Sign in with
Sign up | Sign in
Your question

any plans for "cooled camera detectors?"

Last response: in Digital Camera
Share
Anonymous
March 12, 2005 12:57:01 PM

Archived from groups: rec.photo.digital (More info?)

Hi Everyone,

Astrophotgraphers typically use "cooled detectors" to take long exposures
with CCD's, because the dark-current of the detector is massively reduced
along with the temperature of the detector. One can themo-electrically cool
a CCD, but I believe this requires a lot of current, and the kind of battery
typically stored onboard a handheld camera wouldn't do the job. But what
about a larger battery pack worn on the belt, then wired to the camera?
Could the detector itself be thermally isolated from the rest of the camera
(via thermally non-conductive struts) so that the detector (and not the
entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
one to take very long exposures (half hour long, hour long) without the
concern of detector noise? Are there new solid state detectors out there
with lower thermal noise? Let's say you didn't want to take hour long
exposures, but just two or three minutes long -- one wouldn't have to cool
the detector as much. I'm just thinking that reducing the detector noise
could dramatically improve photographic results, since I read a lot of
complaints about detector noise on digital cameras. Perhaps the concept of
a "cooled camera" is only practical for BIG cameras, or those that don't
have to be moved around much (like something mounted on a large movie
camera, or on the back of a big telescope). But with new miniaturization
technologies, I wonder if cooled CCD's are a possibility, at least in a
larger DSLR? This is a speculative sort of "what if?" question that I'm
raising, and nothing else...

-Scott Speck
kaiju@comcast.net
Anonymous
March 12, 2005 12:57:02 PM

Archived from groups: rec.photo.digital (More info?)

Scott Speck wrote:

> Hi Everyone,
>
> Astrophotgraphers typically use "cooled detectors" to take long exposures
> with CCD's, because the dark-current of the detector is massively reduced
> along with the temperature of the detector. One can themo-electrically cool
> a CCD, but I believe this requires a lot of current, and the kind of battery
> typically stored onboard a handheld camera wouldn't do the job. But what
> about a larger battery pack worn on the belt, then wired to the camera?

For consumer cameras & DSLRs ? I don't see this happening at all :-)

The technology is far too specialized and the cost of implementing it
would be prohibative.


> Could the detector itself be thermally isolated from the rest of the camera
> (via thermally non-conductive struts) so that the detector (and not the
> entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
> one to take very long exposures (half hour long, hour long) without the
> concern of detector noise? Are there new solid state detectors out there
> with lower thermal noise? Let's say you didn't want to take hour long
> exposures, but just two or three minutes long -- one wouldn't have to cool
> the detector as much. I'm just thinking that reducing the detector noise
> could dramatically improve photographic results, since I read a lot of
> complaints about detector noise on digital cameras. Perhaps the concept of
> a "cooled camera" is only practical for BIG cameras, or those that don't
> have to be moved around much (like something mounted on a large movie
> camera, or on the back of a big telescope). But with new miniaturization
> technologies, I wonder if cooled CCD's are a possibility, at least in a
> larger DSLR? This is a speculative sort of "what if?" question that I'm
> raising, and nothing else...
March 12, 2005 12:57:02 PM

Archived from groups: rec.photo.digital (More info?)

Scott Speck wrote:
> Hi Everyone,
>
> Astrophotgraphers typically use "cooled detectors" to take long
exposures
> with CCD's, because the dark-current of the detector is massively
reduced
> along with the temperature of the detector. One can
themo-electrically cool
> a CCD, but I believe this requires a lot of current, and the kind of
battery
> typically stored onboard a handheld camera wouldn't do the job. But
what
> about a larger battery pack worn on the belt, then wired to the
camera?
> Could the detector itself be thermally isolated from the rest of the
camera
> (via thermally non-conductive struts) so that the detector (and not
the
> entire CAMERA) be cooled enough to greatly reduce thermal noise and
allow
> one to take very long exposures (half hour long, hour long) without
the
> concern of detector noise? Are there new solid state detectors out
there
> with lower thermal noise? Let's say you didn't want to take hour
long
> exposures, but just two or three minutes long -- one wouldn't have to
cool
> the detector as much. I'm just thinking that reducing the detector
noise
> could dramatically improve photographic results, since I read a lot
of
> complaints about detector noise on digital cameras. Perhaps the
concept of
> a "cooled camera" is only practical for BIG cameras, or those that
don't
> have to be moved around much (like something mounted on a large movie

> camera, or on the back of a big telescope). But with new
miniaturization
> technologies, I wonder if cooled CCD's are a possibility, at least in
a
> larger DSLR? This is a speculative sort of "what if?" question that
I'm
> raising, and nothing else...
>
> -Scott Speck
> kaiju@comcast.net


There are several interesting threads on this subject on the
sci.astro.amateur newsgroup; they usually degenerate into "film
resolution vs. digital resolution" discussions, but they still have a
lot of good info.

