I am thinking about what the precondenser is doing.

Look at the 2 precondenser models there. During the light phase where the light is going from the precondenser to the fresnel, the light is still diverging, and in that case, it is still obeying the square law. It is merely obeying the square law from a virtual light source location which is altered by the precondenser lens.

Use of a different precondenser with a shorter FL CAN offset this, but with the same precondenser lens, the longer path still loses more light.

If you measure the angles in the 220/precondenser and the 330/precondenser, the angle that the light changes is sharper in the 330 diagram at the precondenser. This indicates the need for a shorter FL lens.

Only collimated light gets around the square law. This is not a spotlight situation. Spotlights use reflectors and/or lenses to get light as close to collimated as possible. Even there, since perfect collimation is impossible it still obeys square law, but from a virtual point, more distant than the actual lamp. This is no different.

Using the same precondenser lens, you have only 2 options to cover the longer FL collimator fresnel. 1. Leave it the same distance from the lamp, and accept that you lose light betweent he precondenser and the fresnel, or 2. Increase the distance from the lamp to the precondenser and lose light due to the square law there.

Changing the precondenser to a shorter FL, I will grant that, in addition to the longer FL collimator can produce the same amount of light, at the expense of a larger virtual arc.

ahh..... I was always assuming that you would have to use different precondensers with different focal length rear fresnels ... so you agree that barring the problems a larger virtual arc might cause, that the 330,, with the correct condenser will produce the same lumens as a 220 rear fresnel and no condenser will? throw in the increased vignetting gained with a precondenser and you've got a winner ...

I don't take increased vignetting as a given either. One of the detrimental effects of a larger arc is worse vignetting, for example.

Also, the closer the lamp gets to the precondenser, the greater the effect of the square law on the light that hits the precondenser lens, which will also be to the detriment of vignetting. Only the use of an aspherical lens can correct this, and there is still a limit to how much correction it can offer without producing very nasty distortion effects. The result is a diminishing returns effect.

I still say that AV is better off just solving the cooling problem, and leaving the optics as is.

huh? The larger the arc the worse the Vig?

Also, the closer the cond to the arc, the more light that gets into the condensor, thus less light going off to never never land. The closer the arc to the condensor, the better, if you have the correct condensor.

Thought I'd throw this in here in case anybody's interested. Now it may seem in this pdf that a longer fl precondenser is better but I think we are not using the lens to collimate but more so converge as much light to the rear fresnel to have it do it's job and collimate more light through the lcd. Depending on the right combination of precon and fresnel will produce better collimation of the most convergence collected from the lamp. Because we are not using a point source here and our arcs are largely elongated we are finding a larger size precondenser is needed. By bringing the precondenser closer to the arc produces a tighter convergence and more light for the fresnel to collimate.

Just my extra 2 cents guys, this is excellent brainstorming.

I was trying to keep it simple. I have always thought that a 330mm Fresnel and a 6" x 120mm fl condenser would be a better match for me or even a 100mm fl condenser. But we don't have one.

To go to a 330 and use the same condenser then max LUX happens with the lamp further from the condenser than it is now, if the lamp is farther away it will be dimmer but I may need to do that to keep things cool with the 2000 watt lamp or the MHI 1200 watt lamps run at 1500 watt or so.

I am loosing about 15% from my heat glass so if I could remove it I will get some of the LUX back.

I did not give all the distanced of everything but I have about 27mm from the lamp center to the back of the condenser lens. If I go to a 550mm front Fresnel I may have only about 15mm from the condenser.

I will try a few different setups to see what I like the best.

The one strange thing I had with this lamp setup was rings of bright spots when I moved the reflector too close and rings of light on the screen when I moved the lamp and condenser too close. There was a very small range that made even light.

With the heat glass, fan, condenser, lamp, reflector, it is kind of hard to get everything exactly right. This is a touchy setup and don't like the sun shining on my hand. And I can’t feel any thing with gloves on.

I have six of the HTI lamps they have a great color and I'm going to sell some of them cheep. They will run just fine on single S106 600 watt ballast at 1100 watts. That’s what I'm using now. You can run the lamps at 900 watts or so and I bet they would last longer.

I have a second MHI 1200 watt lamp coming and the HQI 2000 watt so I'm not going to use the HTI lamps.

Sim: That reminds me I was planning on using a 165mm and a 400mm condenser together for a test, if the second condenser lens is OK then I may place the 400mm lens on my drill press with a small sanding disk and slowly grind a flat spot in the middle of the condenser lens.

I will only flatten the middle 30% of the lens or so.

