Looking for an economical calibration light source at ~ 400nm

[rhd]

I recently purchased a spectrometer from Science Surplus, and have calibrated it using known stable DPSS wavelengths (473, 532, 561, 589 and 593.5). With the spectrometer's default diffraction grating, I can get a range of a bit over 200nm. While the DPSS approach has worked great so far, I'm looking to re-calibrate my spectrometer to cover the 400 to 600 nm range.

This means that I really need to locate a calibration light source, of some sort, in the very low 400s (or even in the high 300s). Can anyone brainstorm a creative calibration light source that would fit the bill? (my bill, by the way, is something I hope to keep to under $30). The light source doesn't need to provide anything other than the low 400 spectral line. I can use a 473, 532, and 561/589/593.5 for the rest.

[slicklasers]

use a 405nm blueray diode...

[DZ]

use a 405nm blueray diode...

Well in a perfect world that would work, but since it's not, a blueray diodes isn't going to work. Not sure what would work best for this. I know there's argon and krypton lines in that region but not sure about dpss.

[slicklasers]

He asked for an economical way... his 532 and 473 will have a shift of +/-5nm so IMO the blu-ray would be the cheapest way at under $100...

[DZ]

532 and 473 will have a shift of +/-5nm

seriously-meme.jpg

Maybe I'm wrong, but I've never heard of DPSS lasers having a drift of +/-5nm. Maybe in the pump diode which results in a loss of power, but not a drift, as far as I know...

[rhd]

seriously-meme.jpg

Maybe I'm wrong, but I've never heard of DPSS lasers having a drift of +/-5nm. Maybe in the pump diode which results in a loss of power, but not a drift, as far as I know...

No, you're not wrong. DPSS doesn't drift +/- 5nm. Lol, I think that notion is the product of the same Chinese laser sellers that will advertise "Portable Blue 532nm 10,000 mW Square Beam Laser 1x 16340". In other words, copy and pasted specs for cheap lasers, that they often indicate a +/- DPSS drift even though that's not backed up by science or reality. If the 808nm pump wanders in wavelength, that only impacts the pumping efficiency. The resulting output wavelength remains incredible stable, with less than 1nm (less than 0.1nm if memory serves) of drift.

A 405nm laser is obviously not the same deal. That will have a wandering wavelength that changes with heat, and from diode to diode. No good as a calibration source. Argon has a strong line at:
358.844021 (which is too low)
440.098598 (which is a little bit high)

Neon was another obvious choice in terms of cheap availability, bit it looks like neon's prominant line is 352.04714 - again too low.

The perfect element would be mercury. There's a REALLY prominent line at 398. Mercury would be fantastic. I'm thinking about this:
http://www.amazon.com/Regent-Mercury.../dp/B000HM9HFU

[DZ]

That's what I thought. I had run a test on a couple blueray diodes for dave at LPF a year or two ago and used 532nm and 473nm as the stage to determine the blueray diodes wavelength, knowing that 532 and 473 should be dead on. If I remember correctly, I had one as short as 402nm and one as long as 416nm.

[-bart-]

I recall that your average nightclub blacklight fluorescent-tube has a ~370 nm peak.
It won't get cheaper than that.
Not sure about the spectral bandwidth, pretty broad, probably like 20nm.

[steve-o]

400 nm LED flashlight ? (Don't know how accurate that would be) http://www.amazon.com/Ultra-Violet-B.../dp/B001VZC5LA
Maybe a noble-gas lamp?

[mixedgas]

+/- 5 Nanometer shift? Someone's dreaming.

532.05 nm is the main peak of a 1.1 nanometer wide doubled 1064.1 nm laser laser line. Ring lasers, such as the Coherent compass, were narrower, but the center frequency does NOT drift from 1064.1

There is NO hobbyist measured drift in the 532 nm wavelength. If you switched to Vandate or ND:YLF hosts there is about half a nanometer shift going to vandate and a 17 nm shift going to ND:YLF. No company is going to make a ND:YLF DPSS unless you request it and pay considerably extra for it. ND:YLF slowly rots on exposure to moisture, anyways.

That wavelength is locked, I mean solidly locked, to the 1064.1 F manifold transition.

946, the source of 473, is even narrower and much lower gain, so it does not shift either.

I did 72 hour spectral runs on Coherent Compasses and cheap hobby 532 nm lasers at the university while looking to see if we could use one to replace a Argon. The compass mode hopped three times in 10 hours. A mode shift is a few 10s of Gigahertz. The CNI mode hopped every few minutes across about 10-20 modes.

