Searching \ for '[EE]:: DIY Digital Camera Sensor cleaning' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/io/sensors.htm?key=sensor
Search entire site for: ': DIY Digital Camera Sensor cleaning'.

Exact match. Not showing close matches.
PICList Thread
'[EE]:: DIY Digital Camera Sensor cleaning'
2007\02\22@063023 by Russell McMahon

face
flavicon
face
The cleaning of sensors for removable-lens digital cameras is an area
replete with risks, expense and mystique. Professional sensor cleaning
companies typically charge $70 for the service ($US and $NZ and
probably $ other too). There are many products available, some
extremely expensive,  to assist you to clean them yourself, and there
are many horror stories from people who have got it wrong.

This superb website explains in detail what needs to be done (which
may not be wholly intuitive), the methods and chemicals and products
available, the dangers, and how a person of moderate aptitude and
dexterity can do it themselves safely and cheaply.

       http://www.cleaningdigitalcameras.com/index.html

Overall the writer inspires confidence that he knows what he is
talking about and is offering sound advice.

An interesting list of available products also gives pros and cons for
each and rapidly put me off one product that I'd seen advertised and
that looked promising.

They also provide links to other sites offering similar advice so you
can judge for yourself.

It would appear that a nearly ideal DIY arrangement, essentially the
process used by most camera manufacturers in-house,  can be achieved
at very low cost with a little work in setting up equipment and
acquiring cleaning materials. (Rubber spatula, correct cleaning
material, pure-as methanol, suitable care, substantial arcane
knowledge).(Complement with surprisingly simple but inobvious air blow
system - and NOT the air blow system that may logically seem most
safe).



       Russell
.

2007\02\22@081105 by Peter P.

picon face
I have strong opinions on this. The facts as I know them: Current digital camera
pixels are between 5 and 10 um per side. A single shaded, blooming or shadowed
pixel is visible to a discerning customer, especially in bright uniform light.
There are 6 to 10 million pixels in a modern camera, with a sensor area upwards
of 130 mm^2. In an ISO class 4 cleanroom with normal airflow there should be one
particle of size >5um on the 130mm^2 surface after 10 minutes of doing nothing
[1]. Just because it is sitting there. The average clean office will achieve
this in 1 minute or less. The average repair time is well over 30 minutes, with
at least 10 minutes of parts out in the open. I consider all current
filter-cleaning methods for DSLRs with removable lens unsuitable for long term
use. This is an ongoing problem, partly addressed by built-in ultrasonic self
cleaners. If you change the lens outside and take more than about 30 seconds to
change it chances are you got a particle on the optical surface. It is
interesting to notice that film SLRs actually clean each frame as it is pulled
in, over a special felt bridge. When this is reasonably clean then it is
possible to say that there will be no dark spots on the picture. When the optics
are judged 'clean' after repair it only means that you cannot see any particles
which are there. One way to alleviate the problem is to have the outer surface
of the filter well out of the focal plane. This does not work out so well with
DSLRs where the room is needed for the mirror and where the lens can produce
more than a cm of depth of field at the sensor plane when stoppered down for
shooting in bright light.

[1] http://www.dycem-cc.com/dycem_basics.html

Peter P.


2007\02\22@115926 by James Nick Sears

flavicon
face
I have dismantled my Sony DSC-P200, both the main body from the back
to clean the sensor and the back of the lens, as well as the lens
assembly itself to clean the internals of the lens mechanism.  I
haven't really done close comparisons between the camera's current
state and the first photos I took with it, but it is certainly
improved over when I started working on it, and I think it's close to
like-new quality again.

With Sony's flat-rate repair costing within a few dollars of what a
new camera costs, it didn't seem that I had much to lose.  Of course
with an SLR that costs 1k+ the equation changes.

Not to mention it was a great experience and really gave me a greater
appreciation of the mechanical engineering/industrial design that goes
into those products.

-n.