This will make the lens a corrective lens with the light on the outside will be condensed but the light in the middle will be unshaped. I think I could get a large increase in the vengeting by doing this.

The surface of the center of the lens that I'm grinding on is not that important. I want the light to be slightly diffused any how and I'm sure I can get a fairly even surface .

jeez, you really are the king of all extreme mods aren't you ?

I think something like this might be interesting.

I'll bite optical ignoramous that I am. I guess the idea of this is to take the hot center and spread it out to the periphery for better ROX? Ya'll start doin' dis I think some go-getter should try to whittle my rectangular pre-con idea out of a block of transparent soap for a REALLY short-lived experiment in optimization!

whittle yeah this is exactly what I was thinking. Either way close or further away (depending on what combination we use i.e. precon or fres) the center is always brighter. Having a concave at the middle of the precon spreads this out, I would think. Of course the abberation would probably be more but I'm depending on the fres to straighten this out, but definately the corners will benefit from it

This should really be moved. It is so OT from AV's plog and there is really good info here that others can also really benefit from.

Actually, I think it is so AV on topic He is king of different tests

It’s the same thing just on the other side of the lens. I have no way to make a smooth curve on the flat side. If you did dish the flat side the size of the lens the cup would have to be small it should work too. I don't have a way to do that with simple grinding tools. A stone with the right curve might work. I can use a flat grinding surface to make a flat spot on the curved side of the lens.

If I ever get started on this I will start a thread.

I generally have conversations with three topics at one and it drives people crazy. I have always liked it that way.

yeah, drives my wife crazy when my kids come over we ahve at least 3-4 conversations going on....ok, we'll leave the "OT" here ...

The main point of this pic was to show that the rear fresnel is not a factor in light collection when a pre-con is used, at least not in the same way that it does without the pre-con . There is just to many variables that influence the final outcome so I deliberately left out allot for simplicity. Maybe I should have mentioned that the pre-con’s diameter and the distance from it to the arc is the same for the 220/330 and the 330/330 examples. I just took it for granted that this was understood, my mistake. This obviously means that the FL of each pre-con will be different.

DJ

I’m not too confident on flattening the pre-con to fix vignetting. I think you’ll end up with two separate light sources so you will get two images at the triplet.
Click to view attachment

Point source and only one surface refraction shown for simplicity.

DJ

Two images are OK as long as the fl is not a lot different. What you really get is the middle of the lens does nothing because it is flat. The light will not be condensed in the middle so will continue to spread out. The outside will condense so there will be light gain.

This no different than what they do with compound shape reflector that have a even power spread, by designed they would make a distorted lamp image but they don't care about the image but the power transfer is more even.

I plan on doing this to a 400mm fl lens any how it is not vary curved any how so the change is not dramatic.

The lamp image is not important but the power transfer is.

Your second drawing just what I want to happen the one ray that would have dirested to the middle of the screen now is going to the edge of the LCD... perfect now the edge will be brighter. In the first drawing the same ray is directed to the middle of the LCD

That’s the point they are very different from each other. It’s a common misconception that if you mealy light the LCD evenly all is good. Any light that illuminates the LCD must come from a single source or close t it. By flattening the lens you have taken its magnification away, as you have said, but you need to realise that now you have two light sources. The real arc and the virtual arc and they are separated by some distance. Take your 6” lens as an example, if you trace the marginal rays back past the edge of the lens a virtual image will form at about 90mm from the plano surface. If you have the arc at say 25mm from the plano surface then this means you would have a 65mm separating the two light sources, not good. I’ve drawn another ray trace that shows where the two separate arc images will form and as you can see a fair amount of light will be lost.

DJ
Click to view attachment

edit:
Just to note, this probably will even the image but it’s probably no better than just blocking a section of the pre-con. So potential gains of the pre-con won’t be as great.

agree

me too..

AV I’m all for experimenting but I also like to have a theory to why something will work or not. The last couple of post are just that, my theory, or a guide. The only reason I posted it was to give you something to look for in your results.
I’ve seen that you have already started the lens grind in the other thread. I’ll be interested in the final outcome, good or bad.

DJ

Actually, I think this is one of the primary things that a precondenser should be able to do for you.

The 'important' thing is to get more light around the boundary, thus reducing vignetting. It's okay to lose some of the light through the center area where there's already a higher value.

Where I'd be concerned is that there's a possibility of a "ring" effect where the corners and the center are relatively even, but there may be a bright ring where effects of both focal fields converge. This would result in a brighter ring in the projection, instead of a smooth transfer. IMO, this would be a lot more visible than the regular vignetting.

