Converting from Lux to Irradiance? (halp)

Invisusira's playground

Moderators: Aergis, Invisusira

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Mon Apr 12, 2010 7:52 am

Sorry about the delay, I had a busy weekend of not stressing out over this D: I emailed my boss about the unsatisfactory results on Friday and he didn't email me back (which I am actually glad about, he tends to just not understand any of this and blame me for things not working right).

I will try contacting the company to see if they can get me any additional information on how the device works. The power spectrum of that light should be pretty easy to find. This seems like the simplest way to go about it.

I know that the relative intensities should be fine but what I am worried about is that I need to get absolute intensities. I also know that since the device is old and not recently calibrated the absolute intensities are going to be off, but as long as I can get all the LEDs to the same intensity that is good.


Purpose/Experiment: expose scorpions to various wavelengths of light (of the same intensity) and monitor their behavior
Method: scorpions are placed in Petri dishes on a glass table, under a PVC pipe container, and monitored from below the table
Light setup: What I am doing is I have these PVC pipe containers, 16cm tall about 11cm diameter, all non-reflective black on the inside, the bottom opening is left open while the top is covered with a black plastic, a hole is drilled in the center of this black plastic top for the LEDs to be inserted into (the 5mm LEDs fit perfectly). So it is pretty controlled in that there is no light other than the LED and measurements are always taken from 16cm away. I have a hole drilled in my glass table to fit a light probe, with the current device I have just been standing the container with the inserted light on top of it (making sure to keep it straight).
I know the light intensity is going to vary depending on where you are measuring under the PVC container so I have been taking the minimum and maximum intensity readings in the container. I tried to get all the LEDs, there are 6 different LEDs, within the same dispersion range (I think they are all 15 to 25 degrees with the exception of one that is around 45).
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby theckhd » Mon Apr 12, 2010 8:20 am

Given your experiment, you might have better luck just buying a cheap silicon photodiode and connecting it to an oscilloscope or accurate voltmeter. Most photodiodes are characterized by a responsivity curve (which may be available online) that would let you convert from current or voltage to intensity.

That said, if PASCO is willing to give you the responsivity curve for their selenium diode, or you can estimate it well enough from the manual, you could get away with using that. Since all you care about is that the sources are the same relative intensity, your original method of multiplying the 300-nm reading by a scale factor would work properly (though you'd still want to estimate the intensity of one of them, I'd assume). It doesn't answer the question of why you're getting higher readings from your calculation than the physicist did with his measurement.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7852
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Mon Apr 12, 2010 10:52 am

Yeah, that is why I wonder if it is alright to use =/ The thing is like 20 years old. I know devices like this need to be re-calibrated every 2-3 years so. I am guessing that it is just old and that is why it is doing this? It also says in the manual that it was meant for relative measurements, but I assume there is a way to properly convert it to absolute (even if it won't be completely accurate).
But I am pretty sure this device isn't sensitive enough to measure down in the range I want, the conversion equation I ended with for UV was .1464*lux/.944516=intensity in uW/cm^2 (.944516 is the k_lambda for 395nm). My device only measures down to .1 lux.

There is also the possibility that I may have done something wrong with the physicist. Here is the equation he gave me to use for each light:
Intensity=voltage*(slope*calibration factor/interval time) and that comes out in uW/cm^2. The slope was the slope of the voltage(x) v photon-counts(y) graph and the calibration factor is what he calculated for me based on the spectrum measurements I gave him. I also don't know if this means anything, but I have written down 1cps= ____x10^-4 uW/cm^2 (it's different for the 400 and the 525), but I don't really know what that is, where I got it from, or why I wrote it down.

As far as getting the exact measurements with this Pasco device go, I just need them somewhere in the ball park. The important part is that I can get all the lights to the same intensity, even if I can't get the exact intensity. The stuff I am doing now is based off a pilot study I did using the intensities I am currently trying to measure (I am adding more lights and turning up the intensity a little). So if I can get the new lights to the same intensity, then I can start my data collection now (data collection is going to take over a month), adjust the intensity based on the current calculated ones from the lux meter (for example, if I want to double the intensity), and try to bug or pay someone from a lab with a spectrometer to check my actual intensities later.