The CMOS sensors used in eg. Canon's dSLR's are supposed to have less
noise, but I'm not the one to really answer that question (and I'm not
going to get involved in a Canon vs. Nikon flame war).

AFAIK, the problem with CCD's is that they have a phenomenon known as
"dark current" from heating of the chip when it is on for long periods
- the upshot of it is that the longer the CCD is on, the more noise
occurs. Cooling the CCD (with stuff like dry ice, Peltier
thermoelectric coolers, liqid nitrogen, etc) significantly reduces
this, so you can get useful hour-long (or more) exposures.

An interesting way around this is to combine several (or many - tens of
thousands even) short exposures into one good long exposure. A freeware
program called Registax does this - I've tried it for some afocal pics
I took through a 150mm reflector of Saturn and Jupiter; it works OK,
especially to reduce noise and increase resolution. It also apparently
works (with the proper equipment) for some deep space objects; they
have to be the brighter ones, though, because the darker ones simply
don't put enough photons out to significantly change a (for example) 15
second exposure.

A cooled CCD camera for astrophotography is expensive, but they're
available. You can get a used older one for US$800-1000 according to a
recent thread I read. I've seen them new for a few thousand dollars -
not out of reach if you're really commited to the hobby, and probably a
bit less than you'd spend for a decent refractor or Mak, to take the
pics through.... I also read something recently about making your own
CCD astrophotography camera, that could be cooled with dry ice - in the
$500-600 range for all the parts.

Good Luck!
ECM
Related resources
March 12, 2005 12:57:02 PM

Archived from groups: rec.photo.digital (More info?)

How about a heat pipe from the sensor to a small ice bucket on the top of
the camera? Perhaps the size of a 35mm film canister with a snap on lid.
Put some ice in there, and your sensor will cool down.

Yes, condensation on the sensor could be an issue. I forget how much you'd
have to cool the sensor to have a useful reduction in noise. But I'm sure
it doesn't have to go to absolute zero.

Problem is, you'd have to add a bag of party ice to your photo kit.

Pete
Anonymous
March 12, 2005 1:53:00 PM

Archived from groups: rec.photo.digital (More info?)

Hi ECM,

Thanks for the info. I suppose I'm wondering, though, if "cooled detector"
technology might ever make it into a name-brand SLR-type handheld digital
camera. Perhaps, for any shutter speed that one would find useful or
typical for DSLR's, the dark current really isn't a factor, so there's no
NEED to cool the camera. And, if one is using the camera for
astrophotography, one could then use a specialized cooled-detector camera,
as you described.

