The Pink Menace! A 70 Watt High-Pressure Sodium-Vapor Arc Cloud Bouncing Spotlight

ScottFHall

Newly Enlightened
Joined
Dec 24, 2009
Messages
47
Location
Merritt Island, Florida, USA
I've got the ballast, wiring, socket, and bulb comprising a 70 watt high-pressure sodium-vapor arc lamp system and a pretty deep 10 inch diameter parabolic reflector all sitting idle here. This thing measures 6000 lumens according to the specs. Yeah, the sort of thing you'd normally just hang over the door of your barn to light your path while doing your part to not further "pollute" the night sky with stray light. The thing is, though, I'm one of those people who actually like light pollution in carefully contained quantities--long, skinny, vertical beams in any color rising straight up to infinity do not offend me at all! So while staring at these parts, I suddenly had this vision of making a custom spotlight capable of putting that gnarly pink circle normally reserved for the ground right up there onto the high, high cumulonimbi! Hence, this new thread: behold the very first pictures of The Pink Menace! A 70 Watt High-Pressure Sodium-Vapor Arc Cloud Bouncing Spotlight!

pinkmenace01.jpg


The rig you see here--the funny little bookshelf object with clamp--is only for doing initial testing of the parts. I took all this out tonight and aimed it above the treeline. It is a completely cloudless night--cold and dry here lately--so I wasn't able to actually try it on a cloud. What I want to do before proceeding, of course, is confirm that this thing will put a circle of heinous pink right on the clouds (and maybe also cast a treelimb shadow onto the clouds, too).

pinkmenace02.jpg


It's a 120 volt AC circuit and will probably stay that way (I don't really want to lug a DC-AC inverter and auto battery around in a backpack with this one). The lamp takes a minute or two to ramp up to full power. Once it does get there, it's very, very bright and it throws out way too many longwave UVs. Even just a minute or two of exposure in one's peripheral vision will make your eyes feel like you've been swimming in chlorine all day. Note to self: put a layer or two of UV blocking glass over the lamp.

The bulb sits nicely down in the deep dish parabolic reflector. It looks to me like the reflector was well designed to throw a tight beam--the only trouble with that is likely the structure of my HPS bulb itself. These lamps have a 2 inch long tubular section inside where the arc happens. Surely, this is not the best configuration for a really tight beam--I think one gets the best result from the smallest possible point source (i.e. a small ball of light as seen within those new HID lamps which have that little spherical arc zone).

So, here are the initial results from below and behind on ASA 1600, 12 megapixels, 1/2 second exposure.

pinkmenace03.jpg


There is a good tight beam emitted--it's just a wee bit dim. It is much more visible to the human eye than it is to the digital camera here--I tried and tried and could not get the camera to "see" exactly as my eyes were seeing the beam. By the way, it's very good at lighting up "dust motes". You've never seen so many dust motes with a white light beam as with this thing! The air looks downright dirty.

The dimness of the beam is no surprise: I believe HPS lamps are designed with the idea in mind that light passing through thin air should not be very visible. On the other hand, light as it strikes larger objects should be very, very visible due to the narrow-band HPS spectrum (which is why we light our roadways with them, right?) The whole result is a very high contrast of dark to light at the price of odd colored pink lighting. In all, though, this bodes well for actually lighting up overhead clouds (my primary objective with The Pink Menace!--the beam is going to go up there with serious dimness and therefore stealth and yet really blast the cloud surface into major visibility--I think). That ought to look really interesting to me and perhaps pretty scary to those off over the horizon who won't know what the they're looking at up in the sky and won't know where it's coming from.

As you can see, there is some pretty significant spill coming out of the reflector. Surely, this is mostly due to the 2 inch long glowing element inside the bulb--light is not coming from a perfect point source and so some of it is going off on overly acute and obtuse tangents. This fact shouldn't mess up my plans to light clouds up, though. I'm already convinced that the necessary candlepower exists in this system and it does have a tight enough beam.

Regarding this major spill, what do you guys suggest other than replacing the bulb and reflector? If I attach a long, flat black interior-colored cylinder to the output end of the reflector, is spill going to be reduced or eradicated? Is that cylinder going to block and absorb the stray photons?
 
Aren't lamps like this designed more for area flood, overhead industrial lighting, warehouses... etc?

Still a killer project though.

Curious what does the ballast consist of?.. step down transformer and a rectifier circuit?
 
