View Full Version here: : Astromodding Canon 550d
Jasp05
11-01-2019, 01:40 PM
Hi guys,
Picked up a cheap Canon 550d and looking to replace the UV/IR Cut filter with the Baader Astro conversion filter.
I have done a fair bit of research on the subject. I've got the Gary Honis guide bookmarked for reference when I take the plunge.
Just want to confirm a few things.
I shouldn't need to remove the LP1 filter. I believe this filter will help with UV/IR cut also. (I image through a refractor, so trying to avoid having to buy a UV/IR filter to install in my imaging train after the mod).
The Baader filter I'm replacing the factory filter with will be enough of a UV/IR cut that I won't need anything else?
Link to the replacement filter - http://www.astroanarchy.com.au/Accessories/FilterCamera_2459213.html
And what glue do you use to secure the new filter in the housing. Online says a 100% silicone glue. But how do you determine how much outgassing the adhesive will give off?
And while I've got the camera body open, I'm considering doing a cold finger mod also.
How do people go about preventing condensation on the sensor and moisture buildup in the camera body itself?
I had considered opening the body completely and just put it all in a sealed box and fill it with argon permanently. and just use the peltier to chill the whole unit. I know this isn't as efficient but would certainly make condensation a non issue.
Also what size peltier module do people use? I find the tec12706 I've got needs a big heatsink to get to freezing temps. (That is with the surface facing open air with no insulation or anything on it. so maybe a smaller heatsink will suffice if it only has to cool a small surface area? )
Any tips or ideas appreciated.
bojan
11-01-2019, 02:39 PM
Hi Aaron,
I modified my 450D (bought it for $10 on ebay as faulty), I simply removed both filters, this is easiest (and cheapest) way of doing it.
For use with Canon FD 400mm f2.8, I have UVIR filter (from Lenoptec (http://www.lenoptec.com/showpro/23.html), it pases halpha un-affected, and it is way cheaper than anything else on the market - they will even cut it to size for your application. I had it cut in round shape to accomodate drop-in filter holder for Canon FD 400mm) (see here (http://www.iceinspace.com.au/forum/showthread.php?t=153419&highlight=lenoptec) the whole story).
If you intend to use it with reflector, you do not need UVIR filter at all, of course.
ChrisV
11-01-2019, 04:19 PM
I did the Gary Honis 550D filter removal as per http://dslrmodifications.com/rebelmod450d4.html . I only removed the IR cut filter - its pretty straight forward with his excellent instructions. I didn't get a DSLR specific UV/IR filter, just used a 2" filter in my scope - so no glueing with mine.
But I wasn't brave enough to do the cold finger mod. Might have reduced the noise in this pic from a warm Dec 2017 night
bojan
11-01-2019, 04:45 PM
This approach gave us more flexibility for use the camera with various equipment and projects (including IR spectroscopy)..
Jasp05
11-01-2019, 04:58 PM
So do you just remove both filters and dont put anything back in there place? Leave the sensor open to dust etc?
And how is the focusing with lenses affected? I image mainly with a refractor but will do some widefield with lenses occasionally.
And how do you manage if you do a Ha narrowband exposure. Do you still need an ir filter with a ha filter? If so how do you get both filters installed between scope, focal reducer / flattener and camera?
bojan
11-01-2019, 05:07 PM
Aaron,
Sensor is not open to dust, it is in it's own ceramic package (with glass lid).
However, the self-cleaning circuit will not work any more (because there will be no low-sticky layer which will be removed with filters), so before attempting to do modification, you should disable sensor cleaning at power-up/down.
bojan
11-01-2019, 05:12 PM
As to focusing, you can remove/replace spacers (between sensor holder and mirror box), so auto focus can work as before (I think they are ~0.3mm thick, just right to compensate for missing filters).
With camera like this (Full-Spectrum Modified, FSM) you can do NB without problems.
I have UV-IR cut filter in the light path (as drop in), L-Pro is inserted (http://www.iceinspace.com.au/forum/showpost.php?p=1296638&postcount=14) in the adapter.
I am guessing NB IR filter can be mounted in similar way.
Jasp05
11-01-2019, 11:31 PM
Is it also true that in order to use my camera lenses for daytime photography or with my lenses I will need to put a clear glass filter in to be able to obtain focus again?
If I were to remove both filters, and just buy a UV/IR Cut filter for astro use, how would it perform at UV/IR photography during the day? And given I would like to do widefield shots with my dslr lenses I'm thinking I'm going to need an internal UV/IR cut filter anyway..
bojan
12-01-2019, 06:38 AM
No, if you adjust the sensor distance with spacers (http://dslrmodifications.com/xsmod041.jpg) I mentioned earlier, autofocus will behave as before.
