Wolverine-Hawkeye Telecine

[Stan Jelavic]

My excel had an error. Here is your picture Bruce with my measurements added. It is pretty close.



I picked the blue series just to see if there is anything here locally but David's lens looks better and we have some good tests from him so it is definitely less risk.
By softer image I believe David meant out of focus a bit (David, please confirm). If overall image is out of focus that would mean that the lens is at the limit of its optical capabilities.
If it is less contrast then yes it would have to do with the lens flare because the angle of light hitting the inside of the lens housing is reflected onto the sensor.
Still working on test clip for the new and old cameras side by side.

[David Brown]

Stan said"By softer image I believe David meant out of focus a bit (David, please confirm). If overall image is out of focus that would mean that the lens is at the limit of its optical capabilities."

Yes, the soft image was overall. It could not be focused. This was when I tried to duplicate the same magnification using the 37UX that I had using the 72BUC. The lens was out of range.

Kamel
The 16mm lens at cnaico (ACH1656MAC)

https://www.aico-lens.com/product/16...ns-ach1622mac/

select the contact us today

https://aicolens.wufoo.com/forms/q16qv3uv193g205/


Sales will reply

Dear Mr. Brown,
So pleased to receive your inquiry.
May I know your company name and website?
Could you please advise the detailed requirements about lens? Then I will check if the ACH1656MAC is suitable or recommend other lenses.

I have no website and no company. Are sales permitted to individuals? (YES)
The lens : 2/3" 16mm f5.6 10mp M12 Macro with IR cut filter.
image source diagonal is 7mm and a 2/3" sensor diagonal (11mm).The image can be cropped in.

They replied saying the 16mm or the 12mm would be a good choice.

Then I had to ask "how do I purchase the lens!"
On Sun, Feb 9, 2020 at 8:30 PM sale3@aico-lens.com <sale3@aico-lens.com> wrote:
Dear Mr. Brown,
The order can be placed directly here. (sale3@aico-lens.com)
The total cost for 1piece ACH1656MAC is $115($85+$30 by Fedex/UPS to United States).
The lead time of 1piece ACH1656MAC f5.6 with IR cut filter will be 3 weeks after receiving payment.
Paypal is acceptable for sample order, pay by Paypal: zhengmiaomiao2011@163.com.
Best regards
Amanda
Aico Electronics Limited
Room 1204, Kairuijinzuo, No 1505 kejiguan Street, Binjiang District, Hangzhou, Zhejiang, China
Tel: 86-571-86619019
Fax:86-571-81069179
Email:sale3@aico-lens.com
Skype:amandazou123456
Tel:18324459426
Wechat:aico18324459426
www.aico-lens.com

[Stan Jelavic]

Thank you David for the info. I will also order one of those.

I did the scan comparison between the BUC02 (left side) and 37ux226 (right side):
https://drive.google.com/file/d/1bCu...ew?usp=sharing
You be the judge.

A few notes:
- the optical magnification on the 37ux226 is set such that it gives 1280x1024 resolution (same as BUC02).
- 37ux226 is 64 bit (12 bits per channel)
- white balance on 37ux226 is set via auto mode one push.

The videos are raw - no post.

[Bruce Davis]

Hi Stan, appreciate the hard work you have put into this (which lens did you use).

The main thing that struck me was the difference in colour and contrast with the first 100 frames, not sure about the so called dynamic range, the blown out areas in the main, remain blown out, but this varies slightly with variations in exposure between the two videos. I am happy to see that the dominating "blue tint" in some scenes with the DFM 72BUC02-ML are gone with the DFM 37UX226-ML.

The DFM 72BUC02-ML camera does have a leaning towards a more blue tint in some scenes (repeating myself sorry) but there does not seem to be much difference in colour with the close up flower shots/frames. Would I be right in saying that the DFM 37UX226-ML camera takes a little longer to adjust exposure between large exposure differences in scenes, or is it because you were using "one push".

Stan were your settings in IC capture more or less the same (I am guessing yes).

Thanks very much Stan, I very much enjoyed viewing and comparing the video.



I emailed Edmund optics re-the 83-943 - 17.5mm FL f/5.6 lens, their reply was that this lens would only be suitable up to a 1/2" sensor size, therefore one might see vignette around the corners, the WD issue (150mm min) can be solved by increasing the distance between sensor and the lens.

Quote "I understand your application is a high resolution one.
Given the limitations (on sensor size), I do not recommend 83-943 as a solution.
Instead, my recommendation would be :-
https://www.edmundoptics.com.sg/p/16...th-lens/41866/
16mm, f/2.8 UCi Series Fixed Focal Length Lens
Stock #33-884
Can easily resolve 200lp/mm
Supports your sensor size
However it is a bit large. S mount but front diameter is 25mm"

I know this is not relevant for us because there is no way this lens would fit, but have posted just as a point of interest (this is obviously a much better and more expensive lens).

Sticking with David's cnAICO ACH1656MAC lens seems like the best option.

Regards - Bruce

[Stan Jelavic]

Good observations Bruce.
The lens used is our standard 12mm from TIS.
I agree with your evaluation:
- ux226 has better colors and better contrast in some scenes
- very similar images in most scenes in this video but no blue tinge
- ux226 has faster frame rates
- ux226 has better resolution for the same optical setup

With the 16mm we should get even better results.
I ran the video on my 24 inch monitor. On the laptop the difference is less noticeable.
Regards
Stan

[David Brown]

Thanks for that test Stan. I know that takes a lot of time to prepare the side by side comparison, especially when changing cameras!