While both are much wider lines then a ion laser or hene laser when free running, there is no significant drift that a CCD spectrometer with 2 nm resolution is going to see.

One they are lasing, low cost DPSS lase over a very narrow band and its stable. You would be hard pressed to measure the ND:YAG lasing curve on anything less then a 3/4 meter spectrometer.

High pressure metal vapor lamps, incandescent lamps, some LEDs and all unstabilized diode lasers shift spectra.
Almost nothing else does.

What you need is a low pressue argon, mecury, or neon lamp. The lines are narrow and fixed. Not a mercury arc lamp, they have band spectra.

Positive column Neon lamps have a weak line at 347 nm, you may see it weakly in a 2$ Radio Shack NE2 lamp if the envelope (Type 80 lead glass) transmits it.

Google Penray lamp. I have a few argon/mercury penray tubes, I may be able to sell you one.

http://www.uvp.com/pdf/Pen-RayLampSpectra.pdf

I can do the glassware and pump a pure argon lamp for you if you need it. So can any neon sign shop.
I can also do krypton, they wont have it on station. Cost would be not cheap, but still cheaper then a pen-ray.

GE AR2s show up on Epay from time to time for about 20$-30$,in fact there are about 10 AR1s on ebay right now.

They will have the spectra of Argon, possibly a few Neon lines from the Penning mixture used as a fill, and maybe a weak, very weak, Barium line spectrum from the cathodes.

The $9.99 175W mercury lamp, which you should buy as a collectors item, will source your lines if you drive it with 10-20 mA. You'll need 2Kv for it to break down, and then Vlamp will drop to about 80 volts. A car ignition coil of the old Ford style is 14.95 at Autozone and if you run it off a Variac, at about 12-24 VAC it should light the lamp. At least thats how I did it as a kid.

Pure Mercury arcs for outdoor lighting just got outlawed, replacement lamps and fixtures will no longer be made. That is why that lamp is so cheap.

Sam has a pic of the HENE sidelight spectrum on the FAQ some place

Here's Helium, this would run off a HENE PSU if you add a ballast resistor. Try 56K or 75K, 2-3 watts.

http://www.homeschoolingsupply.com/u...nt-spthe01.htm

Steve

[rhd]

That's what I thought. I had run a test on a couple blueray diodes for dave at LPF a year or two ago and used 532nm and 473nm as the stage to determine the blueray diodes wavelength, knowing that 532 and 473 should be dead on. If I remember correctly, I had one as short as 402nm and one as long as 416nm.

Ya, that makes sense. You would expect 405s to vary just as much as 445s. And given that my purpose for calibrating in this range is to gain the ability to test 405nm diodes, it would be conceptually inappropriate to use a 405 as a calibration source anyway

I recall that your average nightclub blacklight fluorescent-tube has a ~370 nm peak.
It won't get cheaper than that.
Not sure about the spectral bandwidth, pretty broad, probably like 20nm.

I thought about that - but as mentioned the bandwidth is poor, from 20 to 50nm.

[rhd]

I can do the glassware and pump a pure argon lamp for you if you need it. So can any neon sign shop.

GE AR2s show up on Epay from time to time for about 20$.
Steve

Excuse my double-post. I think I have some post/refresh lag. Re the GE bulbs, this wouldn't work would it? (on account of being AR1)
http://www.ebay.com/itm/GE-W1A-AR-1-...ht_2463wt_1396

The Penray lamp approach looks promising. It seems like Mercury, Krypton, or Argon would be the most ideal for my purposes. Reading the line "UVP invented the Pen-Ray® Lamp and can provide the most experience, history and knowledge in the industry" immediately signals "pricey" in my mind though - am I wrong?

[mixedgas]

The whole GE AR series works off 110 in standard sockets and needs no external ballast resistor.
Go for it.

Steve

[mixedgas]

The Penray lamp approach looks promising. It seems like Mercury, Krypton, or Argon would be the most ideal for my purposes. Reading the line "UVP invented the Pen-Ray® Lamp and can provide the most experience, history and knowledge in the industry" immediately signals "pricey" in my mind though - am I wrong?[/QUOTE]

You betcha, they get serious money for those.

I think I have two penray HGs I picked up at a Hamfest, without the transformers.

Let me go check, I'll post again this evening.

Steve

[rhd]

You betcha, they get serious money for those. I think I have two penray HGs I picked up at a Hamfest, without the transformers.

Let me go check, I'll post again this evening.

