First post and hi-res lens on eBay

Have questions about the equipment used for macro- or micro- photography? Post those questions in this forum.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

First post and hi-res lens on eBay

Post by Bob^3 »

Hi all,
I’ve been lurking here for a number of months and finally joined about a month ago.
I’m a Long-time amateur photographer (40+ years) with a background in science and engineering, currently working in the optical disc industry (CD’s DVD’s and Blu-ray). Up until about a year ago, I’ve been shooting mostly landscapes and macro (up to 2x---all nature subjects) with Nikon D700 and D200 bodies. I’m currently in the process of acquiring the hardware and skills to expanding my horizons beyond 2x. Extensive searches on this forum have sure accelerated my rise up the learning curve! Hopefully, I’m up to speed enough to contribute to the forum.

I have a number of challenging projects in work including an automated x-y-z focus-stacking stage system based on Compumotor motorized (stepper motor) rails with linear encoder feedback (<2um resolution). I’m also assembling materials for a portable, high-speed motorized stacking rail for field use---like the StackShot discussed in another thread, but with a ball screw drive and a rotary encoder on the stepper motor shaft, all from eBay. I’ll start new threads for these with images as the projects evolve.

One time-sensitive item I wanted to mention is a very interesting, JML 21mm f/3.5, fixed-aperture lens currently up for buy-it-now on eBay here. He has 38 of these remaining. The description, sounds way too good to be true:

“This lens is a 21mm f:3.5 objective. Four coated glass lenses in black aluminum case measuring 1.15625" long by 1.0" diameter. Excellent correction of all aberrations for an angle of 35 deg. without vignetting. Resolution in focus is 385 lines/mm when the image of object is in focus (D=6mm), or 500 lines/mm when object is in focus (numerical aperture 1.41). Possible applications include: objective lens for scanners; photographic objective enlarging nearly 10X; objective for a hand-held microscope, enlarging from 30X to 120X (numerical aperture=1.41); lens working with light sensor or emitter as a precise focusing element; or lens for shaping a laser beam. Use with our 25L023 to make a telescope.”

The statements are rather confused and contradictory, (385 lpm or 500 lpm?, f/3.5 or f/1.41?). The only thing left out is the usual “But wait! There’s more!” I asked the vendor where he got the specs and the original purpose for the lens. He just said, try it and let me know. The price was low enough ($10 + $10 shipping), that I decided to pick one up to test. I was simply amazed by the results! Now I sound like the add. :D
Test method:
I’ve found using live-view on my D700 with a 19 inch, HDMI-connected, monitor for lens testing, allows me to quickly compare lenses without the variables (vibration, etc.) added when taking a single image or multiple images and stacking. I use an Edmund Optics, glass resolution slide (positive) as test a target---the line spacing is not close enough to determine absolute resolution. But it is very flat for indicating field curvature and the high contrast edges are good for pixel peeping.
This first post is growing a little long to go into much detail, but my preliminary observations indicate that on the bellows at 8x magnification, the JML lens nicely covers my D700 full-frame sensor, showing high-contrast edge transitions of about 2-pixels at center of frame, and 3-pixels at the extreme corners, with no visible CA! Beats the snot out of my Olympus 20/3.5 macro! I haven’t yet stressed the center-of-frame resolution with my D200. At 10x and above, my Nikon M Plan NA 0.25 objective out-resolves it and has higher contrast. But this little $10 wonder sure gets my vote for best bang-for-the-buck. BTW, it has an unusual mount, no threads just a flange (step). For the moment, I’ve got it stuck to my objective adapter with gaffers tape. I looked on the JML site, but it wasn’t listed, so probably some OEM part. I’ve seen similar flange-mount lenses used on microfiche readers.

I’ll post some images when I get a chance.
Bob in Orange County, CA

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

Post by Bob^3 »

Oops...guess the url didn't stick:

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... K:MEWNX:IT

eBay item number: 180322686893
Bob in Orange County, CA

pierre
Posts: 288
Joined: Mon Jan 04, 2010 12:37 pm
Location: France, Var, Toulon

Post by pierre »

Hi Bob,

Thanks a lot for your post!