On 2/22/07, Peter P. <spam_OUTplpeter2006TakeThisOuTspamyahoo.com> wrote:
{Quote hidden}

> -

2007\02\22@131433 by Peter P.

picon face
James Nick Sears <jnsears <at> jamesnsears.com> writes:

> With Sony's flat-rate repair costing within a few dollars of what a
> new camera costs, it didn't seem that I had much to lose.  Of course
> with an SLR that costs 1k+ the equation changes.

True, but that's not the point. The point is that the SLR design is not
suitable for exchangeable objective service with digital backs unless special
measures are taken (like built in self cleaning).

> Not to mention it was a great experience and really gave me a greater
> appreciation of the mechanical engineering/industrial design that goes
> into those products.

That's true, but few people actually get to disassemble and reassemble a modern
zoom lens barrel without shearing the flat cable(s) in the lens barrel ;-( And
it isn't worth it for most 'point and shoot' cameras like you have. Your camera
retails for about $400. When there is a problem the value drops to less than
half of that. A thorough repair and checkout is hard to do without billing
about $100. So you'd end up paying $100 to make a $150-200 camera work again
(and be worth maybe $250).

Worse, your camera has relatively low requirements compared to a DSLR. That
means that the lens etc is not so critical (because it is a F2.8 max. lens) and
it can only 'eat' dust if the o-ring becomes really clogged. Also the crystal
filter is fairly far off the ccd and thus out of focus most of the time.

Compare with a DSLR that will have a F/2.8 lens but up to 16 times the sensor
area, and whose crystal filter must be very close to the sensor because of the
popup mirror. Whenever you change a lens on that 'gunk' floats in and settles
down.

Maybe there is a way to improve things by putting the crystal filter on a hinge
at the top and bellows around so it is pushed back on the lower side by the
mirror, and pops forward when the mirror springs up. This could keep any dust
on it out of the focal plane, and the bellows could act as an air pump when
compressed back to drive air through a small filter that could blow the front
of the crystal filter clean. And/or put a wiper on the back of the mirror that
wipes the crystal filter every time the mirror is cocked down (or some f the
times - latch solenoid). You read it first here.

Peter P.


2007\02\22@141027 by Jack Smith

picon face

> Maybe there is a way to improve things by putting the crystal filter on a hinge
> at the top and bellows around so it is pushed back on the lower side by the
> mirror, and pops forward when the mirror springs up. This could keep any dust
> on it out of the focal plane, and the bellows could act as an air pump when
> compressed back to drive air through a small filter that could blow the front
> of the crystal filter clean. And/or put a wiper on the back of the mirror that
> wipes the crystal filter every time the mirror is cocked down (or some f the
> times - latch solenoid). You read it first here.
>
> Peter P.
>
>  
Or, perhaps the idea that Nikon used with a film SLR - instead of a
moving mirror, use a 50% transmission mirror to split the light, half to
the prism for viewing and half to the film plane.

In an electronic camera, presumably the 50% mirror would form the front
of a sealed chamber filled with dry nitrogen in which the sensor would  
be placed.

I recognize this just shifts the problem to cleaning the front surfaced
mirror, but perhaps that could be made of material that is more scratch
resistant and also it would perhaps be easier to access. And, of course,
you throw away one F-stop due to the beam splitting.

Jack

2007\02\22@164433 by M. Adam Davis

face picon face
You also have a much greater problem of light leaking into the sensor
area from the viewfinder.

I'm very new to the whole SLR world (Canon 400D/XTi) and am glad that
1) it has some sensor cleaning, and 2) I only have one lens.

I doubt I'm going to have to deal with this particular problem for
some time, but I've been bolstered by a recent post at
http://www.strobist.blogspot.com/
and feel confident that I could do it without damage.

Thanks for the link Russell!

-Adam

On 2/22/07, Jack Smith <.....Jack.SmithKILLspamspam@spam@cox.net> wrote:
{Quote hidden}

> -

2007\02\23@073738 by Russell McMahon

face
flavicon
face
> When the optics are judged 'clean' after repair it only means that
> you cannot see any particles
> which are there.

