OD tables

[Stefano]

I am experimenting with calculating OD tables, with the help of a C code I wrote. This is an example:

 

 

Infrared OD blocking of all stacks with 0.5 mm, 1 mm, 1.5 mm... 4 mm S8612 and Hoya U-360. I had to put a screenshot from Microsoft Notepad because copying and pasting the text directly made it all messy. For example, Hoya U-360 (3 mm) + S8612 (1.5 mm) makes a OD 5.22 stack. I may try other filters if someone is interested, and I have to figure out a better way to display data.

[nfoto]

You really should indicate maximum and minimum OD within a given spectral range of interest. That would be much more informative for the end user..

[Stefano]

In this case I checked between 670 and 730 nm. So it would be better to also put the maximum OD?

[dabateman]

Birna is referring to listing your data in blocks.

So maximum OD blockage from 400nm to 420nm is Z.

Maximum OD blocking from 680nm to 720nm is Q.

 

Figure out the blocks you find most relavent and ranges which aren't too broad. But not too narrow to make the data just look like a scan.

[Stefano]

I think that UV peak transmission could also be important. Then range, thickness... this could really become a software to find the best stack.

[Stefano]

This is the code I used. It doesn't print the result I posted, but comes close.

 

#include<stdio.h>

#include<stdlib.h>

#include<math.h>

#define N 4.5

#define I 0.5

#define H3667 0

#define H3668 0

#define H3669 1.879426

#define H3670 1.06298

#define H3671 0.6216

#define H3672 0.4698

#define H3673 0.412289

#define S867 1.3

#define S868 1.522878

#define S869 1.76447

#define S870 2.01278

#define S871 2.281498

#define S872 2.5406075

#define S873 2.809668

 

int main(){

double OD, max;

double H360[7], S8612[7];

double i, j;

int a;

printf("S8612\n\n0.5 1 1.5 2 2.5 3 3.5 4\n\n");

for(i=I; i<N; i=i+I){

H360[0]=(double)(H3667*i);

H360[1]=(double)(H3668*i);

H360[2]=(double)(H3669*i);

H360[3]=(double)(H3670*i);

H360[4]=(double)(H3671*i);

H360[5]=(double)(H3672*i);

H360[6]=(double)(H3673*i);

for(j=I; j<N; j=j+I){

S8612[0]=(double)(S867*j);

S8612[1]=(double)(S868*j);

S8612[2]=(double)(S869*j);

S8612[3]=(double)(S870*j);

S8612[4]=(double)(S871*j);

S8612[5]=(double)(S872*j);

S8612[6]=(double)(S873*j);

max=10000;

for(a=0; a<7; a++){

OD=(double)(H360[a]+S8612[a]);

if((OD<max)&&(H360[a]>0)){

max=(double)OD;

}

else

max=max;

 

}

printf("%2f |", max);

/*

if((max>0)&&(max<1)){

printf(" ");

}

if((max>1)&&(max<2)){

printf("%c", 176);

}

if((max>2)&&(max<3)){

printf("%c", 177);

}

if((max>3)&&(max<4)){

printf("%c", 178);

}

if((max>4)&&(max<5)){

printf("%c", 219);

}

*/

}

printf("\n");

}

 

system("PAUSE");

return 0;

}

[Cadmium]

The chart represents minimum out of band optical density (not maximum OD). You don't need to know maximum OD, just minimum which defines any tendency of a leak.

What you might want to know are three things.

1) Minimum OOB OD

2) The nm of the above

3) The in-band width, cut off delineating in and out of band.

[Stefano]

So, for example, I may define the transmitted region as the region where the transmission is higher than OD 3, calculate its width (I may also do FWHM), peak transmission (%), peak transmission (nm), "peak leak" (minimum OOB OD) (OD), peak leak (nm), total thicknesses (mm). Then I can search for the best stack, for example by finding all stacks with OD between 5 and 6 (5.something), that use at least 1 mm or 2 mm of U-glass (depending on the type of glass, to be sure there is no VIS leak), that are below a certain thickness (for example 5 mm) and with maximum peak transmission. I may also filter away stacks with too much S8612 (they will not have great performances at shorter wavelengths). I have to distinguish between violet leaks and red-infrared leaks. That's a lot of work to do, especially with Schott filters (I have to enter data manually for every nm in my regions of interest).

