The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Strange double image a7rII w Leica APO 50

dchew

Well-known member
I'm sure there is a reasonable explanation for this but it escapes me. The other day I was hiking around Mont Blanc. Although the weather was mostly uncooperative, I did get up at 4 am during a break in the clouds and took a few nightscapes. This one is not very sharp, and upon closer inspection it looks almost like a double image. Here is the full frame:



And here is a 100% crop. If you look at the star trails you can see two parallel lines for each one, brighter above, slightly dimmer below.



First of all, please don't judge this on my lack of fine tuning noise reduction. I am new to this camera and have yet to find my comfort zone above base ISO. Here are the stats:
RSS TV24 w/ just the swivel base; no ball head
Sony a7rII
Leica APO 50 summicron Asph
f/4
2 sec time delay
10 sec exposure
ISO 800
- 1 ev exposure

This cannot be camera shake because both star trails are continuous for each star. I think the camera would have to be oscillating quite fast between two identical positions for the entire 10 second exposure to produce this by shaking, no? I double-checked EXIF in Rawdigger - image stabilization was off. To confuse the issue more, here is another photo a few minutes later with the Loxia 35 f/2:



And again a 100% crop.



No double image. So whatever it is it was associated with that camera/lens combination. But I cannot get it to repeat the problem! I've taken several images this morning to verify the APO 50 is indeed still as sharp as you would expect with no "double images".

It might be possible that I did not have the lens shade extracted. The moon was overhead but nowhere near the edge of the frame.

Any thoughts?

Dave
 

dchew

Well-known member
Here is a 100% crop of a shot this morning testing the APO 50 with the a7r. No issues...



Dave
 

pegelli

Well-known member
How about an inadvertent but very small camera move after 4 or 6 sec into the 10 sec exposure. That would explain the double star trail with one being slightly brighter then the other.

Edit: oops, senior moment, that would put a zig-zag in the star trail and not double it up, so my "theory" is incorrect.

I'll keep thinking
 
V

Vivek

Guest
It is shake from somewhere. Two exposures within that 10s.
Either that or aliasing.

Dave, Gorgeous scenes and they are not helping my AA50 lust.
 

dchew

Well-known member
It is shake from somewhere. Two exposures within that 10s.
Either that or aliasing.

Dave, Gorgeous scenes and they are not helping my AA50 lust.
Thank you Vivek. One thing I forgot to mention: I have four shots taken right after each other at different ISO's (because at 4am after a long hike I couldn't remember what best ISO to use...). All of them have the same double image.
:(
 

dchew

Well-known member
How about an inadvertent but very small camera move after 4 or 6 sec into the 10 sec exposure. That would explain the double star trail with one being slightly brighter then the other.

Edit: oops, senior moment, that would put a zig-zag in the star trail and not double it up, so my "theory" is incorrect.

I'll keep thinking
At first I thought the same thing but you are right. It would jump during the trail and make a "Z" pattern. I'm still stumped. Vivek's aliasing comment is possible, but it looks to me like the entire image is this way, not just the stars. I could be wrong about that though...

Dave
 

Guy Mancuso

Administrator, Instructor
Could actually not engaged that lens or the aperture jumped into place. It might be aperture ring
 

mjm6

Member
This may be an example of interreflections in the sensor stack...

Try shooting a night shot again of the full sky. See if the offset is related to the angle from center axis, and operates radially. That won't prove it is interreflections, but it will prove it is optical rather than an odd offset that is happening.
 

dchew

Well-known member
Could actually not engaged that lens or the aperture jumped into place. It might be aperture ring
I was wondering if I bumped that or the focus ring, but it is in all four shots and I think that would also look like a zig-zag.



This may be an example of interreflections in the sensor stack...

Try shooting a night shot again of the full sky. See if the offset is related to the angle from center axis, and operates radially. That won't prove it is interreflections, but it will prove it is optical rather than an odd offset that is happening.
Interesting. I was thinking some sort of reflection would do it, but the sensor itself didn't occur to me. It seems to me it could only be reflections or IBIS gone whacky.

I will have to wait until I travel home to do more tests, so perhaps early next week.

Thanks for all the suggestions so far...

Dave
 

k-hawinkler

Well-known member
Quote:"I double-checked EXIF in Rawdigger - image stabilization was off."

SteadyShot was really switched off, right? :)
At 4 AM I would forget that. :-(
 

Annna T

Active member
I was wondering if I bumped that or the focus ring, but it is in all four shots and I think that would also look like a zig-zag.


Interesting. I was thinking some sort of reflection would do it, but the sensor itself didn't occur to me. It seems to me it could only be reflections or IBIS gone whacky.

I will have to wait until I travel home to do more tests, so perhaps early next week.

Thanks for all the suggestions so far...

