Problems with manual focus (Canon EOS)

Affiliation
American Association of Variable Star Observers (AAVSO)
Mon, 07/18/2022 - 15:29

As I have posted before, I am doing a feasibility study on using a Canon DSLR camera to detect exoplanet transits, and am still trying to get good data on very low amplitude variable stars. A problem I have had consistently is that the focus rings on my lenses (35-150mm and 55-250 mm) work only intermittently. Typically when the image is out of focus, it is possible to focus it, but once focus has been achieved, it is usually impossible to back the focus ring off to defocus slightly. *Sometimes* I am able to do this, but more often it works only until very slightly defocused, and then stops working altogether (until it starts working again for unknown reasons). Last night's data was ruined because of this - the central pixels of my target star are saturated because I could not defocus quite enough.

Has anyone else encountered this problem? The fact that it fairly consistently works when out of focus but then clings tenaciously to focus once achieved makes me think this behavior is a design "feature", perhaps something in the firmware, and not a failure mode requiring service.

Affiliation
Variable Stars South (VSS)
Problems with manual focus (Canon EOS)

Liz,

Do you use a computer to control your camera? If so, what software do you use? If not, the following may possibly help.

DSLR Assistant is software to control a DSLR camera from a Mac. There are, I gather, many features, but the one most relevant to your problem is focus control, which is claimed to "Adjust lens focus precisely ...." I have seen one screen shot that shows what seem to be fine, medium and coarse focus buttons (allowing focussing in both directions), presumably clicked, or clicked and held, to activate them.

You said that a night's data was ruined because the focus was not appropriate and some pixels were saturated. If you control your camera with a computer you can display a test image and check for saturation before capturing the definitive images.

Please just ignore anything in the above which you already know or do.

When I was doing DSLR photometry, I always used a computer (or a smart phone app) to control the camera. I use a Windows PC, and thus different software from the above.

Roy

Problems with manual focus (Canon EOS)

Hi Liz,

Like Roy, I used software to precisely control lens focus back when I was doing DSLR photometry. The software was BackyardEOS which I found to be very good.

Manual focus was very hit and miss, with very tiny movement required between good and bad defocus. One of my first lenses had quite bad lag when changing the direction the focus ring was moved.

I would strongly recommend getting a quality fixed focal length lens instead of the zoom lenses you mentioned. I used a Canon 200mm f2.8L lens which was very good.

Good luck with your feasibility study. Cheers,

Mark

Affiliation
American Association of Variable Star Observers (AAVSO)
Thanks Mark, but Backyard…

Thanks Mark, but Backyard EOS apparently works only on Windows. Unfortunately the camera was purchased on a one-time Faculty Development grant at my college, so I can't just order another lens for it on the fly. Even sending the camera in for service would be complicated. I have read in many places that zoom lenses tend to drift in focal length, but I have had no trouble at all in that regard - I use my 55-250 fully zoomed in. Once or twice during a session I check it manually to make sure it hasn't backed off, and the focal length in the headers is always rock solid at 250mm. The only problem is with manual focus as I described in the OP - not that it's too touchy, but that once optimal focus is achieved, it often stops working so that defocus is impossible. I say "often" because in my previous two sessions I managed to get it just right after all. It acts like an intermittent failure of some kind - but it could be that the ring goes to sleep when manipulated in certain ways and that it's designed to work that way. Unfortunately no one on any Canon forum has been able to give me an explanation that's consistent with my experience.

Liz

Affiliation
American Association of Variable Star Observers (AAVSO)
DSLR Assistant

Thanks Roy, this looks promising - IF manual focus responds any better to software control than it does to physically rotating the focus ring. The cabling would require an adapter as the camera does not have a USB port and the Mac does not have an HDMI port... but it's worth a try if the firmware update I did last night doesn't fix the problem. I'm going to try again today. One thing I discovered yesterday is that the focus ring apparently doesn't go to sleep after capturing focus on nearby objects (at least it didn't yesterday), so I need to try it with objects at infinity. We had over an inch of rain yesterday, so I haven't had a chance to take it outside until now.

Liz

Affiliation
American Association of Variable Star Observers (AAVSO)
It's too soon to say for…

It's too soon to say for sure, but the firmware update MIGHT have fixed the problem. As far as I can tell, manual focus now works as advertised: it is inactive until the shutter button is depressed partway. Once you hear two soft beeps the focus ring will work until it isn't moved for a few seconds, then it goes inactive again. Doesn't matter if it's focused on infinity or closer (it will not focus past infinity, but I think continuing to turn the focus ring in that direction counts as "motion" for the purpose of keeping manual focus active).

The going inactive is a bit annoying but not a real issue as long as it holds the focus I've set throughout the session... and of course, as long as it continues to work predictably.

I think I will grab DSLR Assistant and try it anyway - being able to check images for proper focus could save hours of wasted work. If I can download a test image directly in the field and check it for saturation in AIJ that would be even better.

Thanks,

Liz

Affiliation
American Association of Variable Star Observers (AAVSO)
So I tried DSLR Assistant…

So I tried DSLR Assistant last night at Stellafane. I wanted to use it precisely so that I could control the image focus *before* capturing the image. It looks like that requires to press the "Start Live View" button. I tried that indoors in a well-lighted setting, and it works like a charm. But I was skeptical because Live View displaying on the camera's LCD monitor does NOT show even the brightest stars, even with the exposure time set to long enough that the captured image will show stars down to 9th magnitude or even fainter (I haven't tried to display Venus, but the dimmest object I've been able to display in Live View is the Moon).