[dabateman]

So, for example, I may define the transmitted region as the region where the transmission is higher than OD 3, calculate its width (I may also do FWHM), peak transmission (%), peak transmission (nm), "peak leak" (minimum OOB OD) (OD), peak leak (nm), total thicknesses (mm). Then I can search for the best stack, for example by finding all stacks with OD between 5 and 6 (5.something), that use at least 1 mm or 2 mm of U-glass (depending on the type of glass, to be sure there is no VIS leak), that are below a certain thickness (for example 5 mm) and with maximum peak transmission. I may also filter away stacks with too much S8612 (they will not have great performances at shorter wavelengths). I have to distinguish between violet leaks and red-infrared leaks. That's a lot of work to do, especially with Schott filters (I have to enter data manually for every nm in my regions of interest).

 

Nah.

Just buy a filter, get a good camera and take some photographs. Seems like too much time at the computer and not enough time holding a camera.

 

Leaks can be good, offer some colour. Too tight a filter is just too monochromatic.

[Stefano]

Leaks can be good if you want them. I would try a Hoya U-340 4 mm thick filter alone to have approximately equal amounts of UV and IR. A normal 2 mm thick U-filter is basically IR with a touch of UV.

 

My ideal filter should have OD 5 IR blocking, at least OD 3 at 400 nm, and be under 5 mm if possible. Hoya U-340 (3 mm) + S8612 (1.5 mm) may be the best for my tastes. 3 mm is a strange thickness for U-360, but it should be possible to have it up to 4 mm (like U-340).

[Stefano]

If anyone is interested...

 

All stacks using Hoya U-360 and S8612 that have IR blocking between OD 5 and OD 6. Thicknesses between 0.25 mm and 4 mm, with 0.25 mm increments.

 

Hoya U-360 (0.250000) + S8612 (2.750000)

 

Hoya U-360 (0.250000) + S8612 (3.000000)

 

Hoya U-360 (0.500000) + S8612 (2.500000)

 

Hoya U-360 (0.500000) + S8612 (2.750000)

 

Hoya U-360 (0.750000) + S8612 (2.250000)

 

Hoya U-360 (0.750000) + S8612 (2.500000)

 

Hoya U-360 (1.000000) + S8612 (2.000000)

 

Hoya U-360 (1.000000) + S8612 (2.250000)

 

Hoya U-360 (1.250000) + S8612 (2.000000)

 

Hoya U-360 (1.250000) + S8612 (2.250000)

 

Hoya U-360 (1.500000) + S8612 (2.000000)

 

Hoya U-360 (1.750000) + S8612 (1.750000)

 

Hoya U-360 (1.750000) + S8612 (2.000000)

 

Hoya U-360 (2.000000) + S8612 (1.750000)

 

Hoya U-360 (2.000000) + S8612 (2.000000)

 

Hoya U-360 (2.250000) + S8612 (1.750000)

 

Hoya U-360 (2.250000) + S8612 (2.000000)

 

Hoya U-360 (2.500000) + S8612 (1.750000)

 

Hoya U-360 (2.750000) + S8612 (1.500000)

 

Hoya U-360 (2.750000) + S8612 (1.750000)

 

Hoya U-360 (3.000000) + S8612 (1.500000)

 

Hoya U-360 (3.000000) + S8612 (1.750000)

 

Hoya U-360 (3.250000) + S8612 (1.500000)

 

Hoya U-360 (3.250000) + S8612 (1.750000)

 

Hoya U-360 (3.500000) + S8612 (1.500000)

 

Hoya U-360 (3.750000) + S8612 (1.500000)

 

Hoya U-360 (4.000000) + S8612 (1.250000)

 

Hoya U-360 (4.000000) + S8612 (1.500000)

 

This is very raw, I have to filter out all stacks which are too thick, or use too thin filters. This is just a proof of concept.