Oh yeah, these lights are meant to be floods but I thought I'd give it a shot since I read that the visibility potential of HPS as it falls on objects is much higher than white light. Seems like a good candidate for putting a much brighter spot on a cloud than we usually see even if some of the light out of the reflector is going to get spilled.

I can't find the exact catalog number ballast/ignitor out there on line but I think it's just like this one...

http://www.1000bulbs.com/images/PDF/S0070-02C-111-specsheet.pdf

That's about $25, the bulb costs about $20, and the parabolic reflector is about $10. Not bad.

Thanks for the vote of killerness. Much appreciated :tinfoil:. We'll see if this one is a go or not once some nighttime clouds roll in.
 
Now this is what being a flashlight guy is all about - nice project.

You are dead-on about the need for UV protection - don't delay about dealing with this.

It is useful to compare what you are attempting and other methods of shining a beam on the clouds.
- Your setup looks like it will have less loss from passing through the air
- The "typical" reflectors for that application are more like 5 - 10 ft dia
- Carbon arc is the more common approach - mostly for the high surface brightness, but yours can work.

From your beam shots, esp. the indoor one, it looks like you are loosing about 50% of the light to spill - pretty common.

Your idea of using a long, shiny pipe / tube to narrow the beam will work ( I used a shotgun barrel once) but it is largely just lossses, not that supplemental to throw.

It is not easy to align, but you can also add an aspherical lens to the setup. It will help capture some of that spill and direct it forward. The lens needs a focal length approx the dia of your reflector. This approach does not make for nice looking beams, but can enhance throw - with a lot of work. More or less, it is a band aid to having too small of reflector dia.

It might be helpful to look at the home made 50in dia reflector project that is also posted in the homebuilt section.
 
Well, I'm sure this can be made into a cloud-bouncer, but you'll need a bigger reflector. That 2" element is not giving you extra spill, it's diffusing your throw over a wide area -- we call that flood. A 2" element in a 10" reflector will give you a similar beam angle to a 2mm element (XP-G) in a 10mm reflector (small AAA lights), or an MC-E in a P60 reflector.

As for the claim of higher reflective brightness, that's not particularly convincing; no doubt certain surfaces reflect sodium yellow-pink better than white light, especially yellow surfaces, but is there any reason to suppose clouds are among them? And the reason sodium lights are so popular has a lot to do with lm/W and the lobbying efforts of astronomers who prefer narrow-band light pollution so they can filter it out; I've never heard them recommended for less scattering in air. Not that it couldn't be true, but I'd be pretty surprised. IMO expect similar performance to white light in this regard.
 
HarryN, thanks--very helpful info. I have seen your 50 inch and it's excellent work. As yet, I've just tried building a narrow wedge parabolic reflector--about 20 inches. No pics here but it looks sort of like a little St. Louis arch and shoots out 1000 watts of halogen light quite effectively in a rectangle beam. I can make those fast and maybe this would work with my HPS bulb here, too.

Benson, I'll consider what you say for sure and let you know what kind of results I get. It's definitely an experiment--I haven't found too many references at all to people trying out HPS in the forums here.

Thanks a lot to both of you.
 
Hi Scott - while I would love to take credit for that 50 in reflector, that is the work of LightSward. Here is a 36 in project he is contriving: http://www.candlepowerforums.com/vb/showthread.php?t=251723

I took down the pictures of my projects as there were too many copycats of my side x side designs. Some are still in long term progress though. Most of my current efforts are for an LED drop in the the Streamlight SL-20x family.

It will be interesting to follow your project evolution.
 
HarryN, ah yes--a different guy made the 50. Got it. I somehow forgot that.

LED work--now that is cool technology. The big wave of the future it seems. I've got a couple around the house here. I'd love to find out exactly what's going in in that famous light that induces nausea. That would be a fun one to copy--life of the party/the ultimate practical joke kind of thing :cool:

So I took the HPS rig out again tonight, let it warm up for 15 minutes to be sure of max power, and tried it straight upward with the 2 foot cylinder attached this time (as I did inside the house earlier today to obtain that latest pic on the ceiling). Outdoors, that cylinder definitely decreased the spill situation and increased the visible contrast from darkness to the light of the beam. With the naked eye (well, sunglasses-covered eye to cut some of those UVs), I could much better resolve the visibility of the beam as it rose on up to infinity. It was visible from a distance--30 feet or so--and yet more visible from directly behind the source. The shape of the beam taper is quite gradual as I'd hoped--the beam starts out at 10 inches wide and spreads at a very slow rate as it rises up in the sky. That bodes well for hitting a cloud with a circle that's remained small enough: a sort of intangible variable but there it is. I know the 2 foot long cylinder is causing some of this gradual taper effect and yes, this cylinder-adding business is certainly not the most elegant and efficient way for me to operate this HPS lamp. As you guys are suggesting, probably 50% of the power is going astray which is a shame. I think I can assume that 3 feet or 4 feet of cylinder is going to correct the issue yet even more. So this is good for now--until a cloud floats by in a day...or two...more? There's still not a single one in the sky! We've got a cold front here, dry, low pressure (or is it high pressure? Can't remember)