Calculator is here (https://www.optical-calculation.com/calculations/longitudinal_shift_induced_by_a_pla ne_parallel_glass_plate.php).
on Astrosurf webpage (http://www.astrosurf.com/buil/350d/350d.htm) they say 350d needs to be adjusted by 0.8mm
External and internal UV_IR CUT filters will behave the same... You will still need to play with white balance for daylight photography.
Rocolax (that I tried to use) will be OK for daily photography (but it is not good for Halpha, it blocks it).
Jasp05
13-01-2019, 08:02 PM
Well I've removed the LP2 filter and left the LP1 intact.
I never replaced the filter with any other glass. Going to take some test shots tonight.
Only hiccup was my dslr lenses are no longer able to achieve focus which i kind of expected.
bojan
13-01-2019, 09:30 PM
Are you referring to autofocus at some distance or infinity?
It should be possible to do manual focus (Bahtinov mask helps a lot here!) with liveview on bright star...
I am using 3D printed viewer for that (post #3 (http://www.iceinspace.com.au/forum/showthread.php?t=145306&highlight=things))
And, just today I bought swabs for sensor cleaning (https://www.ebay.com.au/itm/NEW-10PCS-APS-C-CCD-CMOS-Sensor-Cleaning-Swab-Kit-For-DSLR-Lens-Digital-Camera/323063177309?ssPageName=STRK%3AMEBI DX%3AIT&_trksid=p2060353.m2749.l2649).. (I have some dirt on the glass that doesn't want to go away with air blower), they may be helpful for you as well.
Jasp05
13-01-2019, 10:23 PM
The focus seemed to be off at any distance. using the stock lens (18mm-55m) I tried to focus on some text on a book which didn't seem to ever get into sharp focus.
It's outside running on my ED72 taking some shots of the flame and horsehead nebula.
I did a quick test exposure on M42, which blew me away as to how much more detail I can get in one 60sec sub after the mod.
I reckon one 60sec "Modded" sub is worth at least 10 Unmodded subs.
The camera seems to be fine whilst focusing at the telescope. (I don't yet have a bahtinov mask but use liveview) so that's the main thing. I'm only imaging at 357mm FL anyway, so that's still a pretty wide field even if my dslr lenses aren't usable.
bojan
14-01-2019, 07:03 AM
I am pleased to hear this exercise was succesful :-)
So it is autofocus? ... When asking you about focus, I was actually suggesting you should be able to use your lenses in manual focus mode without problems...
As mentioned earlier, you could also try to remove/replace the spacers (between mirror box and sensor board) to sort autofocus without Badder glass (which is too expensive IMO).
And as you noticed, even if the modded camera is used only for astro photography, it is still worth the effort :)
Jasp05
14-01-2019, 10:29 AM
Definately is worth the mod. Even if it took me a few hours due to being "over-cautious".
The focus I was playing with manually by hand. I must admit I only really looked through the view finder while trying to focus and I never took a photo and reviewed it. (Was running out the door for dinner).
And my 550D never had any shims on it... I questioned that in the instructions and looked everywhere in the body for some. but none were there. The sensor is mounted on springs which apply a fair bit of stiffness when screwing everything back together so maybe that was enough for my camera... (Wouldn't of thought they would space each camera on their individual characteristics??)
One thing I will say that is by leaving the LP1 filter in-tact I don't think I will worry about getting a UV-IR filter anytime soon. My stars are slightly bloated compared to my unmodded 1200D, but not to a point that it bothers me. (You have to compare side by side images to really notice).
But then again I'm still pretty new to this, (Just passed the 12 month mark), so maybe I just haven't developed enough of an aversion to these things yet.
bojan
14-01-2019, 10:50 AM
Interesting... I had spacers/shims on my 450D and removed them, not that I needed to because I do not use this camera for everyday photograpphy, I have another one for that (also obtained for couple of $ on ebay as faulty).
Factory focus fine adjustment is done in Autofocus unit which is on the lover part of the mirror box, and it is done individually for each camera and then fixed.
Anyway.. clear skies!
Merlin66
14-01-2019, 06:29 PM
Aaron,
Coming in late....
I removed both filter to maximize spectroscopy response.
To use standard lenses I added an Astronomik CLS clip in filter - works very well.
The dust shake/ Anti alias filter you have retained does provide pretty good UV-IR cut.
ChrisV
14-01-2019, 06:41 PM
Well done Aaron. Mission accomplished
1. Filter removed
2. Camera still working. Bonus!
Jasp05
14-01-2019, 11:58 PM
Thanks Chris. :D
I must admit I was pretty nervous when it came time to hit the on switch and find out if I just made myself an expensive paper weight..