Well the grind was not as deep as I would like. I have a lens with a 400mm fl to start with. I was thinking of using two 400mm lens because well I have 4. The overall fl would be around 180mm (that what a simple test showed.)

I might grind some more later.

This is still way better than simply blocking the light. Most of the light going through the flat spot will still hit the edge of the Fresnel so there will be vary little total loss.

Ideally we would make a compound lens but this is something I can't do.

What happens with an optimized PJ reflector? They have to redirect some of the light that would go to the middle to the edge or they would not get the light pattern they have.

I now have two new lamps, a second MHI 1200 lamp and a HQI 2000 watt lamp. I'm having a hard time getting the 2k to start with the ballast I have now but I will get it running it just may take some time.

I have talked about the MHI lamp before but I only had one and if I want it to be my main lamp I needed a spare now I have two so for now these are the lamps for the PJ.

The last pictures I took with the HTI lamps looked washed out so I spent some time looking for the problem. It was some of the adjustments on the LCD and a lot of light leaks so I got out the tape and fixed most of them.

I did some contrast readings with two lamps. One is a 500 watt elliptical studio light with a 3.5" reflector the other is the MHI 1200 watt lamp running at about 900 watts.

The 500 watt studio light is called a smith victor.

I did 3 LUX readings across the screen and got 24 88 18 The SV lamps were always hot in the middle. I took the contrast reading from the middle of the screen with the cool program from paladin (Thank you for the work)

The LUX in the middle was 74 and black was 7 stray light just off the screen was 3 lux. One thing about an ellipse reflector is there is no stray light.

I then installed the HMI lamp and heat shield

The LUX across the screen was 103 198 194

Then the contrast was 188 white and 20 black

The stray light was 22. Yes there is more light bouncing off the back than was than from the screen.

I did block a lot of the light leaks so after fixing the LCD settings the screen looks great.

Click to view attachment

Click to view attachment

My cracked LCD has now formed a slight blob in the middle and flickers from the vibration front the fans but keeps on working.
And with a 500 watt lamp on next to the screen.

Click to view attachment

I don’t think contrast is a problem at 900 watts and may be fine for a lot higher power. I will up the power to the MSI lamp to around 1400 watt soon. And do another test.

To me, it appears that the blues in that picture are weak, though whether this is the camera or projection, I can't tell.

Overall, the brightness that you're getting is very impressive.

The lamp is underdriven and new, it looked green but that will change as it breaks in. I am going to run it a little more than full power tonight amd adjust for max brightness at the edge.

The white screen with the new lamp. Looks green.

Click to view attachment

That would explain it. I took another look, and the reds also looked weak, but it didn't "jump out" as much. This would make sense with the green lamp colour.

I am now resizing the pictures at 500x 380 so you can post two side by side and it will still fit on a 1024x768 screen. Maybe?

I decided to remove the antiglare on my cracked NEC LCD at first I though adding some heat might make it faster but after a few hours I think all it did was dry out the power towels faster.

Click to view attachment Click to view attachment

Click to view attachment

It took about 8 hours before it loosened up it was still tricky to not take up the polarizer I used a razor and a magnifying glass to make sure. I think it could have gone longer but it did come off in one piece. The glue was all over and would not wipe off but I put the papertowles on for a few a little bit and it all came off.

The crack is filling in with oil and the black spot got much larger from moving it around.

I did to the ultimate sin I unplugged the controller cable when the LCD was on! I was not vary careful with the FFC either so now the LCD is dead.

Click to view attachment

I did jiggle some wires and got some picture on the good half.

Click to view attachment

What I really wanted to see is how much brighter it was wit out the antiglare.

I have not moved the lamp front the last test from last night. I had 103 198 194 average=165 lux last night.

Tonight with the antiglare removed I have 114 259 218 LUX for an 197 average. I got about 16% increases in LUX.

I wish I knew if unplugging the LCD killed it or the movement of the ffc’s or the water?

I might have a spare LCD soon.

Those streaks look like FFC damage to me.

Naw, the one I broke did the exact same thing. I rubbed mine, and got some areas back, but it was pretty much toast on the side that can't get electricity or whatever it is, to it. It was the movement of the lcd that did it. It will be your test lcd You may be able to find the same lcd, with a bad contrast button or something like the other one you had, and use the control board from this, if it is exactly the same one.

Wow AZ, I hadn't read this topic yet! Great work on PJ so far I'm going to read up and get myself up to speed.

I'd put my money on FFCs or water or possibly ESD?? I've knocked the controller to LCD cable off mine more than 10 times while running and it's still working fine.

Well there is that gigantic CRACK in the middle too. This was just for fun to see how it goes but I almost had a great picture.

