It sounds logical to place the fan at the top, and make the slot at the bottom of the box for inflow. However this would work only for those who plan to mount the box from the top. Like most of them, if you plan to keep the box on a platform, the slot will be blocked and would not be able to draw any air.

Also, it looks like having a slot at the top will not be effective in this case, as the fresh air will simply enter from the slot and exit from the fan. without really cooling your enclosure.

Thirdly, leaving a tempered glass gap at the bottom and keeping the slot as well as fan at the top will again not be a good idea because the path of airflow between these 3 points is changed twice, and will lose the suction force.

Looks like Brainchild had well pondered over these while designing the box

You would always want the gap of the tempered glass opposite of where you have the slot so all the air passes by the lcd before being drawn out of the box. Brain has a good plan that has served him well, but if you would like to experiment feel free.

I appreciate all your replys.. maybe I didnt word the question right... I'm not at all doubting brains ideas...hes the man! What I was wondering was would it work better to mount the fan where indicated in the design, but reverse it so its blowing air in, and letting the heat escape thru the top vent. That way the air flow path will still be crossing the lcd. the only possible negative effect I see on this situation could possibly be that the heat from the lamp was passed along the LCD... I don't know.. maybe that wouldnt be an issue. I may experiment with that prior to placing the LCD in the exposre and take some temps... I'll keep it posted. Before then if anyone has any other thoughts, feel free!

Later!
Hed

I have read here on the forums where it's been tried, and pushing cool air in is not as efficient as pulling hot air out by a pretty wide margin.

Now, with one pushing and one pulling -- that would definately help, but the noise of 2 fans can be a bit excessive to some people. The common solution there is to include variable speed fans and cut the RPMs down to decrease the noise. You also have to make sure you aren't pushing cold air on your bulb.

Mounting the fan(s) on isolating "shock mounts" also helps with the noise factor -- those little suckers can be quite loud when you mount them directly to what amounts to a large wooden echo chamber.

Also keep in mind that when using 2 fans you can get also some funky harmonics going, so making them variable speed can help with that as well.

I have also seen some temperature controlled PC fan circuits on the market -- the hotter they get, the faster they spin -- not sure if they would work in this application though -- the heat in the projector exceeds the heat inside a PC by a good bit.

If heat is a major concern in your build, you may want to consider the added expense of a solid state ballast -- they run much cooler, and can be included in your design as an external component since they are fully enclosed and heatsinked.

~DM~

Hi folks,

Extract air from the box, do not push air in. If you want to push air in, then create a forced air-cooling system. Place a fan blowing inside the intake and a powerful fan pulling air out of the light chamber.

Five things to remember (I might be missing some points but this will help)

1: Keep your ballast cool, run the fan behind your ballast to insure it has proper airflow.

2: Make sure your intake slot is just the right size, not to big or small.

3: The intake needs to be positioned properly; you need to seal the front chamber and the light chamber. (The air is pulled down remember, think about it logically)

4: Light equals heat, so if you draw hot air over the LCD then add heat building from the intense light; you will fry your LCD.

5: Room temp does matter, if you are in a hot house, you will need to increase your airflow.

Cheers,
Lee

I would like to add that the original design was driven by placement issues with the old gear as well as efficacy. Since I've moved over to the tubular HQI and the electronic ballast I've relocated my fan to the top of the enclosure which has further dropped the temp in the box. If you can conceive of a way to do this I recommend you do so.

xanderphillips, you are correct for the most part.

Sealed box, and filter clothe on the intake.

You now have reduced the fan to a vacuum, removing its total CFM capacity. Even if you could enlarge the slot (Or add slots), you will still be under a small vacuum inside the chamber itself. If you use forced air you start to reduce the over all pressure of the vacuum inside the box thus relieving some of the pressure on the fan.

Two things, you improve over all airflow, and you can extract air from a cool air source with the forced airside.

I have tried this on hot nights, and it does improve the temp.

Note: What brain suggested about the fan at the top of the box would be a better modification. You are then extracting the heat more efficiently.


Cheers,
Lee

Hi,

Just a question, will the schematic diagram i have made work sufficiently, because this is how i am going to wire it up.

Thanks,
adam

I plan to route my air from the bottom of the projector and out the top at a 45º angle.

Rorshach - i thought you where just kidding when you said you where going to make a pj like the one above - thats awesome -

Yep, I will FINALLY! start on it soon. Waiting on my father-in-law to return from vacation as he has a nice wood working shop. Watch for a Rorshach thread soon!

