The Canon EOS and Photography Forums

As Roger Clark shows at the end of the article at http://www.clarkvision.com/articles/microadjustment/index.html, microfocus adjustment varies with camera internal temperature.

I had this in the back of my mind for some months, but didn't put the information to use until today. Here in western Oregon, the winter weather usually doesn't vary much in temperature during the course of a few hours or a day, protected as we usuall are by a nice thermally insulating cloud layer much of the time: there are days when the variation from max to min is less than 5 F. In the last month or so we've been in an uncharacteristically sunny spell and I do essentially all of my bird photography from my car, so the interior is heating up more than usual and I now have to adjust the microfocus periodically during the day or I will get runs of images that are out of focus with a Canon 7dii and EF600/4LISii plus 1.4x teleconverter (effective focal length about 1300mm).

I probably have run into this problem before with some of my other long lenses over the years, but, since my photo gear is stored in my car in an unheated garage and I use little heat in the car for comfort and have the window(s) open for long periods when I'm actually working on a bird, the ambient temperature doesn't vary that much in winter and I never gave it any thought.

All of a sudden my car has now gets a lot warmer than the outside air as the morning progresses and the camera is also getting warmer and I'm getting lousy focus! I had previously adjusted the microfocus to a level of -1 or -2, but today I had to go to -5 to get things back to acceptable focus by 10AM.

Clark's graph shows a 7 unit change in the adjustment for 45 deg F change in camera temperature and slightly less ambient temperature change.

Note that this does not apply to using LiveView, since it is reading directly off the sensor no matter what the temperature is, but I often switch between the two methods and now have to be concerned that the excess sensor heating caused by using LiveView for any length of time will disrupt the focus adjustment when switching to phase-detect AF via the viewfinder.

I guess I'll have to learn to use Clark's quick AF microadjustment technique after all.

If you shoot from a warm car into cold air you'll get a lot of turbulent air flow. The hot air will flow out of the car and in front of the lens. This is a separate effect from focus shift, but one which will soften images significantly.

To get the optimum optical quality, the lens needs to be at the same temperature as the air. Normally people just ignore this, but it's a very well known phenomenon to amateur astronomers. Some telescopes can take hours to equilibrate and they won't perform well until they do. There's absolutely no way to observe with ultimate resolution with a telescope from a heated observatory and I'd guess the same is true for photography form a car that's not at the same temperature as the outside and with a lens that's not thermally equilibrated.

The air inside the lens also needs to be at a uniform temperature or optical quality will again suffer due to convection cells being setup. Longer lenses with larger internal volumes are most affected.

http://www.lcas-astronomy.org/articles/display.php?filename=temperature_changes_and_optics&category=general
http://www.telescope-optics.net/turbulence_error.htm

Temperature causes a focus shift with long lenses. This is well known and the reason what most telephoto lenses will focus "beyond infinity" to allow for the effects on focus of temperature. The focal length of the lens will change slightly. It's not surprising that this might upset AF and AF microadjustment since phase shift AF is a "drive to" process, not a feedback process. Live View is a feedback process, even with assistance from dual phase sensitive pixels, hence it will typically get better (but slower) AF accuracy then the normal reflex phase shift AF system.

You can easily see these effects if you can put an eyepiece on a long lens and use it as a telescope at high power. Images that otherwise look stable in the viewfinder of in Live View can be seen to be shimmering and shaking as thermal air currents both inside and outside the lens distort the view.

The effect is shown in the video near the bottom of this page - http://www.bobatkins.com/photography/tutorials/astrophotography.html

Hmmmm, makes me wonder how I ever managed to get critically sharp images!

But, then I remember that I never had a serious problem with a properly functioning 600mm lens. I expected to get 90+% sharp images with a cooperative subject and a decent amount of light and almost all from my car in recent years. All of my problems with soft images have been due to a poor lens or with low light and active subjects.

Also, I would think from your description that the worst possible condition would be with the lens in full sun when it would heat up dramatically above ambient air temperature and send shimmering waves of heat right off the front of the lens hood in addition to the internal  convection cells. And how about those big black Nikon lenses? Heat must really wreck their performance.

Aren't you overstating this a bit?

Overstating? Maybe, though I have certainly seem the effect a number of times when shooting at focal lengths over 500mm. I remember if being quite noticeable when shooting from a car with the heater on in winter. It's also quite noticeable on shots of the moon, where even when using a fixed manual focus setting, you can see variations in sharpness from shot to shot due to atmospheric turbulence.

It's not an effect I'd typically worry about under normal shooting conditions where the subject is reasonably close. However for long distance shots atmospheric turbulance sets a real limit on image sharpness. It's hard to say what the distance is where you start to notice it, but it's certainly less than a few hundred yards when conditions are bad, and it's just about always there on terrestrial shots of a mile or more. It's one reason why longer and longer lenses become less and less effective for high magnification shots at long distances. Even with a perfect lens and neglecting things like size, weight , cost, tripod stability etc, a 500mm at 100m will give you better results than a 5000mm lens would at 1000m. A 50mm lens at 10m might be even better!

I think the point is that when you have a perfect lens and perfect focus and a perfectly stable tripod, there can still be effects that will lower image sharpness due to temperature and turbulence. There are nights that amateur astronomers might as well stay indoors because "seeing" is so bad. Professional telescopes are put on the top of high mountains to try to get above as much of the atmosphere as possible and so reduce image degradation.

Can't really comment on the difference between black and white lenses, though I do remember a test that Popular photography did many, many years ago when they found that black lenses/cameras were indeed quite a lot hotter then white lenses.cameras in direct sunlight. White cameras never really caught on (Minolta once made one). It wasn't a small effect either, maybe 30 degrees difference, though I don't know the exact conditions. Could have been noon in Death Valley...though more likely it was done in New York.

OK, now we're really off into the stratosphere! To the moon, even.

I know you've made the point about having lots of disturbed air between the lens and subject before and it makes a lot of sense, but do you really photograph subjects as small as the largest birds at distances of several hundred yards (meters) and expect exhibition grade images?

And, of course the moon is tough to photograph *through MILES of atmosphere*.

This is way off course from my initial message about the effect of camera temperature on microfocus adjustment settings which has nothing to do with thermal cells inside large, almost completely empty reflector telescopes; the heat pouring out of a car with the heater on in winter and shooting through huge expanses of atmosphere.

Regarding those hollow reflectors: if the internal air circulation is such a problem, why not partially evacuate them? Industry makes huge mirrors, they can make large pieces of glass to seal the front, like photographic reflector lenses have. Caulk-up all the mounting fasteners and around the eyepiece, attach a little hand or foot operated bellows to suck some of the air out just before a shot and there you go! Of course it wouldn't be quite that easy, the pressure of the evacuated interior will distort the whole structure, but that's just a design element that needs to be addressed in the planning stage.

A 50mm lens would certainly be better, but at 10 meters? For sparrows? I don't have the field skills or physical ability to photograph any kind of birds that way, I'd like to see *your* images. And, also regarding distances to subjects, I think in terms of feet up to about 200, or so, and then forget it! Beyond that range even large hawks are in the "record shot" category, not to be confused with high-quality images. Even birds as big as Great Blue Herons only become part of a landscape at distances in that range and beyond.