I still may go to full power once just to find the LUX. I got outbid on the cracked ACER LCD to replace the bad controler on the acer I have. That LCD is still good.

I got back the LG 1400 to 1 contrast LCD today and I posted it in the trading post. I looked at it a lot closer and it does have a lot more color and contrast than the crappy NEC I have been using.

I am thinking of the ASUS glare screen. some one is going to strip one soon.

vvebsta :Thanks for reading I hope you enjoy.

AZ where did you get your test pic of the girl? Can I use it too?

Did you do an ANSI reading on your latest power up?

It has been a long time since I got it. It does say DIYPro on it. It was 345k but I made it 100k for posting. You can just right click it.

Click to view attachment

I am going to go ahead and use the LG 1752TX 17” LCD. It has a great picture and some one said a real 700 to 1 contrast.

I was looking at the cracked NEC and the half the was not working looks to be from the crack spreading not the ffc’s . It is still junk and I may still try to find a really cheep 17” LCD for testing. I have an Acer with a contrast problem that might be fixed with a new controller.

After next week I should be able to spend some real time on the PJ.

I did a 1300 watt test with no LCD in the condenser lens thread.

I also have a 1000 watt HPS ballast coming for the 2000 watt HQI sports lamp.

Hey AV,

What triplet are you going to use? I know you really liked Simul8r's 18" opaque projector triplet lens.
I couldn't remember if his triplet would crack if exposed to the heat?
Especially, by your Fusion light source.

In your opinion what is the best triplet to use. I was going to buy the pro lens kit, but would like some advice
from a LL Genius. I'm going to wait to your pj is finished before I build mine.

Hey any chance your going to use a precondensor, even with all those lumens?

The triplet is far from the lamp especially if it is like AV's so there wouldn't be a worry of it ever cracking. I would be more concern with the LCD especially if he's using a precon. Efficiently cooling it will help a great deal or you will see the crystals begin to darken. With AV's bulb you definately need a high temp precondenser or some serious cooling in the works.

Kryptonian: The pro lens is the best lens for focus and throw. It also cost the most.

I do not have perfect focus with the 450mm fl Buhl lens, it is fine for movies but you can see the soft focus on the windows desktop.

The 450mm x 128mm triplets should be a little brighter than the pro triplet but you lamp choice will tell you what triplet to use. Short arc lamp like the ceramic or SD400 go pro HQI or T15 lamp go 450mm. Always use a condenser lens.

Read Simulator's plog and maybe the HQI thread. Some of us are nuts and make it a lot more work than it has to be but we have solved the brightness problem really( read the condenser lens thread and the over clocking lamps and ballast thread) and the new small ceramic lamp might be great for some compact folded designees.


"In your opinion what is the best triplet to use. I was going to buy the pro lens kit, but would like some advice
from a LL Genius. I'm going to wait to your pj is finished before I build mine."

I would not wait for me I'm slow. Send be a PM if you want any ideas.

Back in action. I have gotten everything out of the way and can now spend some real time on the PJ. I have all the parts I need.

For those that have forgotten.

The lamp is the Philips Broadway 1200 watt MHI lamp running on two S106 ballast at full power.

The triplet is the American Optics 450mm x 120mm lens.

The Fresnels are the 600mm pro Fresnel and the 3dlens 330mm Fresnel.

The reflector is the 8" Colortran Fresnel spotlight reflector.

The condenser lens is a 6x9 from an old Century elliptical spotlight.

The LCD is the LG L1652TX Flatron 1400 to 1 contrast which is really a 700 to 1 panel

I will have to redo my light box because the first one is 1" too large to allow full adjustment once inside the box. It will work with the 330mm fresnel but I might go to a longer fresnel setup if I ever get the 2000 watt lamps.

I plan on finishing the basic box and sled this week and have a function PJ by next Sunday.


I finished the cooling section on the base tonight. I'm mounting the two ballast in the bottom and the caps in the bottom but the neither will go next to the lamp. I added a set of wheels to the bottom of the base so I can move it around. They a little bouncy so I might try a different kind.


The next few days are going to be rapid build so I will keep the camera batteries charged.

Click to view attachment

The hole is for a 120mm fan on the inside of the box.

Click to view attachment

looks solid,, reminds me of a giant 25 inch diameter telescope I made a few years ago,, it rolled on heavy castors

I striped the 17" LG panel tonight. All went well The hard part is just geting the shell to split after you do that the OSD controls stay in the frame so don't just lift the plastic shell off.

Click to view attachment

The controler box comes off in one peice just unplug all the conectors and remove the tapeand it will let you flip it to get to the last two conections.

