Here are my projector plans. I'm going to be using a 15" 16ms LCD, I dont know what brand yet. This will be made of 1/2' MDF, I'm going to use the basic lens kit. Most of it will be built in my dad's workshop, as he has some pretty nice equipment (like a tablesaw, a mitre saw, a radial arm saw, drill press, routers, lots of fasteners, and workspace). I'll do all the electronics stuff here.

I'm not really certain how I'm supposed to do the keystone system. Where should the lens pivot? Also, does this have to have fine adjustment (like a screw drive) or can it just be a "loosen the friction nut and move it then retighten" kind of deal.

My focus system is one that I have been thinking of using, but isnt permanent. Basically, it is like a box in a box design without the box. There is an assembly that slides on rails inside the box and can be tightened from the outside with knobs. The front of the panel will have a hole cut out to let the lens move in and out of it.

The cooling system will consist of 2 120mm computer cooling fans attahced to a rheobus. They will be built into the control panel. I'l lfind some speaker mesh to act as a light filter and as an air filter on the intake side.

I'll be using a 400 watt bulb and a big coil ballast, but I'm not sure what I should get and what is compatible. I'm building this for as cheap as possible, so anything that can cut costs without sacraficing much of the usability would be preferable.

Lastly, has anyone concluded whether or not it would be possible to use lexan for the UV screen? If this lightens it any or makes it cheaper than glass it would be very helpful.

Thanks for your interest!

The ballast will get too hot there. The keystone could affect the air inlet.

Nice design, looks good.
Brain has some good points about cooling, maybe you could mount the ballast behind the box on the outside.
Lexan, at least the XL10 seems to be the prefered UV shield you just need to make sure the cooling is good enought to avoit bowing.
2 x 120mm fans seems like plenty of cooling so you just need to work on you cooling paths.

Would there be any benefits to heatsinking the ballast? I could move it lower I guess, perhaps below or beside the lamp to avoid some of its convection. I tried to position the fans above the lamp tor remove as much of this hot air as possible.

Also, what is the ideal lamp to use in a compact space? Is it the E28 lamp or whatever its called? Also, I think I have a soup ladel that is pretty large, but there is a slight problem with it. It has 5 small holes drilled in the bottom to allow liquid to flow through, but I'm not sure if this will affect anything. Does the reflection sphere have to be flawless to get a good picture?

Why are you setting up the fans to pull air through instead of pushing it through?

With where you are placing your fans, you will have LARGE amounts of light leaking from your PJ. Your ceiling will be lite up like a parking lot killing your contrast on your screens picture.

I would also point out that the amount of air passage space (or lack of) with this design will "starve" your fans. I would have the air inlet at the top and allow air to pass "down" in both front and back of the lcd, then out the "back" of the PJ. Similar to the LL design. Give it more than the designed .5" (at least 1") of a gap for air to pass under, as that will be much too small a gap for 2 fans.

A 120mm fan is better than 12 square inches of air passageway. The design calls for about 7sq in. underneath the panels. Not enough, especially for 2 fans.

.02

I'm not going to have the fans just wide open up there, there is going to be a heatshielding system that will block out most of the light, and they will also have a mesh grille to keep fingers out and light in. I'm not really concerned much about the cooling because I'll be using 38mm thick fans. These pull air with far more pressure than standard 25mm fans, so it should be able to draw air through the slit easily. Also, Air velocity has far more effect on cooling potential than air quantity. If there is very little room between the lenses and the LCD, you will have a higher air velocity with the same fan configuration and therefore the LCD will run cooler.

I know a lot about thermodynamics, I think I can find an adequate solution.

As for the keystone correction, I have see it done both ways but wasnt sure what is the best way to do it. Can the lens be hinged anywhere? Which way produces the best picture?

EDIT: A 120mm fan has approximately 100mm squared of actual fan area. The motor hub and the outer edges must be subtracted.

Also, has anyone ever liquid cooled a PJ? I mean, has anyone ever submerged the LCD in between 2 lenses filled with glycol? This is done to cool CRTs on my projection TV, I was wondering if it would be possible in a PJ, sealing all the elctronics of course.

Actually, it is a pretty good idea to push cold air in instead of pulling hot air out. This is because cold air is denser. So a 120 cfm fan pushing cold air in will cycle more air mass than a 120 cfm fan pulling hot air out. The volumes will be the same, but the densities will be different. This will have a beneficial effect on cooling.

I think if some of the other concerns (such as light leakage) are adequately addressed then this is a more than sufficient solution.

Forget that last response. I came back to this thread and just read the last few posts and misinterpreted the layout. It seems you will be pulling hot air out regardless of option.

The shift in the density of cold/hot air is utterly negligible, as is the shift in the density of the air due to higher pressure inside the box.

Airflow over heated surfaces > *.*

Isn't that what I said? I said it's "better than 12 sq inches", which is around 88mm square.

I took in account for the motor hub and was being generous. 100mm square is more than 14 square inches.

2x120mm fans is 28 sq inches. With a 7 sq in air space, you will be creating a heavy vacuum.

Then you would know that a "vacuum" is less sufficient at cooling than "pressure".

Anyway, no need to be defensive (that's kind of how it sounded ). Hey, you just asked for your layout to be crituqued... If you don't like to be critiqued, you shouldn't ask to be.

Another idea to block the light from going through the fans is using louvers. I'll post some pics on how I did my PJ to give you some ideas.

I have about zero light leaking from my PJ.

Anyway, gotta get back to work.

Cheers!

Not much of a vaccum really; an average 25mm 120mm can only produce 3mm h20 of pressure. Even a fairly powerful 38mm fan can only create around 10mm h20 of static pressure. This is not enough to create a vaccuum.

