Polaroid 43.53mm F/6.6

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

Moderators: Pau, rjlittlefield, ChrisR, Chris S.

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Polaroid 43.53mm F/6.6

Post by kaleun96 »

Picked up this lens from BMI surplus in as-is condition. No threads on either end so I mounted it in some polymorph plastic with a 30.5mm male thread providing the mold for the female threads.

Image
Image

The lens is used in the Polaroid ProPalette 7000/8000 series film recorders. They claim 4000 and 8000 line resolution respectively but I assume this is more analogous to DPI than lpmm?
I saw some prices floating around for these retailing at $12,000USD or so if I recall correctly. Nowadays you can find them for $100 in working condition.
Some more info can be found here:
http://smscommons.newschool.edu/relab/w ... 8045FS.pdf
https://www.meyerinst.com/html/polaroid/pp7000.htm

For some tests I put it up against my Scanner-Nikkor ED 7 element lens at approximately 1.3x and 2.3x. The lighting setup was identical and only slight differences in extensions between the two lenses.

These are all stacked images, about 6 or 7 images per stack for both 1.3x and 2.3x tests.

Scanner-Nikkor ED @ 1.3x
Image

Polaroid 43.53 @ 1.3x
Image

Scanner-Nikkor ED @ 1.3x [100% centre crop]
Image

Polaroid 43.53 @ 1.3 [100% centre crop]
Image

Scanner-Nikkor ED @ 1.3x [100% bottom left crop]
Image

Polaroid 43.53 @ 1.3 [100% bottom left crop]
Image


The Scanner-Nikkor seems to have better contrast and looks good in the centre but the Polaroid is much better at the edges.
I also think the Polaroid is lacking contrast primarily due to flare. As I shot the Scanner-Nikkor first, I didn't want to change the lighting setup to try and reduce the light reflecting back into the lens of the Polaroid.
An earlier test had shown me the Polaroid can do much better so I'm confident it is largely lens flare, more on this below.

Scanner-Nikkor ED @ 2.3x
Image

Polaroid 43.53 @ 2.3x
Image

Scanner-Nikkor ED @ 2.3x [100% centre crop]
Image

Polaroid 43.53 @ 2.3x [100% centre crop]
Image

Scanner-Nikkor ED @ 2.3x [100% bottom left crop]
Image

Polaroid 43.53 @ 2.3x [100% bottom left crop]
Image


Similar story to above, Nikkor is better with the contrast but poor at the corners. I believe the flare is once again a problem here for the Polaroid, especially considering how much better the corners look than the centre.
The Polaroid is sharp across the whole field of view and seems to control aberrations better than the Nikkor too.


To reinforce my point about the lens flare troubling the Polaroid, here is a 100% centre crop from an earlier 2.3x test of this lens. It's the same location as the centre crop above (note the two small divots).
Image

Aside from the ruler being much dirtier, you can notice how much better the contrast is and there's no washing out of the details!

Hopefully these tests were a bit more consistent and useful than my last ones. Feedback and questions welcome!

Lou Jost
Posts: 4581
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Good test, very good find.

I read the brochure you linked to, and maybe that resolution figure represents the number of lines across the whole image, not per inch?

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Post by kaleun96 »

Lou Jost wrote:Good test, very good find.

I read the brochure you linked to, and maybe that resolution figure represents the number of lines across the whole image, not per inch?
Thanks Lou.

I've come across a few other bits of information on the resolution:
"The Polaroid HR 6000 was purchased in 1999. It has twice the resolution (4096 lines/inch)" - The HR 6000 being a cheaper and maybe older model of the ProPalette.
http://w3.marietta.edu/~biol/macro/pfilmrec.html

That would be 160 lines per mm, not sure if that equates to line pairs per mm though. Then of course double that for this lens, which is used in both the ProPalette 7000 (4000 lines/inch) and 8000 (8000 lines/inch).

Lou Jost
Posts: 4581
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

Again, I am not sure those figures are lines per inch, even though the writer of that last document used that phrase. The Meyer Instrument documents say things like "4000-line 35mm slides" and "It produces professional-quality 35mm slides at 4,000 lines of resolution" and "Image quality: Continuously variable resolution up to a maximum of 4096 horizontal by 2732 vertical lines". The scanner's software also says "4096", without any abbreviation for the units.