I have ordered one to try too.


Best Regards,
Pierre

AndrewC
Posts: 1436
Joined: Thu Feb 14, 2008 10:05 am
Location: Belgium
Contact:

Re: First post and hi-res lens on eBay

Post by AndrewC »

Bob^3 wrote:...

I have a number of challenging projects in work including an automated x-y-z focus-stacking stage system based on Compumotor motorized (stepper motor) rails with linear encoder feedback (<2um resolution). I’m also assembling materials for a portable, high-speed motorized stacking rail for field use---like the StackShot discussed in another thread, but with a ball screw drive and a rotary encoder on the stepper motor shaft, all from eBay. ...
Sounds like a kindred spirit :)

How are you planning to control them ? When I first started with automated stages I put them on closed loop positioning but to be honest I rarely use it now. The kinds of loads I was moving were well within the capability of the motors (small Nema 17 sizes) and even microstepping at low torque I wasn't losing steps so tend not to bother with it now.

As of now I run positioning out of a PIC based controller (has nice quadrature decoders built in) but I'm very slowly moving over to a VB interface.
rgds, Andrew

"Is that an accurate dictionary ? Charlie Eppes

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

Post by Bob^3 »

Hi Andrew,
I’ve been following the development of your setup with great interest. I think you are correct with regard to absolute feedback on stepper-motor driven stages---in general, for this application, it is probably overkill, at least at the focus step sizes requirered for magnifications of 10x or less. However, the x-y stage I was able to pick up on eBay for a good price happened to have 1um resolution, Renishaw linear encoders, built-in. Also, the controller I’m using (Parker Compumotor 6K2) does have encoder inputs. So, there was really no extra effort to implement this option.

That said, there were additional factors I considered. First, the source and history of the motorized stages available at a reasonable price, perhaps best described by the phrase ‘eBay is like a box of chocolates…’ :D It doesn’t take much damage, particulate or wear on bearings or drive screws to make a stage stutter. Also, micro-stepping does have many trade-offs, as was discussed in an earlier thread. Factors such as the mechanical load (horizontal or vertical orientation), lead-screw or ball-screw pitch, whether or not a reduction gearbox is used between the motor and the screw and the required minimum focus-step distance (based on maximum intended magnification), will determine whether or not open-loop micro-stepping is reliable. Absolute feedback from an integral linear encoder, used in combination with micro-stepping, should be able to minimize or eliminate many of these potential issues.
Bottom line, even if I don’t decide to make the encoder output part of the feedback loop, I can still use the controller’s ability to monitor and display the absolute position to verify that the stage is moving the correct distance.

I, also, have been testing microcontroller (pic and arduino) based stepper controllers and will probably use one of these on the portable rail I mentioned where size, weight and ability to run on battery power are important factors. For the studio, I wanted a controller and motor drives that didn’t require low-level programming (hey, I’m a hardware guy and don’t particularly enjoy programming). After much research, I chose all Compumotor and Daedal components (divisions of Parker Hannifin) because they are readily available on eBay, are very robust, plug and play and relatively easy to configure. Most importantly, Parker is the only manufacturer I found that maintain a full set of documentation on discontinued parts on their web site, including manuals, sample programs and a programming aid called Motion Planner (all for free download). The 6K series controllers have another feature I find very usefully called “teach mode”. This allows the controller to record the movement of the stage as you jog to the start and end positions for the stack using a joystick. In practice, I set the desired distance per image in microns (or number of steps), jog to the furthest focus point on the subject, press a button to record that position, jog to the nearest focus point, press a button, then press the start button to run the stack. The controller has battery-backed ram so it can be used as a stand-alone system once programmed with a pc through either the serial or network port.
Bob in Orange County, CA

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

Post by Bob^3 »

Posted: Wed Apr 14, 2010 4:01 am Post subject:

--------------------------------------------------------------------------------



Hi Pierre,

I would be interested in knowing your impressions of this lens. Maybe I’m hallucinating after sniping on eBay for so long, but the copy I have seems to clearly out-resolve my full-frame (24x36mm) D700 (8-micron pixel size) sensor corner to corner from 7x to over 8x magnification. At 5x-6x and 9x-10x it’s very useable with only slight softening in the extreme corners.

