26 replies to this topic

[elakrab]

I have always wondered if my sensor (my camera was modified by someone else back in the day, he marked the adjustment screws but who knows) is tilted, or perhaps the bayonet, but somehow convinced myself it was always the lenses I had (mainly the 70-200 f4) and the redcat. 

 

But surprise, today I tried the Samyang at f2 and it showed a much heavier tilt. Since I am now turning into the sensor as my main suspect ... how would you proceed to diagnose it and solve it? Does anyone know anyone in Berlin (or well, Germany) who would be up to fix it? 

[GalaxyPiper]

Here is some information for you:

 

https://thinklucid.c...nsor-alignment/

 

Does it affect any pictures? Alignment aberration in photographs?

 

If not, then it may be made that way on purpose.

 

Or it might be the lenses. If it is the sensor, you might have to send the camera in to the manufacturer, which means Japan.

 

Canon then will tear the whole camera apart and fix anything it finds.

Edited by GalaxyPiper, 23 October 2019 - 10:20 AM.

[elakrab]

It does affect my pictures, a lot

 

Corners

 

 

Eccentricity

 

 

FWHM

 

Edited by elakrab, 23 October 2019 - 10:20 AM.

[elakrab]

And a flat field

 

[GalaxyPiper]

That looks more like very bad collimation. A Sensor tilt would have all the stars elongated in one direction, not in multiple directions.

Cheap lenses can have this affect too.

Or a mirror that has it's adjustment for collimation screws too tight. This would have the effect of what you are seeing, as different parts of the mirror heating at different rates or straining against the adjustment screws.

Rent another Canon 6D and see if anything changes, if it does, it's the camera.

If not, it's the lens path.

Eliminate what is "not" the problem first.

Edited by GalaxyPiper, 23 October 2019 - 10:44 AM.

[elakrab]

Well, it was a Samyang 135mm f2 lens.

[tkottary]

That's spacing issue.

[elakrab]

That's spacing issue.

 

Mmmm, in a lens? And with that totally non-radial pattern? 

[GalaxyPiper]

Mmmm, in a lens? And with that totally non-radial pattern? 

I think he might mean that the "T" mount is too long or too short if mounted to a telescope.

Or the lens is for a EOS 4/3rds camera and not a full sensor camera. I'm not familiar with Samyang lenses.

If you don't have the correct lens to take advantage of the 20.2 megapixel CMOS sensor, the "Sweet" spot it too small to take advantage of that!

This would also give you the aberration...

Or the sensor set too far back on the optical plane.

You need to make sure you have the proper lens that works with that camera. Canon has F mounts, EF mounts, EF-M mounts...

Also, I think we were all assuming you were using this on a Telescope, instead of just a long lens, I don't think that was made clear.

More information the better...

 

And of course, Clear eyes, and open skies. (or is that the other way around?) 

Bryan.

Edited by GalaxyPiper, 23 October 2019 - 01:08 PM.

[maxmir]

Do you have another lens to try? 

 

Max

[elakrab]

It is a proper EF lens, no adapters of any kind but using its native mount.

And yes, I do have other lenses. The canon 70-200 shows some tilt, but in that case I am confident it is tilt because it is just one corner. Here it looks like a more weird pattern

[elakrab]

Ok, the lens is coming back to amazon. It was pretty unfortunate but I think there is something weird with it.

Edited by elakrab, 23 October 2019 - 11:47 PM.

[GalaxyPiper]

Ok, the lens is coming back. It was pretty unfortunate but I think there is something weird with it.

You can take it, and your camera, to a dedicated professional camera store, and get their opinion about it.

They won't charge you, or, they shouldn't.

Edited by GalaxyPiper, 23 October 2019 - 06:52 PM.

[whwang]

Never try to use a Samyang lens to diagnose your camera.  Samyang is the least trusted brand in quality control and many many reports are there about bad collimation on Samyang lenses.

 

To properly test whether there is a sensor tilt, you first need good lenses, better those without OS.  Then you can shoot stars at zenith.  The zenith part is very important, as it can help to eliminate (or greatly reduce) the potential plan in the lens mount..  Then if you see anything asymmetric in the four corners, you can say for sure there is a tilt in the lens or in the camera.  If several good lenses show the same symptom at zenith, then you can say the tilt is in the camera, not the lens.

[mmalik]

Can you dump your 6D and start fresh? I'll be testing a similar setup soon, SpaceCat and a7III.... Please let me know how exactly are you testing for sensor tilt (...your graphs above)? Please explain in detail so I can try replicating the same procedure. Regards

Edited by mmalik, 23 October 2019 - 10:42 PM.