[Stefano]

Removing filters below 1 mm:

 

Hoya U-360 (1.000000) + S8612 (2.000000)

 

Hoya U-360 (1.000000) + S8612 (2.250000)

 

Hoya U-360 (1.250000) + S8612 (2.000000)

 

Hoya U-360 (1.250000) + S8612 (2.250000)

 

Hoya U-360 (1.500000) + S8612 (2.000000)

 

Hoya U-360 (1.750000) + S8612 (1.750000)

 

Hoya U-360 (1.750000) + S8612 (2.000000)

 

Hoya U-360 (2.000000) + S8612 (1.750000)

 

Hoya U-360 (2.000000) + S8612 (2.000000)

 

Hoya U-360 (2.250000) + S8612 (1.750000)

 

Hoya U-360 (2.250000) + S8612 (2.000000)

 

Hoya U-360 (2.500000) + S8612 (1.750000)

 

Hoya U-360 (2.750000) + S8612 (1.500000)

 

Hoya U-360 (2.750000) + S8612 (1.750000)

 

Hoya U-360 (3.000000) + S8612 (1.500000)

 

Hoya U-360 (3.000000) + S8612 (1.750000)

 

Hoya U-360 (3.250000) + S8612 (1.500000)

 

Hoya U-360 (3.250000) + S8612 (1.750000)

 

Hoya U-360 (3.500000) + S8612 (1.500000)

 

Hoya U-360 (3.750000) + S8612 (1.500000)

 

Hoya U-360 (4.000000) + S8612 (1.250000)

 

Hoya U-360 (4.000000) + S8612 (1.500000)

 

 

Removing stacks thicker than 5 mm:

 

Hoya U-360 (1.000000) + S8612 (2.000000)

 

Hoya U-360 (1.000000) + S8612 (2.250000)

 

Hoya U-360 (1.250000) + S8612 (2.000000)

 

Hoya U-360 (1.250000) + S8612 (2.250000)

 

Hoya U-360 (1.500000) + S8612 (2.000000)

 

Hoya U-360 (1.750000) + S8612 (1.750000)

 

Hoya U-360 (1.750000) + S8612 (2.000000)

 

Hoya U-360 (2.000000) + S8612 (1.750000)

 

Hoya U-360 (2.000000) + S8612 (2.000000)

 

Hoya U-360 (2.250000) + S8612 (1.750000)

 

Hoya U-360 (2.250000) + S8612 (2.000000)

 

Hoya U-360 (2.500000) + S8612 (1.750000)

 

Hoya U-360 (2.750000) + S8612 (1.500000)

 

Hoya U-360 (2.750000) + S8612 (1.750000)

 

Hoya U-360 (3.000000) + S8612 (1.500000)

 

Hoya U-360 (3.000000) + S8612 (1.750000)

 

Hoya U-360 (3.250000) + S8612 (1.500000)

 

Hoya U-360 (3.250000) + S8612 (1.750000)

 

Hoya U-360 (3.500000) + S8612 (1.500000)

 

 

Keeping only the thinniest S8612 combinations (at the same Hoya U-360 thickness), done by hand:

 

Hoya U-360 (1.000000) + S8612 (2.000000)

 

Hoya U-360 (1.250000) + S8612 (2.000000)

 

Hoya U-360 (1.500000) + S8612 (2.000000)

 

Hoya U-360 (1.750000) + S8612 (1.750000)

 

Hoya U-360 (2.000000) + S8612 (1.750000)

 

Hoya U-360 (2.250000) + S8612 (1.750000)

 

Hoya U-360 (2.500000) + S8612 (1.750000)

 

Hoya U-360 (2.750000) + S8612 (1.500000)

 

Hoya U-360 (3.000000) + S8612 (1.500000)

 

Hoya U-360 (3.250000) + S8612 (1.500000)

 

Hoya U-360 (3.500000) + S8612 (1.500000)

 

 

Then I want at least OD 3 at 400 nm:

 

Hoya U-360 (2.750000) + S8612 (1.500000)

 

Hoya U-360 (3.000000) + S8612 (1.500000)

 

Hoya U-360 (3.250000) + S8612 (1.500000)

 

Hoya U-360 (3.500000) + S8612 (1.500000)

 

So the above four stacks are the only four you can make using Hoya U-360 and S8612, in thicknesses ranging from 1 to 4 mm with 0.25 mm increments, that have OD between 5 and 6 (in the 670-730 nm region), are no thicker than 5 mm, use the least amount of S8612 and block 400 nm violet to at least OD 3.

 

The best one, since they all have the same S8612 filter, must be the first one:

 

Hoya U-360 (2.75 mm) + S8612 (1.5 mm).