This may become my go-to lens for this difficult to cover magnification range, at least on full frame. The working distance at 7x is over 10mm. It has a perfectly circular fixed aperture stop installed---a quick look at bokeh looks very good. I haven’t measured field flatness, yet. But at 7x on my glass resolution slide, it looks fine.

If the stated specs are true, at 500 lpm (2-micron), it matches the required resolving power 4:1 ratio to deliver a diffraction limited image to an 8-micron pixel size sensor. However, it may be a little soft on a 4-5 micron crop-sized sensors or a 21-24 megapixel full frame sensors.

Regards,

Bob
Posted: Wed Apr 14, 2010 4:01 am Post subject:

--------------------------------------------------------------------------------

Hi Bob,

Thanks a lot for your post!

I have ordered one to try too.


Best Regards,
Pierre

Bob in Orange County, CA

rjlittlefield
Site Admin
Posts: 23561
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

Bob,

I agree with your guess about microfiche reader. These lenses can be highly optimized for the specific application and they work very well exactly at their design point. I am intrigued by your observation that it gets soft at the corners at 6X and 9X but is sharp corner to corner at 7x-8x.

The best lenses I know personally for this magnification range are the Olympus 20 mm f/2 and 38 mm f/2.8 bellows macro lenses, see HERE and follow the links. But they are pretty expensive. If this $10 wonder comes even close, it will clearly win the bang-per-buck contest.

I've ordered one, for curiosity.

--Rik

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

Post by Bob^3 »

Hi Rik,

I was hoping either you or Charles would be willing to pick one of these up and test it. Don’t know, maybe I got the first proof of principle prototype that they spent $100k on to custom polish. I grabbed a second copy for backup and to see any sample variation.

Unfortunately, I don’t own the Olympus 20mm f/2 or 38mm f/2.8 for comparison. But compared against my Olympus 20mm f/3.5, the $10 JML lens has slightly better resolution at center frame and much better at the corners. Both lenses show very-low chromatic aberration, corner to corner.

To clarify the term “soft corners”, I think the JML lens is very useable from at least 6x to 9x (better at 6x than 9x). At 9x, it’s looses some resolution all over as would be expected due to diffraction (effective f/35 if the f/3.5 spec is correct). At 10x an above the Nikon M Plan NA 0.25 is clearly better. I haven’t tested below 6x, limited by the minimum extension on my PB-6 bellows.

Like the Olympus 20mm/3.5 (I have the multi-coated version), it does appear to be a bit prone to flare and loss of contrast, at least as I’ve been testing it with a glass resolution slide target, with 100% back lighting (using the slide copy attachment for the PB-6 bellows). The working distance is comparable to the Nikon M Plan 10x objective. However, the 1 inch diameter and flat (non-tapered) working end might make lighting somewhat challenging.

I have some images I’ll host on my PBase site when I get a chance.

Rgards,
Bob in Orange County, CA

pierre
Posts: 288
Joined: Mon Jan 04, 2010 12:37 pm
Location: France, Var, Toulon

Post by pierre »

Hi Bob,

Not received yet....

Quite impressive project of yours !


Regards,
Pierre

Bob^3
Posts: 287
Joined: Sun Jan 17, 2010 1:12 pm
Location: Orange County, California

Post by Bob^3 »

Here's some test images from the JML 21mm f/3.5 lens from eBay compared to the Olympus 20mm f/3.5 macro, taken with the Nikon D700 on bellows.
The subject was an Edmund Optics glass resolution slide. The lines are 1.0 mm spacing for easy scale reference (width of each black or white band is 0.5mm, 3 cycles=8x on D700 sensor).
RAW files converted in Adobe Camera Raw CS4, No sharpening applied at any stage, just exposure and black levels set for no clipping. Note that the crops are 500% not 100%.
I got the magnification slightly different for the two lenses, but this made little difference.