Dave
I have had the same kind of double startrails with the A7r. I think it is a camera/sensor movement. It is very short, but strong (think a jump), which would explain why there are no zigzag to see. IMO, the 2 sec. shutter delay isn't enough to prevent motion. Using a remote (either the remote app on the iPad, or the cable remote), took care of the problem for me.
 

jlm

Workshop Member
so the length of the star trace is rotation of the earth during the exposure, and the double trace, roughly perpendicular to that, is the issue?

jm
 

henningw

Member
I would have guessed a filter issue. Did you use a filter?

I've seen exactly this with a filter which had the front and back planes not perfectly parallel. This used to happen on filters from the 70's and earlier very frequently, with Tiffen and Harrison&Harrison having hardly any perfectly plano-parallel filters for sale. In recent decades this has been much improved, but not perfected.
 

dchew

Well-known member
Quote:"I double-checked EXIF in Rawdigger - image stabilization was off."

SteadyShot was really switched off, right? :)
At 4 AM I would forget that. :-(
Yeah. I actually have Memory Dial 2 programmed for a tripod, and stabilization is set off in that mode, and I did check the EXIF data. Cuz yeah I would have forgotten that too!

I have had the same kind of double startrails with the A7r. I think it is a camera/sensor movement. It is very short, but strong (think a jump), which would explain why there are no zigzag to see. IMO, the 2 sec. shutter delay isn't enough to prevent motion. Using a remote (either the remote app on the iPad, or the cable remote), took care of the problem for me.
Thanks Anna,
I suppose that is possible, but I don't understand why it doesn't appear in the 35mm shots. I will test a longer delay when I get back and take some star photos!

so the length of the star trace is rotation of the earth during the exposure, and the double trace, roughly perpendicular to that, is the issue?


jm

John,
Actually parallel, but I think that is what you meant, so yes.

I would have guessed a filter issue. Did you use a filter?

I've seen exactly this with a filter which had the front and back planes not perfectly parallel. This used to happen on filters from the 70's and earlier very frequently, with Tiffen and Harrison&Harrison having hardly any perfectly plano-parallel filters for sale. In recent decades this has been much improved, but not perfected.
No filter. But I still wonder about a reflection between the lens and the sensor, especially since it does not look like it is in the 35mm shot, although the trails themselves are smaller because of the wider view so it is harder to see for sure.
 
V

Vivek

Guest
Dave, One possibilty is the shiny M mount of the AA 50 being exposed to the sensor. The RF coupling in this lens isn't shiny but the sensor could still see the M mount if the adapter is not baffled properly. The Loxia has a good baffle.

Check the rear of the adapter with the AA 50 mounted.
 

dchew

Well-known member
Dave, One possibilty is the shiny M mount of the AA 50 being exposed to the sensor. The RF coupling in this lens isn't shiny but the sensor could still see the M mount if the adapter is not baffled properly. The Loxia has a good baffle.

Check the rear of the adapter with the AA 50 mounted.
Good idea Vivek, it is a Novoflex adapter and there is definitely some silver surfaces showing through with the adapter mounted.

Update:
I am beginning to think either Vivek or mjm6 are correct. I've been concentrating on the right side of the image because there are fewer clouds on the right. But actually the few stars on the left side do not have the double image on the ISO 100 image (30 sec exposure).

So I re-opened the ISO 3200 shot in Rawdigger and compared it to the ISO 100 shot. The ISO 3200 shot was 3.2 seconds so there are almost no trails. Sure enough there are double points on the left side, but the fainter point is down and to the right, instead of down and to the left like on the right side of the image. You can't see it on the left side of the ISO 100 shot because the star trails cover up the double image on that side. [If anyone followed that I will be amazed!]

As near as I can tell, in the ISO 3200 shot there are double points in a radial pattern. Looks like next week is test week! I have the 55 FE at home and will do some direct comparisons.

Off to bed here in Switzerland and flying back tomorrow.

Dave
 

Paratom

Well-known member
I have had something like you mention with the Leica T when taking images with some candle lights.
My interpretation was that it could be an internal reflection between sensor and lens.
I have only experienced it in one situation, so I dont worry too much about it.
Does it happen more often in your case?
 

akclimber

New member
I think you're simply seeing star trails. You were shooting with a high pixel density camera using a fairly long lens (50mm) with a fairly long exposure time (10 seconds) at a latitude of 45.8 degrees. Generally speaking, at that latitude, to be safe, your maximum exposure time to avoid star trails should be 500/focal length, or in your case here, right at 10 seconds. I do recall tho that that formula may be based on film, not digital sensors, and do know from experience that the higher the pixel density the shorter the exposure time at which star trails will be apparent. So in this case I really do think it's just a matter of exposure time showing star trails. It doesn't look like mountain ridges show the effect, which I think bolsters my attempt at a reasonable explanation :)

Cheers!
 
Top