yiwee
Feb 10 2007, 03:29 PM
Hi Guys/gals?!
Fantastic resource here. However cant seem to find anything on measuring the focal length of a Projection Lens. I have a Projection Lens taken out from an unused OHP but cannot plan my diy projector without this information.
Help please...?
thanks!
sfij
Feb 10 2007, 10:04 PM
so take a 2D image (preferably an overhead), project it through the lens to the wall. move the object until you get a sharp image on the wall. You have the object-to-lens distance, lens-to-wall distance.
from here:
tenzip
Feb 10 2007, 11:52 PM
One thing you can use to do this is a lamp with a low-watt clear bulb, like 60 or less watt. Move lamp and triplet until you get a sharp image of the lamp's filament on the wall, then measure distances.
DAZZZLA
Feb 11 2007, 08:24 AM
QUOTE (sfij @ Feb 11 2007, 09:04 AM)

so take a 2D image (preferably an overhead), project it through the lens to the wall. move the object until you get a sharp image on the wall. You have the object-to-lens distance, lens-to-wall distance.
from here:

Although this is a good way to get a FL of a lens, it won’t be accurate to find the exact FL of a multi-element lens like a triplet. The problem is that the thin lens equations don’t take into consideration the distance between the lens elements so there will be some error. But it will be close.
To get a bit closer you can use the object and image distance you have suggested but also use a known dimension marked on the 2D image, perhaps 100mm or so, and then measure how much magnification has occurred. Then with some extrapolation between the two equations you can calculate the FL. It takes quite a few measurements though.
DJ
sfij
Feb 11 2007, 12:54 PM
QUOTE (DAZZZLA @ Feb 11 2007, 09:24 AM)

Although this is a good way to get a FL of a lens, it won’t be accurate to find the exact FL of a multi-element lens like a triplet. The problem is that the thin lens equations don’t take into consideration the distance between the lens elements so there will be some error. But it will be close.
To get a bit closer you can use the object and image distance you have suggested but also use a known dimension marked on the 2D image, perhaps 100mm or so, and then measure how much magnification has occurred. Then with some extrapolation between the two equations you can calculate the FL. It takes quite a few measurements though.
DJ
DAZZZLA your method is far way better. other thing is, that of course, several measurements should be taken, i.e. 30-60, then compute mean value and deviation. if deviation is low, the mean value is somewhat accurate
yiwee
Feb 14 2007, 03:32 PM
Hey..You guys are great. Thanks very much for the help..now just need to understand it and then try it out. I've tried the wall method, i think its much easier altho not as 'accurate' ,but what the h*** we're suppose tweak tune and adjust until its perfect., right?
Now, how do I measure the FL for an UNsplit fresnel? So if I dont have that measurement, how do I position the unsplit fresnel vis-a-vis the light source? I will be putting it before the lcd right?
Since my fresnel set came from an OHP, its stuck together, I'll go with unsplit design and see how it goes.
Brilliant Farmer
Feb 15 2007, 08:20 PM
So useing the formula 1/s1+1/s2=1/f
If s1=144" and s2=14"
then should it be 1/144"+1/14"=1/f
so 1/f=0.078373015873015873015873015873016?
and f=12.759493670886075949367088607598?
or does 1/f=0.012046485260770975056689342403628?
and f=83.011764705882352941176470592564?
Did I miss something? Am I not getting this?
sfij
Feb 15 2007, 10:14 PM
QUOTE (Brilliant Farmer @ Feb 15 2007, 09:20 PM)

So useing the formula 1/s1+1/s2=1/f
If s1=144" and s2=14"
then should it be 1/144"+1/14"=1/f
so 1/f=0.078373015873015873015873015873016?
and f=12.759493670886075949367088607598?
or does 1/f=0.012046485260770975056689342403628?
and f=83.011764705882352941176470592564?
Did I miss something? Am I not getting this?
I feel You have used the diagonal of your LCD and the diagonal of the projected screen. The formula was about LCD-to-lens distance and lens-to-wall distance. You may use DAZZLA's method if you already have the diagonals. Let's Have your LCD's diagonal
DLCD and your your projection's diagonal
Dproj. And let S is the distance between the LCD and the wall. From here:
f=S*DLCD*Dproj/((DLCD+Dproj)*(DLCD+Dproj))Important note.
DLCD,
Dproj,
S should be measured in the same units, i.e. both of them should be millimeters, inches or feets - and you will get the focal length according to.
tgreenwood
Feb 16 2007, 09:18 AM
QUOTE (Brilliant Farmer @ Feb 15 2007, 02:20 PM)

So useing the formula 1/s1+1/s2=1/f
If s1=144" and s2=14"
then should it be 1/144"+1/14"=1/f
so 1/f=0.078373015873015873015873015873016?
and f=12.759493670886075949367088607598?
That's correct. 12.75 in focal length.
QUOTE (Brilliant Farmer @ Feb 15 2007, 02:20 PM)

or does 1/f=0.012046485260770975056689342403628?
and f=83.011764705882352941176470592564?
Nope. Where did these numbers come from?
Tgreenwood
pramod.shinde
Feb 18 2007, 04:23 PM
1/s1 + 1/s2 = 1/f is very true.
easier would be take the projection lens or the fresnel in daylight (preferably at noon) and try to get a clear image of sun on the ground and then measure the distance of the lens from ground. it is very easy but the results are not accurate but acceptable. donot focus it on a flamble object, like dry grass, paper or wood you might end up burning it.
GIGAFAN
May 3 2007, 06:12 AM
Just to say that fresnel lenses from OHP, are ALWAYS(i havent seen any other) combination of 220 for back, and 330 for front fresnel, the only diference is output lens(triplet, or a single).Fokus for certain size is determined by the distance of the lamp from the first fresnel.Usualy (90%)lenses for OHP are adjusted for close distances, so it,s hard if you want to make smaler picture projection on longer distances(more than 3-4 meters).
ChuckL
Apr 8 2009, 02:43 PM
I need to also measure the FL of my triplet. I have has trouble from day one with corner focus. I suspect that the lens is about 380mm. I was told that this lens was 17.5( 450mm). I am using it with a 550 for field and if the lens is 380 then it is a mismatch and would explain why the lens isn't working well at the corners. Here is my issue. I got out of school 30 years ago and I don't understand the formula that is above. This was called new math when my kids were in school and I didn't understand it then. Can somebody break it down for me?
Using the above example...
1/s1+1/s2=1/f Is this fractional?
If s1=144" and s2=14"
then should it be 1/144"+1/14"=1/f (1/158)? How does this equal 1/F below?
so 1/f=0.078373015873015873015873015873016
and f=12.759493670886075949367088607598 How do you get this?
Thanks
ChuckL
Apr 8 2009, 04:03 PM
This is what I did for math using the above example. The numbers look right.
1/144+1/14=1/f
I multiplied the the first fraction by the second one denominator...
and the first one by the second one's denominator
14/2016+144/2016=1/f
14 + 144=158
158/2016=1/f
158 / 2016 = 0.078373015873015873015873015873016
2016 / 158 would = 12.759493670886075949367088607595 Would this be the conversion from mm to inches?
Would this be MM? Would measurements for1/s1 and 1/s2 be from the center of the triplet?
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please
click here.