[elakrab]

Never try to use a Samyang lens to diagnose your camera.  Samyang is the least trusted brand in quality control and many many reports are there about bad collimation on Samyang lenses.

 

To properly test whether there is a sensor tilt, you first need good lenses, better those without OS.  Then you can shoot stars at zenith.  The zenith part is very important, as it can help to eliminate (or greatly reduce) the potential plan in the lens mount..  Then if you see anything asymmetric in the four corners, you can say for sure there is a tilt in the lens or in the camera.  If several good lenses show the same symptom at zenith, then you can say the tilt is in the camera, not the lens.

 

Thank you. Makes sense, at least the other lenses I have don't show such extreme aberrations in the corners. Also if I remember correctly there are days for which my redcat was showing really symmetrically and nice stars over the field (but not always, some days it shows a very bad tilt) so maybe the sensor is not bad? Anyway it is a f5, so not directly comparable to a f2 lens. I need to do more test and try pointing at the zenith to discard flexures in the camera mount (which are they, but it is hard to identify their impact).

 

Anyway, I have asked Amazon for another unit and I will check it. If it is better I keep the samyang as the good units are known to deliver very nice images even fully open, but if it gives such horrible images I will just return it and live with the Canon 70-200 f4 (worse on-axis but not terrible off-axis). I am definitely not buying a $500 f2 lens to close the diaphragm to f6. .

Edited by elakrab, 24 October 2019 - 12:01 AM.

[elakrab]

Can you dump your 6D and start fresh? I'll be testing a similar setup soon, SpaceCat and a7III.... Please let me know how exactly are you testing for sensor tilt (...your graphs above)? Please explain in detail so I can try replicating the same procedure. Regards

 

Why would I dump the 6D? I mean, the sensor tilt, if it is confirmed to be that and not something like some part flexing or miscolimation of some lenses, could be eventually corrected by someone with an optical bench and enough patience.

[mmalik]

Why would I dump the 6D?

Old and outdated tech given today's standards; I've used 6D ages ago...

[whwang]

Sure, 6D is not as good as recent Nikon/Pentax/Panasonic/Fuji cameras that use Sony CMOS, but 6D is still better than the newer 6D2, and it doesn't eat stars.  So I would say keep it as long as it's not broken.

[whwang]

Also if I remember correctly there are days for which my redcat was showing really symmetrically and nice stars over the field (but not always, some days it shows a very bad tilt) so maybe the sensor is not bad? 

 

My redcat is somewhat like that.  In my limited experience using it, sometimes it show perfectly good stars in all four corners, and sometimes is has somewhat worse stars in some corners (under D800's resolution).  This could be caused by the camera mount and different pointing directions of the scope, or caused by lens shifting in the scope.  I haven't used it frequent enough to tell better.  But to be honest, all my lens/scopes with prices from ~$1,000 to $10,000 show more of less this kind of behavior.  I think under toady's high-resolution sensors, there are no perfect lenses or scopes.

[elakrab]

Old and outdated tech given today's standards; I've used 6D ages ago...

 

Well, If only I had the money ... I would get a more recent camera, peltier cooled, to replace my 6D, a FSQ106 or one of those hyperbolic systems to replace my scopes, a harmonic mount to replace the AZ Gti, a camper van to go to dark sites, move back to lovely skies in Spain . But I'm a broke postdoc and academia unluckily gives a limited budget for goodies that is in continuous fight with the budget for food and bills . 

[GalaxyPiper]

Well, If only I had the money ... I would get a more recent camera, peltier cooled, to replace my 6D, a FSQ106 or one of those hyperbolic systems to replace my scopes, a harmonic mount to replace the AZ Gti, a camper van to go to dark sites, move back to lovely skies in Spain . But I'm a broke postdoc and academia unluckily gives a limited budget for goodies that is in continuous fight with the budget for food and bills . 

Now you understand why most amateur astronomers are in their late 50's early 60's and older. Because we didn't have the money then either, and had to suffer and dream with the limited 60mm refactors or 4.5 inch Newtonian's we had at our disposal.

Most of us didn't have the money when we were younger.

That's why is behooves you to join an Astronomy club that has bigger better equipment purchased for the club that you can use.  

Edited by GalaxyPiper, 24 October 2019 - 12:39 PM.

[elakrab]

I must be the least lucky person on Earth ... this is the second Samyang 135

 

https://imgur.com/a/J2t4tKN

[elakrab]

Here are the corners

Attached Thumbnails

• 

[whwang]

To be honest, I wouldn't be surprised if this is a Samyang lens.  But also to be fair, have you ruled out sensor tilt of your camera using other more trusted lenses pointing at zenith?