Regarding photography of this beam: it is still a problem compared to my Sunforce 25,000,000 HID cp spotlight. I got just about the same beam visibility results with this HPS system on the digital camera that I did last night and as it was last night, these results do not well match the way my own eyes are seeing this beam in realtime. So I'm posting no pics tonight--nothing would be gained beyond the text info I've already given here.

Pleasanton? I remember Pleasanton. I think if I were there these days, I'd already be up on Mount Hamilton over in San Jose with this HPS gear of mine. Guaranteed access to clouds!

--Scott (Sunnyvale resident, 1995-7)
 
Harry, a couple more things, please: your discussion of adding an aspherical lens is very interesting. I'm confused, though: do you mean 1) a spherical reflector with a focal length of ten inches that would kick light back at both the lamp and main reflector or do you mean 2) a transparent flat backed, round topped lens--a lighthouse lens kind of thing? I think I've seen a good candidate for that lately at Walmart--some kind of flat backed/round topped magnifier.

You've mentioned focal length--what about diameter size? Full 10 inch or some smaller increment?

I have been pondering the lighthouse lens thing some in recent days--efficient or not, those old glass lens structures just look awesome!
 
Hi Scott- it seems we should do an email on some items.

Lenses are complicated, but my simple mind has suffered though enough in this area that I consider myself partially trained by fire – but not an expert by any means.

The lens you might normally use is a 1/2 ball or 1/2 sphere, generally referred to as "spherical". This tends to collimate the light coming from the source. (Flat side toward the source.)

As it turns out, each wavelength / color of light is bent slightly differently, so in a spherical lens (1/2 ball), only one wavelength is really focused properly, and the others less so.

If you squish that 1/2 ball a bit so it is slightly flattened, then it is now "aspherical". As it turns out, this largely corrects for this multi wavelength focus problem, so most modern lenses tend to be aspherical instead of spherical. There are also coatings that help, but that is another subject. Most of the people on CPF use this type of aspherical lens in their setups.

Several 100 years ago, the optics guys figured out that most of the focusing was being done in the surface, not in the bulk of a lens. This means that the entire core" of the ½ ball is just wasting light by absorption. One solution is to hollow it out.

A further solution is to now take that hollowed out ball, slice it into rings, and make them all in one plane. This is now commonly referred to as a "Fresnel" lens, and is the basis of traditional lighthouses. It is very enlightening to visit one, especially the older ones. These things throw like crazy. I think wikipedia is a good source of info on this.

The downside of using an aspherical lens for a light is that normally, you are only projecting the light that is coming from the source in the general direction of the lens – the rest is wasted. In a lighthouse, this did not matter, as the beam rotates and shoots from multiple directions, but in your light, it does matter.

Ideally, you would want a reflector to get "most of the light going in the right general direction", and then use a lens to focus it. I didn't invent this approach; I saw it on the headlights of a 1989 BMW, so it has been around at least that long. Nonetheless, this turns out to be difficult, but not impossible, especially if you are willing to experiment (buy things that eventually don't work out) This is the approach I took on the project "Breeze" in my sig line, since it only had about 12mm dia. to work with.

Once again – more or less, you are using a 2 stage optics setup to effectively increase the diameter of the reflector. This is not cheaper or better, just potentially smaller.

The focal length you need is dependent on the setup, but in general you need a deep reflector and a lens at least as large dia as your reflector. The lens distance is varied until the beam has the general characteristics you want – and will never be perfect. The focal length you buy will be limited to about 1x the diameter just because that is the nature of them.

Just so I make it really clear – this approach can work, but the reflector and lens are now a "system" and depend on each others exact shape and interaction. It takes a LOT of experimental tries to make it work, and obtaining special shape reflectors and focal length optics is a painful and expensive. You can save yourself some sanity by just buying some really deep, large reflectors.
 
