You mean half the size of course. 8)
That's right... I'll need as big as I can get for my XHP700 array. ;)
I'm using the OP to throw out ideas/cries for help, etc...
Currently, I'm debating the following;
* Decision; I was planning on doing a Recoil type setup, & this is still my preferred option.
To get any type of real concave mirror in the size I want (~30 cm), is very costly, so I'm currently thinking of finding an appropriate size dish/bowl, & coating it with a layer of adhesive mirror sheet.
Thoughts?
---------------
* Driver help; Will one zener modded BLF DD driver v1.0 (older style FET) be able to handle 4 * XHP70's in parallel? Or should I use two separate drivers?
--------------
Find a domed surface, and use it as a fiberglass mold? I saw a video on youtube where a guy made a concave solar collector mirror out of fiberglass. He had an old tank (I think it was a water heater tank). He used the bottom of the tank as a mold to form the shape of his dish. He used fiberglass resin with reinforcing fiber mesh to make the actual dish. After he removed it from the tank, he cleaned it up and smoothed it out with a sander. Then, he used metallic tape to make the interior surface reflective.
One simple option is to use a ball or a balloon for your shape mold, like some have used for home-built integrating spheres. You can surely find a ball large enough. You’d probably want a ball that is twice or more your finished diameter, so that you’re not making a whole dome in 30cm, but only part of a dome. With a ball, you could just use canned spray-foam around the outside to get the shape. Or you could still use epoxy or fiberglass resin.
Another possible variation might be that you could make a dome shape with cardboard or something, and epoxy inside it to smooth it out. This would be the most complicated and time consuming way. Doing multiple layers and sanding in between them, you could probably do fairly well. The better your starting shape is, the less work it will be. I don’t know how good you are at fabricating, but it might work.
At the moment, I'm on the lookout for a pre-made(or close to pre-made) solution for the reflector dish. I hadn't factored in Making one yet, lol.
As for the driver/s, I think I have answered my own question by looking over the FET spec. sheet, & DJozz's XHP test graphs.
Unless I'm reading things wrong, it looks like I would pushing things by running two XHP70's off the same driver, let alone four... 
Well, for the best throw and beam profile, you’re better off using only one emitter anyway. Why did you want to do four?
The reflector will have to be parabolic 
One made with a ball ie: round, would only reflect all light back to the led.
Cheers David
The top candidate so far is a satellite dish, which I would have to cut down to size.
As for more than one emitter? ; I'm after big lumens :D
Well, only if the LED were positioned at the focal point of the curve. Look at how integrating spheres work. They don’t focus the light back to the LED. They are spherical, yet they scatter the light evenly in every direction. The position of the light source influences where the curve will focus the light back to.
If the led is not at the focus point the light will scatter, thats why reflectors are parabolic, all light is reflected at a known angle, light from the source must be reflected at different angles at different places of the reflector so they all come out the front at the same required angle.
Put a led pointing in at the focal point of a parabolic dish with edges at 45 degree’s and all of the light emitted will be focused by the reflector
Cheers David
Reflectors in flashlights are getting hit with light from behind. Or, you could say, from the back edge of the parabola. The light is travelling in a specific way and the parabola forms it a specific way in order to focus a beam outward. Here, we’re talking about a light source pointing toward the front of the curve, not from the back. That makes things a bit different. If you think about a perfect sphere, and a point source at the center, all the light is focused back to center, as you said before. If you move the point source, the light will still be focused but not to the same place or at the same angle. That is because the angle that the light hits the curve is no longer perpendicular or 90 degrees to the surface of the curve.
I just did a quick online search and found this webpage. The “Lesson Three” talks about concave mirrors and how light reflection works with them. I was unable to find specific mention of a distinction between parabolic vs spherical surfaces. But, the way the text was worded suggests that they are assuming a spherical mirror shape.
Edit: Ah, here we go, another website says this:
So, you are perfectly correct, except that under certain circumstances, I am also correct.
Sorry for the thread hi-jack Fmc, it finishes here
nuff said
Cheers David
A google search for parabolic reflectors, turns up some stuff in the 12" diameter range and less than $50.
I know it's not AU links, but I would think there's stuff like this over there too. Maybe not.
Hell, it's almost cheaper if I buy one and send it to you. What a difference in price!
Good thing you did post this link. Otherwise, FmC might not know that the satellite dish is a no-go for this project, because it is off-axis.
I wonder if you could shape a parabolic mold negative(think beehive shape) then use it to cast a Fiberglas and epoxy reflector. Do your best Richard Dreyfus imitation. Whom knows? You might even be able to pick it up. Kind of like the way integrating spheres are done but in a parabola.
Holly…
Can’t wait to see all the progress and final product!
Here’s a parabolic mirror on ebay that is in bidding. As of 1:00 PM EST, it has one bid and is at $9USD. Listing says it will ship to Australia. Has free shipping.
Edit: It is 40.5cm size.
68 cm focal length. At that distance you would need an intermediate lens to get the light onto the surface.
Thanks for the discussion, guys.
Believe me, I spent a lot of time searching on-line for something suitable, with a reasonable price.
I'm choking on forking out for four XHP's, never mind ~$150 for a reflector...
I have some "97%" reflective flexible adhesive mirror squares in the mail, so I'm really just looking for a suitable shape to coat with them, although a 'ready to go' reflector would be nice.
Explain to me why I can't use the satellite dish if it's tilted on an angle, or if the emitters are off-center?
:beer:
I assume you have seen