A few members have been experimenting with pre-condensers so here is a spreadsheet I wrote that may help. I can’t guarantee how accurate it is but it should give a reasonable starting point.
Change the values in the white box and the result will be in the green circle.

Page 4 was written by RymGB


DJ



Credits:
This spreadsheet is a culmination of many peoples work in this thread. Thanks should go to RymGB, dajyn, Simul8r, Mark and everyone else that I forgot to mention.

edit: Added a simple 2D arc angle calculation to compare different combinations

Great work!

Just a quick question.
Is there a theroretical limit to how much extra lumens a pre-condenser can give a projector?
In an optimal setup, how much extra do you think you could get: 50% - 200%?

Thanks.
There is a limit but to calculate it is difficult. Even our resident mathematician, Rox, is having a hard time determining it.

DJ

ok, I've used this and found some solutions for some lenses etc..but do any of these numbers have to do with if the light from the precondenser covers the entire fresnel? not sure what the measurement in H13 (fres diag) means....

H13 is the size that the fresnels will need to be to adequately cover the diagonal of the LCD at the distance it is in front of the fresnels. It then uses this value to calculate the angles and position the pre-condenser. If that's what you are asking.

DJ

I should mention that to find the “Center of Triplet to LCD” value use my focal calc program. Perhaps if I get enough time I may be able to integrate it into a separate page

so , looking at the two maximuns(arc magnification and collection) what is the relationship between the two, in the sense of....its a better system if they are relatively close to each other(ie 45mmm for mag and 51mm for collection)?

maybe a better question to ask is are there any numbers on here(assuming that the lens will work, ie the plano-arc distance is greater than the radius of the bulb) that indicate better performance? I know you can't calculate the percentage light increase(if any), but to compare two lenses, any numbers to watch to see which would be better relatively speaking? say inputting the pyrex altman 6x12 and the altman 6x16.....

The “Maximum Arc Collection” values are only experimental. I was trying to Combine Snell’s law with thin lens theory. The “Maximum Arc Magnification” seems to be the most accurate at the moment.


Arizonavideo PMed me with a simular question about comparing different combinations. The R2 (L36) on Page2 under “Pre-Con Refraction Angle and Arc placement” is the angle of light collection that you could use. You would need to use it wit the “Plano to Arc value” to get a reasonable comparison.

I wasn’t going to upload this spreadsheet so it is nowhere near completed. In other words it’s not user friendly. I’ll see if I can add a quick calculation to the spreadsheet to make it easier.

DJ

I’ve uploaded the modified version.

DJ

thanks, well, trust me, we really appreciate the work that goes into this kind of thing thanks for doing this for us... and to RymBG and daiyn and everyone else contributing to the precondenser work, and we can't forget that old troublemaker Sim

Thanks for reminding me of the main contributors. I add a credit to the first post.
DJ

Hey guys, not understanding what these numbers are meaning. Here are the things I filled in
Triplet 129mm width
Length 156mm
center of triplet to lcd 576mm

Lcd 381mm
Distance to fresnel from lcd 40mm

fresnel, sandwich
light/fresnel 330mm
lens/fresnel 550mm

arc 71mm

what condensor will work? I have both a 220 and a 330mm fresnel, so i can switch if needed. Thanks Chad

An arc of 71mm is allready to large to fit though the triplet so there is little hope of using a pre-condenser. Unless you use a longer rear fresnel.
Sorry I didn’t reply sooner, I didn’t realize your question was here.

so, been using this for a while , modeling different combos..... and have a question... does a reflector have any effect on a precondenser system, in terms of distances(arc to plano)? or just in terms of brightness as in a non-precondenser system....

also, is there a way to figure out the...ummm... arc magnification/angle collected if you move the plano from the calculated maximum? for example...
modeling a 17" system
input 431 for lcd diag, 550 for triplet - lcd, 220/600 for fresnels and a 6"(dia)x9" fl altman pyrex... gives 170 degrees collection but arc - plano is 4mm so is there a way to calc what the angle collected is if you put the plano at 24mm?
and as a side note do you know what "white plate" glass is?

The reflector is treated much the same as in a standard set up. It should be large enough to cover the pre-condenser so a larger section of the sphere would be needed. I used the pro reflector which isn’t quite large enough but it still managed to work just not efficiently as it should.