Thanks,
Scott

"ECM" <thedeepabyss@whoever.com> wrote in message
news:1110642456.113366.80740@o13g2000cwo.googlegroups.com...
> Scott Speck wrote:
>> Hi Everyone,
>>
>> Astrophotgraphers typically use "cooled detectors" to take long
> exposures
>> with CCD's, because the dark-current of the detector is massively
> reduced
>> along with the temperature of the detector. One can
> themo-electrically cool
>> a CCD, but I believe this requires a lot of current, and the kind of
> battery
>> typically stored onboard a handheld camera wouldn't do the job. But
> what
>> about a larger battery pack worn on the belt, then wired to the
> camera?
>> Could the detector itself be thermally isolated from the rest of the
> camera
>> (via thermally non-conductive struts) so that the detector (and not
> the
>> entire CAMERA) be cooled enough to greatly reduce thermal noise and
> allow
>> one to take very long exposures (half hour long, hour long) without
> the
>> concern of detector noise? Are there new solid state detectors out
> there
>> with lower thermal noise? Let's say you didn't want to take hour
> long
>> exposures, but just two or three minutes long -- one wouldn't have to
> cool
>> the detector as much. I'm just thinking that reducing the detector
> noise
>> could dramatically improve photographic results, since I read a lot
> of
>> complaints about detector noise on digital cameras. Perhaps the
> concept of
>> a "cooled camera" is only practical for BIG cameras, or those that
> don't
>> have to be moved around much (like something mounted on a large movie
>
>> camera, or on the back of a big telescope). But with new
> miniaturization
>> technologies, I wonder if cooled CCD's are a possibility, at least in
> a
>> larger DSLR? This is a speculative sort of "what if?" question that
> I'm
>> raising, and nothing else...
>>
>> -Scott Speck
>> kaiju@comcast.net
>
>
> There are several interesting threads on this subject on the
> sci.astro.amateur newsgroup; they usually degenerate into "film
> resolution vs. digital resolution" discussions, but they still have a
> lot of good info.
>
> The CMOS sensors used in eg. Canon's dSLR's are supposed to have less
> noise, but I'm not the one to really answer that question (and I'm not
> going to get involved in a Canon vs. Nikon flame war).
>
> AFAIK, the problem with CCD's is that they have a phenomenon known as
> "dark current" from heating of the chip when it is on for long periods
> - the upshot of it is that the longer the CCD is on, the more noise
> occurs. Cooling the CCD (with stuff like dry ice, Peltier
> thermoelectric coolers, liqid nitrogen, etc) significantly reduces
> this, so you can get useful hour-long (or more) exposures.
>
> An interesting way around this is to combine several (or many - tens of
> thousands even) short exposures into one good long exposure. A freeware
> program called Registax does this - I've tried it for some afocal pics
> I took through a 150mm reflector of Saturn and Jupiter; it works OK,
> especially to reduce noise and increase resolution. It also apparently
> works (with the proper equipment) for some deep space objects; they
> have to be the brighter ones, though, because the darker ones simply
> don't put enough photons out to significantly change a (for example) 15
> second exposure.
>
> A cooled CCD camera for astrophotography is expensive, but they're
> available. You can get a used older one for US$800-1000 according to a
> recent thread I read. I've seen them new for a few thousand dollars -
> not out of reach if you're really commited to the hobby, and probably a
> bit less than you'd spend for a decent refractor or Mak, to take the
> pics through.... I also read something recently about making your own
> CCD astrophotography camera, that could be cooled with dry ice - in the
> $500-600 range for all the parts.
>
> Good Luck!
> ECM
>
Anonymous
March 12, 2005 1:53:01 PM

Archived from groups: rec.photo.digital (More info?)

Scott Speck wrote:

> Hi ECM,
>
> Thanks for the info. I suppose I'm wondering, though, if "cooled detector"
> technology might ever make it into a name-brand SLR-type handheld digital
> camera. Perhaps, for any shutter speed that one would find useful or
> typical for DSLR's, the dark current really isn't a factor, so there's no
> NEED to cool the camera. And, if one is using the camera for
> astrophotography, one could then use a specialized cooled-detector camera,
> as you described.
>
> Thanks,
> Scott

The modern DSLRs are doing multiple things to reduce dark current.
For example, the Canon DSLR, like the 20D, 1D Mark II, turn
off electronics during a long exposure to reduce heating of
the sensor. You can do exposures of ten to 20 minutes at
high ISO and get images with less noise than equivalent
speed film with these cameras. Then combining multiple
exposures reduces noise further, making the dark current
effectively a non issue.

>>>kaiju@comcast.net

>>The CMOS sensors used in eg. Canon's dSLR's are supposed to have less
>>noise, but I'm not the one to really answer that question (and I'm not
>>going to get involved in a Canon vs. Nikon flame war).
>>
>>AFAIK, the problem with CCD's is that they have a phenomenon known as
>>"dark current" from heating of the chip when it is on for long periods
>>- the upshot of it is that the longer the CCD is on, the more noise
>>occurs. Cooling the CCD (with stuff like dry ice, Peltier
>>thermoelectric coolers, liqid nitrogen, etc) significantly reduces
>>this, so you can get useful hour-long (or more) exposures.
>>
>>An interesting way around this is to combine several (or many - tens of
>>thousands even) short exposures into one good long exposure. A freeware
>>program called Registax does this - I've tried it for some afocal pics
>>I took through a 150mm reflector of Saturn and Jupiter; it works OK,
>>especially to reduce noise and increase resolution. It also apparently
>>works (with the proper equipment) for some deep space objects; they
>>have to be the brighter ones, though, because the darker ones simply
>>don't put enough photons out to significantly change a (for example) 15
>>second exposure.