I'm not sure if buying new equipment is an option. We were looking in to a radiometer but that would have been $2000 and that is out of the price range (my boss is wanting to get out of light research after this study is done). Today I proposed the idea of paying someone to come in and check them for me though.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby Arcand » Mon Apr 12, 2010 12:59 pm

So the ultimate goal here is to get stung and develop scorpion powers, right?
Arcand
Moderator
 
Posts: 4525
Joined: Fri Mar 02, 2007 11:15 am

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Mon Apr 12, 2010 1:08 pm

Arcand wrote:So the ultimate goal here is to get stung and develop scorpion powers, right?

First I need to get these lights right, then it's just a simple matter of making a laser beam that can harness the energy of the radiation source to blast the scorpion with, and then I can get scorpion powers.
Science is hard :(
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Mon Apr 12, 2010 1:40 pm

I've managed to find a curve for the tungsten lamp at 2700K, not the best but I can use it to make the estimates. But what I am wondering is how this method is really different from the previous one? Should this one enable me to measure the UV?

Theoretically what I am already doing should work, the big problem is that I have no way to check if it is working right :( I would like to try the back-converting way though and see if that proves to be more accurate.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby theckhd » Tue Apr 13, 2010 7:58 am

mew wrote:Yeah, that is why I wonder if it is alright to use =/ The thing is like 20 years old. I know devices like this need to be re-calibrated every 2-3 years so. I am guessing that it is just old and that is why it is doing this? It also says in the manual that it was meant for relative measurements, but I assume there is a way to properly convert it to absolute (even if it won't be completely accurate).
But I am pretty sure this device isn't sensitive enough to measure down in the range I want, the conversion equation I ended with for UV was .1464*lux/.944516=intensity in uW/cm^2 (.944516 is the k_lambda for 395nm). My device only measures down to .1 lux.


The fact that it's old doesn't necessarily mean anything. I have photodetectors in our lab that are 20 years old, and they work just fine (though they're mostly silcon, I'm not sure if selenium is less stable).

mew wrote:There is also the possibility that I may have done something wrong with the physicist. Here is the equation he gave me to use for each light:
Intensity=voltage*(slope*calibration factor/interval time) and that comes out in uW/cm^2. The slope was the slope of the voltage(x) v photon-counts(y) graph and the calibration factor is what he calculated for me based on the spectrum measurements I gave him. I also don't know if this means anything, but I have written down 1cps= ____x10^-4 uW/cm^2 (it's different for the 400 and the 525), but I don't really know what that is, where I got it from, or why I wrote it down.

It's hard to trace out exactly what he did, because it depends on his equipment. He basically converted photons per second to energy and divided by detector area to give you that formula. But the area and detector responsivity are encapsulated in the slope portion, so I can't really tell you anything useful about it.

mew wrote:As far as getting the exact measurements with this Pasco device go, I just need them somewhere in the ball park. The important part is that I can get all the lights to the same intensity, even if I can't get the exact intensity. The stuff I am doing now is based off a pilot study I did using the intensities I am currently trying to measure (I am adding more lights and turning up the intensity a little). So if I can get the new lights to the same intensity, then I can start my data collection now (data collection is going to take over a month), adjust the intensity based on the current calculated ones from the lux meter (for example, if I want to double the intensity), and try to bug or pay someone from a lab with a spectrometer to check my actual intensities later.

I think the PASCO device will be fine for making sure the light sources are the same intensity. You just have to multiply the reading by the spectral response curve given in the manual - it looks like it's only 50% efficient at 400 nm, and around 85% efficient at 525 nm. So if you multiply your 400-nm measurements by 0.5 and your 525-nm measurements by 0.85, you should be able to compare them directly.

mew wrote:I'm not sure if buying new equipment is an option. We were looking in to a radiometer but that would have been $2000 and that is out of the price range (my boss is wanting to get out of light research after this study is done). Today I proposed the idea of paying someone to come in and check them for me though.

