Modified EOS 450D with strange "band"
Posted 08 November 2012 - 10:34 PM
Got it home, and promptly disassembled it and removed the IR Cut filter, per Gary Honis' fine webpages on the topic. Didn't encounter any problems.
Took it out for its first time on the scope tonight, and got some really strange behavior. Below is an example.
These are 300" exposures, at 1600ISO, of NGC 7662, in a heavily light polluted red zone. I was expecting the red tint to all the images, given the camera mod...but out of 10 such exposures, 9 had this apparent "void" of IR at the bottom right. It's really quite dramatic...
The really strange thing is that 1 of the 10 showed no trace of the band...
I tried different exposure times and ISO later, on M45. Shooting 30" exposures at ISO800, out of 30 exposures, 28 had the band, 2 did not.
The problem appeared seemingly randomly, regardless of ISO or exposure time.
Anyone have any ideas?
Posted 08 November 2012 - 11:17 PM
Is it possible that during reassembling the camera that maybe light is penetrating from the cable connections on the side?
It also looks like the mirror was up (live view) when these pictures where taken, just a guess.
Hope you can find the root cause.
Posted 09 November 2012 - 07:09 AM
No, not in a total dark area. High light pollution as I mentioned.
Interesting idea about the cable connection on the side. I'll have to look.
As for the mirror being up, I don't think so, but I'm not sure to be honest. I shot this one the same way I always have my 1100D. Mirror lock is disabled in the settings, virtual mirror lock enabled in BYE, shot in the "Imaging" mode of BYE.
Appreciate the suggestions, I'll check on some of those things, see what I find.
Posted 09 November 2012 - 09:51 AM
Well, this morning I spent some time in the garage, as "dark" an area as I can get. Ran all kinds of long exposures with lenses, with the scope, various ISOs, mirror lock on and off, and so on.
I can't seem to come up with any sort of consistent settings to reproduce it. It happened MUCH less frequently this morning, only once every 8-10 shots now. But it seems to happen in Frame/Focus mode, Imaging mode, Mirror Lock on or off, any ISO, any exposure time.
The only thing I noticed was that it NEVER happened with a standard lens...only while mounted to the scope. I've double checked the T-Ring, and the optical train, can't find anything. I shot 20 shots with my 1100D, the band never appeared.
I'm STUMPED at this point...
Posted 09 November 2012 - 11:08 AM
You said you had virtual mirror lock enabled in BYE. If this is the case I'm 99% sure this is the culprit. Turn it off.
Virtual mirror lock uses the live view mirror up function to flip the mirror a few seconds before an exposure. As per the warning provided with this feature on some camera model this will lead to amp glow and I'm 99% sure this is what you are getting.
In 2.1 I will be supporting in-camera mirror lock with only the camera supplied usb cable for all DIGIC 4 and up camera model.
Posted 09 November 2012 - 11:21 AM
Posted 09 November 2012 - 11:25 AM
2 ea at 30, 60, 120, 180, and 240s, ISO 1600. Mirror lock is off.
I'll report back when this is done.
Posted 09 November 2012 - 11:46 AM
I suggest that the edge of something (the reflex mirror or part of its assembly) is occasionally getting stuck or is loose and unstable (slightly moving about) and thus a small area is occasionally remaining in the optical frame to cast a small shadow.
With the T ring in place and not connected to the scope why dont you face down the front of the camera and observe what is happening with the mechanics when the shutter is fired. If the mirror assembly is loose or slightly out of position you may see it jiggle when the mirror lock is on or when off it may not always reaching its proper resting position. The amount the the object is out of position looks very small so you may have to sight carefully from a variety of off axis angles to see any item eclipsing the edge of the sensor.
Posted 09 November 2012 - 12:21 PM
Tonk, I originally thought the same thing last night, and did exactly as you describe...that was, in fact, my very first troubleshooting step. I can't see anything obvious, but of course if it IS mechanical the discrepancy could be so small as to escape notice during a simple visual inspection.
Guylain has been dead on about such things in the past for me, and appears this time (so far at least) he may be on to something.
I wonder, though, if it's a combination of both, so to speak? Perhaps Tonk's on the right track, and it's a mechanical issue with the mirror, that only presents itself in the raised live-view position? Who knows....
In either case, I appear to be able to reliably take exposures now with no issue...well, 10 of them anyway. I'm going out to a dark site tomorrow night, and will take both cameras with me, with hopes of using the 450D for the evening. I'll report back Sunday with results.
Thanks for the help and insight so far, guys!
Posted 09 November 2012 - 04:49 PM
I thought that myself after your comments last night, Guylain...but as I said in the previous post, it happened again with mirror lock on or off this morning, didn't seem to matter.
I think you are mixing mirror lock and mirror up. The "virtual mirror lock" in BYE means "mirror up" as it use the live view mirror up function to raise the mirror. Turn virtual mirror lock off and never look back
In 2.1 I will support the in-camera mirror lock with only the use cable for all DIGIC 4 and up camera models.
Don't discard what Tonk said, he's rarely wrong when it comes to these things and I have found his advice to be life savers in the past.
Hope this helps,
Posted 09 November 2012 - 06:51 PM
Two problems here - you can't search CN for matching images - and I can't remember what the resolution was (if there was one). This is now bugging me!
Posted 09 November 2012 - 07:33 PM
I think you are mixing mirror lock and mirror up.
I don't think so, as I thought I knew the difference, and what I thought I knew agreed with what you said. I believe it was simpler than that...I thought I tested this morning with Virtual Mirror Lock disabled, and didn't.
Don't discard what Tonk said
Wouldn't dream of it. Have the other threads open in other tabs as well.
Hope this helps
Advice from you nearly always does.
OK I got bugged so went hunting ...
Some threads that might be related regarding mirror shadows OR misuse of liveview:
Tonk, can't thank you enough for tracking those down. Have them open in other tabs now, and intend to digest them before leaving tomorrow. Thanks tons!
Posted 09 November 2012 - 07:43 PM
Same weird shadow thread - however no resolution was posted. Just me an others wittering on about mirror shadows and Jerry closest to the probable cause. However Bredan has just posted the most likely cause and resolution ...
Posted 09 November 2012 - 07:51 PM
Yep spot on ...anti vibration pause and "virtual" mirror lock issue. I and others have seen exactly the same thing.
Right - so thats the cause and solution . Nice spot Brendan!
From APT - worth repeating here in case Bredan's link dies at some point in the future ...
If you want to set AntiVibration pause to 0s, make sure to set "Mirror Lockup" to OFF.
If you want AntiVibration greater than 0s and to use PC operated shutter cable, you must set "Mirror Lockup" to ON.
Posted 09 November 2012 - 09:45 PM
Posted 12 November 2012 - 06:50 PM
Once again, really appreciate the detective work, gang!