Hey Brain!..
I may try that. DO you have the air vent still located at the top, or did yo7u locate that somewhere else too?

Still on top.

How much room do I need behind my projector fan? Because the back of my projector will be sitting pretty close to a wall, I will have a couple of inches of space. So I need to determine whether to put the fan on the back or the side.

P.S. Rorsach, very cool looking design!

Adam, as a fellow newb I thought I would respond to your post. Your drawing looks good to me except I am unsure why you have a 240V fan. In other countries do they carry 240 V through one line? If you live in the US then you need two lines comming in to run 240V equipment. What you show is proper for a 120V fan.

Wouldn't it be easier , just to install a 172mm 120Vac fan , with approx 240cfm ? Anyone ever use one of those ?

Should I put my fans at the top of the box???

Not needed.

What would be the best fan for cooling that won't run me out of the room with a loud noise? Are you guys sure one fan will suffice? I keep reading about people who box runs hot with one fan. Should I add two fans? Where exactly would I put "breathing holes" in my box if they needed it??

Hundreds of boxes have been made that do NOT run hot with a single 120mm fan following the recommendations in the guide.

If a design requires you to mount lcd electronics on the lamp side of the enclosure, are there any other precautions that need to be taken other than covering them with flashing?

Sufficient airflow.

I have a cooling question.

I am wondering what the pros/cons are for using two separate air-paths. Specifically, if the tempered glass completely sealed the lamp off from the lcd/fresnels and the cooling was set up with one fan/vent to cool the lamp area and one fan/vent just to draw air through the gap between the LCD and the tempered glass.

I'm aware there would be more holes in the case to watch for light leaks, but my thinking is the fans could be quieter, since neither air-path would have tight bends or restrictions in it.

I'm new, so if already covered, I'm sorry..

Did anyone ever consiter this types of units for as a cooling aid?

http://www.allelectronics.com/cgi-bin/cate...=775&type=store

It could work. If you try it let us know.

i was woundering about that reading through this whole post, if anyone has used a peltier cooler, you can get them really thin and if you wanted it to become cooler all you have to do is mount heatsinks on both side and add a fan on the hot side, ive seen this done to a styrofoam cooler were they were able to drop the inside temperature odown to -3 degrees celcius from 24 degrees above, thats about 74 degrees Fahrenheit, but some caution the hot side does get HOT but i think acceptable in this application, have acouple smaller ones or a single bigger one and add some heat sinks and still continue to run a fan and im sure it will cool it down more, might even be able to get it cold enough to be able to run it without a internal fan, but dont quote me on that one, there really neat to play with

i do know the tecnical aspects behind peltier coolers but i think that most people would just get confused

Hope this helps a bit

oh forgot to add for the amount that it would probably cost you to run some its not really worth it when you could just add another fan, but these would probably help in places where its consistantly hot

I was thinking just putting the unit at the front of the box, drawing cool area though
the lcd and lens back towards the fan..

Brain,


Why 1/8 firewall, why not 1/4 or 1/2 inch, would there be significant loss at that stage..

The LCD receives sufficient cooling if designed to my specifications.

I'm thinking of designing a water cooled system like here. Any comments?

The primary object of cooling your projector is to keep the LCD from overheating from radiated heat from the bulb (invisible IR radiation). The bulbs themselves are designed to be in enclosed fixtures (there are probably ventilation holes). so cooling the bulb is not as important as keeping the LCD cool.

Now having a layer of water may do well to filter IR radiation and help keep the LCD cool (I don't know for sure, but I think water absorbs UV and IR). But water has a different refractive index to air, place a spoon in a glass of water and look at the side, the spoon will be bent (then again maybe "There IS NO spoon" ). So the water may bend the light so its not at the focal length of the condenser fresnel. Also, there's the question of algae growth in the system, and is the water perfectly clear?

Lastly, water and high voltage do not mix well if a leak in the system happens to form. The arc is started with a very high voltage pulse to create an arc and uses high voltage to maintain the arc - that's what the ballast is for. Let alone water an 120v

Food for thought, though the idea is kind of cool

Ed

At >£400 a throw ?

There's no way I'm spending £400. I'm thinking of designing a fixed system similar to theirs with internal reflector and ports on the side for optical coupling for fiber optics for lighting other than for the projector.

I'm using GE's MVR250/U that outputs 16K lumens. I know the envelope filters UV radiation. And I know that water helps filter that too. I'm not sure about IR though. I ran a test and found that most of the heat is convected and rises. Very little was radiated to the black panels if any as it stayed the same as ambient. So I'm guessing the envelope filters IR too.

