Hi all,
I upgraded my scanning rig with a NEX7 APS-C body and a Minolta Dimage Scan Elite I scanner lens.
Rationales and details can be found here: https://www.arnogodeke.com/Blog/Using-a-scanner-lens-for-DSLR-scanning
Rationales and details can be found here: https://www.arnogodeke.com/Blog/Using-a-scanner-lens-for-DSLR-scanning
Great that you’ve written all this up in so much detail both here and on your blog. I hadn’t realised that the 5400 scanner lens was so particular with respect to magnification, though it makes sense that it would be I suppose. If you were to fill your 24MP sensor with the 35mm negative rather than using just 20MP is the fall off in the corners quite apparent?
Good question. I will try further when I find time and report back. I do need to play around with it more to find the optimal configuration, since I simple enforced the lens to work at its design value but I’m not sure what happens when I slightly deviate from that and fill the frame. I’m currently struggling with the alignment and am puzzled why I need to create a few degrees of apparent misalignment of the camera with respect to the negative to obtain sharpness across the entire frame. Perhaps something is still somewhat off. The bottom line, however, is that indeed the lens is very good for this purpose. My Micro-Nikkor 105/2.8 worked very well also, and would work even better on an APS-C sensor and provide more DOF rendering alignment less critical, but it’s rather bulky and heavy so I rather get the tiny scanner lens working. More later!
Tilt and shift can be used to prevent reflections like e.g. seeing the lens front reflected from the negative. Whether Minolta used this tech in their scanners, I can’t say. Tuning the alignement (mirrors can help to do that) should be the first step anyways.
Thanks for your reply, you’ve probably seen that Belinda was/is a great advocate for this lens on here, she still comes on from time to time. Perhaps she may be able to help.
Great to see the Minolta 5400 lens mounted in your setup Arno. I recently acquired a Scanner Nikon ED 7 element lens from a Coolscan and currently are getting parts from RAF Camera to fit it. Will post photo and results later.
If you haven’t seen this yet, it is quite interesting, see at end 5400 lens covers full frame!
Yes, I had been in contact with Belinda and she helped out but also claimed that the limited DOF was a bit hard to work with.
Yes, I had seen this. 1:1 (and 1.2x by the late Robert O’Toole) were confirmed to work well, so I could use my 16 MP FF DSLR as well, but insisted on using a Sony APS-C since 24 MP, easily adaptable, and small, compact, and light. Not least of all it releases my main DSLR from scanning duties. So I need to operate it reversed, around 0.67x magnification. I will do as Harry suggested above and see what happens if I used it at 0.67x as opposed to 0.57x (5400 design value). In forward direction at 0.67x it’s too far from the 1.76x deign value and the corners are very fussy that way. Will update my findings when I find more time.
Update: I just tested the thing covering full-frame, i.e., at 24 MP and, hence, now at 0.67x reversed, corresponding to 1/0.67 = 1.49x in forward direction. I see no apparent difference. It’s better if anything: I’m now touching line pairs 0-2, or 62 lp/mm in the center. Corners seem similar as at 0.57x, perhaps even better, and I no longer have a need to some small angle between the negative plane and the sensor. Perhaps re-mounting the lens has aligned it a tad better. Great! So now it’s at 24 MP with a tad higher resolution overall it seems.
Now alignment of the sensor with the negative becomes really critical to optimize it. @Digitizer was mentioning the mirror. I was never so successful at that, but saw a picture somewhere with “drops” or “dust” on the mirror and some homogeneous distortion outward from those if OOF. I don’t remember where I saw that though, but I need a proper (optical) way to align the negative and the sensor now since so far I was doing that mechanically…
Another update:
Did another mechanical alignment and then a mirror test > Seemed fine now. Overall shot SOOC in B&W, black and whitepoint adjustments only:
Resolution in center at 200%:
Top left, top right, bottom left, bottom right, all at 200%:
Center at 200% with “detail” slider in the “detail” window in the LRC developer module maxed out to 100%, which slightly improves things a tad more:
I believe this is as good as it gets for this lens: Around 62 lp/mm center and 55 lp/mm in the corners. Bottom left corner seems just a tad behind the rest, but not by much. I did my best focusing (zoomed in a lot in the center). Happy camper!