 

Linear:

 

 

Diabatic:

 

 

How can I save those graphs as images? Those above are edited screenshots.

[rfcurry]

Stefano,

 

To save those graphs as images, I make any edits to the graph titles in Excel, then I click on the center of the graph to select the entire graph area and CTL-C to copy. I paste the result into a freeware program - usually PhotoFiltre - by "Edit>Paste as new image". That gives me a clean 842 x 642 px image.

I hope that helps.

[Stefano]

Just tried on Paint, seems to work. Thanks!

[Andrea B.]

What I want to know is this: the transmittance for each of those stacks having OD5-OD6.

[Stefano]

What I want to know is this: the transmittance for each of those stacks having OD5-OD6.

All of them? The first list?

 

I can assume the peak is between 360 and 390 nm (probably between 360 and 380), and enter the transmission values of S8612 and Hoya U-360 at those 3 wavelengths, to calculate the peak (and even the peak wavelength).

[Stefano]

This is the list for all stacks with components thicker than 1 mm between OD 5 and OD 6 (they can be thicker than 5 mm and so on). I removed all stacks which used thicker S8612 at the same Hoya U-360 thickness (I kept the thinnest S8612 filter for every Hoya U-360 thickness), done by hand. I checked with two random stacks and the numbers match.

 

Hoya U-360 (1.000000) + S8612 (2.000000) ti(max) = 66.913792 % at 370 nm

 

Hoya U-360 (1.250000) + S8612 (2.000000) ti(max) = 64.809148 % at 370 nm

 

Hoya U-360 (1.500000) + S8612 (2.000000) ti(max) = 62.770701 % at 370 nm

 

Hoya U-360 (1.750000) + S8612 (1.750000) ti(max) = 62.914178 % at 370 nm

 

Hoya U-360 (2.000000) + S8612 (1.750000) ti(max) = 61.112891 % at 360 nm

 

Hoya U-360 (2.250000) + S8612 (1.750000) ti(max) = 59.590213 % at 360 nm

 

Hoya U-360 (2.500000) + S8612 (1.750000) ti(max) = 58.105473 % at 360 nm

 

Hoya U-360 (2.750000) + S8612 (1.500000) ti(max) = 59.059301 % at 360 nm

 

Hoya U-360 (3.000000) + S8612 (1.500000) ti(max) = 57.587789 % at 360 nm

 

Hoya U-360 (3.250000) + S8612 (1.500000) ti(max) = 56.152942 % at 360 nm

 

Hoya U-360 (3.500000) + S8612 (1.500000) ti(max) = 54.753845 % at 360 nm

 

Hoya U-360 (3.750000) + S8612 (1.500000) ti(max) = 53.389608 % at 360 nm

 

Hoya U-360 (4.000000) + S8612 (1.250000) ti(max) = 54.266022 % at 360 nm

[Stefano]

If you remove the last two stacks, you obtain all stacks below 5 mm. This is the same list as the third one in post #12.

[Andrea B.]

Very interesting! Thank you, Stefano.

 

I have been using U-360 x 2.0 + S8612 x 2.0. But based on your results, I think I will experiment with the U-360 x 1.0 + S8612 x 2.0 even though it shifts the peak to 370 nm. I'm curious to see how much difference there might be between the two stacks in actual field use.

[Cadmium]

U-360 1mm + S8612 2mm will be slightly more visual violet/blue. Not a huge difference, goes up to about 405nm @ OD3.

I prefer graphs to tables myself, shows everything all at a glance.

 

[Stefano]

I know Andrea doesn't mind a bit of violet in her UV photos. I do instead.

[Cadmium]

Here is the difference. The slight difference in false color is accompanied by faster exposure time.

For example between these two shots there is 2/3 of a stop difference.

Left: U-360 2mm + S8612 2mm = 3s, f/11, ISO 200.

Right: U-360 1mm + S8612 2mm = 2s, f/11, ISO 200.

[Stefano]

That's your Moon U stack, right?

[Andrea B.]

If only someone could explain to me exactly where violet becomes ultraviolet !!!

[Stefano]

I like the 400-700 nm definition for visible light, since it has round, easy numbers. I consider a 405 nm “blue ray” laser violet, as well as the 404.7 nm mercury H-line. Wikipedia says that the famous 365.4 nm I-line is violet, while I strongly consider it to be UV.