For the newer version with double the resolution, the Polaroid brochure describes the image quality this way:
"Image quality:
Any resolution up to a maximum of 8192 horizontal x6702* vertical pixels. *Maximum vertical resolution (in pixels) varies withfilm/camera format: 5460 for 35mm film, 6370 for 4x5"film, and 6702 for 120 film (medium format 6x7 cm."

To me this indicates that the 8000 line resolution figure is in lines (pixels) per picture height for 6x7 cm film. That would be 4000 line pairs per picture width, not per inch.

Still the results look really good!
Last edited by Lou Jost on Thu Mar 28, 2019 6:39 pm, edited 1 time in total.

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Looks like a real gem!

What were the working distances at the magnifications you tested?

Edited to add: I assume that the f6.6 is effective aperture at its intended magnification. If it were f6.6 nominal, it could not beat the SNED lens, which I believe is f2.6 nominal.

Second edit: can you measure the exit pupil and thereby calculate the nominal aperture of the lens? Just place a ruler across the back of the lens and read the diameter of the aperture from a good distance away. If it's f6.6 nominal, it should measure ~6.6mm across.

Hmm, I think I've discovered the secret of this lens! It's aperture is the square root of its focal length. What magic! :wink:

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

I found a "8k series ProPalette Repair Manual" which has some more information. Apparently the 43.35mm lens was only used for scanning 35mm film. A 101mm f6.7 lens was used for 4x5, and a 74mm f5.5 for 6x7.

The resolution is listed as 8192x5460 for 35mm using the 43.35mm lens, so it seems the sensor is 8192 pixels wide (an 8k line sensor, hence the name of the scanner series). I'd bet the pixels are either 5um or 7um, and the sensor is either 41mm long (and mag is 1.13), or 57mm long (and mag is 1.59).

Here is the link to the manual:

https://www.manualslib.com/manual/84336 ... selected=2

Edited to add: Hmm, looking more into the theory of operation, it does not look like this scanner has a line sensor at all. I'm still reading the theory to see how it works. Interesting! And new to me!

Further edit...egg on my face...this thing is not a scanner, but a digital printer. Digital data drives a monochrome CRT, whose output is sent through a color filter, and is focused by the lens onto a film camera. The CRT is 7" diagonal (maybe 4"x6") , so magnification is very low, around 0.24.

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Post by kaleun96 »

Lou Jost wrote: To me this indicates that the 8000 line resolution figure is in lines (pixels) per picture height for 6x7 cm film. That would be 4000 line pairs per picture width, not per inch.
That makes more sense and so for 35mm film, at 36mm across, we'd be looking at ~111lpmm?
ray_parkhurst wrote:Looks like a real gem!

What were the working distances at the magnifications you tested?
Glad you think so too! For the price (~$25 + shipping) I'm pretty happy at least, cheaper than I paid for the SNED.

From memory the working distance was about 45mm at 2.3x and 55-60mm at 1.3x. Extension for those was then maybe 100mm at 2.3x and 60mm (34mm of extension plus the width of a M42-Sony E adapter) at 1.3x, not including the sensor to flange distance (18mm).
ray_parkhurst wrote: Edited to add: I assume that the f6.6 is effective aperture at its intended magnification. If it were f6.6 nominal, it could not beat the SNED lens, which I believe is f2.6 nominal.

Second edit: can you measure the exit pupil and thereby calculate the nominal aperture of the lens? Just place a ruler across the back of the lens and read the diameter of the aperture from a good distance away. If it's f6.6 nominal, it should measure ~6.6mm across.
Exit pupil diameter looks bang on 6.6mm
Definitely noticeably smaller exit pupil than the SNED, here's a photo for the curious:
Image
ray_parkhurst wrote: Edited to add: Hmm, looking more into the theory of operation, it does not look like this scanner has a line sensor at all. I'm still reading the theory to see how it works. Interesting! And new to me!

Further edit...egg on my face...this thing is not a scanner, but a digital printer. Digital data drives a monochrome CRT, whose output is sent through a color filter, and is focused by the lens onto a film camera. The CRT is 7" diagonal (maybe 4"x6") , so magnification is very low, around 0.24.
This is pretty much the same process I went through a few weeks ago when researching the lens :lol:

Once I realised it was a recorder and not a scanner I gave up on trying to quantify what I might expect from the lens, too many unknowns there for me. Interesting that it does well at almost 10x its working magnification (if it is approx. 0.24x).