It shouldn't take a $2000 raidometer to measure the intensity of your light source - in fact it should be relatively cheap and easy to do so. A simple photodiode could cost less than $10 depending on how accurate it needs to be and what wavelengths it needs to work at.

Are you working at a University? If so, and there's a Physics or Optics department, and in particular someone who works with lasers, there's probably someone with a simple power meter or detector that would do the job. If your set-up is as portable as I'm imagining it, it would only take 30 mins to an hour to take data and find the absolute intensity of your light source. Hell, if you're anywhere near Rochester NY, you could swing by and I'd be happy to do it myself. But the type of equipment I have access to isn't all that uncommon, so any research group that does a decent amount of Optics research should have a photodiode you can use.

If that's really not an option, I can walk you through the numerical calculation you'd want to do to estimate the absolute intensity using the numbers your PASCO device is giving you. You'd need to digitize or estimate the response curve of the spectrometer (from the manual), the spectrum of a 2700K tungsten source (probably just the blackbody equation), and the response curve of the eye (tabulated and available all over the place). It wouldn't take long, but would only be as accurate as you think the spectrometer readings are.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7852
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Tue Apr 13, 2010 11:38 am

theckhd wrote:I think the PASCO device will be fine for making sure the light sources are the same intensity. You just have to multiply the reading by the spectral response curve given in the manual - it looks like it's only 50% efficient at 400 nm, and around 85% efficient at 525 nm. So if you multiply your 400-nm measurements by 0.5 and your 525-nm measurements by 0.85, you should be able to compare them directly.

mew wrote:I'm not sure if buying new equipment is an option. We were looking in to a radiometer but that would have been $2000 and that is out of the price range (my boss is wanting to get out of light research after this study is done). Today I proposed the idea of paying someone to come in and check them for me though.

It shouldn't take a $2000 raidometer to measure the intensity of your light source - in fact it should be relatively cheap and easy to do so. A simple photodiode could cost less than $10 depending on how accurate it needs to be and what wavelengths it needs to work at.

Are you working at a University? If so, and there's a Physics or Optics department, and in particular someone who works with lasers, there's probably someone with a simple power meter or detector that would do the job. If your set-up is as portable as I'm imagining it, it would only take 30 mins to an hour to take data and find the absolute intensity of your light source. Hell, if you're anywhere near Rochester NY, you could swing by and I'd be happy to do it myself. But the type of equipment I have access to isn't all that uncommon, so any research group that does a decent amount of Optics research should have a photodiode you can use.

If that's really not an option, I can walk you through the numerical calculation you'd want to do to estimate the absolute intensity using the numbers your PASCO device is giving you. You'd need to digitize or estimate the response curve of the spectrometer (from the manual), the spectrum of a 2700K tungsten source (probably just the blackbody equation), and the response curve of the eye (tabulated and available all over the place). It wouldn't take long, but would only be as accurate as you think the spectrometer readings are.
Unfortunately I am no where near NY :P
I'm trying to get in contact with the laser labs at my university to see about power meters. One said theirs is in use for like 12 hours a day but they gave me another to call, I left a message.
I asked someone about the photodiode and they said you hook it in to a voltmeter (I have one of those) but to get the absolute intensities I would have to get a calibrated power meter, which is expensive. But they did also say that my photometer should be fine for relative intensities.

Is the CIE curve the same thing as the response curve of the eye? I've already done the estimates on that one as well as my device's curve, so I just need the black body one :D Is that "Planck's law of black-body radiation"?

Hopefully that lab gets back to me on the power meter though.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby Arcand » Tue Apr 13, 2010 12:19 pm

Is it okay to be using a human-eye luminosity function for this?
Seems like you'd need a scorpion luminosity function.

(This sounds like another of my dumb jokes but I'm actually serious.)
Arcand
Moderator
 
Posts: 4525
Joined: Fri Mar 02, 2007 11:15 am

Re: Converting from Lux to Irradiance? (halp)

Postby theckhd » Tue Apr 13, 2010 12:36 pm

mew wrote:Unfortunately I am no where near NY :P
I'm trying to get in contact with the laser labs at my university to see about power meters. One said theirs is in use for like 12 hours a day but they gave me another to call, I left a message.
I asked someone about the photodiode and they said you hook it in to a voltmeter (I have one of those) but to get the absolute intensities I would have to get a calibrated power meter, which is expensive. But they did also say that my photometer should be fine for relative intensities.

