Jump to content
UltravioletPhotography

Lens testing with a spectrophotometer


rfcurry

Recommended Posts

I have a spectrophotometer that I use for filters. Does any forum member use a spectrophotometer for testing UV transmission of lenses? The test chamber of my unit is large, it would be easy to configure it with a support cradle for lenses. I wonder, though, whether it is reasonable to expect that the narrow band of light that is input, will be of sufficient height on the output sensor to register with any accuracy.

 

Any recommendations gratefully accepted.

Thanks.

Link to comment
It all depends on the intensity of your source, a 30W Deuterium source is typical of many spectrophotometers (photomultiplier tube) and should be plenty. Just try it and see.
Link to comment

Unlikely, a typical spectrophotometer is designed for use with non-refractive and non-scattering sample cuvettes, essentially optical flats.

 

About 30 years ago I routinely conducted in vitro sunscreen transmittance assays. The substrates were various roughened materials, to replicate skin surface topography, which combined with the scattering properties of the emulsions themselves necessitated the use of an integrating sphere. For example, making a transmittance measurement of a diffuser, rather than filters as you do, would present similar issues.

 

My old in vitro sunscreen transmittance spectrophotometer had an integrating sphere accessory but there are other considerations. The ISO standard for lens transmittance in addition to requiring the use of a sphere, also calls for a double monochromator and other specifics on beam collimation and beam diameter relative to apparent lens pupil. One significant problem is that a sphere can attenuate by ~90% so you signal is 1/10th of what it would be and a spectrophotometer loses ~1 OD of dynamic range.

Link to comment
Unlikely, a spectrophotometer is designed for use with non-refractive and non-scattering sample cuvettes, essentially optical flats.

 

Not all spectrophotmeters are designed for cuvettes otherwise there would be no need for a $90K benchtop spectrophotometer. A spectrophotometer was used daily for over 30+ years (literally 8hrs a day) for analyzing rough and cut gemstones, both transparent, translucent and opaque.

Link to comment

My spectrophotometer operator's manual states:

The Sample Compartment dimensions (115 mm wide x 112 mm high x 169 mm deep) allow ample room for

positioning of awkwardly shaped solid samples. You can also quickly and easily remove the standard base-

plate to add even greater room and versatility.

and I have taken that advice. I do use some cuvettes when testing solutions for their UV transmission characteristics; however, I also load it with 52mm filters. It is quite a workhorse, I use it constantly, now that I have finally sourced OEM halogen lamps for the Vis and NIR. I don't look forward to replacing the Deuterium lamp (very expensive).

 

Thanks.

Link to comment

Not all spectrophotmeters are designed for cuvettes otherwise there would be no need for a $90K benchtop spectrophotometer. A spectrophotometer was used daily for over 30+ years (literally 8hrs a day) for analyzing rough and cut gemstones, both transparent, translucent and opaque.

 

Certainly, I should have said a "typical" spectrophotometer.

Link to comment

Thanks Shane and John for this help and for responding so quickly to my ping.

Reed, let us know how it all turns out!!

Link to comment
I am not familiar with your spectrophotometer, one thing I would suggest is to define it's dynamic range. If it is zeroed in ABS mode and you block the sample beam with a completely opaque material where does the ABS or OD top out to a flat line?
Link to comment

My machine is a Hitachi U-1500 ratio-beam spectrophotometer with a range from 200-1100nm. It is not a dual-beam with a reference sample, instead the light is split and both beams go to different detectors, but only one passes through a sample area.

 

John, here are the machines operating parameters.

 

The top ABS is 3.000.

 

Photometric Range:

ABS = -0.05 to 3.000

%T = 0 to 300 percent

CONC = 0.000 to 9,999

 

Photometric Accuracy

± 0.002 Abs (0 to 0.5 Abs)

(measured with NIST 930D filter)

± 0.004 Abs (0.5 to 1.0 Abs)

± 0.3%T

 

Photometric Reproducibility

± 0.001 Abs (0 to 0.5 Abs)

± 0.002 Abs (0.5 to 1.0 Abs)

± 0.15%T

 

Does that help at all?