No, amateurs are doing this quite effectively and getting
results within a factor of 2 to 4 of state of the art
cooled systems. The main cause of this factor is the
limited bandwidth and transmission of the filters over the DSLR
sensors, and not the performance.

Roger
Anonymous
March 12, 2005 2:07:44 PM

Archived from groups: rec.photo.digital (More info?)

On Sat, 12 Mar 2005 09:57:01 -0500, "Scott Speck"
<speck82@comcast.net> wrote:

>Hi Everyone,
>
>Astrophotgraphers typically use "cooled detectors" to take long exposures
>with CCD's, because the dark-current of the detector is massively reduced
>along with the temperature of the detector. One can themo-electrically cool
>a CCD, but I believe this requires a lot of current, and the kind of battery
>typically stored onboard a handheld camera wouldn't do the job. But what
>about a larger battery pack worn on the belt, then wired to the camera?
>Could the detector itself be thermally isolated from the rest of the camera
>(via thermally non-conductive struts) so that the detector (and not the
>entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
>one to take very long exposures (half hour long, hour long) without the
>concern of detector noise? Are there new solid state detectors out there
>with lower thermal noise? Let's say you didn't want to take hour long
>exposures, but just two or three minutes long -- one wouldn't have to cool
>the detector as much. I'm just thinking that reducing the detector noise
>could dramatically improve photographic results, since I read a lot of
>complaints about detector noise on digital cameras. Perhaps the concept of
>a "cooled camera" is only practical for BIG cameras, or those that don't
>have to be moved around much (like something mounted on a large movie
>camera, or on the back of a big telescope). But with new miniaturization
>technologies, I wonder if cooled CCD's are a possibility, at least in a
>larger DSLR? This is a speculative sort of "what if?" question that I'm
>raising, and nothing else...


Cooled detectors have a few issues in practice.

First, Peltier coolers require a good deal of
power. Don't forget, they *move* heat, they
don't actually remove it. The moved heat still
needs to be dissipated.

Secondly, there's the issue of condensation on
the sensor itself. I'm not sure how that's
dealt with.. maybe by blowing a stream of dry
nitrogen over the sensor.

But as you can see, there are practical issues
to be dealt with.


rafe b.
http://www.terrapinphoto.com
Anonymous
March 12, 2005 2:42:26 PM

Archived from groups: rec.photo.digital (More info?)

As for condensation on the sensor, what if the CCD itself was in a small
vacuum-sealed chamber, thereby eliminating the condensation problem?
Vacuum-sealed display/detection devices have been around for a long time
(CRT's in TV's), so that technology is proven. I like the idea of the ice
bucket... Too bad you couldn't stick a couple'a beers in there, too. :-)
-Scott

"Pete" <nobody@nowhere.com> wrote in message
news:hyvpw4npx2yn.1a3trl1axv1fz.dlg@40tude.net...
>
> How about a heat pipe from the sensor to a small ice bucket on the top of
> the camera? Perhaps the size of a 35mm film canister with a snap on lid.
> Put some ice in there, and your sensor will cool down.
>
> Yes, condensation on the sensor could be an issue. I forget how much you'd
> have to cool the sensor to have a useful reduction in noise. But I'm sure
> it doesn't have to go to absolute zero.
>
> Problem is, you'd have to add a bag of party ice to your photo kit.
>
> Pete
Anonymous
March 12, 2005 4:47:40 PM

Archived from groups: rec.photo.digital (More info?)

Scott Speck wrote:
> Hi Everyone,
>
> Astrophotgraphers typically use "cooled detectors" to take long exposures
> with CCD's, because the dark-current of the detector is massively reduced
> along with the temperature of the detector. One can themo-electrically cool
> a CCD, but I believe this requires a lot of current, and the kind of battery
> typically stored onboard a handheld camera wouldn't do the job. But what
> about a larger battery pack worn on the belt, then wired to the camera?
> Could the detector itself be thermally isolated from the rest of the camera
> (via thermally non-conductive struts) so that the detector (and not the
> entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
> one to take very long exposures (half hour long, hour long) without the
> concern of detector noise? Are there new solid state detectors out there
> with lower thermal noise? Let's say you didn't want to take hour long
> exposures, but just two or three minutes long -- one wouldn't have to cool
> the detector as much. I'm just thinking that reducing the detector noise
> could dramatically improve photographic results, since I read a lot of
> complaints about detector noise on digital cameras. Perhaps the concept of
> a "cooled camera" is only practical for BIG cameras, or those that don't
> have to be moved around much (like something mounted on a large movie
> camera, or on the back of a big telescope). But with new miniaturization
> technologies, I wonder if cooled CCD's are a possibility, at least in a
> larger DSLR? This is a speculative sort of "what if?" question that I'm
> raising, and nothing else...
>
> -Scott Speck
> kaiju@comcast.net
>
>
I did see a couple of example pictures from a camera that was allowed to
sit outside in cool weather one night for a while. The pictures taken
when the camera had cooled about 30 degrees were SIGNIFICANTLY less
noisy. I was rather impressed, but haven't seen anything to quantify
this effect for any particular camera.


