EDAboard.com | EDAboard.de | EDAboard.co.uk | WTWH Media

LED Slide Projector

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics AUS - LED Slide Projector

Goto page Previous  1, 2

keithr0
Guest

Wed May 02, 2018 11:45 am   



On 4/3/2018 8:07 PM, Clifford Heath wrote:
Quote:
On 03/04/18 17:48, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 03/04/18 08:36, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 02/04/18 09:58, Computer Nerd Kev wrote:
I've been looking into whether it would be practical to replace the
incandescent bulb in an old 35mm slide projector with a high power
COB LED, such as the Cree CXA3070 series which is about $60 from RS.

http://www.cree.com/led-components/products/xlamp-leds-integrated-arrays/xlamp-cxa3070


The aim is to replace a 500W "DAY/DAK" bulb (or whatever the 240V
equivalent is), as shown here:
http://www.replacementlightbulbs.com/lampdaydak.html
According to that page, the target is therefore around 12,500
Lumens. A little less than the Cree can get to, but two 6,000Lum
Cree LEDs might be an option too.

The total intensity doesn't matter. What matters is how
wide the emitting surface is for the given intensity.

Don't the lenses in the projector take care of this?

No, a lens cannot focus a beam to a finer angle. If the
original beam is 1 degree wide, a perfectly focussed
lens will produce a 1-degree wide image of the source.

what should I be attempting to match if not Lumens?

The intensity per square mm. That's why HID lamps are used
in the first place. Every pixel in the image is a miniature
image of the source, so far as is achievable.

I believe you're pointing out that the area over which the
light is emitted has to be similar in order for the total
light output to be relevant. The datasheet for the CXA3070
LED indicates that this area is within a 23mm diameter
circle. The filament arrangement in the original bulb
measures about 10x10mm, but I figured (perhaps
optimistically) that the incandescent bulb would be less
directional than the LED so it would be a good enough
match. Admittedly I'd forgotten about this by the time
I started considering two LEDs, which I'll concede
wouldn't work.

And just in case we're straying off-topic, a reminder
that I'm talking about a 35mm photographic slide projector,
not a digital projector. I don't believe the original
bulb (in the 50+ year old machine) is a HID type.


If you put a pin-hole in a piece of paper and hold it up
to the sun, you don't see a pinprick of light on the ground.
You see an image of the sun. The size of the image depends
on the distance; the angle of the light from the two sides
of the sun is about half a degree. If you add a lens in the
hope to produce a pin-prick of light, you'd fail. Every
point on the surface of the lens is receiving light from
all of the sun, and the rays that hit each point stay spread
across the whole half-a-degree no matter what system of
lenses and mirrors you use, because optical systems are
*always* reversible. So the only way to get a beam that's
narrower is to throw away most of the light by using only
the light that passed through *two* pinpricks. That gets
you a beam of light that comes from just one small part
of the sun. Just not a very bright one.

The exact same argument applies to your projector, whether
it's projecting 35mm slides or through an LCD or DLP system.
The wider your original beam angle from the source, the
less resolution (focus) of the projected image. To get a
bright, sharp image, there's no substitute for an emitter
of exceptionally high intensity - a very hot VERY small
spot.

Clifford Heath.


That doesn't make sense, what you want is the light source focused to
give an even illumination over the whole area of the slide. The lens on
the other side of the slide is what you need to focus to a point.

keithr0
Guest

Wed May 02, 2018 11:45 am   



On 4/4/2018 9:04 AM, Computer Nerd Kev wrote:
Quote:
Clifford Heath <no.spam_at_please.net> wrote:
On 03/04/18 17:48, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 03/04/18 08:36, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 02/04/18 09:58, Computer Nerd Kev wrote:
I've been looking into whether it would be practical to replace the
incandescent bulb in an old 35mm slide projector with a high power
COB LED, such as the Cree CXA3070 series which is about $60 from RS.

The total intensity doesn't matter. What matters is how
wide the emitting surface is for the given intensity.

Don't the lenses in the projector take care of this?

No, a lens cannot focus a beam to a finer angle. If the
original beam is 1 degree wide, a perfectly focussed
lens will produce a 1-degree wide image of the source.

what should I be attempting to match if not Lumens?

The intensity per square mm. That's why HID lamps are used
in the first place. Every pixel in the image is a miniature
image of the source, so far as is achievable.

I believe you're pointing out that the area over which the
light is emitted has to be similar in order for the total
light output to be relevant. The datasheet for the CXA3070
LED indicates that this area is within a 23mm diameter
circle. The filament arrangement in the original bulb
measures about 10x10mm, but I figured (perhaps
optimistically) that the incandescent bulb would be less
directional than the LED so it would be a good enough
match. Admittedly I'd forgotten about this by the time
I started considering two LEDs, which I'll concede
wouldn't work.

If you put a pin-hole in a piece of paper and hold it up
to the sun, you don't see a pinprick of light on the ground.
You see an image of the sun. The size of the image depends
on the distance; the angle of the light from the two sides
of the sun is about half a degree. If you add a lens in the
hope to produce a pin-prick of light, you'd fail. Every
point on the surface of the lens is receiving light from
all of the sun, and the rays that hit each point stay spread
across the whole half-a-degree no matter what system of
lenses and mirrors you use, because optical systems are
*always* reversible. So the only way to get a beam that's
narrower is to throw away most of the light by using only
the light that passed through *two* pinpricks. That gets
you a beam of light that comes from just one small part
of the sun. Just not a very bright one.

