Kevin_A

Here is the Nikon (Sigma 40/1.4) project for a rainy week.

First, I removed 3 screws from the Nikon bayonet, keeping one screw on so the contact/cables don't get disturbed.

I then attached the first of 2 adapters that were made by the web machine shop at the vacated 3 screw locations:

This plate fits into the rubber weather seal (the Samyang is probably not weather sealed; I know my Samyang EOS mount is not). This is why I cannot go directly to a larger diameter plate.

The second (larger) of the 2 adapters is bolted on the one above:

This particular one has tapped M2.5 four holes each at R31 and at R30. R30 happens to be what the lens spanner holes of the ZWO M54M-M48F-2, for example, while R31 are the holes for the ZWO EFW "bolted" configuration (which PrecisePart can build).

And here is the ZWO M54M-M48F-2 bolted to the second plate above:

The ZWO plate had to be modified by countersinking (90º metric) the original two holes. You can see the countersunk screw heads above, to clear any M54 device.

Finally, here is the Sigma 40/1.4 (I had set the iris to f/2.4 before doing all of this) connected to the ASI585 (for illustration -- wrong backfocus, of course). Remember that I had replace the original T2 rings on my ASI585 by a PreciseParts one with M54 hole and the same thickness as the original.

No bayonet for me! :-)

In the meantime, the fixed iris for the Samyang 135/2 (using the retaining ring from the ZWO fisheye) is all glued and ready for me to gather courage to open up the second Samyang :-).

This photo shows the iris assembly from the learning Samyang, and should just plop into the second Samyang.

In the photo, the original bladed iris is sprung closed (now we know the iris spring is in this assembly). To use the fixed iris, I just need to open up the aperture stop completely. I chose a fixed iris for f/3.3, so after the Frankensteining, I won't be able to have faster apertures. But I can choose smaller apertures by closing up the aperture ring to something smaller than f/3.3, as long as I am willing to tolerate the 18 radial spikes on bright stars. Given my preferences, I will probably never close it beyond f/3.3. And the lens become less exciting when it is slower. At f/3.3, it is equivalent to a 40mm astrograph.

The project to place a filter drawer/EFW before the flattener of the Baby-Q is also ready for testing. Just waiting a few days for the clouds/rain to go away. Can't wait untill I can change filters on the FSQ from indoors and not have to go outside to change backfocus.

Chen

Kevin_A I had a fairly clear night 2 days ago so I tested my 135mm at f2.8 with the exact backfocus shims inserted.

OK, this is looking down the Samyang 135/2.0 lens barrel from the front element.

Fixed aperture is controlled by the ZWO fisheye lens that sacrificed its retaining ring for a better purpose.

This is with the Samyang's aperture control closed up to f/8:

Whew.

We'll find out what I screwed up when I point it at the stars. I will at least try a normal terrestial image tomorrow when there is daylight. Raining just moments ago :-). If I screwed up badly, I will just buy another one, and use its native iris *spikes", but with M54 instead of bayonet. If I didn't screw up, this might make it the best Samyang 135 on earth :-).

BTW, the front element of this second Samyang has two thin brass shims. The first one did not.

There are not many steps. I left most of the back part alone. Just had to remove up to the aperture control ring. Had to remove the cosmetic focusing scale shell, though, to expose those three focuser attachmentscrews (in nylon). If I have to do it again, I will make sure they don't fall out, but with the screw retracted a little so that the rear group barrel can slide in and out. But leave one screw out of the plastic, so that we can line up one screw to its mating threaded hole in the rear group.

Definitely don't think 99% of the unwashed masses should attempt this. They can't even level a SeeStar :-).

BTW, with the EOS bayonet removed, I had some 3.8mm to 4.3mm of backfocus left after installing the Askar backfocus adjuster. set to 18mm I am guessing the actual numbers since I am just using a tree 70 yards away to judge, and using geometric optics to calculate the difference to infinity focus.

BTW, foliage was a terrible muddy brown with the AR window of the cooled ASI585 (AR window). But I am not installing any internal filters, since I don't trust even the IDAS filters to work at f/2 or f/2.4.