I might try a test in the 0.7x range to see if it improves even more or is about the same.

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

I assume the SNED lens is from the 4000 scanner, correct?

My earlier comment about f2.6 was regarding the SNED from 8000 scanner. I'm not sure what the aperture of the 4000 is.

I'm hoping folks can start calling the SNED lenses by their models...maybe the ones from the 4000/5000/IV/V could be called SN4000, and the ones from the 8000/9000 models called SN8000?

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Post by kaleun96 »

ray_parkhurst wrote:I assume the SNED lens is from the 4000 scanner, correct?
I only know it's from the Super Coolscan/Coolscan range and is the 7-element design.

The 8000s are the 14-element design lenses, right? I mentioned mine was the 7-element design as I thought there are only two variations of this lens (7 or 14 element) but if there are more variations I can try and find out what model scanner it came from.
I'm not sure what the aperture of the 4000 is.
Robert has it listed as f2.0 nominal on his page for the SNED 7-element lens.

ray_parkhurst
Posts: 2750
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

kaleun96 wrote: This is pretty much the same process I went through a few weeks ago when researching the lens :lol:

Once I realised it was a recorder and not a scanner I gave up on trying to quantify what I might expect from the lens, too many unknowns there for me. Interesting that it does well at almost 10x its working magnification (if it is approx. 0.24x).

I might try a test in the 0.7x range to see if it improves even more or is about the same.
You may also want to try it reversed.

I always wondered how they printed my presentation slides when I presented a paper at a forum or conference. Now I can imagine them being printed in one of these Rube Goldberg contraptions/engineering masterpieces.
kaleun96 wrote:
The 8000s are the 14-element design lenses, right? I mentioned mine was the 7-element design as I thought there are only two variations of this lens (7 or 14 element) but if there are more variations I can try and find out what model scanner it came from.
No worries, I believe the lens is the same from all 4 models of scanner.

Olaf G
Posts: 11
Joined: Wed Jan 09, 2019 11:37 am
Location: Germany

Post by Olaf G »

The 7-element Scanner-Nikkor ED @ 1.3x looks strange to me. I use it on full frame @ 1.2x and it performs very well at the edges. Was it mounted in the intended way, i.e. with the paint dot towards the sensor?

Lou Jost
Posts: 4581
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

I agree with Olaf, the Scanner Nikon should have performed better, especially given the huge difference in aperture.

RDolz
Posts: 89
Joined: Mon Aug 28, 2017 9:32 am
Location: Valencia (Spain)

Post by RDolz »

I totally agree with Olaf and Lou.

But it is true that in telecentric mode, mine, gives different quality for each direction. It would be interesting to compare the quality of your Coolscan with the point of paint towards the sensor and with the point towards the object.

See http://www.photomacrography.net/forum/v ... ht=#241781
Ramón Dolz

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Post by kaleun96 »

Olaf G wrote:The 7-element Scanner-Nikkor ED @ 1.3x looks strange to me. I use it on full frame @ 1.2x and it performs very well at the edges. Was it mounted in the intended way, i.e. with the paint dot towards the sensor?
Interesting! It is mounted with the dot towards the subject as I saw that in some of Robert's photos on his website. I'll try it the other way and repeat the test.

I also have it mounted with a Raf adapter so that side of things should be ok.

kaleun96
Posts: 195
Joined: Sat Oct 20, 2018 3:47 pm
Location: Stockholm, Sweden

Post by kaleun96 »

You guys were right! Mounting the lens with the dot towards the sensor gave much better results. I haven't bothered with crops etc but if you would like to see them I can post them.

The short of it is that the SNED in this configuration performed significantly better than the Polaroid at both 1.3x and 2.3x (in the centre). I'm still seeing bad results in the corner at 2.3x so will do a further test. Something isn't right there.

I also quickly tested a newly acquired Magnagon 4/75 at 1.3x and it was close with the SNED. Since the working distance is so different between the two, I'm unsure if the differences are primarily due to lighting or the lenses. Will do more testing with it!

Post Reply Previous topicNext topic