--
Ron Hunter rphunter@charter.net
Anonymous
March 12, 2005 6:41:16 PM

Archived from groups: rec.photo.digital (More info?)

Ron Hunter wrote:

> I did see a couple of example pictures from a camera that was allowed to
> sit outside in cool weather one night for a while. The pictures taken
> when the camera had cooled about 30 degrees were SIGNIFICANTLY less
> noisy. I was rather impressed, but haven't seen anything to quantify
> this effect for any particular camera.

EOS DIGITAL Astrophotography Guide:
http://web.canon.jp/Imaging/astro/index-e.html

M31 Andromeda Galaxy
EOS 20D @ ISO 800 12x5min
4" Skywatcher Newton f4,5:
http://www.schweifstern.de/images-pages/deepsky/m31_20D...

Canon vs Nikon DLSRs for Astrophotography:
http://www.astropix.com/HTML/M_DAP/M150/M150.HTM

Canon 1D Mark II Second Light
http://www.astropix.com/HTML/M_DAP/M10/M10.HTM

Digital Astrophotography
http://www.astropix.com/HTML/M_DAP/TOC_DAP.HTM

Dark frames analisys
http://usuarios.lycos.es/rbarbera/noise/astrod70.html

Canon 10D Digital SLR Camera for Deep Sky Imaging
http://www.hostultra.com/~ghonis/051803D10.html

ClarkVision Photography: Astrophoto 1 Gallery
http://www.clarkvision.com/galleries/gallery.astrophoto...

Astrophotography Signal-to-Noise with a Canon 10D Camera
http://clarkvision.com/astro/canon-10d-signal-to-noise


web search will find hundreds. thousands of other sites.

Roger
Anonymous
March 12, 2005 8:35:26 PM

Archived from groups: rec.photo.digital (More info?)

You mean the peltier cooler?These were used for extreme overclocking with
pentium 3s...when the cpu speed was still measured in MHz.They need huge
power supplies, something like 120 W(extra)120 W in 12 V is 10 amperes,quite
a large current.Extra batttery packs wouldn't do, mains power is
needed.That's another reason why even a 1000 euro laptop is never as
powerful as my 700 euro celeron, complete with cd r/w drive, 17"crt and
printer.See www.checkmate.gr, it's in english anyway.
--
Tzortzakakis Dimitriïs
major in electrical engineering, freelance electrician
FH von Iraklion-Kreta, freiberuflicher Elektriker
dimtzort AT otenet DOT gr
Ï "Scott Speck" <speck82@comcast.net> Ýãñáøå óôï ìÞíõìá
news:n7OdnTyJr7qjnq7fRVn-gw@comcast.com...
> Hi Everyone,
>
> Astrophotgraphers typically use "cooled detectors" to take long exposures
> with CCD's, because the dark-current of the detector is massively reduced
> along with the temperature of the detector. One can themo-electrically
cool
> a CCD, but I believe this requires a lot of current, and the kind of
battery
> typically stored onboard a handheld camera wouldn't do the job. But what
> about a larger battery pack worn on the belt, then wired to the camera?
> Could the detector itself be thermally isolated from the rest of the
camera
> (via thermally non-conductive struts) so that the detector (and not the
> entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
> one to take very long exposures (half hour long, hour long) without the
> concern of detector noise? Are there new solid state detectors out there
> with lower thermal noise? Let's say you didn't want to take hour long
> exposures, but just two or three minutes long -- one wouldn't have to cool
> the detector as much. I'm just thinking that reducing the detector noise
> could dramatically improve photographic results, since I read a lot of
> complaints about detector noise on digital cameras. Perhaps the concept
of
> a "cooled camera" is only practical for BIG cameras, or those that don't
> have to be moved around much (like something mounted on a large movie
> camera, or on the back of a big telescope). But with new miniaturization
> technologies, I wonder if cooled CCD's are a possibility, at least in a
> larger DSLR? This is a speculative sort of "what if?" question that I'm
> raising, and nothing else...
>
> -Scott Speck
> kaiju@comcast.net
>
>
March 12, 2005 9:41:27 PM

Archived from groups: rec.photo.digital (More info?)