JML 21mm f/3.5, full frame, 7.5x, Nikon D700
Image




JML 21mm 500% center crop, 7.5x
Image




JML 500% corner crop, 7.5x
Image




Olympus 20mm f/3.5, full frame, 8x, Nikon D700
Image




Olympus 500% corner crop, 8x
Image




Olympus 500% center crop, 8x
Image

These are my first images posted...hope they come through OK.

As I processed these, I noticed the Oly has more CA than I noticed with Live View.

Regards,
Bob in Orange County, CA

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

That's interesting, Bob.
Was the Olympus at full aperture?
I suppose it might improve on stopping down
Was it the latest/most coated one? I believe there were a couple of variants.

I find on my D700 that a sharp edge, with the most bestest of lenses, takes three, or slightly more, pixels on the sensor. (It would be nice if the low-pass filter could be easily removed!)
That's about what I see in yout blow-up, so I'd agree, the cheapie does seem to be doing better than could make much difference.

The effective aperture would be (7+1) x3.5 = 28, which according to the common references, should be starting to have some effect, with a D700 sensor. The Oly 20 F2 could be better. Whether or not you and I would see it at this magnification, I don't know! As you suggested earlier, those with smaller pixels could benefit.
I have a sample currently being held up by the Icelandic Impediment. It'll be interesting to repeat something like your tests.

I wonder what these little lenses cost new!

morfa
Posts: 554
Joined: Mon Oct 12, 2009 2:14 pm
Location: Stockholm, Sweden
Contact:

Post by morfa »

Thanks for bringing this interesting lens to our attention Bob! I'd love to see some test shots of a "real" subject.

The first image shows a large, diffuse blue flare at the center of the image.

Do you always get this with this lens? Have you tried shading the front element?

I'm asking because I've had trouble with some lenses producing this kind of centered flare – often bluish or purple in color. If the lens has a variable aperture you can vary the size of the flare (stop down and it becomes small and more distinct) but it's always perfectly centered. It tends to appear to a higher degree when using flash and/or when there is something very bright in the frame. Some lenses show this behavior to such degree and frequency it puts me off using them.

Anybody else recognize this? Ideas on what causes it?

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

John,
Anybody else recognize this? Ideas on what causes it?
I've experienced it when experimenting with different optical arrangements in microscope trinocular heads. My guess is it has something to do with reflections off the low-pass filter in front of the sensor and then reflected back from one of the lens elements. I know that in some cases manufacturers have re-designed lenses with the digital sensor "reflection problem" in mind... particularly by paying closer attention to lens element shapes and better (and more extensive) lens coatings.

Have you noticed this more with older lenses (with presumably less efficient coatings), or with lenses that have noticeably low curvature elements?

mgoodm3
Posts: 273
Joined: Mon Sep 08, 2008 8:50 am
Location: Southern OR

Post by mgoodm3 »

I see this kind of flare issue with quite a few lenses when shooting a glass slide using reflected light. A dark piece of cloth over the front of the slide often helps. Also helps to back the light off a bit.

morfa
Posts: 554
Joined: Mon Oct 12, 2009 2:14 pm
Location: Stockholm, Sweden
Contact:

Post by morfa »

Thanks Charles,

Yes, I've only seen this with older ("pre digital") lenses.

Light bouncing back and forth between the rear element and sensor filter has actually been the most likely explanation I've come up with for myself, and I almost mentioned it in my last post.

My main reason for thinking this has been that one lens that shows this behavior is the (otherwise excellent) Tamron 90mm f2.5 (MF, adaptall version). I've tried the later "DI" version and it does not have this problem at all. I remember reading somewhere that the "DI" series were designed/modified to be used with digital sensors. Perhaps this is one of the manufacturers you are thinking of as well. The adaptall version has a perfectly flat rear element facing the sensor – I don't remember looking at the rear element of the "DI" version.

Post Reply Previous topicNext topic