I judge their "dirtiness" by the results on an image blown up on a
screen until I can easily discern each pixel. You can take test shots
of a neutral surface and then enhance the contrast and use an
appropriate aperture to increase the detectability, but so far I have
found that the nasties become obvious enough when viewing real-world
images at high magnifications. Or, worst case, simply when showing the
displayed images to other people.

I intend to acquire the means to do top class sensor cleaning at a
close to zero price. The lint free "Kimwipes"* (tm) wipes which are
recommended as suitable for this purpose are available for under
$US0.01 each in volume and I am abut to check the availability and
pricing of high purity methanol. A suitable rubber tool and a suitable
manual blower will complete the outfit. The end result should be a
workable system up to the best in performance and costing a very small
amount per clean compared to commercial alternatives. We'll see.



       Russell


Despite their being recommended by experts for sensor cleaning, the
Wikipedia Kimberley Clark page suggests Kimwipes may be a damage risk
in this application. Some more research needed.


2007\02\23@075430 by Peter P.

picon face
M. Adam Davis <stienman <at> gmail.com> writes:

> You also have a much greater problem of light leaking into the sensor
> area from the viewfinder.

I think that a very high resolution viewfinder (not screen) can replace the 'R'
in DSLR and use the main imager for VF picture service in a low resolution mode,
as is used in many point and shoot cameras. That would allow room for reasonable
out-of-focal-plane dust catcher arrangements in the body in front of the main
imager. It's all about user requests. Users want SLRs so users get SLRs.

See the links from 'Alternatives to the 35mm-based DSLR' for some solutions:

 http://en.wikipedia.org/wiki/Digital_single_lens_reflex_cameras

Olympus/Panasonic already use the liveview system instead of 'R' in DSLR. Or
rather, it's 'R' with a 'twist' (the optical path to the second sensor is
interesting).

Peter P.


2007\02\23@093243 by Russell McMahon

face
flavicon
face
> I think that a very high resolution viewfinder (not screen) can
> replace the 'R'
> in DSLR and use the main imager for VF picture service in a low
> resolution mode,
> as is used in many point and shoot cameras.

I have argued for this for a long time.
However, the viewing screen (= EVF) would have to be better than the
best provided to date.
I have a KM 7D dSLR and a KM A2.
The A2 has arguably the best EVF ever made with a true VGA 640 x 480
resolution in the tiny display.
Optically it's not up to the true optical path of the DSLR.
Having magnification assists critical focusing greatly and in some
situations it may in fact "out focus" the DSLR but its still not a
nice way to see your pictures.
A really top level camera would need both optical and EVF capabilities
OR a much better EVF.

I'm idly thinking about adding a CCD camera to the 7D viewfinder at
some stage to give me LCD screen live view as an option. This will be
truly horrid compared to the optical view and doesn't do quite a few
things that a true EVF does but would give me remote viewfinder access
and general preview - in some cases much valued.

>It's all about user requests. Users want SLRs so users get SLRs.

Users tend to be stupid :-).
They used to use dinosaurs so they ask for more dinosaurs.
What they REALLY want is as above (dual system or superb EVF) but they
don't know it yet.

> See the links from 'Alternatives to the 35mm-based DSLR' for some
> solutions:
>
 http://en.wikipedia.org/wiki/Digital_single_lens_reflex_cameras
>
> Olympus/Panasonic already use the liveview system instead of 'R' in
> DSLR. Or
> rather, it's 'R' with a 'twist' (the optical path to the second
> sensor is
> interesting).

Alas, the implementations to date are horrendously flawed. When used
in EVF mode the Olympus has about a 1 second shutter lag as they
pointlessly swap modes before taking the photo. Utterly unacceptable
in almost any real world photo situation.

While my older now dead KM 7Hi has a far worse EVF than the 2 models
on A2  (7 7i 7Hi A1 A2) the 7Hi is the better camera overall. Shows
that progress doesn't always cause progress.

