until they match and save the calibration into the flash memory of the spectrometer (after saving the original), I am now using a new OOptics 600 um cable, same result. I was asked if I'm using a cosign corrector or diffuser, nope.

Edit: This was a solution.

Here's where I bought from with a custom range and 25 um slit: http://aseq-instruments.com/ Under $1000.

I highly suggest migrating down to a 100u Cable, say 62/125...

Next up:

Calibration wavelengths do come from here:

https://physics.nist.gov/PhysRefData...ines_form.html

So I stick in Mercury (HG) I, from 360 to 600 and get flooded..

So lets jump to here:

https://spegroup.ru/upload/wikifiles/379695.pdf

Here are the strong lines of Hydragyrum:

365.015
404.656
407.783*
435.833
546.074
576.960
579.066

From your screenshoy

Your 436.6 should be 435.833
Your 405.3 should be 404.656
Your 580.3 should be 579.066

Looks like you need to do the polynomial calibration. In which case I'd get my hands on a HeNe and a Hydrogen lamp before proceeding to do the high end. Reason being their spectra is a lot simpler then an Argon lamp.
Do you have any gas lasers around?

Polynomial calibration needs Excel and a plug-in for the math, or you can do it with Spectrum Studio. How you switch to raw pixels in your software, to find your calibration, I don't know. I'd be very very angry if my unit arrived that far off.

If that's brand new, their calibration tech either sent you the wrong file or is a total newbie.


Here is Hg:

https://media.sciencephoto.com/image...ry_Spectra.jpg




From my notes based on the Mise and Practice of the BIPM Standards Lab...

RED HENE:
632.816 nm in air

GREEN HENE:
543.365 nm in air

YELLOW HENE:

593.932 nm in air (Pros use a vacuum path for calibration, for um, reasons)

Argon Ion:
Dominating Blue line (488)
487.98634 nm (AIR)

Hydrogen H alpha is 656.28 nm in Air.

Usually I find the lines from a lamp or laser source with a monochromator, pcaom, or dichroic filters.

A quick check is an older fluorescent lamp where the HG emission shines through the phosphor, then only the strong HG lines are visible and you have no doubt.

A handheld transmission grating really helps.

See attached..


Steve

Quote:

---

Originally Posted by Laser57

Update: ASEQ recommended shifting the Spectrum+ software scale (in this case, to the right as it's reading too high consistently on all lines) until they match and save the calibration into the flash memory of the spectrometer (after saving the original), I am now using a new OOptics 600 um cable, same result. I was asked if I'm using a cosign corrector or diffuser, nope.

Edit: This was a solution.

Here's where I bought from with a custom range and 25 um slit: http://aseq-instruments.com/ Under $1000.

---

Yes. This is the way to correct for that offset. Once the grating in these spectrometers is dialed in during manufacture they're usually locked down with epoxy. Any further wavelength scale calibration is essentially a pixel offset for the detector.


Steve's got some good advice in his reply, too.

Count Funkula knows what he speaks...

Yeah, I forgot to warn you. The grating position looks fine. The offset is best fixed in software. Do not try moving the grating.

Your choice is sliding the scale, if their software permits that. Or a Polynomial Cal using multiple sources, math, and a very, very, good wavelength table. Any place at the edges in the table, where you do not have a source, would result in errors at the ends of the scale. For most people, when self calibrating, that is at the red end, which is why I keep a specialized hene with multi-line optics around. Neon lamps of the indicator variety make good sources for orange-red-IR, but the issue is indexing them as their spectrum is very complex.

Steve

Thanks guys! Steve, I copied your post and saved to a document as well as downloaded all of your attachments. I do have a UV Pen-Ray and HeNe laser, I will look to get others if I find them at a reasonable price. I need to study all of this and will likely be back to ask more, you did a lot for me with this, appreciate it!

IDK what's up with this brand new spectrometer being so far off when the published resolution is .35 nm, I will try to find a 100 um cable too. I was hoping I was doing something wrong, ASEQ answered the 600 um cable I'm using should be OK but it's not the one they use, they have their own offering for that.

I never thought of moving the grating or adjusting anything, only software. I have a trial copy of Spectra+ from LaserTack, still learning how to use it, unable to understand it well enough. I suppose Ocean Optics/Ocean Insight spectrometer software would work too, but when I downloaded it and tried, not able to communicate with this spectrometer yet, however, I don't know what I'm doing yet either, this is all new to me.