Focus can be aligned with something like this too:
It’s meant to be displayed on a screen. I suppose that it should work as a print or transparency too. The pattern should be symmetrical when things are properly aligned.
BEWARE: staring at the image can make you feel dizzy.
Note: Original image size is 1000 x 1000 pixels
Mark D Segal describes the Zig-Align & Versalab methods here:
@Digitizer : Ah, yes, I have that permanently on my desktop. It’s great for lens collimation and works by creating interference patterns. Not sure if it will work for alignement in our intended purpose, but I’ll give it a shot. Perhaps I can simply show it on my phone and mount that…
@Harry : Yes, I read that some time ago, and was intrigued by the zig-zag but the prices are a tad high for what it does. Perhaps we can be smarter and cook something up ourselves (…).
Updates: I played around with various options for alignment and found it most accurate to use Vlad’s target’s “Siemens stars” in combination with focus peaking: One moves the negative or camera back and forth, and observes focus peaking happening. The focus peaking indication should happen at all 4 Siemens stars in the corners exactly simultaneously. When this happens, alignment is perfect. The process is described on Vlad’s pages (film4ever.info). This process in the end worked best for me. I updated my blog with the most recent findings and results (Using a scanner lens for DSLR scanning: An ideal solution? - arnogodeke) so as to not further pollute this thread.
I have not noticed flicker with mine. Have you fully confirmed that it does this with various power supplies?
How has this setup been treating you? Worth it?
I have just gotten a Scanner-Nikkor from an 8000 and needing to figure all the steps to get it attched to my camera (R5 for slides or S1R for high res mode). It is a bit daunting getting this solved.
When I first bought the scanning Nikkor (also from a CS8000) I went through RAF camera to get an adapter made and spent approx. 100$ getting a two-part adapter made. The first one is a Thorlabs SM2 thread that clamps onto the lens and the second is a long SM2 thread that the first connects to with a standard M42x1 photographic thread at the end. I did this for versatility because I had been reading Robert O’Toole’s website and liked the idea of using a common interface such as SM2. I wanted to be able to use other lenses on the same setup.
I recently bought a lot of M42 extension tubes, some of which have a replaceable female thread, i.e. they have a male M42 on one end and the female thread is an insert with a v-groove that you can replace. To my surprise the clamping diameter is EXACTLY that of the nikkor’s OD. Now I just use that as an adapter. This is what it looks like:
Here you can see it in action with a long and ugly lens hood I 3D printed. The RAF camera is still mounted to the lens but I haven’t used it since.
You WILL need a lens hood and good flocking btw. And a very sturdy macro rail. It’s not easy to focus this thing.
This is how i fine tune my alignment exactly. 1-Mirror to rough it in. 2-Vlad’s targets to fine tune with my leveling feet on my holders.
ok, all great info. thank you! Was also looking at Robert’s (RIP, his website CloseUp is incredible) mounting solution.
I have a NiSi macro rail, and getting a new copy stand this week - Polaroid MP-4 (cheap and sturdy) or a Kaiser Copylizer Exe.cutive with RS1 head and which amazingly has a light up base great for contact sheets and test photos (if it doesnt flex under load). But I digress.
Any specific M42 tubes you could point me at? Sounds economically efficient AND useful.
As for the hood you made, is lens that sensitive to ANY stray light?! Either way, good info to have as I have read about this lens being good but not much about its practical use.
Then the only thing left is getting the spacing right for roughly 1:0.8 to 1:1 range. I see your bellows, does that feel like the best solution? I am worried about all the extra weight to properly mount and focus this lens. At a certain price point a Sigma 70 Art or 105 Art has to suffice!
… or one of https://www.venuslens.net/
They have been doing some nice lenses at reasonable prices lately.
Phil
Sorry for my late reply (again).
Yes, I recently found several USB power supplies while tidying up and tried them out. It’s best to use a power supply from Apple, as someone has already written here. At the moment I’m using power supplies from Apple and DJI.