Harry, thanks: great info! That's completely clear and ramps my knowledge up immediately. You're basically talking about the same reflector with lens system that I found in an old Argus slide projector I dissected once. This compound approach is really interesting. I have built a number of telescopes, a large lens-based solar furnace, and have played around with some fresnels, so I think I can probably work this out.

I just tried putting an 80 mm achromatic telescope objective lens that I have here in front of the lamp and reflector. The focal length is about 10 inches. Not enough diameter to match my reflector, I know, but this immediately caused a tightly focused circle of light to be cast on the wall across the room (accompanied by plenty of spill due to the mismatch of diameters). Very cool arrangement--I'll definitely look into this approach more.
 
P.S. I've researched it more and this approach you're talking about is also what goes on in old fresnel stage lights which consist of a lamp with a spherical reflector behind it able to move up and rack focus to a fresnel lens at the front--a variable focus spot is thereby created. The cylinder sidewalls holding it all together absorb the stray photons which are simply dissapated as waste heat. This seems to descend from early lighthouse technology through early searchlight technology and then into fresnel stage lights which in many cases today still appear to be structured in the same sort of somewhat inefficient form.

Overall, I like this antique approach a lot as a solution to pursue with my small 10" parabolic reflector holding a 70 watt HPS lamp. I figure I can build a cylinder/heat sink to get and set optimal beam focus from the combination of rear mounted reflector, lamp, a front mounted sheet of clear UV plastic or glass, and a front mounted fresnel (as near to 10" diameter as I can find). I can mount that cylinder readily like a dobsonian telescope along with a lazy susan bearing to get full altitude and azimuth coverage for cloud bouncing.

So this may well be the resolution of The Pink Menace! project. We'll see and pics will follow, of course.
 
fresnelspotlight.jpg


Here is some progress on my decision to complete The Pink Menace! as a spotlight with a parabolic reflector corrected for the cleaneast, straightest beam via use of a fresnel condenser lens.

What you see here is A) the 70 watt HPS lamp in its 10 inch parabolic reflector. It is temporarily positioned here so you can see how it will rear mount with a few screws and tabs onto B) an 11 x 11 inch ventilated 3/4" plywood back board. Also show here is C) the opposite 11 x 11 inch 3/4" plywood front board which has been cut to provide as much light passthrough as possible of a rectangle-shaped fresnel lens D). These two boards comprise the top and bottom of a 6-sided wooden box enclosure that will set the lamp center axis and the fresnel lens 15" apart (the distance which seems optimal because it creates the most linear output of rays--this basically results at low start up power in the visible image of the bulb and the whole circular reflector being projected like a camera obscura in good focus on the wall of the room).

A piece of UV ray blocking glass will also be in the optical path. The sides of the box will have big circles of 3/4" plywood attached so that the whole thing can be nested into a freely alt-azimuth capable dobson telescope style mount. The ballast will be mounted out of sight under a box that will serve as the base of the dobson mount.

Yes, the whole design is a compromise analogous to old fashioned fresnel stage spotlights and early searchlights. As is often the case, I've got the urge to re-build some obsolete technology. And to turn an HPS flood into a spotwith various easily and cheaply acquired bits and pieces.

In all, I should be putting a bright pink spot on clouds pretty soon.
 
Oh GEEZ!!! I can't understand half of whats been posted in this thread.... like you guys are speaking Klingon or something. But WTH... I am enjoying your project immensely. Your enthusiasm is infectious.

thanks for sharing this DIY.
:popcorn:
 
Sorry Kramer5150--pictures are worth 1,000 words, after all, so I do keep posting pics, too. Thanks for your continued interest in this "difficult" pink (or sometimes yellow/orange as captured in photos) light. I know most people do dread sodium!

Here's the latest pic regarding the actual rate of spread of the beam over distance. This shape reminds me of the dumbbell nebula--it's caused by the fairly well focused image of the glowing arc within the bulb being projected out by the optical system. Thanks to HarryN once again for the detailed tutorial on how to condense a flood down into a spot with a corrective lens!

HarryN, send your postal mailing address and full name in a private message if you would. I figure I'll arrange to have a honorary doc in Optics & Photonics sent to you for your in depth assistance. Zero tuition and fees :thumbsup:

hotspotsmaller.jpg
 
Hi Scott,

Thanks for the offer. PM sent.

Frankly, what I really need is someone to hire my son. He lives in the Port Charlotte, FL area, has a BA, knows business really well, etc. He is a free thinker but will work hard.

Anyone hiring him (full or part time) also will receive free flashlight advice AND even a free custom flashlight - :twothumbs
 
Top