You could calculate it manually using trigonometry but if you did move it from 4mm to 24mm the pre-condenser would refract the rays so that they wont cover your LCD, you’ll end up with a image that only has a circle of light in the centre. Go to sheet6 on the spreadsheet and try changing the angle value and you should see what happens graphically to the outer ray.


The only reference I know is in the link you posted in my plog

DJ

ok, looks like a 6x12 is the best option available(by which I mean you can find one easily) for a 17" monitor then... unless we can figure out a way to get a 6x9 only 4mm from the arc!

edit: sorry...menat to say in my situation it looks the best so far, 131 degrees (220/600- 35mm arc, 24mm from arc to plano which I figure is right at the bulb surface or a couple of mm away, hopefully pyrex can handle that?)

You could try separating your fresnels. Or use a 330mm rear fresnel

DJ

well, corrrect me if I'm wrong(happens a lot ...sigh) but the angle collected is probably the best way to determine which setup should be brighter , right? if I seperate the fresnels(I think it has to be like 150mm to work), the angle collected is 128 degrees with a 6x9 330, therefore its less light than 220 with 6x12(131)..or am I missing something?

does using a 220 or 330 make a difference in the amount of light collected in a precondenser system? I know that sounds like a stupid question, I realize its huge difference in non-condenser system, but I thought in a precondenser system, as long as the fresnel is covered by the light from the precondenser it didn't...do you know what I'm trying to say?...
in other words, if you have 2 light systems one with 220 fresnel and one with 330, if the angle of light collected by the condenser is the same(and all else is the same except the condenser and fresnel) they will be equal brightness, yes? no?

ok, i'm lost...will a pre-condenser work with a standard build? 15" LCD with all standard LL lenses and built to LL spec. or is a pre-condenser only good if you have pro lenses?

I too would like to know this.
I already have the std lens package.
I bought it before I was interested in the pre-condenser option.
If it can be done, what would suggestions be for the pre-condenser that would be the best option for us.
Thanks

for that answer, check out Simul8rs original precondenser thread:

http://www.lumenlab.com/forums/index.php?showtopic=9656&hl=

he was originally using the standard triplet I believe, though I'm pretty sure you will have to use another 317mm fresnel in the rear in place of the standard 220.... try pluggin some numbers into dazz' calculator( start with 62mm for the triplet width(LL standard) and 317 for the rear fresnel and start changing things! (the arc-plano distance must be > 24mm to be feasible if using the T15 style bulb, that works out to a little past the surface of the bulb, though realistically, > 28mm is better- dazzla, please correct me if I'm wrong)

As much as i love a good 42 page thread.... i don't have time to read all of that...i just finished reading the whole Diffused Light Engine thread.. (all 15 pages took me about three days ). good stuff.. but not proven (yet) it went from using cow filters (i mean BEF filters ) to now talking about light tunnels...

i bring this up because i'd like to know more about both but maybe they are to "bleeding edge" for my skills. I'd help out and test things but I'm lost in the math...

anybody got an optics for dummies book?

well, you could always just duplicate Sims entire setup...seriously...in his new PLOG he pretty much tells you measurements and everything, plus he's really helpful .... and his results cannot be argued with ....
but theres always the standard way too....lots of examples, lots of good results...

Dazzla: heres a question I have been wondering about ...
all things being equal(lcd size, lcd--> triplet distance, lamp, triplet, front fresnel...), in a precondenser system, is the amount of light only dependent on the arc collection angle? for example,I can use the 4.5x6.5 shed condenser with a 317 rear fresnel or a 6x12 with a 220 fresnel AND if I change the fresnel gap and the lcd-fresnel gap to make the arc collection angles the same , will they produce the same lumens?

edit: also, just because I can be slow sometimes.... arc means the center of the arc, and plano->fresnel means the flat side of the condenser to the rear fresnel?

For pure light collection theoretically they both should be the same. Although the setup with the larger gap between the fresnels will loss a little bit of light around the edges but that could be compensated with a slightly larger collector fresnel and move the LCD forward. The ideal set-up would have no pre-condenser at all and use a fresnel with a very small FL. I think the biggest advantage in using a pre-condenser is light distribution, a work around for the inverse square law. So the 330 fresnel in this respect would be better because of the physical distance the rays have to travel before they enter the fresnel. Using the 220 fresnel would mean that any sperical aberration that the pre-condenser is introducing would not have as good an influence as the 330mm. You can look at it this way: If we used an extremely long rear fresnel and a pre-condenser to collect the same light collection angle as a short FL rear fresnel the inverse square law could be over corrected. We could end up having the corners of our image being brighter than the centre. This is all theory and yet to be proven so the difference between the 220mm and the 330mm may not make a big difference.