I just ordered a couple of generic 82mm didymium filters to place on the lens' filter thread (people use them for MIlky Way photos to remove the sodium city lights). It may not cut the UV, though, so I may have to invest in a 82mm B&W or Zeiss UV filter too. Any cheaper glass will ruin the good lens.

Hoya has a 82mm that will cut both UV and IR, but I fear even good Hoyas will ruin a good lens. But I am willing to try.

Using a cheap filter is like putting a $20 SVBONY filter on a $4000 astrograph.

With the QHY 10mm thick EOS adapter, I would have 16.5mm worth of backfocus, thus being able to use just the last 0.5mm of the backfocus adjuster (18mm +/- 2mm). Might work, and if so, just barely. With the thinnest ZWO EOS adapter, I would be 3mm short, which means there is no way to use the backfocus adjuster.

I can't install a camera rotator with the backfocus adjuster in place. But I can do without it when trying to measure backfocus as long as the tilt is small.

Chen

Kevin_A

"First light" cooled ASI585MC and Sigma 40/1.4 at f/2.4 (aperture was set earlier using the Nikon/Sony a7 adapter). 47 dB gain, 770 ms exposure duration.

I looked out the window, saw some clearing in the clouds, so I pointed the Sigma 40 with the ASI585 out the window (yes, cheap glass :-). You can see some reflection from room light.

If you look just to the left of and near the top of the tall tree on the center right, you can see what looks like a tiny patch of H-alpha. So, I cropped a part to see what I am looking at, and plate solve at astrometry.net.

The red patch looks like Lagoon Nebula. And looks like it caught Trifid at the top right corner of the crop, too :-).

Nothing is properly focus, backfocus is probably totally bogus (never tried it on stars). No tracking, no guiding. Lens is just sitting on a WO eq wedge being used as an alt-az "mount." But hey, first light is first light :-).

We are in astronomic darkness, so the bright clouds are simply illuminated by lights from Portland.

Chen

Kevin_A

Moved futher in altitude to clear the treetops, and I see this (scaled to post a smaller image):

Siril shows this (original pixel size, not scaled):

Heh, this is with no tuning of the backfocus, or even proper focusing of the stars. But remember that this is just the teeny tiny 585 sensor size.

I have the Hungarian wide angle Bahtinov mask, but not worth trying from behind a glass window :-).

Chen

Kevin_A i think that this new approach will be far superior than an stepdown ring aperture mask!

Aperture masks are actually OK with long (and simple) lenses. A simple 40mm f/1.4 should have an aperture of 28.5mm. But if you look at the Sigma 40, to compensate for everything, like light falloff, closer focus than infinity, small geometric distortion, etc, the designers go wild, and ended up with a front element glass of a whopping 72mm in diameter. Thats 2.5x the size of a simple lens. The 135mm Samyang is not as bad, but a 28mm or 14mm camera lens is even worse.

In addition to the light falloff problem, an aperture mask may not even give you the correct f-number. That, you can tune though, I believe, by watching the ADU of a star at the center of the field, i.e., using T-numbers like the movie makes use.

I think the zillion-element lens designers today just let the computer model the "best" place to place the iris.

But, here is a specialty lens (for the Leica M series no less) where the iris is place at the front of a lens:

https://www.japancamerahunter.com/2012/02/ms-optical-perar-28mm-f4-super-triplet/

[The Leica M has a flange focal distance of 27.8mm. So this lens will actually work for the 17.5mm astro cameras. Want to try it? :-) A little slow, "optimal" seems to be f/8.]

Compare that to the objective of a triplet astrograph that is designed just for one distance -- infinity. There you will see focal length/aperture to be close to the f number.

That is why I have never liked camera lenses. They are not optimized for infinity focus. I'll bet most lenses are designed to work best at 10 or 20 feet. Lenses like the Canon 100/2.8L is probably optimized for even shorter distance, like portrait distances.

Takahashi epsilon series will give f/2.8 type speeds, but nothing shorter than 180mm focal length. The FMA135 is pretty junky, unfortunately; plus Askars, as usual, are slow. So, we are really stuck suboptimally with camera lenses for short focal lengths.

