24-105 F4 L and dust...

[davebell]

Up until today I prided myself on the fact that my 5D "sensor" (filter) was

still clean after multiple lens changes and a few thousand shots. I am very

careful changing lenses, i.e. indoors only and use a blower to clean the lens I

am about to attach. Anyw, all was well until I did some macro shots of flowers

today and noticed all sorts of gunk. This turned out to be a combination of dirt

on the flowers but also a fair bit of dirt on the sensor, visible from F5.6,

clearly visible at F11 and even more so at F22.

 

I used a combination of a blower and the Eclipse wet cleaning method with the

original Eclipse fluid (after brief research), not Eclipse fluid II (or whatever

they call it). Anyway, all is good save for a minor spot still visible at F11

(only after auto-level in CS2 does it pop out).

 

So finally to my question. I very recently acquired a 24-105mm F4 L lens which

is the only lens I use which extends when zooming. All other lenses are either

primes or the non-extending 17-40mm F4 / 70-200mm F4. So does anything have

experience of or suspect the 24-105mm sucking in dirt and depositing it on the

sensor filter? This is my only theory at this stage.

 

Thanks in advance for comments!

[peter_houtmeyers]

I think the 5D sensor is more to blaim than the lenses used on the body. I have heard the complaints about the 5D sensor being a dust magnet frequently on my different forums.A bigger sensor seems to attract more dust than a smaller. I use this 24-105 L IS on a 30D and a 1d body without the dust problems you seem to have.

[deselby]

Not only does the 5D sensor seem to attract a lot of dust, but the camera body seems to have plenty of nooks where dust can accumulate. I would think it unlikely that your lens will be sucking in dust and popping it back onto the sensor - although I don't have the 24-105, I would have thought that the rear element would be fixed, is it?

 

I'm not careless with my own 5D, but I do have to clean the sensor semi-regularly (I gave up on the viewfinder a long time ago), and it's always a little bit stressful. I use the sensor swabs and the eclipse fluid, and it's generally acceptable (some small minor spots in the corners) at f/22 after a couple of cleans.

[mark u]

Dust will be much more visible when doing macro work than when photographing landscapes because the effective aperture is increased over the marked aperture through the influence of the bellows factor (1+M) where M is magnification. At life size, the effective aperture is twice the marked one (possibly higher if you also take account of the pupil magnification factor, which often acts in the same direction with a macro lens).

[PuppyDigs]

The 24-104 L has dust skirts on the nested barrel as well as an o-ring seal on the mount.

So it's not likely to be drawing in much dust. However the backpressure from zoom and

focus actions may be disturbing dust already in the mirror box.

 

I have a 5D, 20D and 10D and don't find them any different in terms of attracting or

showing

dust on VF or CMOS. However it is well known in recent years Canon has been less than

careful with dust and factory leavings in the 5D and 30D bodies. Some bodies have

shipped

with "factory installed dust" in nooks and crannies. Normal use, especially the plunger-like

movement of zoom elements, easily disturbs this dust and it's redeposited on VF and

CMOS.

 

The best way to be rid of this pest is a preemptive cleaning of the mirrorbox (mirrorbox

cleaning kits are sold at B&H). Thus the "factory installed" won't be disturbed and

redeposited

on VF and CMOS.

[frank_dzambic]

I keep seeing this come up, about lenses that change length acting as a bellows and sucking in air and then blowing out the back into the camera. I don't believe it for a minute. Granted this is just my personal opinion, but I don't see how it's possible. Looking at the back of my 24-70, it's solid. There are no holes anywhere for air to escape into the camera. It's a solid EOS mount, with a solid piece of glass in the middle. I think if air is escaping anywhere when the lens is changing length, it's escaping out of the barrel of the lens towards the front. I just can't see how it's possible for any air to escape from the back.

 

I guess one way to finally prove it would be to maybe build a clear plexiglass box, and then put an EOS mount on one wall, then mount a lens to it, then fill the box with smoke, then extend and retract the lens fully and see if it disturbs the smoke in the box at all. But my personal opinion again is that I just don't see how it's possible for any air to escape from the back of the lens.

[frank_dzambic]

Well isn't that interesting. I guess I'll have to retract my earlier opinion. I just held the rear element of my 24-70 right next to my ear and "pumped" the lens to full extension and back a few times and I can definitely feel and hear air coming out of the back of the lens. That's amazing. I wouldn't have expected that in a million years. Why wouldn't they design it so the air escapes from the front rather than the back I wonder? So I guess it is conceivably possible that the lens could blow dust or dirt from the air into your camera.