But here are some images from the first night out. After having a quick play processing them I do notice the star bloat more now. But overall it's not too bad IMO.
The DSLR temps were pretty high last night unfortunately. (LCD and liveview definitely heat things up).
But overall I'm extremely happy with the end result. I've attached a few before and after images. Note the exposure time before the mod and after. all subs @ 60sec 800 ISO.
Jasp05
17-01-2019, 12:06 AM
Just got around to testing my normal lenses on the modded camera.
Turns out autofocus gets pretty close to being spot on when viewing through the viewfinder. But its still off.
However if I turn on live view and use it to autofocus it's bang on! I've tried this on my Nifty Fifty lens, and a 55-250mm lens.
I'm not sure how or why that works but it does. I'll try and post an image or 2 tomorrow showing the difference.
Have I just been extremely lucky to have autofocus still work after the mod without any shims or replacement filter?
Ukastronomer
17-01-2019, 01:00 AM
:) :) :)
bojan
17-01-2019, 06:49 AM
In live view mode the focus is evaluated via sensor (and firmware... probably by FFT) (https://fstoppers.com/education/difference-between-dslrs-viewfinder-autofocus-and-live-view-autofocus-195881), not by autofocus phase sub-system which is separate and works through small mirror (behind/below the main one) and sensors placed at the bottom of the mirror cage.
Jasp05
20-01-2019, 09:17 PM
I feel I may have jumped the gun a bit with my "achieving focus with my lenses through live view".
Although it works for daytime (not at infinity), I had a play tonight and it seems I don't have enough back focus to achieve a sharp focus at shorter focal lengths.
On my 18-55mm for example I can't get focus at 18mm, but I can at 55mm.
My 55-250mm I can't get focus at 55mm but can at 250mm.
So it seems the sensor spacing needs some adjusting to compensate.
Jasp05
30-01-2019, 08:17 PM
Seen as i have to adjust the sensor spacing after modification to enable focus with my lenses again. How do you go about testing while making adjustments?
Should only need the back cover off to make adjustments. So i guess i could take test exposures after making an adjustment?
And given there are 3 screws how important is having no tilt on the sensor?
How would i test for sensor tilt also?
bojan
30-01-2019, 08:26 PM
Another approach to get infinity focus is to adjust lens mechanics.
Adjusting tilt is tedious job, I would start with micrometer like this one (https://www.ebay.com.au/itm/Universal-Car-LCD-Display-Digital-Tester-Depth-Gauge-Tyre-Tread-Brake-Pad-Tool/253967669474?hash=item3b21a718e2:g: xWEAAOSwC1Fb4T2W:rk:2:pf:0).
CCD inspector (http://www.ccdware.com/products/ccdinspector/) may help to evaluate and indicate where the problem is..
The final adjustment can be done by placing thin shims on the camera flange (cigarette paper, or aluminium foil) to compensate for the residual tilt
Jasp05
30-01-2019, 11:51 PM
Would it be worth trying to make an artificial star for testing fwhm once adjustments have been made?
Or can i just rely on focus from examining the image after a test shot is done. At least the star test would give me a way to measure if focus is better?
bojan
31-01-2019, 07:22 AM
Probably worth it... this is (I believe) how manufacturer adjust those things (not artificial star, but definitely test pattern is used).
(For example, I have a problem with one of my lenses (Samyang 500mm F/6.3), it suffers from coma in frame corners, while centre is OK (and it is only my specimen that shows this problem, others are OK ).
I plan to use red laser module and another catadiopric lens (MTO 1100) as colimator, so my artificial star will be at optical infinity. Then I will try to adjust coma by moving the back lens group and checking the image quality using live view.)
Test shot is easiest way to determine if focus is OK or not.
Maybe you should use autofocus at distant object while it is still light, then switch to MF and wait for a dark to take star test?
At some point when modifying my 450D I was thinking about checking the distance (and tilt) of the sensor with some sort of microscope at high magnification, this arrangement should have a very narrow focus depth, but there was no need for that, tilt was acceptable, and I am using that camera exclusively for AP.
Another way to test tilt would be to try to use laser beam reflected from sensor glass cover (assuming it is parralel with sensor plane).
Something lik ethis one: http://electron6.phys.utk.edu/optics421/modules/m5/Interferometers.htm
I have laser distance module that uses self-interference with reflected beam, thus detecting the change in distance.. it is VERY sensitive, resolution is well below 1um).. To use it for that purpose, first I need to locate it (somewhere in my cubboards), then I need to make flange adapter.... this is a project for future.
vBulletin® v3.8.7, Copyright ©2000-2024, vBulletin Solutions, Inc.