Interesting webpage, took their 300D apart. See what's inside, without
voiding your warranty ;) 

http://ghonis2.ho8.com/rebelpeltier.html (IR mod)

This could be inexpensive as used 300D are becoming common...
Anonymous
March 13, 2005 2:39:32 PM

Archived from groups: rec.photo.digital (More info?)

rafe bustin <rafeb@speakeasy.net> wrote:

> Secondly, there's the issue of condensation on the sensor itself.
> I'm not sure how that's dealt with.. maybe by blowing a stream of
> dry nitrogen over the sensor.

I think the trick used in digital camera backes is to have a very good
condensation detector so that as soon as the onset of condensation is
detected the peltier cooler is turned off. Combine that with a fan,
and you can do a fair bit of cooling without needing dry nitrogen.

Andrew.
Anonymous
March 15, 2005 6:53:14 AM

Archived from groups: rec.photo.digital (More info?)

rafe bustin <rafeb@speakeasy.net> writes:

>Cooled detectors have a few issues in practice.

>First, Peltier coolers require a good deal of
>power. Don't forget, they *move* heat, they
>don't actually remove it. The moved heat still
>needs to be dissipated.

Figures I've seen say that the Peltier dumps about 5 times the heat out
the hot side as it absorbs from the cold side. You can get about 20
degrees C drop across one Peltier cooler, so for more temperature delta
you stack multiple Peltier modules in series thermally. Because of the
above, each layer in the stack needs about 5 times the heat transfer
capacity as the layer above it. So even though the CCD itself doesn't
output much heat, by the time you get a stack of Peltier modules 3 or 4
high, you're dumping a *lot* of heat out the heat sink (and you need a
lot of input power).

You also need feedback control of the temperature, if you expect it to
be stable for long exposures.

>Secondly, there's the issue of condensation on
>the sensor itself. I'm not sure how that's
>dealt with.. maybe by blowing a stream of dry
>nitrogen over the sensor.

That would work, but then you need a nitrogen tank.

I once did some work with a Photometrics cooled CCD camera. Their
approach was to seal the CCD and the Peltier stack into a vacuum
chamber, so there was no water vapour to condense on it. The chamber
had a window at room temperature, so there was no condensation problem
on the window. The camera head had a small port on the side so your
could re-pump the vacuum every year or two.

As you might guess, it was not small, and it was AC line powered.

Dave
Anonymous
March 15, 2005 6:55:47 AM

Archived from groups: rec.photo.digital (More info?)

Pete <nobody@nowhere.com> writes:
>Yes, condensation on the sensor could be an issue. I forget how much you'd
>have to cool the sensor to have a useful reduction in noise. But I'm sure
>it doesn't have to go to absolute zero.

If I remember correctly, the dark current is cut in half every time you
reduce the temperature by about 8 degrees C. So ice should be good
enough to get you about 5X lower dark current compared to room
temperature (20 degrees C). Certainly worthwhile for long exposures.

Dave
Anonymous
March 15, 2005 7:13:16 AM

Archived from groups: rec.photo.digital (More info?)

Dave Martindale wrote:
> Pete <nobody@nowhere.com> writes:
>
>>Yes, condensation on the sensor could be an issue. I forget how much you'd
>>have to cool the sensor to have a useful reduction in noise. But I'm sure
>>it doesn't have to go to absolute zero.
>
>
> If I remember correctly, the dark current is cut in half every time you
> reduce the temperature by about 8 degrees C. So ice should be good
> enough to get you about 5X lower dark current compared to room
> temperature (20 degrees C). Certainly worthwhile for long exposures.
>
> Dave
I have seen example pictures which certainly do bear out your statement.
Just leaving the camera out in 0F temps for an hour made VERY
significant improvement in the perceived noise level at high ISO. Of
course, that would probably not be good for the batteries...


--
Ron Hunter rphunter@charter.net
Anonymous
March 19, 2005 2:19:56 AM

Archived from groups: rec.photo.digital (More info?)