An ideal pre-condenser, if we could make one, would be an aspherical meniscus lens with its surface corrected to introduce more sperical aberration instead of less as is done on normal aspherical lenses. Another addition to this hypothetical lens would be to make it a hybrid meniscus/fresnel lens. So the centre portion would be the meniscus and the outer edge could be fresnel prisms used in their total internal reflectance mode simular to the original light house fresnels,

yes

DJ

thanks for the info, too bad we don't have any lens makers
I thought there probably wouldn't be much difference between a 220/330.... as for light distrbution, I know that using a precondenser increases my vignetting % on the order of 45-50%(from the 40s to the 60s) which is a very noticeable difference.... so you think if we used say a 550/550mm fresnel combo (assuming we could find a good precondenser) the vignetting would be even better?
and while (as I believe you have stated) a precondenser may not increase lumens for a 17", it would increase the vignetting, yes?
.... and just as curiosity, if you had your ideal precondenser, how much better do you guess it might be than a simple plano-convex?

I think so but there are a lot of variables to take into consideration. There is only one way to find out .

ok, been playing around some more and doing some testing and have come across something which in retrospect is fairly obvious, but it never occurred to me...
I've tried a couple of different setups based on the calcs and have noticed that the brightest picture is not where the "optimal" solution is.... mostly the distance from the plano to the rear fresnel.... I've tried a couple of different setups(by which I mean all the equipment is the same, condenser etc) but the distances are different and in both cases, when setting it up according to the calculator, there is the least amount of spillage I have ever seen in my pj (good thing!) ...but I end up moving the lightbox(including condenser) about 2-3"(from 11.5" to 9" last time) closer to the rear fresnel to get a huge amount more light(center lux went from 46--> 71).... and there is definitley spillage at the triplet(arc sticks out about 1/2" past the triplet, at the triplet), but the brightness is way higher... vignetting went down, from memory, lux looked to be in the 30s on the corneres of the 46 lux center reading so = > 70% vignetting... while at the higher lux reading vignetting was in the mid 60s or so....
I think I was just paying too much attention(when fiddling with the calculator) to the max arc mag and collection angle and not watching how far the plano got from the rear fresnel...
I'm guessing that for maximum light that there is some optimal distance from the rear fresnel to the plano, right? just like in a non-condenser setup... and that once the arc/plano gets further than the focal length of the rear fresnel away that the actual light that hits the rear fresnel is lessened... assuming that the arc can fit into the triplet that is...

I know the placement of each piece affects the placement of the other pieces, but in general, but would you say.... given 2 solutions with the same "X" degrees of light collected, same bulb, the one with the rear fresnel closer to the plano will be brighter?

You’re pretty much on the money with your observations. My criterion was to fit the entire arc though the triplet and have the entire arc visible to the fresnel. When you move the arc and pre-condenser closer two things are happening: The virtual arc produced by the pre-condenser is being magnified more by the fresnel combination so some light will be lost as spillage around the triplet. At the same time the rear fresnel is collecting more overall light but at the expense of only receiving partial views of the arc.
In this pic you can see that the virtual arc is position at the fresnel focal plane. I’ve drawn three rays one from the top of the arc(purple), one from the middle(green) and one from the bottom(light blue). At this position the fresnel can see the entire arc or in other words the corners of the fresnel are being lit by equal amounts of light from the top, middle and bottom of the arc. The problem is in this approach is that some of the arc is over shooting the fresnel, these sections are only partial views of the arc that I didn’t believe would be useful for evenness. So when you move the arc and pre-condenser closer you are in affect collecting more light but the centre will become brighter because it has equal view of the entire arc. And the corners will only be lit by the centre and one end of the arc or if you move it really close the corners will be lit by only the end of the arc. Of course at this close a distance the fresnels efficiency will be very low. I hope that made sense

