Re: concave mirror with hole

"redbelly" <redbelly98@xxxxxxxxx> wrote in message
news:1157653257.541645.301560@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

This "image" will be just 1 x 1 pixels, since you have a single
detector element. My whole point was that you don't need a really
good, spatially resolved image. You just need to focus as much of the
power as possible onto a 1 mm^2 area. Aberations become less of a
concern than, say, designing a camera or microscope lens where the goal
is to produce excellant, undistorted images.

Hi,
I was virtualy agreeing with you on that point its not the acuracy of the
image,
but in trying to get a small spot that is initself effectivly an image of
the small laser spot on the target.
however with the focul length very much shorter than the distance to target
the perfect image would be a pin point,
clearly such a perfect image is not needed.

My original suggestion was to
1. Find a supplier for the mirror you want, without the hole, and then
2. Find somebody who can cut a hole in the mirror, without ruining the
mirror. You're best bet is to find a local glass shop that can do
this.

There were lots of interesting sugestions wich I thank you all, I found a
few cheap spherical glass mirrors on ebay,
and some good off axis aluminium types from scientific places wich were
expensive,
however the plastic lamp reflector has proved to be an extremly good idea,
even a very simple mock up prototype was easy to make/drill etc and has
worked considerably better than I was hoping for,
for now at least.

the metal mirror I tried,
wich I later found a good way to test by puting a simple led torch some
distance away from it,
forms a clearly visible quite bright image but a rather large one, so
most
of the light isnt captured by the 1mm^2 detector.
I just did it with a piece of .15 mm *** steel and a ball pein hammer
and
a flat metal surface,

Even though I said the image needn't be as good as, say, in a camera, I
can't imagine this method being suitable. How do you get a curved
mirror using a flat surface? And even if you had a curved surface, a
hammer will not produce the smoothness you need.

Wel like I said it takes some skill, its like an art, if you bash the metal
against the flat surface with a hammer it will squish out and expand where
you hit it, if you do it carefully this causes it to deform into a dish
shape, even on a flat anvil, if your able to work out how much it needs to
deform and where to start etc you can make all sorts of shapes, once you
start geting the right shape you concentrate on smoothing out the hammer
marks by hitting it more gently.
you need some proper metal bashing hammers the right size and shape, (wich
havnt been used for banging in nails !)
and tons of practice and patience.
a polish with brasso made it quite shiny, you cant realy see my hammer marks
exept where I was a bit heavy handed,
but like I said it would be time consuming to get it good enough.
I cant realy show a picture I dont think it would show up the curvature.
My dad and brother were both mechanical engineers so I picked up quit a bit.

Yes, it is possible to grind and polish your own mirror, if you really
want to. If it were me, I would just buy it. For example, from
Thorlabs in the US:

I just wouldnt have the patience, it takes like 100s of hours lol.

D = f = 25 mm, 42 $US or 44 Euro, part # CM254-025-G01
D = f = 50 mm, 59 $US or 62 Euro, part # CM508-050-G01

Other diameters and focal lengths, plus a reflectance curve, are at

http://www.thorlabs.com/NewGroupPage9.cfm?Guide=10&Category_ID=138&ObjectGroup_ID=1161


Thanks, some nice mirrors there at a reasonable price, im not sure what the
focul length im using now is but its very short compared to the diameter, I
think the only type with such short FL are torch type reflectors wich works
great,
coated plastic seems quite good enough, especialy as it should be cheap.

They also list lasers, however its a shame they dont spec how fast they are,
Im trying to find a faster laser but I cant find such details in the specs
for (cheap) visible 5mw types.

thanks
Colin =^.^=


.

Quantcast