You wouldn't need a calibrated power meter, the photodiode would work. Most photodiodes are specced at a certain Responsivity (in Amps/Watt), and the spectral efficiency curve (essentially how the Responsivity varies with wavelength) should be easily available. So you'd have to measure the current (or voltage) output of the device, convert that to Watts, divide by the area of the detector (should be in the specifications), and multiply by the spectral efficiency at that wavelength to get an absolute value.

Of course, an expensive power meter that does all the thinking for you is nice to have, if one's available. You don't really need anything too fancy though, this hand-held power meter is the type I was thinking of. There are a variety of more and less expensive options though. In particular, you could get a calibrated photodiode for under $200, though for your purposes you could just get the regular uncalibrated versions and assume that the specs are accurate (the deviations won't be significant for your purposes anyway). If you don't want to build the read-out circuit yourself, you could go for the pre-packaged versions that don't cost too much more.

And note that ThorLabs will be massively overpriced if you're just buying a simple diode since they're a reseller. You can probably get the individual photodiodes for less than a tenth of what they're selling them for if you go through a wholesale vendor or manufacturer directly. Edmund Optics might be a little better for this - for example you can get a normal silicon photodetector for $19. You'd probably want one of the ones with a UV-enhanced response just to make sure you can measure the 400-nm sources, which bumps it up to around $50.

mew wrote:Is the CIE curve the same thing as the response curve of the eye?

Yes.

mew wrote:I've already done the estimates on that one as well as my device's curve, so I just need the black body one :D Is that "Planck's law of black-body radiation"?

Yes. Tungsten lamps are incandescent bulbs, so they should have a standard blackbody emission spectrum.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7852
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Wed Apr 14, 2010 8:06 pm

Sorry about the delay.
I contacted a laser guy and he let me use their power meter :D This seems like it will work just fine. I went in today and we did some basic measurements but I scheduled time to go in tomorrow and do the actual measurements that I need.
I'm also going to compare that meter to my lux meter at the same time to see if it is accurate and how exactly it is measuring things. The measurements I took in lux and converted (using the method you taught me) was way off from what his meter was measuring (mine were approximately twice as high). Now I wish I had written down what model his meter was so I could look up its sensitivity curve. It said something meaning "broad spectrum" or "full spectrum", so I think that implies that it is equally sensitive to all wavelengths but I want to check on that. I can always get the name of it tomorrow.

Thanks again for all the help! My boss is wanting to move away from behavior studies so he won't be working with lights anymore, but I will keep those devices in mind and let him know about them in case he does want to do more light stuff in the future.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Thu Apr 15, 2010 7:18 am

I didn't think of this until this morning...
So my lux measurements, converted to uW/cm^2 were about twice as much as the ones taken with the power meter (I'm going in to take some more accurate ones today and possibly compare the lux meter to the power meter readings). However, my lux measurements were done through a plastic diffuser but the power meter measurements were taken from inserting the LED straight in to the meter. Which is really weird that the lux measurements were about twice as high since the diffuser really reduces the intensity.
Since distance is a factor (mine were taken from 17cm away, the power meter was 1cm, the laser guy estimated (based on ... I forget the word for it but if the light follows standard diffusion patterns) what the power would be at 17cm. So that's what I want to go in today to do, actually measure from 17cm away and hopefully get some more measurements.

I think it would be nice to have our own power meter in the lab though.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby theckhd » Thu Apr 15, 2010 1:42 pm

You really want to take the power measurements at the proper distance. If he took the measurements at 1cm and estimated what they'd be at 17cm, then they're probably wrong. Since he's a physicist, he probably didn't do a proper radiometric calculation of what to expect. In fact, he may not have ever seen one done - it's not often covered in standard physics courses.