In article <n7OdnTyJr7qjnq7fRVn-gw@comcast.com>,
"Scott Speck" <speck82@comcast.net> wrote:

> Hi Everyone,
>
> Astrophotgraphers typically use "cooled detectors" to take long exposures
> with CCD's, because the dark-current of the detector is massively reduced
> along with the temperature of the detector. One can themo-electrically cool
> a CCD, but I believe this requires a lot of current, and the kind of battery
> typically stored onboard a handheld camera wouldn't do the job. But what
> about a larger battery pack worn on the belt, then wired to the camera?
> Could the detector itself be thermally isolated from the rest of the camera
> (via thermally non-conductive struts) so that the detector (and not the
> entire CAMERA) be cooled enough to greatly reduce thermal noise and allow
> one to take very long exposures (half hour long, hour long) without the
> concern of detector noise? Are there new solid state detectors out there
> with lower thermal noise? Let's say you didn't want to take hour long
> exposures, but just two or three minutes long -- one wouldn't have to cool
> the detector as much. I'm just thinking that reducing the detector noise
> could dramatically improve photographic results, since I read a lot of
> complaints about detector noise on digital cameras. Perhaps the concept of
> a "cooled camera" is only practical for BIG cameras, or those that don't
> have to be moved around much (like something mounted on a large movie
> camera, or on the back of a big telescope). But with new miniaturization
> technologies, I wonder if cooled CCD's are a possibility, at least in a
> larger DSLR? This is a speculative sort of "what if?" question that I'm
> raising, and nothing else...
>
> -Scott Speck
> kaiju@comcast.net

Peltier junctions are grossly inefficient. The problem is not only
providing power, but getting rid of the heat they produce. Their
efficiency plunges with increased temperature deltas so it's critical to
keep their hot side well cooled. High resolution CCDs are big power
consumers too. You might have to pump away 50W with forced liquid
cooling. For all that effort you might as well circulate alcohol
chilled by dry ice and forget about the heat pump.

Canon seems to have solved the noise problems with their CMOS sensors.
Even hour-long exposures have a manageable amount of noise. Because of
the very low power drawn by CMOS sensor and the entire camera, you could
probably chill the sensor by piping cold decompressed gas in through a
hacked telescope mount.
Anonymous
March 19, 2005 11:09:34 PM

Archived from groups: rec.photo.digital (More info?)

Paul Allen wrote:

> Decades ago, there was an article in Sky & Telescope on the construction
> of a nitrogen-cooled film astrocamera. They guy may have constructed
> the camera from scratch, or he may have used an old 35mm body. I don't
> recall. He was doing prime-focus work, so the 'scope was his "lens".
>
> Enough liquid nitrogen for a night's shooting was put in a small
> styrofoam cooler with a snug-fitting lid. A flexible plastic tube
> conveyed boiled-off nitrogen gas to the film chamber where it blew
> directly onto the film. It worked nicely to defeat reciprocity
> failure in long exposures.
>
> You could argue that needing a source of liquid nitrogen is a
> cumbersome requirement. The guy who wrote the article had little
> trouble acquiring the stuff from a welding supply outlet, although
> there was considerable expense involved in the process. Possibly
> dry ice in the cooler would work as well as liquid nitrogen?
>
> Surely some inventive soul with access to a machine shop will modify
> their D70 to accept a cryo-cooling input tube and post their results
> here? :-)
>
> Paul Allen

I am not aware of an LN2 camera in Sky and telescope, but there were
numerous dry-ice cameras. The king of amateur astrophotography
was Evered Kreimer in Prescott, Arizona. He popularized his
dry ice cooled camera and many others built them (this was
the late 1960s to 1970s). I built one too (mine didn't
work very well). Evered's photos were (and still are)
stunning; he used a 12.5-inch telescope to get them. Today,
amateurs are doing better with uncooled DSLRs and 3 to 4-inch
telescopes. Some amateur astrophotographers are putting fans
blowing on their backs of DSLRs to remove heat. This is perhaps
less needed with the latest DSLRs which turn off electronics
during a long exposure. If I remember right, a few have
even put Peltier coolers on the backs of their cameras (especially
those who work in warm environments). This will work if you
can plug into power, which is generally true with telescopes
(even a car battery when remote). Noise in DSLRs is very
temperature dependent. Here are some noise measurements on a Canon
10D as a function of ISO, exposure time, and temperature:
http://clarkvision.com/astro/canon-10d-signal-to-noise

Roger
!