[uriah]

Frank Dzambic wrote "Why wouldn't they design it so the air escapes from the front rather than the back I wonder? So I guess it is conceivably possible that the lens could blow dust or dirt from the air into your camera".

 

Frank, the 24-70 is weathered sealed which means the front is sealed from moisture and dust. The part that mates to the camera has a weathered sealed ring around it. As long as the lens is attached to a weathered sealed body not much is going to get inside in terms of moisture and dirt.

[glenn nk]

By applying some knowledge of physics, it becomes apparent that the 24/105L (I have one) is drawing dust into the 5D:

 

An externally zooming/focusing lens changes volume. Since nothing in the lens can significantly change volume (metal, plastic, glass, air), then the lens must draw in some air in order to zoom out.

 

The 24/105 lens is weather sealed (try to slip a thin piece of cardboard between the zoom ring and the barrel - it will soon encounter an O-ring seal).

 

Watch the rear element of the 24/105 when it's zoomed in and out. When zoomed out, the rear element moves forward (away from the mirror box). The volume of air that is immediately behind the rear element must increase as the rear element moves forward (away from the camera).

 

Since the lens is sealed, and the air must come from somewhere - that "somewhere" is through the camera body. Air contains dust particles (the 5D is not sealed).

 

It's not the 5Ds fault, my 30D lets in air the same way - it simply has to - physics.

 

Since I?ve been using my 100 f/2.8 Macro almost exclusively for the past three months, I?m experiencing much less sensor dust.

[jon_austin]

"Anyway, all is good save for a minor spot still visible at F11 (only after auto-level in CS2 does it pop out)."

 

You'll *never* get your sensor (filter) perfectly clean, and even if you could, it wouldn't stay that way. If the only spot(s) you have after a cleaning are invisible in your test shots unless / until you apply auto levels in your image editor, you've done a great job of cleaning, and have nothing to worry about (for a couple of weeks or so).

 

Incidentally, I've read all the posts about the "factory-installed" crud inside recent models of Canon dSLRs, and I'm happy to report that my recently-acquired 5D's sensor (filter) is virtually spot-free.

[sim]

<em>"Since the lens is sealed, and the air must come from somewhere - that 'somewhere' is through the camera body. Air contains dust particles (the 5D is not sealed)"</em>

<p>

So where does the air come from if the lens is used on a body that is sealed?

[robin_sibson1]

What Sami said. Claims about weather sealing of lenses that change length when zooming, like the 24~70 and 24~105, probably need to be treated with a certain amount of caution; they may be reasonably weather-resistant, but that doesn't mean that the lens-body combination is airtight. With the lenses that move the front element slightly within the overall physical length of the lens, like the 17~40, Canon claim only that they are weather-sealed once a filter is in place.

 

I regularly use my 100~400 on a 20D under 'safari' conditions in South Africa, and don't experience any problem that regular attention to the sensor with a blower won't fix. Any problems with a 24~70 or 24~105 are likley to be minor in comparison.

[PuppyDigs]

Robin is exactly right: weather resistant lenses are not airtight or waterproof. They simply

have gaskets on switches and the mount and a skirt on the nested barrel. They resist the

entrance of water droplets and small amounts of dust but do not by any means totally

prevent it. The main point of air leakage on the 24-105 is the front extension barrel. If it was

totally airtight it would not easily move during zoom operation. It has to vent air otherwise

the AF motor would never have enough torque to focus.

[mark u]

I suspect that the flapping mirror is much more responsible for distributing dust around the mirror chamber than the effects of zooming and focussing a lens. If you want to reduce the impact of dust, try consistently using MLU, giving enough time for the dust to settle after the mirror lifts. Of course, that is inconvenient for most kinds of shooting.

[glenn nk]

As pointed out, weather sealing is not the same as waterproof, and furthermore, waterproof is not the same as air tight.

 

If weather sealed cameras and lenses were waterproof or airtight, there would be no lens that could be focused or zoomed unless these operations were completely internal.

 

It's really quite elementary - if the volume of the camera/lens combination changes, air must get in and out. The problem is in determining how and where the air gets in and out.

 

The comment by Mark U is interesting - the flapping mirror likely does more to redistribute dust in the mirror box than anything. In second place is the shutter curtain, which considering the speed at which it moves, is certain to cause air movement.

 

Particulate matter that is small enough to be almost invisible to the naked eye is dust, and will float in the air for long periods of time. So personally, I wouldn't put too much faith in "dust settling down".

 

Almost by definition, dust is the stuff that floats in the air for hours, not milliseconds.

[davebell]

Thanks to all for your interesting comments!