Gone is the lovely 80 MB or so buffer which gave high speed shooting
at full res over many frames and crisp inter shot times when needed.
(Bride walking up aisle is the classic).
Resolution increased from 5MP to 8MP on the same sensor size and lens
so now noise is 8/5 worse or more. Very noticeable. I'd trade down
from an A2 to a 7Hi if I found one at OK price.
A2 video is better (Almost VGA at 30 fps).

Crossing a 7D with a 7Hi could produce an awesome hybrid - DSLEVFR.



       Russell





2007\02\23@103625 by Mircea Chiriciuc

picon face
       As a test I use to take a picture of my TFT monitor displaying a
white sheet on photoshop (just get it all white). I take the picture
with the zoom (if the lens has zoom at full wide and manual "defocus"
to ininite. Take the picture in aperture priority and close it as
much as you can in order to get the long exposure time. Move the
camera in a small circular movement as it's exposing. I then open the
picture in photoshop, convert it to grayscale and apply autolevels on
it. The dust will be very clear now.
       To clean the sensor (better said the filter) put the camera on
mirror lock-up (with AC adapter connected or with full charged
batteries and clean the filter using a rubber spatula and a lint free
wipe with a few drops of methanol on it or Eclipse. Wipe gently the
filter from left to right and back without lifting the spatula from
it. Mount the lens and your done. But be as quick as you can, because
methanol evaporates very quik.
       Do the test photo again and see the result.
       Since I have tried the blowing method also, I can say this. It may
work, but in a dust free environment. I only managed to get more dust
on the filter and firmly atached to it, because of the force of the
air current from the blower.  Do a test photo before and after
blowing the filter and you'll see what I'm talking about.

       Good luck and my best regards to you all,

       Mircea Chiriciuc

2007\02\23@103639 by Mircea Chiriciuc

picon face
       As a test I use to take a picture of my TFT monitor displaying a
white sheet on photoshop (just get it all white). I take the picture
with the zoom (if the lens has zoom at full wide and manual "defocus"
to ininite. Take the picture in aperture priority and close it as
much as you can in order to get the long exposure time. Move the
camera in a small circular movement as it's exposing. I then open the
picture in photoshop, convert it to grayscale and apply autolevels on
it. The dust will be very clear now.
       To clean the sensor (better said the filter) put the camera on
mirror lock-up (with AC adapter connected or with full charged
batteries and clean the filter using a rubber spatula and a lint free
wipe with a few drops of methanol on it or Eclipse. Wipe gently the
filter from left to right and back without lifting the spatula from
it. Mount the lens and your done. But be as quick as you can, because
methanol evaporates very quik.
       Do the test photo again and see the result.
       Since I have tried the blowing method also, I can say this. It may
work, but in a dust free environment. I only managed to get more dust
on the filter and firmly atached to it, because of the force of the
air current from the blower.  Do a test photo before and after
blowing the filter and you'll see what I'm talking about.

       Good luck and my best regards to you all,

       Mircea Chiriciuc

2007\02\23@152519 by M. Adam Davis

face picon face
The latest high end Canon (1DS MK III, IIRC) apparantly has mirror
lock up live view using the actual sensor.

I'm curious about this, because the argument against this setup has
always been that the metering and focusing sensors were cut out of the
image path (they reside in the pentaprism of most SLRs), not to
mention the sensor is designed differently since it has a 'real'
shutter compared to simpler digicams.

Of course these functions can be done using the main sensor, but then
the sensor has to be designed very differently, or read out completely
very quickly and repeatedly (then processed, and downsized for the
viewfinder).

This camera has two Digic III processors in it, and can take 10 frames
per second for up to 110 frames, though, so I suspect they've simply
thrown a lot of processing power at the problem.

Wonder if it'll give you a real time histogram on display.

The next step will be a "true" DSLR that has a movie mode.

-Adam

On 2/23/07, Russell McMahon <apptechspamKILLspamparadise.net.nz> wrote:
{Quote hidden}

> -

2007\02\23@165632 by William Chops Westfield

face picon face

On Feb 23, 2007, at 6:30 AM, Russell McMahon wrote:

> I have a KM 7D dSLR and a KM A2.
> The A2 has arguably the best EVF ever made with a true VGA 640 x 480
> resolution in the tiny display.