Anyway, not a big problem for me, since I don't do widefields. The surgery on the Samyang, and the conversion to no-bayonet are just educational and engineering challenge. I'll be back to my Baby-Q and leave the camera lenses for meteor showers. The Sigma would have been really useful for the Aurora, where I had to use my All Sky camera (Samyang 8mm Fisheye with a ASI294MC; not too many people use a ASI294 as an all-sky camera -- they are terrible for DSO work, but not a bad camera for an all-sky camera :-)

I will probably try out the Sigma 40's backfocus on the next clear night, then test at the Samyang iris after that, and then the "filter between Petzval and flattener" on the Baby-Q after.

I have a feeling that the Sigma 40 is actually better than what the pundits say. I had suspected that the astro hobbyists poo-poo'ed it because of its price; and want to see for myself. I am not looking for price vs performance, I am looking at pure performance regardless of price. Cheap enough to give it a test (can also go to Goodwill if I don't like it).

Kevin_A

ASI585MC revisted... since I want smaller pixels than the ASI2600.

I started to look at the total camera noise for each exposure duration ("shutter priority," ha ha).

Top row is the read noise (eyeballing from ZWO's grophs, which they probably took from where the sun does not shine, just like their camera windows).

Next row is the dark noise at -10ºC in terms of e-/second.

Under that are the dark current noise based on this row, for 1 sec, 10 sec, 60 sec, 120 sec and 180 second exposures.

(10 dB gain is a linear factor of 3.16 (== sqrt(10)) by the way, 20 dB is 10x, and 25.2 is close to 18x linear gain.)

Next, I sum the read noise with the dark noise -- assuming the noise are statistically independent, so total noise is sqrt(N1 * N1 + N2 * N2). Remember: we are taling about summing noise, so must use RMS instead of just adding the two noises.

The * numbers are read nois limited (i.e., read noise is so large compared to dar current noise, so dark noise is irrelavant).

Notice that one should never use this camera with gain = 0, except perhaps with 300 second type exposures (not kind to Bortle 6). Total noise at gain 0 is consistently high at, and read noise limited to, 7 e-.

Now, if you want to use 60 second exposures, by all means use HCG at 25.2 dB gain. The read noise and dark current noise balances out at around 30 second exposures. For 30 seconds, read noise = 1.1 e- and dark current noise = 1.08 e- for HCG, so it is "optimal" in some sense. I think I will try that (i.e., HCG at 30 seconds).

If you want to use longer exposures than 120 seconds, HCG is not good. The dark current noise (multiplied by the large gain) is going to eat you alive. The total noise at 120 second exposure is actually about equal for HCG and non-HCG at 20 dB gain.

At 180 second exposures, 10 dB gain actually has way less noise than HCG (4.9e- vs 6.6e-). So, 180 second at gain 10 dB might be another nice pair to use.

Short summary: to control the noise in this noisy camera, at -10ºC, use HCG for up to 120 second exposures (30 second exposure is where read noise == dark current noise). Then switch to 10 dB gain for exposures longer than that.

Native read noise (gain 0) is just crazy high, at 7 e-; so all lucky imaging will have this much noise per frame. Even the 2600 (a pretty noisy camera) has read noise around 3.5e- at 0 dB gain. What we see in practice seems to reflect that. And you can't average it out by using longer exposures either, since the dark current noise is also pretty high. Bright galaxy camera.

Chen

Kevin_A i

OK, here is the color with the ASI585/Sigma 40 from a 82mm stack of the Hoya UV/IR-cut filter and a BreakThroughPhoto (SF based company) didymium filter.

The city light reflection from the clouds should be good for testing with vs without didymium filter tonight.

Chen

Kevin_A the true achilles heel of that camera is the low FWC

I think it is a short exposure camera. It is really more suited for the EAA folks.

HCG at 10 seconds has really low noise (see my earlier table). The 15.2 dB extra gain from your 10 dB is a factor of 5.7, so the HCG exposure value at 10 seconds is pretty much like your 60 seconds at 10 dB gain. If the FWC is OK at 60 sec/10 dB, then this (HCG with shorter exposure) may be a better combo since it clips at the same place, but has quite a bit less noise, and 6 times less total integration time.