Thanks.

Link to comment

Hi Reed

I have a cheapy USB spectrometer that I use to test many things, I mostly use Sunlight as the light source, as that is what I am mostly photographing under.

http://www.science-s...s/spectrometers

 

You have sparked my interest to look for a Deutrium light source again. I found this one in Australia.

http://www.spectrola...ccessories.html

"This power supply will operate the 10 volt deuterium lamps from us and any other manufacturer.

Suitable part numbers from our list of 10 volt deuterium lamps include:

D002, D004, D005, D007, D008 and D011"

http://www.spectrola...rium_lamps.html

I guess I will try the D002. My spectrometer doesn't need much light to work with. My guess is these lamps are 3watts, 10v at 0.3A & with a 1mm opening for the UV to shine through ? I would mount this lamp in a metal tube holder with a 1/8" hole opposite the lamps 1mm opening, for the optic fiber end to fit into.

Col

Link to comment

Col,

I just reviewed the specs on the Science-Surplus CCD spectrometers in your link above.

The shortest wavelength attainable with the gratings listed is 310nm. Since most of the advantage provided by Du is at lower wavelengths, you might find more utility in a small Xe source.

You likely have already seen Enrico's primer on continuous UV sources which describes the use of Xenon automobile headlamp retrofit kits you might consider.

 

PS

I want to build one but I do not want a HgXe lamp and all the headlight kits I can find are labeled to contain Hg.

I don't trust online sources, they seem to call everything Xenon.

Link to comment

My machine is a Hitachi U-1500 ratio-beam spectrophotometer with a range from 200-1100nm. It is not a dual-beam with a reference sample, instead the light is split and both beams go to different detectors, but only one passes through a sample area.

 

John, here are the machines operating parameters.

 

The top ABS is 3.000.

 

Photometric Range:

ABS = -0.05 to 3.000

%T = 0 to 300 percent

CONC = 0.000 to 9,999

 

Photometric Accuracy

± 0.002 Abs (0 to 0.5 Abs)

(measured with NIST 930D filter)

± 0.004 Abs (0.5 to 1.0 Abs)

± 0.3%T

 

Photometric Reproducibility

± 0.001 Abs (0 to 0.5 Abs)

± 0.002 Abs (0.5 to 1.0 Abs)

± 0.15%T

 

Does that help at all?

Thanks.

 

 

Sure, I recall you mentioned what it was somewhere, I just never had hands on one of those.

 

With a max ABS of 3 the minimum possible percent transmission is of course 0.1%, and transmittance of 0.001 or 1.00E-03, not really zero.

The accuracy and reproducibility of %T are a bit higher than this because linearity is lost when approaching the noise limit.

 

I know you know this, but others who do not may benefit from our discussion. :)

Link to comment

>... but others who do not may benefit from our discussion.

 

Thank you. I follow your discussion and hope to befit from your explanation. Even if I can't take part in an active way ...

Link to comment

Col,

I just reviewed the specs on the Science-Surplus CCD spectrometers in your link above.

The shortest wavelength attainable with the gratings listed is 310nm. Since most of the advantage provided by Du is at lower wavelengths, you might find more utility in a small Xe source.

You likely have already seen Enrico's primer on continuous UV sources which describes the use of Xenon automobile headlamp retrofit kits you might consider.

 

PS

I want to build one but I do not want a HgXe lamp and all the headlight kits I can find are labeled to contain Hg.

I don't trust online sources, they seem to call everything Xenon.

 

Hi John

The automotive headlight kits call their lamps 'Xenon', because Mercury is a dirty word now.

Col

Link to comment
My glass blower told me that his hand crafted xenon and argon lamps always contain traces of mercury. Otherwise ignition voltage is excessive high.
Link to comment

Yes, ignition pulses are very high voltage, sometimes tens of thousands of volts.

Years ago the company I was with used a 6.5 kW Xe-arc lamp that would reboot nearby computers when you ignited it.

We had to build a Faraday cage for it! :)

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...