And it's got viewfinder magnification, too.  Still, when I dug out
my old film SLR, I was truly shocked how much easier it was to
manually focus than my A2.  Much brighter and cleared in dim light.
Little microprism or split prism on the focus screen was a big help
(and seems absent even on DSLRs for some reason.  Sigh.)  (and the
A2 autofocus is not one of it's strong points...)  I can certainly
see the attraction of retaining the optical viewfinder even in a
digital camera...

BillW

2007\02\23@172027 by Peter P.

picon face

>110 million pixels and counting ...

http://www.dalsa.com/news/news.asp?itemID=252

These are the people who make the 22MP sensor for the new Mamyia cameras.

Peter


2007\02\23@195527 by James Nick Sears

flavicon
face

On Feb 22, 2007, at 1:10 PM, Peter P. wrote:

> James Nick Sears <jnsears <at> jamesnsears.com> writes:
>
>> With Sony's flat-rate repair costing within a few dollars of what a
>> new camera costs, it didn't seem that I had much to lose.  Of course
>> with an SLR that costs 1k+ the equation changes.
>
> True, but that's not the point. The point is that the SLR design is  
> not
> suitable for exchangeable objective service with digital backs  
> unless special
> measures are taken (like built in self cleaning).

So then the equation changes even more.  Agreed that I wouldn't go  
poking around inside a high-end camera without a solid idea of what I  
was doing.

{Quote hidden}

I don't understand what you're arguing.  I said the camera wasn't  
worth paying to have cleaned.  You're saying the camera wasn't worth  
paying to have cleaned.  I think you are underestimating your  
audience here.  It was a relatively straightforward disassembly/
reassembly that I would have to guess the average person on this list  
(assuming some preselection among the members on this list toward  
hobbyist abilities) would be skilled enough to execute.  The most  
difficult part of the whole process was mustering the courage to pry  
the outer cover off, which was held in place by a few drops of adhesive.

>
> Worse, your camera has relatively low requirements compared to a  
> DSLR. That
> means that the lens etc is not so critical (because it is a F2.8  
> max. lens) and
> it can only 'eat' dust if the o-ring becomes really clogged. Also  
> the crystal
> filter is fairly far off the ccd and thus out of focus most of the  
> time.

Both the dust on my sensor and the dust embedded within my lens  
assembly were very visible on my photographs.  It had degraded to the  
point where the choice was between cleaning it or trashing it.

Again I'm not saying anything about cleaning an SLR as I have no  
experience with SLRs.  But for point and shoot cameras, a set of  
small screwdrivers and some patience can get you well on your way.

-n.










'[EE]:: DIY Digital Camera Sensor cleaning'
2007\03\15@071834 by Rolf
face picon face
Russell McMahon wrote:
> The cleaning of sensors for removable-lens digital cameras is an area
> replete with risks, expense and mystique. Professional sensor cleaning
> companies typically charge $70 for the service ($US and $NZ and
> probably $ other too). There are many products available, some
> extremely expensive,  to assist you to clean them yourself, and there
> are many horror stories from people who have got it wrong.
>
> This superb website explains in detail what needs to be done (which
> may not be wholly intuitive), the methods and chemicals and products
> available, the dangers, and how a person of moderate aptitude and
> dexterity can do it themselves safely and cheaply.
>
>         http://www.cleaningdigitalcameras.com/index.html
>
>  
.....

Just to further endorse this method. I have been using it for a while
based on the above exact web page. I have had no problems on my Nikon
D70s. I use Eclipse and PEC pads, with a piece of 1/32" balsa wood (from
my model airplanes), that has just the right amount of flex in it, and
has been cut to the exact width of my sensor.

I do a fair amount of photos at f16 and more, and for these it is
critical to get the sensor clean, else the "Dust Bunnies" can be really
obvious.

Rolf

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...