DJ

thanks for the explanation, actually yes it does make sense...I'll have to think about it some
more still though:)... and as for eveness, I've found that once you get into the mid-upper 60s the vignetting is, to my eyes anyway, pretty close to perfect so losing 10 points (when I moved the arc-plano closer to the rear fez) was not noticeable..hmmm but we all know how the camera exxagerates the vignetting, maybe I'll rustle up a camera again and see if its noticeable
well, its back to the calculator I guess, this time with a closer eye paid to the plano-frez distance AND the light angle collected..still trying to decide if a 6x12 or 6x9 pyrex lens(with a 330 rear frez) would be better with the 17" setup... so far I think the 6x9 is winning.... any thoughts on that?

is there a way to calculate the size of the reflector so it covers the precondenser?

I need some help with this calculator

What means "arc" ?

So

I have 12 cm diameter GLASS plan-convex lens : Focal length 20-22 mm(dont know exactly) : edge thickness around 3 mm

I have 300*400 mm fresnel one 220 fL another 330 fl : for 17" LCD sandwich method

I have this objective: http://www.diy-beamer.com/DE/store/comersu...p?idProduct=155

SO what will be the distance from 220 fl fresnel to precondenser lens ?

What about sandwich method?

Dazz,

On page one of the pre-condensor worksheet, in cell I-31 (Arc Collection Angle), does that refer to the angle from the center to the edge (57 in the picture attached), or the angle from the edge to the edge (114 in the picture attached)?

Thanks!

Dazzla:

What is the best way to hack the spreadsheet to compute a split-fresnel setup? Make Sheet1!D9 ("Distance to fresnel") negative? Offset D5 ("Center of Triplet to LCD")?

Thanks for your good work.

As more and more people are trying condensers in their setups, I'm wondering if it would be helpful to update this thread a bit and maybe get an uber-guru or two (ahem, Dazzla) to break down the basics of condenser selection (as it relates to lamp and reflector size, etc.).

But first, I am hoping to get a clarification on the spreadsheet, as it currently stands. I found this thread over on diyaudio, in which a "Guy" explains a graphical way to calculate positions for reflector and condenser. Using this method gives me a 50mm arc to plano distance and a 90mm plano to fresnel. These figures more or less agree with the calculator on page 4 of the spreadsheet (if used after un-protecting), but using the main page gives me 25mm arc-to-plano and a 100mm plano to fresnel. Obviously experimentation is required and the calculator is just a guideline, but I'm hoping someone can explain what factors or calculations make the two answers different.

Pertinent data for my calculations:
Lamp: 8mm arc
Rear fresnel: 165mm FL, 228mm diameter (same diameter LCD, 9" diagonal, 16:9)
Condenser: SS 6.5" (165mm) FL, 4.5" (114mm) diameter

Also, how does one calculate the "n" value for a condenser?

The graphical method discussed by Guy uses the thin lens definition. Where the lens is treated as theoretically flat and spherical aberration is not included. It also doesn’t take into consideration the length of the arc. There is another thread over there where me and Rox were discussing another graphical method that included the arc length but it only used thin lenses as well. A similar method is used for the calculator on page 4 by RymGB except it uses maths to find the answers. That’s why you see a similarity in results. The focal calculator uses Snells law to trace the rays by refraction. This method is a bit closer to reality.

n is a constant relating to the refractive index of the lens material. I think by memory I set it as default to borosilicate.

It's not a simple task, I'll think about how I could condence it.

114°

I understand (well, not entirely, the optics I mean ) and hope I didn't come off as too demanding. It's just that having partial or imperfect knowledge is sometimes more frustrating than complete incomprehension

Also, the advice given over on diyaudio is often the exact opposite to what is found here. Ace over there says get a 70mm precond with 70-80 FL and those numbers don't even work in the calculator. Perhaps much of this is related to the fact the LL specializes in 15-17" projectors and diyaudio is more into the small guys.

Ace specialized in 7” LCDs that has different FL than used for a larger LCD. If you input the specs for a 7” LCD, 220 rear fresnel, 330 front etc you will find than the focal calc will give you a meaningful result
DJ

DAZZZLA, am i right, that te actual version calculates for the "unsplit" setup? what difference can be forseen if i'd like a split fresnel setup?

Could you help me determine which pre condenser i should use and where the fresnels go for my setup unfortunately i just cant figure out the calculator.