Unfortunately, my impression is that DSLR Assistant's Live View feature merely redirects the camera's Live View to the computer screen. Since no stars are visible on the camera LCD monitor, none are visible on the computer screen either. Obviously if this is the case, then DSLR Assistant is worthless to me. An even more serious problem is that with the camera mounted on my SkyTracker, once a star is about 45* above the horizon in the eastern or northeastern sky, it's impossible to aim at the star by peering through the viewfinder unless one's head is very small - I just can't squeeze mine in between the back of the camera and the polar scope. So that's another reason I was counting on DSLR Assistant, since Eps Cep is now high enough in the sky by the end of astronomical twilight that it's no longer possible to aim the camera in the usual way.

So my question is: am I missing some feature of this software? Some sensitivity setting? Or something else? I didn't bother trying to adjust the ISO because it doesn't seem to affect Live View's actual sensitivity on the camera, though it definitely affects the apparent lighting of the Live View image in bright daylight. It appears that all the lighting effects in Live View are simulations done with post-processing, and have nothing to do with sensitivity.

Has anyone here actually used this software? (Unfortunately I cannot use Backyard EOS as I don't have a Windows laptop... :()

Thanks, Liz

Affiliation
Variable Stars South (VSS)
Test images fromDSLR Assistant

Liz,

I use different software, but always control my caneras using it. I just take test shots for correct focus with relatively short exposures of just a few seconds on higher ISO settings. The image is displayed immediately on the computer. When the focus looks good, take another test shot at your chosen final ecposure (ensure that the ISO is back to where it should be) to check for saturation.

Controlling the camera from a computer is just so much easier than controlling it manually, even if you focus the lens manually. However, I suggest that you trial the focus control in DSLR Assistant indoors. It will depend on just how it works, but if, for example, you take the focus to infinity, then find you can defocus with (say) 2 or 3 clicks of the focus button to get just where you want to be, you MAY be able to reproduce your focus position from session to session. That's how  focus control works with my software.

Roy

Affiliation
Variable Stars South (VSS)
Test images fromDSLR Assistant

Liz,

I use different software, but always control my caneras using it. I just take test shots for correct focus with relatively short exposures of just a few seconds on higher ISO settings. The image is displayed immediately on the computer. When the focus looks good, take another test shot at your chosen final ecposure (ensure that the ISO is back to where it should be) to check for saturation.

Controlling the camera from a computer is just so much easier than controlling it manually, even if you focus the lens manually. However, I suggest that you trial the focus control in DSLR Assistant indoors. It will depend on just how it works, but if, for example, you take the focus to infinity, then find you can defocus with (say) 2 or 3 clicks of the focus button to get just where you want to be, you MAY be able to reproduce your focus position from session to session. That's how  focus control works with my software.

Roy

Affiliation
American Association of Variable Star Observers (AAVSO)
Test images from DSLR Assistant

Hi Roy,

I'll certainly try focusing from within DSLR Assistant... but that won't help me in capturing science images in a pretty large swath of the northern sky. It sounds like I can't hope to use the software to guide my aiming the camera, only in setting exposure parameters once I've found the target, by taking test images prior to starting a science imaging run.

I wonder why they couldn't do more with their Live View function. I would think they could have simulated an actual gain function, since even over short time intervals a measurable number of photons must be incident on the sensor. Disappointing, but it is what it is.

Liz

Affiliation
Variable Stars South (VSS)
Finding DSLR targets without Live View

Liz,

I don't use Live View. I aim the camera at the approximate location of the target, take a short test exposure, and plate solve it (using Point Craft within my capture software), which displays the RA and DEC of the centre of the image. I adjust the mount incrementally, plate solving each image, until the target field is recognizable, and then fine tune the position of the camera visually. The plate solving is blind (the software is not 'informed' of the approximate point position). Each solve usually takes only about 15 seconds, sometimes a bit longer, because the FOV is wide.

All this is because my Star Adventurer model does not have goto capability.

It takes time, but since I have only one target per night it's not an issue for me personally.

Roy

Affiliation
American Association of Variable Star Observers (AAVSO)
DSLR Targets without Live View

Hi Roy,

That sounds very time-consuming, but I guess it's worth a try. ASTAP I believe uses local databases to plate solve, and is very quick when given approximate coordinates, so it may be doable that way. Visual fine tuning using the viewfinder wouldn't be possible, but taking multiple test exposures might be workable, if slow and cumbersome.

Then comes the other problem, if I've done multiple adjustments after polar aligning the mount: after a certain point, it's impossible to double check the alignment without removing the camera as the camera blocks the polar scope!

Thanks,

Liz

Affiliation
American Association of Variable Star Observers (AAVSO)
Of course! The problem is…

Of course! The problem is keeping it, as even small jostlings can spoil it and require redoing the alignment. But it's not a matter of taking short cuts: if the camera blocks the polar scope, there's nothing I can do with my SkyTracker. That is a problem even now with Eps Cep starting about 90 minutes after sunset.

Liz

Affiliation
American Association of Variable Star Observers (AAVSO)
Focus using DSLR Assistant

So I tried this indoors today. At least with my camera, the manual focus controls in the app are enabled ONLY in Live View mode. So I have to activate it, even though I can't use it. But using the computer to adjust the focus with test images should be doable, if slow and cumbersome.

Now I just need to wait for a clear night, a rare and precious item here in Vermont. :(

Liz