I should be able to design a collimating optics engine for the lamp. Most of the diffraction occurs from the photons flowing through dissimilar materials (while your underwater the spoon appears the same). It's somewhat relative to electrons flowing through dissimilar materials. I'll have the tempered glass in front flat.

I want so much to store instead of wasting that heat. I'm thinking of using peltiers to pump the heat to a hot water storage tank. Algae growth is something I hadn't thought of. I'm thinking distilled water and isopropyl alchohol to deter algae.

I've been an electronics engineer for 19 years and I'm pretty much a Jack of all trades, but maybe I'm trying to make it more complex than it needs to be. I should probably just K.I.S.S. it (Keep It Simple Stupid).

Can anyone comment on the amount of heat a projector vents into a room? Will a movie or two turn raise the ave temp of the room 3 degrees or 30 degree?

It all depends on what the outside temperature is and how big the room is and how many people are watching the movies.

Its a 400watt bulb so....It will add about 400watts to the room!
You can buy small plug-in room heaters. usually they are in the 800, 1200, watt range. They will easily warm a medium size room up in cold weather.
One person puts out about 100watts while watching TV - (150 if its and action flick)

Cheers,
Kirk

Seems like hot rooms would be a problem, especially in the summer. What is the heat output like on commercial projectors?

Metal halide bulbs are more efficient in relation Lumens per Watts.
As I know, Watts is, how many electricity are converted in heat in a device. Does a matter, what it is. If it has the some amount of watts will produce the same amount of heat.
Probable you will not feel the same with just a bulb in your room. Because the projector has fan which will force hot air, to move across your room.
So, I believe. How much it will increase the room temperature, and it will, depends more of your room than the projector.

All bulbs do not produce same amount of "heat" per watt. They put out the same amount of energy (radiation) per unit of time. That radiation can be in the visible spectrum, or it can be in the infrared spectrum. It could be in other EM bands such as gamma, microwave, various radio bands, etc. etc., but any commmerial light should have negligible output in those bands due to safety as well as fed. regulation.

A lamp with higher lumen output per watt will necessarily produce less radiation in the infrared band thus less "radiant heat". Or so the powers that be say.

R

Actually, watts are a unit of power, not electricity. This could be electrical power, mechanical power, dissipation of heat or radiation of photons per unit time, and so on.

Two lamps with different efficiencies could well generate different amounts of heat at the same wattage.

As a ridiculous example, if Lamp 1 converts 99% of the electrical power it uses into electromagnetic radiation in the visible spectrum, then a 400W bulb would be radiating heat at a rate of 4W. If Lamp 2 converts 1% of its electrical power to visible light, and the rest to heat, then it will be radiating heat at a rate of 396W.

When comparing lamps of the same type, your assumption that wattage is related to heat is probably valid. But for comparing different types, I really don't think that it is.


Hope this helps,
Phil.

Cooling the lamp area would be prudent. It can get hot enough to cook your mdf/wood. Even though the lamps are designed to place in closed environments they are usually done so with enclosures that dissipate heat much better than the “PJ” (metal.. usually aluminum) It is also generally regarded that these lights are not going to be in close proximity to human touch (i.e. Gym, telephone pole, rooftop). And that the threshold of heat tolerance on components in a “fixture” is much greater than the components of the “PJ”. Even if the Lamp box is lined in metal (good Idea) it makes for better design if you draw the heat as far away from electronic components as possible. It would also help with heat dissipation in the room. As for airflow, you need to have an open source to draw air from.

After reading this I sound about as clear as mudd, but I think you get the picture.

R

I understand what you're saying, and i more or less figured that section would still need some sort of ventialtion, especially due to the proximity to the fresnel.

but my (and i think steve woods) question still stands i think, about making the 2 halves isolated from each other and simply cooling the lcd cavity seperately rather then the whole back half of the projector as one unit. i would think that would be much more efficient in keeping temps way down?

which does bring up another issue i have, i've seen designs here where people attach the lamp fresnel to the tempered glass and also where they seperate the two (by a notable amount).

are the properties of the fresnels such that they can take the heat that the tempered glass is passing on to it (when attached together)? or is that why some people are seperating them?

ALSO....one other thing while i'm babbling, the UV Filter, where does it go? and is it supposed to be the size of the lamp fresnel?