Chen

@"Kevin_A"

Sigma 40/1.4 at f/2.4, APS-C frame,Alioth at center:

Center and right looks good. Let side shows severe tilt (an almost identical tilt exists for both the M54M-M48F-2 and the PreciseParts M54 adapters, so the tilt is farther on the lens side). It looks like a moderately "simple" tilt -- the left side of the frame is too far from the image plane:

Tuck tape might be able to fix it.

However, look at the top right and bottom right corners. Not bad.

Now get this: very low level diffraction spikes with the lens closed down from the fully open f/1.4 to f/2.4 here.

Here is a comparison of the Samyang 135 and the Sigma 40 irises:

Samyang 135/2 at about f/4 (smaller than my fixed iris):

Sigma 40/1.4 closed down to f/2.4:

You get what you pay for, again.

But this is good news. I won't have to open up this lens at all (scary at f/1.4 when putting it back together :-).

Tomorrow, I will first try going back to the original Nikon bayonet. If the tilt goes away, I can just use it as is, with the QHY Nikon adapter (10mm thick). Unlike the ZWO Nikon adapter, the QHY adapter has 18.7mm of available backfocus (the 10mm QHY EOS adapter only has 16.2mm of backfocus because the EOS flange distance is only 44mm).

Perhaps I can incorporate a push-pull tilt adjuster in that set of plates from the web machine shop. I will look into that -- should not be very difficult. I think that is the best thing to do.

Chen

Kevin_A have you tested the WO P68 yet and fine tuned the backfocus?

Yeah, no problem. Don't know what the Cloudy Nights people were complaining about. The backfocus just has to be tuned to within 0.1mm. Even 0.2mm off is a bit questionable. It is a very fast scope.

Only complain I have is that after fine tuning backfocus, the focal length came out to be 270mm. Plate solves don't lie. So this is not a f/3.8 scope as advertised, but more like an f/4. Yeah. Don't sound as good now, eh? As usual, never trust people trying to part you from your money by offering something that sounds cheaper than it should be.

They may have simply taken a number that is 4% less than the designed focal length to make the advertised specs look better, and that way, they can't get sued, since a lawyer can weasel their way around 4%. I want to see what the designers think the focal length is :-).

The fact that 270mm focal length gets you almost precisely f/4 makes me think there is some marketing hanky panky. They just don't think hobbyists will measure that carefully. Their shills don't.

Except for the high end equipment makers, this hobby is full of people bending rules. That is why things like Takahashis and 10Microns are worth every penny they ask for.

Ironically, I once got a fortune cookie that said: "Good things are not cheap. Cheap things are of no value." :-)

Chen

Kevin_A

Quick test with shimming the Sigma 40 (@f/2.4). I added 0.2mm right in the direction of the pole, and this is what the aberration inspector shows now:

The tilt is better, but still there, and it is more obvoious now that it is not just up-down error, but the bottom right is worst.

But if the aberration inspector output is what the tilt produces, I probably don't have to do anything about it :-).

I will play a bit more with the tilt, but this lens is as good as the photo world lens reviewers make it out to be. Especially when it is straight off-the-shelf with a decent iris. It really shows that the Japanese do pay more attention to detail when it comes to optics (thank you, Ed Demmings). Even thier irises are cleaner.

The center of the frame is 2º directly north of Alioth, that is why there is no bright star in the center panel of the aberration inspector. I was trying to place the entire dipper part of Ursa Major inside the frame.

Notice the difference in color of Dubhe vs the color of Alioth, even though I am using a didymium filter. This is 180 seconds at 10 dB gain with your favortite APS-C camera.

The Hungarian wide-angle 100mmx100mm Bahtinov is working well too. Very bright (very high resolution mask deposited on photographic film and sandwiched between plastic plates), and no missing diffrraction spikes all the way to the corners of the frame (mosaic of small copies of the Bahtinov pattern).

Chen

Kevin_A

Moved the 0.2mm tilt adjuster (that 3D printed stuff from Canada :-) to about the corner of the sensor, and I get this:

Still better. I will try 0.3 mm worth of tilt tomorrow, and a tad less rotation to the corner of the sensor. (That's going to be quite a few layers of the 0.075 mm worth of Tuck tape!)

I'll bet most people have more tilt than this :-).

Chen