The exact same argument applies to your projector, whether
it's projecting 35mm slides or through an LCD or DLP system.
The wider your original beam angle from the source, the
less resolution (focus) of the projected image. To get a
bright, sharp image, there's no substitute for an emitter
of exceptionally high intensity - a very hot VERY small
spot.

Between the bulb and the slide there is a diffuser (as well
as a couple of lenses), so the light hitting the slide
isn't an exact image of the source. I've been playing
around with the removable inner lens assembly using some
torches and it would seem that a slightly wider source
would still illuminate the slide well enough.


You don't want an image of the light source, you want even illumination
over the whole area if the slide

Quote:
If a wider beam angle reduces the resolution, then should
I be able to approximate a larger source beam angle by
using a much shorter projection distance and observing
the decrease in image clarity?

The beam width of the illuminator has absolutely to do with the
resolution of the image, the quality of the projection lens is what sets
the resolution of the projected image.

Computer Nerd Kev
Guest

Wed May 02, 2018 11:45 pm   



keithr0 <user_at_account.invalid> wrote:
Quote:
On 4/4/2018 9:04 AM, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 03/04/18 17:48, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 03/04/18 08:36, Computer Nerd Kev wrote:
Clifford Heath <no.spam_at_please.net> wrote:
On 02/04/18 09:58, Computer Nerd Kev wrote:
I've been looking into whether it would be practical to replace the
incandescent bulb in an old 35mm slide projector with a high power
COB LED, such as the Cree CXA3070 series which is about $60 from RS.

The total intensity doesn't matter. What matters is how
wide the emitting surface is for the given intensity.

Don't the lenses in the projector take care of this?

No, a lens cannot focus a beam to a finer angle. If the
original beam is 1 degree wide, a perfectly focussed
lens will produce a 1-degree wide image of the source.

what should I be attempting to match if not Lumens?

The intensity per square mm. That's why HID lamps are used
in the first place. Every pixel in the image is a miniature
image of the source, so far as is achievable.

I believe you're pointing out that the area over which the
light is emitted has to be similar in order for the total
light output to be relevant. The datasheet for the CXA3070
LED indicates that this area is within a 23mm diameter
circle. The filament arrangement in the original bulb
measures about 10x10mm, but I figured (perhaps
optimistically) that the incandescent bulb would be less
directional than the LED so it would be a good enough
match. Admittedly I'd forgotten about this by the time
I started considering two LEDs, which I'll concede
wouldn't work.

If you put a pin-hole in a piece of paper and hold it up
to the sun, you don't see a pinprick of light on the ground.
You see an image of the sun. The size of the image depends
on the distance; the angle of the light from the two sides
of the sun is about half a degree. If you add a lens in the
hope to produce a pin-prick of light, you'd fail. Every
point on the surface of the lens is receiving light from
all of the sun, and the rays that hit each point stay spread
across the whole half-a-degree no matter what system of
lenses and mirrors you use, because optical systems are
*always* reversible. So the only way to get a beam that's
narrower is to throw away most of the light by using only
the light that passed through *two* pinpricks. That gets
you a beam of light that comes from just one small part
of the sun. Just not a very bright one.

The exact same argument applies to your projector, whether
it's projecting 35mm slides or through an LCD or DLP system.
The wider your original beam angle from the source, the
less resolution (focus) of the projected image. To get a
bright, sharp image, there's no substitute for an emitter
of exceptionally high intensity - a very hot VERY small
spot.

Between the bulb and the slide there is a diffuser (as well
as a couple of lenses), so the light hitting the slide
isn't an exact image of the source. I've been playing
around with the removable inner lens assembly using some
torches and it would seem that a slightly wider source
would still illuminate the slide well enough.

You don't want an image of the light source, you want even illumination
over the whole area if the slide


That's what I was implying by stating that there is already a
diffuser in the projector.

Quote:
If a wider beam angle reduces the resolution, then should
I be able to approximate a larger source beam angle by
using a much shorter projection distance and observing
the decrease in image clarity?

The beam width of the illuminator has absolutely to do with the
resolution of the image, the quality of the projection lens is what sets
the resolution of the projected image.


I'm happy to accept that.

But this was all resolved earlier. what I was really hoping for
was some convenient approach to mounting a suitable LED and its
heatsinking to an exisiting projector without turning it into
a week-long project.

It's a bit of a moot point now anyway because I decided
such a solution probably didn't exist and bought two NOS
filament bulbs from America (yes, 240V) for about the same
$60 as it would have cost me for one of the LEDs. They're
rated for approx. 25hrs. So that's about 50hrs, plus however
long the old one lasts, of viewing before I have to look for
more. That's probably long enough really.

--
__ __
#_ < |\| |< _#

Goto page Previous  1, 2

elektroda.net NewsGroups Forum Index - Electronics AUS - LED Slide Projector

Ask a question - edaboard.com

Arabic version Bulgarian version Catalan version Czech version Danish version German version Greek version English version Spanish version Finnish version French version Hindi version Croatian version Indonesian version Italian version Hebrew version Japanese version Korean version Lithuanian version Latvian version Dutch version Norwegian version Polish version Portuguese version Romanian version Russian version Slovak version Slovenian version Serbian version Swedish version Tagalog version Ukrainian version Vietnamese version Chinese version Turkish version
EDAboard.com map