TIA
andy

Jerseyjohn,
I've thought about that metal casing design before and I'm convinced it would help significantly with the cooling issue, however knowing how hot these things get you would have to surround your metal case with some sort of insulation (something like lattice) which would prevent objects or fingers from getting burned. And they would get burned if they touched the area surrounding the light. I think a decent design using this concept would be to cut a bunch of slats in the sides of the box surrounding the light section and put metal flashing on the inside wall so that the area of the slats would be exposed to the air.

It also might be of interest on this thread. I did a two hour temperature test of the M59 coil core ballast when it was not in the box (not heated in any way by the lamp) and found that it's temperature stabilized to around 160F This is fairly hot by itself and gets a bit hotter with it inside the box. And as Brain said, the ballasts stay warm for a long time after they are turned off. They are a huge chunk of iron and absorb a large amount of heat and give it up slowly. I'm wondering if the hot ballast will cause any heating circuits to cut back on. In my design, the fan isn't next to the ballast but the airflow runs past the ballast before it goes to the lamp and out of the box. Also I was wondering what the heat characteristics of the electronic ballast are, I'm guessing because they're lightweight that they cool down fairly fast. I also did a temperature test of the area immediately surrounding the ballast and I got this graph: The time is in minutes [edit] notice that the temperature hadn't yet leveled off even after two hours running, I had to run off and didn't want to leave it on. I'll do another one soon to try to find the stable temperature of my setup.[/edit]

I hope this is interesting or at least slightly useful to someone. I did this test today. My fan is loudish but it has a speed control so I wanted to find out how low I could turn it and still get efficient cooling. My ballast was my main concern since I've burned one out already. The ballast gets up to 160 F by itself (not heated by lamp) so I wanted to make sure the air blowing past it was at least 20F cooler than that. This test was run for a total of 8 1/2 hrs. Probably the main point of interest for everyone is that it takes roughly 200 minutes or just under 3 1/2 hours for the temperatures to completely stabilize inside the projector. The temperature probe was placed in the wires very close to the ballast. Based on my physics knowledge, it would take about the same amount of time (200 minutes) to cool back down to room temperature. It will actually be a bit longer since things are thermally isolated inside the box and it would be shorter than 200 if the fan is left on.

interesting pickle, i would guess that your intake hole is restricting the fan since changing the speed did nothing.

thank you for the graph, i was too lazy to make one but certainly interested in what happened inside the box, especially after i turned the lamp off.

Not to re-open an old can of worms... but I'm curious about using flashing in the light box. (Yes, I've done a ton of reading here...) and I have to ask if flashing is really necessary. Some seem to swear by it, some don't seem to use it. My first blush guess is if you flash the lid, that may be all you need. What about the electronic ballast? Can you escape flashing around it? Flashing it would block the air flow to it I would think, and I would assume if you've done your airflow properly, you can probably avoid flashing altogether.

Okay... enough of my redundant rambling... any opinions?

Okay... it just hit me... my real concern! I'm in the middle of building a 17" vertical design... most people report their lids getting warm... but with a vertical, that means all the heat will be headed up towards the heat shield (Not the Phaser shield... but the Lexan...)

actually when I turned the fan down, the temperature stepped up about 5 degrees each time I turned it down. The time scale on the graph is so compressed it's hard to see the steps but the step profile is the same as the large one at the beginning. I only started to turn the fan down once the projector temperature stabilized which happened around 200 minutes at a temp of around 125. and each time I turned the fan down, I waited for the temp to stabilize again before it continued decreasing fan speed.

Andygolf,
I think that flashing should not be necessary, there are many who believe (and I agree) that flashing the ballast would improve its longevity however I have experienced a ballast meltdown (I don't think anyone else has) and I think part of the problem was the the flashing I used was actually touching the ballast. Heat transfers faster through solids touching solids than solids touching gas (air). The problem with my setup in part I believe was that the ballast was touching the flashing too much so while the flashing provided a shield from the light itself, it also provided a became a "new skin" for the ballast and in that respect didn't help at all since it was just transfering the heat directly to the ballast again and probably slowed the cooling some too. I think I also had a faulty ballast. In my new setup I'm using a heatshield material similar to what people put up in their car windshilds to protect from heat a sort of aluminum bubble wrap. I expect this to work very well and initial tests show that it does. In short, I'd say that there is going to be a long term difference in performance using the flashing but it's not necessary for the projector to work correctly.

Interesting work picklejones. What is the CFM of your fan at max? It looks like you might save as much as 10-15 min of cool down time if your fans were at max CFM. At which step along the way were you satisfied with the fan noise? Any estimated CFM for that?