Thanks! My bet is that you'd probably want to sell them for a bit more than I'd have to spend on this project, but I'm still interested to see what you come up with.

Re: the AR bulbs - I think I'd still run into the problem of basically have strong lines in the 350s, which is too low.

It sounds like what I probably need is a low-pressure mercury lamp. There's a REALLY strong line at 398, which would be perfect.

[mixedgas]

Price is right...

http://www.ebay.com/itm/GE-W1A-AR-1-...ht_2463wt_1396


I may have to cal a spectrometer next week. This just got interesting on a personal bias.

SO:

http://web.mit.edu/8.13/8.13d/manual...tion-lamps.pdf

Nice argon lines listed there at 394 and 404. Probably weak, but you'd need to look at the NBS (now NIST) online spectral book to be sure.


Re the PenRays...

Um, 10$ + shipping is going to be too much?

Steve

[rhd]

Price is right...

http://www.ebay.com/itm/GE-W1A-AR-1-...ht_2463wt_1396

Steve

Before we go too far down the Argon route, can I ask for clarification on a few points?

1) Am I correct that Argon would give me a strong spectral line at 357/358 and then another strong line at 440 (My source btw is: http://physics.nist.gov/PhysRefData/ASD/lines_form.html)

2) That bulb says AR-1. The lines at 357/358, and 440 are AR-II. Does that imply that the AR-1 bulb won't have those lines? Or is this an incorrect interpretation?

3) 357/358 would be below my calibration range, but 440 wouldn't. If I calibrated a spectrometer using 4 coefficients (4 spectral lines), those being 440, 473, 532, and 561, would it be inappropriate to use this spectrometer to subsequently measure 405nm diodes? (since I wouldn't have had a calibrating spectral line near to the 405nm wavelength)

EDIT:
The Argon lines at 394 and 404 are REALLY weak. The realtive intensity of the 394nm and 404nm Argon lines are 204 and 288. I'm not sure what that is relative to, but for comparison, the Argon ilnes at 357nm and 358nm have a relative intensity of 12,000+ and the 440nm line is 8,000.

So I don't think an Argon source would be usable for the 394/404 line as calibration. Hg on the other hand, has a VERY strong line at 398nm - a relative intensity of 10,000,000 which appears to be it's strongest line in the 350 to 450nm range.

$10 would be no problem at all Is it Mercury?

[steve-o]

Too bad they can't make a xenon or mercury-neon laser. There's some fun frequencies in there

[-bart-]

Too bad they can't make a xenon or mercury-neon laser. There's some fun frequencies in there

Or a low pressure sodium vapor laser, with corresponding efficacy, that would be the day !

[mixedgas]

Or a low pressure sodium vapor laser, with corresponding efficacy, that would be the day !

Xenon lases, neon-mercury mixes lase. They just lase pulsed. I've never gotten my hands on a pulsed xenon, they were made for resistor trimmers. I've seen them, but always get outbid.

Mercury lasing was what Bridges was doing when he found Argon lases.

AR1 is simply the part number of the lamp, it is not a ion spectrum, just a glow spectrum. GE drives argon quite hard in the lamp. Can't you just lengthen your CCDs integration time and pick up the weak lines?

IF not the AR1 lamp, why do I just not carefully pack up a penray and loan it to you.
Steve

[steve-o]

They make xenon lasers Steve? Wow. How fast can they pulse? How many lines and what wattage outputs?

[logsquared]

not sure if this is an ar1 or 2 but you can have it if you want it.bulb1.JPGbulb2.JPG

[mixedgas]

Sodium does lase, in the IR, diode pumped, quite well. In fact its hush-hush these days.
Diode pumped metal vapor lasers are making a comeback, just so far they are all IR.

The the lower of the two sodium yellow "D" lines lases like crazy, however, you need a dye laser to "seed" it.

Steve

[mixedgas]

IF RHD does not take that AR lamp, send me a PM.

Steve

[Xytrell]

Not a mercury arc lamp, they have band spectra.

High pressure lamps (like the amazon one listed) are still narrow enough, but they require expensive (>$50?) ballasts



It's the ultra-high pressure ones (like in DLP projectors) where it gets too broad:



I recommend any fluorescent lamp you already have in your house. It will have a mercury H-line at 404.7nm as well as a G-line at 435.8nm. They're easy enough to pick out especially if you were to zoom in on that area. Phosphor emission doesn't usually happen in this area because it wouldn't contribute much to the brightness of the lamp.



I picked up one of the ebay argon lamps just now so I'll measure that when it gets here.