Chris

Spectragryph is excellent if your hardware is supported.
Had a trial copy for solving an academic spectroscopy problem, it rocks.

There is an orphaned program I will link to tomorrow that has a calibration runtime in it, it probably will not talk to your unit but manual data entry will generate coefficients if your software supports file generation.. However I would think that since your under warranty, that if you give the factory the pixel vs line data, they might make a new file for you. I'm assuming your still overseas, and can't ship it back.

Getting to 0.35 nm accuracy is tough with a hobbyist environment..

I doubt the 600 u cable is hurting you too badly. You can run without the fiber and check the error.

Some Henes have a Raman "ghost" line at 650.1 pumped by 632.8 , you might want to take a peek and see if it's there in the output beam. It's great for calibration but may only be a few microwatts, the spectrometer might just '
see" it with a long integration time.

Rarely red Henes have a microwatt of so of a 640 nm laser line, and the sidelight through the glass of the tube has a beautiful, known, spectrum of narrow line width Helium, Neon, and intermediate lines.

Steve

I downloaded Spectragryph, appears LaserTack has rebranded that as their own, under license. However, I cannot get Spectragryph to connect, it looks identical to LaserTack's Spectra+ though, but the German version costs a lot more. The ASEQ spectrometer I have is also being sold under the LaserTack branding at twice the price, maybe because he does a better calibration.

I just got done editing my post.

No clue on Lasertack. I'm doing good if I can keep up with what OO and BWTEK are doing.

I have a pile of BW Teks I need to calibrate this weekend. Hence my recent interest in calibration.

keep in mind anything DPSS is usually 1.1 nm wide, so 532 561, and 472.1 can be poor choices to calibrate from.

Argon, HeNe, Krypton lasers, and low pressure discharge or glow lamps like your argon/mercury unit are a good idea.

HID ie high pressure mercury, high pressure sodium lamps, and HiD/ metal halide have beautiful spectra, but pressure broadening turns their lines to mush and makes band spectra.

The exceptionaly rare low pressure sodium streetlamp has a beautiful sodium doublet without much pressure broadening.
especially when warming up. I miss those, but they got rare when HID got better in the US. They are the really long soft yellow street lamps, was dominate in Europe and Africa.

Steve

I found some low pressure sodium lamps on ebay and bought a box of them last year, but haven't purchased a fixture with a ballast transformer to fire them up, I need to find something. Edit: Here's a seller on ebay with some low pressure lower wattage bulbs and a lot of other interesting light producing sources in Ukraine: https://www.ebay.com/str/vannadiy This is the seller I bought my last bulbs from: https://www.ebay.com/itm/116056147129


I have a HeNe producing half a mw I will measure today and see what I find. I have a bunch of those BW units from a Raman medical device, if that is the spectrometer you are working with, I have those to work on someday too.

Thank you for the heads up on DPSS, I didn't know they weren't suitable due to having too wide a line width.


Chris

If your trying in the dark and have nothing, DPSS is fine if you find its center and ignore a weak line tucked in along its 1.1 nm bandwidth.


Steve

The sales representative, owner, engineer or whatever his position is responded to my email question why the unit is not calibrated, I received the following response:

"The spectrometer was calibrated with our fiber and our diffuser (cosine corrected adapter SAM905_AD1). I can not grantee the accuracy with
another vendor optical fiber."

Is this reasonable? I responded if I needed specific equipment in order for the calibration to be valid this should be called out on the sales page, or provided. It wasn't. Anyone have an opinion on this? Is his statement reasonable or something they should have put on their sales page or the gear included?

http://aseq-instruments.com/LR1.html

Quote:

---

Originally Posted by Laser57

The sales representative, owner, engineer or whatever his position is responded to my email question why the unit is not calibrated, I received the following response:

"The spectrometer was calibrated with our fiber and our diffuser (cosine corrected adapter SAM905_AD1). I can not grantee the accuracy with
another vendor optical fiber."

Is this reasonable? I responded if I needed specific equipment in order for the calibration to be valid this should be called out on the sales page, or provided. It wasn't. Anyone have an opinion on this? Is his statement reasonable or something they should have put on their sales page or the gear included?

http://aseq-instruments.com/LR1.html

---

use a neon light from like a power strip. You’ll have more lines than you know what to do with. Unless that fiber is magic non linear stuff or really long I can’t see how the fiber affects wavelength. Focus yes.

Seems to me this guy is throwing out excuses of why this instrument is inaccurate.