8 inch diagonal lcd, 220, 330 fl fresnels unsplit, 320 sta ll triplet
150 double ended hqi osram.

Use this as a starting point:
Click to view attachment


Then try different values for pre-condenser FL and diameter. Try to get the Arc Collection Angle as large as possible but watch that the Plano to arc value doesn’t get to small.

DJ

what is BFL? How do i know if my plano to arc value is to small?
Also why does center of triplet to lcd show 345mm when the triplets focal is 320mm?
Where is the plano point? the numbers above seem to give the best results as per your suggestions. So this means the condenser goes at 59mm from center of arc or 17mm? And the condenser of choice would be 110mm FL and 90mm diameter?

BFL is Back Focal Length and in this instance it’s used as a theoretical value. Technically BFL is the distance from the first principal plane to the FL. There is also front focal length that is a distance between the second principal plane to the front FL. Effective Focal Length is sort of an average between the two planes as well as some distance between them. Don’t worry about using it, it’s only there for my reference when I was writing.



If the distance is so small that it is imposable to move the center of the arc close enough to the Plano surface of the pre-con. EG. The lamps outer envelope is 20mm diameter but the arc center needs to be 5mm from the Plano surface, 10mm would be the closest you could get it.

The triplet's FL is only a value that describes its power. To create an image the triplet will be further from the object than its FL. Placing it at the FL will not create an image. Have a read of the this and this.


Plano isn’t a point, it’s the flat surface of a PCX (Plano ConveX) lens.


17mm


Not necessarily so. The diameter and FL could be different and still give the same collection angle. It all depends on what lenses you have available.

Plano isn’t a point, it’s the flat surface of a PCX (Plano ConveX) lens.
17mm
Not necessarily so. The diameter and FL could be different and still give the same collection angle. It all depends on what lenses you have available.
[/quote]

Thanks for your explanations however you lost me on this last reply. I thought the purpose of the calculator was to figure out what optics to use for best results. For example I have access to a 225mm condenser that is 88 mm in diameter. Will this perform just as well? If so how will i know before buying, making box etc? How will I know where the condenser fresnel will go? Again I thought the job of the calculator was this.
An object beyond the focal length = Inverted real image

An object beyond the focal length = Inverted real image. So in an unsplit configuration this means the triplet must be placed further then the focal length of the lens which explains that you placed 345mm for a 320 triplet in the chart. How did you determine this distance?

Also do i measure this distance to the center length of the triplet lens or to the back glass of the triplet?

Actually, there is only one thing I need to know. That's the distance between my condensorlens and my fresnel. But for this calculater I need to fill in variables I do not know.

To calculate the distance fresnel-condensor, the only thing you need to know is the FL of the fresnel and the condensor, the distance between the ARC and the condensor and the ARC (as far as I know).
Is there any calculation possible so I know where to place my condensorlens so that I have a parallel light beam from the first fresnel?

I use:

Fresnel, Focal Length = 220 mm
Condenser Lens , Diameter = 110mm, Focal Length = 220mm
ARC = about 30 mm
Distance between ARC and Condensor lens = about 2 mm

Well I think I should have looked better. I found this:

http://www.lumenlab.com/forums/index.php?showtopic=9849

This calculater tells me the distance is about 180 mm.

Alright, so can someone give me a hand with using this spreadsheet? I've been trying to get it to work, but I'm not sure I'm using it right and a couple of the numbers are giving me errors instead of answers, so I figure someone more knowledgeable then me might be able to help me out.

Here's my numbers:

220 rear fresnel, 550 front fresnel
17" LCD
18" Beseler (roughly 190mm long, 119 mm wide if the ebay seller measured right)
4.5" condenser lens, 6.5" EF
LCD - Triplet is 22", using focal calc
LCD - front fresnel is 3/4"
rear fresnal to LCD is 1/2"
Had to estimate the arc length because I'm not sure, bulb is an Ushio: http://www.atlantalightbulbs.com/ecart/10E...UHI-S400DDUSHIO

So can anyone tell me where I should place my condensor in relation to the rear fresnel and where the bulb should be in relation to that?

I'm trying to go from start to finish on this in the next week (I have it off for the holidays and I'm handy with tools), so I'd like to get started tomorrow with cutting MDF if I can get this figured out.