The LED is not a true point emitter, so it won't give a uniform radiation pattern (which is what I'm assuming he used). The radiation pattern will depend a lot on the LED packaging. Some of them come with little molded lenses on the end so that they're much brighter when you look directly at them than they are from the side. Others have a square packaging that causes funny aberrations in the far field.

The plastic diffuser will also change the radiation pattern in ways that will be difficult to guess at or model. Your best bet is to measure with the power meter under the exact experimental conditions - in other words, put the power meter where the scorpion would normally go, and make sure the diffuser and everything else is set up the same way it would be in the actual experiment.
"Theck, Bringer of Numbers and Pounding Headaches," courtesy of Grehn|Skipjack.
MATLAB 5.x, Simcraft 6.x, Call to Arms 6.0, Talent Spec & Glyph Guide 5.x, Blog: Sacred Duty
User avatar
theckhd
Moderator
 
Posts: 7852
Joined: Thu Jul 31, 2008 3:06 pm
Location: Harrisburg, PA

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Thu Apr 15, 2010 7:24 pm

theckhd wrote:You really want to take the power measurements at the proper distance. If he took the measurements at 1cm and estimated what they'd be at 17cm, then they're probably wrong. Since he's a physicist, he probably didn't do a proper radiometric calculation of what to expect. In fact, he may not have ever seen one done - it's not often covered in standard physics courses.

The LED is not a true point emitter, so it won't give a uniform radiation pattern (which is what I'm assuming he used). The radiation pattern will depend a lot on the LED packaging. Some of them come with little molded lenses on the end so that they're much brighter when you look directly at them than they are from the side. Others have a square packaging that causes funny aberrations in the far field.

The plastic diffuser will also change the radiation pattern in ways that will be difficult to guess at or model. Your best bet is to measure with the power meter under the exact experimental conditions - in other words, put the power meter where the scorpion would normally go, and make sure the diffuser and everything else is set up the same way it would be in the actual experiment.
I think I am just going to remove the diffusers from the experiment all-together. Even with the diffusers there is still variation in what the animal sees.

So pretty much what I did today was put the power meter next to the Lux meter and took readings from both meters at maximum intensity of the light, gradually moving the light source away. What I found was that the Lux meter wasn't detecting until about 4-5 inches away because the intensity was too high before that distance and that the power meter stopped detecting the light at about 5-6 inches away because the light at that point became too weak (in my experimental setup the light is 6.7 inches away), which is weirding me out a bit. I did, however, take power measurements for various voltages for each light. So I am hoping to get some a linear equation from plotting those points (the device is supposed to be linear).
I'm going to spend all day tomorrow doing math and making graphs to try and see if I can salvage anything out of those distance measurement trials. He did just do some quick ratio conversion to assume the intensity at __ distance based on (I forgot the word again, I wrote it down this time but it's in my office) dispersion, but at the same time he was saying that he imagines it is not like regular light and is an in-between for lasers and regular ((insert that word here)) dispersion. But you are right, he just kind of was like "Well, this is 1.7cm away, so let's estimate that 17cm away will be 1/10th of this." or something like that (it might have been 1/100th).
I have been taking all my measurements from the point of maximum intensity (pretty much the spot directly in front of the beam/bulb). But that doesn't solve the problem of not knowing the intensities at 6.7 inches.

I bought my LEDs from here (I probably should have linked you this a long time ago) http://www.superbrightleds.com/cgi-bin/ ... htm#fivemm so I know the "angle" of each but I don't really know how to (if possible) use that for any sort of calculations. I mean, I know how to do geometry and trigonometry but I don't know the specifics of (if possible) applying it to light dispersion.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

Re: Converting from Lux to Irradiance? (halp)

Postby mew » Fri Apr 16, 2010 6:44 am

Isotropic.
ImageImage
User avatar
mew
 
Posts: 1903
Joined: Sat Sep 27, 2008 5:42 pm
Location: US

PreviousNext

Return to Arkham Asylum

Who is online

Users browsing this forum: No registered users and 1 guest

cron

Who is online

In total there is 1 user online :: 0 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 380 on Tue Oct 14, 2008 6:28 pm

Users browsing this forum: No registered users and 1 guest