Fans also work more effectively drawing air out rather than pushing air in. The only time a fan is used to blow on something is when something needs a directed flow of air to stay cool. CPU fans usually blow in for this reason because axial fans create a cone of airflow out of the exhaust.

I'm an avid computer overclocker and am well versed in the methods of cooling electronics. I have a hand made liquid cooling setup in my computer using my own waterblocks on some parts. I know what I'm talking about.

I started looking towards blowers (squirrel cage fans) for cooling this unit for the light leakage reason: an axial fan will allow light through because air flows right through it. A blower turns the air and blows it out centrifugally. This turn makes it possible to block off all possible leaks for light. These blowers also offer much higher static pressures than axial fans, but take up more room. I guess it is a trade off, but if I can engineer it in there I might go that route.


Back on topic: I'm still waiting for a response about the keystone mechanism. Is it better to have a top hinge? Also, should there be a fine adjustment on this or can it be set by hand and secured in place?

Center hinge in better IMO because it keeps the focal point centered.

When you hinge at the top (or bottom) it tends to swing the focal point off center affecting brightness. That was the case I was having.

Thats what I wanted to hear. Thank you very much for that concise response

not to hijack the thread, but WOW, nice brightness Agent with that overhead light on!!!

Bought the fan, I just saw it and noticed the fins overlap. This is good because it means light will not be able to pass through as easily. Also, it is very thick and should push good airflow even while undervolted. Pressure is going to be good as well. This fan just seemed ideal to me.

Using PV=nRT we can calculate the density difference between the cold side air and the hot side air.

n=PV/(RT)

therefore

n1/n2 = T2/T1 assuming pressure and volume moving through the fans stays the same. Therefore n1/n2 = 311K/294K (assuming 70F cold side and 100F hot side) = 1.06

This equates to a 6% increase in density if you put the fan on the cold side instead of the hot side.

Assuming convective cooling is proportional to density, this would equate to approximately 6% more efficient cooling.

Is this "utterly negligible" - probably not. Is this large enough to justify going through significant design changes to place the fan on the inlet - probably not.

No, you are wrong. A if the air mass is denser, then the fan will be working harder to push it through. This will slow down the revs and reduce airflow. It may look like 6% better cooling on paper, but many other factors will bring this down a lot. There are too many variables to simply say, "yeah, this is 6% better when you do this".

This is a chart compiled by a forum member at Procooling.com. His data is used in the testing of waterblocks and liquid cooling systems, but the same physics are being applied here:


According to this chart, when fans are pulling air through a heat exchanger, the heat exchanger operates at a lower temperature. Pushing air through results in an increase of heat.

Uhh, don't see it. Can't....Read....Chart.

Could you explain some of the terms.

Thanks.

If what you say is true (which I take with a grain of salt) then pushing the air is better.... Why? Dense air can absorb more heat.

I'm not a scientist, nor do I speak n=PV/(RT). But you don't have to be a scientist to know denser = more mass = absorbs more.

We're really littering this guys thread too. There's a sticky for "COOLING THE BOX". (Hey, I can't help it, the topic is all caps. )

Cheers!

Sure Thing:
Basically what we are looking at is a chart compiled to show the benefits of using a shroud on a radiator and different fan configurations in a computer liquid cooling system: Both signifies a fan pushing air through and a fan pulling air through the radiator together. Pull signifies a fan pulling air through the radiator, and Push signifies air being pushed through the radiator. What this chart shows is that there is an overall reduction of temperatures when pulling air through a radiator rather than pushing air through it. The same physics are at work in the projector, so what this means is that pulling air through the projector is more effective than pushing air through with the same fans and cooling layout.

My computer is using much of the theory quantified in this chart. I think I'll attach a picture of it sometime, it is pretty rediculous.

Here is the simplified chart showing just what we need, notice the overall lower temperatures when pulling air through a system:

If I may take a stab at this.. I believe the reason the "pulling air via a shroud" works better (in these cases) is only because you get "even" air flow across the radiator. "blowing" air into a shroud to cool it isn't as affective as some areas will get a Lot more air than other areas. If you put your hand on the other side of the radiator, you'll feel more air flow in some areas vs. others.

The area where the air velocity is highest (directly below the fan) will have less cooling affect per air mass because the air mass is not absorbing as much heat per air mass (if that makes sense).

It is even air flow across the radiator allows the most transfer of heat. Has nothing to do with push or pull.

The same thing will happen in a LCD monitor though...You dont want to only cool a cone shaped pattern of it, especially if you dont want the extra noise of several small fans.

Pulling air through guarentees even flow over the entire LCD and therefore cooler operation.

That chart is primarily there to disprove the belief that less air density = less cooling in projector applications.

This could actually go on and on like the 1000+ post count "reflector" thread. *shrug*...

Anyway,This is not really comparing apples to apples. There is quite a bit of difference in the surfaces we're talking about.

LCD surface is smooth thus the air will spread across it fairly eaily. "Evenly"? No, you will have some surface getting more air than other portions.

However, a radiators surface is not flat (smooth). When the air hits it, it's going to go through it there... it's not likely to spreadout because the way it's vented, plus you are blowing through it, not across it.

Anyway, like I said, we're just beating up a dead horse. I'll leave it at that.

One known fact. Use Fans... Cool Your PJ. Looks Good. Last Long Time. Where you aim your fans is where you aim them. If someone tells you what to do with your fans, you can just tell them what they can do with THEIR Fans!

hehe. Good luck with the PJ noodle.

p.s. Don't over-clock it!