Click to view attachment

Click to view attachment

With the controler boad now free that leaves the back light You have to remove about 6 small screws and then the frame is held togather with built in clips that need to be pried open with a snall flat blade screwdriver. It's fairly easy.

Place the flat blade here.

Click to view attachment

I bent back the top and bottom slightly with my fingers so the clips would not relock.

Remove the frame and its free.

Click to view attachment

I really like the fact that the panel does not have the small ffc on the side like most panels, this will make mounting it far easier.

All in all the strip was fast and easy with a perfect pannel for mounting.

The controler and inverter can stay in ther box and it makes for easy mounting.

Awww yah! Welcome to the LG club!

I liked the way the LG lookes I just wanted to try the high glare screens to save the risk of the strip.


I remade the light box today it now will fit in my box I still have to mount the lamp holder and reflector mount and add the vent and fan holes.

Click to view attachment

Click to view attachment

Click to view attachment

I made the back curved and it only has two screws for easy access to adjust the lamp and reflector. I had a job call and had to stop working for today.

Glad to see you getting this project under way AV!

Thanks Lucky.


Well things keep getting in the was slowing me down but it doesn't help when I remake the light box 1" too narrow for the lamp.


I had to go back to the scrap yard for some more aluminum and I ran in to some S51 ballast. They are the advance brand and look to have low hours. So I bought them. There was some one who wanted to test a HMI575 wit the S51 so I may keep them for a while.

Click to view attachment

I will have to remake the light box one more time. I won't be done this week but will be close.

If any one wants old light box it is free for the asking. You would need the small 150 watt ceramic lamp to fit inside. The 250 watt HQI fc2 might fit too.

are you planning on using tempered glass for a heat shield or just have the condensor/light box setup cool itself and not worry about extra shielding for the lcd? (also are you going to have a sheet of lexan between the lamp and lcd?)

I have 4 types of heat glass. A 1/4" thick 7" x5.5" hot mirror from the 1000 watt opaque pj , a 10 x12 tempered glass plate, a 10" round heat glass from a Colortran spotlight that is two pieces, and a piece of lexan.

Clear glass lets almost all the heat through, the hot mirror is just a little too small and the Colortran heat glass you can see the line between the two half and the lexan may still melt. I don't have a perfect heat shield yet.

I have done all the testing with the Colortran glass right in front of the condenser lens. Some times the seam looked bad and some times I could hardly see it. It did reduce the heat to safe levels.

I have not done a full scale test of the final cooling system which will be three 38mm x 120mm fans each pushing 105 to 140 CFM. They will be mounted to a duct on top of the LCD facing inside the box in a u shape.

The problem I have with the air flow is I have not planed on having the extra lexan or glass behind the rear Fresnel. With out this I may have a problem with it melting because I can't force air across the outside of it with out the lexan.

I may use the 10 x12 tempered glass to make air flow over the rear Fresnel but it will have to be over an inch away because it is not larger than the LCD.

It is easy to make changes to the heat shield so only time will tell.

I finally got to spend some quality time with the table saw and router so I now have a box. I knew it would be large and it is.

Click to view attachment

Right now it is 8" extra long incase I go with a totally different light setup for the 2000 watt lamp. The 2000 watt HQI makes even more heat so using a longer rear Fresnel will keep things cool. By using a 550mm/550mm Fresnel and a 6x9 and a 6x16 condenser lens everything is perfect again. I did a quick test setup and it looked OK. I did pay for three 2000 watt HQI lamps I hope they make the trip safely.

The box turned out OK it is fairly light with the sides made out of 5/8 plywood and the bottom and top made of 1/4" hardboard. I need to make a latch for the rear of the top. The top fits so well I don’t know if I will even add a foam seal.

Click to view attachment

The cooling system does not need the lid on to cool the LCD and I planed it this way. With the 1200 watt HMI there would be no way to adjust the lamp with the top off before the LCD turned black.

I have lots more little things to do both too the light box and the pj box but next is the LCD sled and cooling system.

Click to view attachment

That is the third light box and I finally made it the right size! I though I had lots of extra room but the more I look at it I had only a ½” of space on the top and bottom If I have to move the light box all the way back. The main problem is the 8” reflector I needed room to get to the adjuster on the bottom so it needed to be a little taller but then the box would be too large. The 6” Colortran reflector would make things a lot easier or don’t do a Hass box.

Great to see that your making progress AV. What's her overall length?

It's actually a 16 foot catamaran. He just hasn't gotten the other skiff and the sails
worked out yet.

But seriously, awesome work AV.