I say it was a draw. The resolution was equal. The lens is a limiting factor for the 37UX226. I didn't keep score on the exposure or color. Some scenes looked better on one side, then the other. Pretty close and I think either one could be tweaked to improve the differences. I can't explain the differences in the first 100 frames. That would clearly go to the 37UX226.

I assume you had the HDR switched on. There are a variety of lighting changes on your film. The auto settings worked. Auto exposure can fail thru a fade-in, but it looked ok. That's good because as much as we believe we want to do scene by scene settings manually, it's tedious!

Just don't forget the 16mm lens paired with the -BUOC02 will need to be physically adjusted if scanning S8, then R8.

[Stan Jelavic]

Hi David, agree that the two videos are close but still on my monitor I could see on three of four instances where the ux226 has better contrast, In one case with the castle walls with the clock the BUC02 had better contrast.
The first 100 frames had different white balance and looked too dark to me. So I hit the white balance button once.
The resolution in both videos is the same. I used 10mm spacers for the ux226 which puts the sensor at the higher level than the BOC02 and consequently lower optical magnification but same pixel resolution. The idea was just to test the colors, contrast and dynamic range and did not want resolution to enter the equation.
One possibility is that since the UX226 sits higher the lens flare is lower and hence better contrast. Not sure. But hopefully the 16mm will be even better.
Not too concerned about the 16mm and BUC02 because do not plan to use that lens with it, but it is a valid input for other people. It is also one of the advantages of the UX226 i.e. should work ok in both configurations without cumbersome changes.

[Stan Jelavic]

Here are some examples where ux226 shows better performance:

Less lens flare, better contrast:

No red shadows:

Better resolution - check the wall details:

Less blue tinge:


UX226 can run at 2FPS scan rate with very dark scenes without smear. BUC02 needs to be slowed down to 1FPS
or run at higher gain.

[Kamel Ikhlef]

Hi Stan, Very nice day today: I just received the v12 kit at last.
Thank you, thank you very much.
I haven't tried it yet, but I trust you did a great job, very neat.
I will be able to test on the same basis as the forum team.
Thanks again.

[Stan Jelavic]

That is good news Kamel. Took a little while.
Happy testing
Stan

[Stan Jelavic]

Ordered 16mm lens from aico-lens.com
It will take some time before it gets here. In the meantime have some ideas on how to improve the takeup tension by counting the frames.
The takeup tension can be set to higher after power up. Then a frame counter can be implemented that controls the tension so that the capstan is always engaged but the pull is not as strong after first 20 feet of film.
The capstan operation can be changed to turn off the stepper if no motion is detected and turn the sonalert on.
Since the takeup switch is not used in run mode, it can be used to turn this feature on.
the switch can be left in the off position in units that do not have capstan,

[Bruce Davis]

Hi Stan, (just some thoughts) I had a similar idea, with counting of frames and varying the take up pulses as the take up reel spooled film diameter increased. While the diameter is small a longer pulse would be required and as diameter increases pulses can be shorter.

I was not aware that you could control the torque of the take up reel/motor so that there would be a constant tension on the film take up. As the start diameter is smaller than the end diameter, in this case the torque on the take up real would need to be LESS at the beginning and MORE as the reel spooled diameter increases.

For this to work reasonably well/accurately one would need to calculate how many frames are taken up per revolution as the diameter increases (per revolution) and the torque that needs to be applied to the reel for the tension on the film take up to remain approximately the same through out the changing diameter.

All this would be great, but what happens when one stops capturing images part way through a reel and the Hawkeye is turned off, for what ever reason. Does the chip retain the frame count (extra memory chip)? Can the number of frames be reset if one decides to restart a reel?

Great ideas Stan - thanks

Regards - Bruce

[Stan Jelavic]

Thanks Bruce. Good questions.
Yes, the torque can be controlled by varying the pulse width. In V12 the width can be controlled very precisely. In previous versions you cannot go below a certain minimum torque which is pretty high.
Based on some of my tests and tests published by David it may be required to increase the initial torque for the first 20 feet of film and then leave it at the value that we currently have with V12.
If you stop in the middle of the reel the first 20 feet may have some jitter depending on how the film is routed.
Another alternative is to turn capstan off for 20 feet by turning the takeup switch off. The takeup off condition can be detected and that can turn off this new takeup feature and the torque would run low.
The purpose of this feature is to reduce the jitter but also to keep the capstan engaged for the 1st 20 feet so the two issues are tied together and that is why the takeup switch can be used to control both.

[Bruce Davis]

Hi Stan, Quote "tests published by David it may be required to increase the initial torque for the first 20 feet of film" not sure if I am misunderstanding the terminology being used here.

If I remember correctly David stated that there was not enough take up travel (rotational distance) on the take up reel at its minimum diameter (the dia when there is very little film on the reel). This to me describes that not enough "on TIME" is provided to the take up motor, this is totally different to my understanding of "more torque" (more tensional pull on the film).

Regarding the capstan I am not clear on your description. As long as the film is rotating the capstan motor is passive and does nothing (at least on my Hawkeye) so I do not understand how this ties in with the take up. Only if the film stops (a jam or broken sprocket holes) will the capstan activate and try to pull the film through.

Regards - Bruce

[Stan Jelavic]

Hi Bruce,
The motor torque is proportional to the on-time. So by controlling the on-time duration we control the torque. If the torque is to small the film will have slack and will fall off the capstan which will cause capstan to time out. So that is the connection. Too much torque will cause jitter.
Since the takeup switch is not used in run mode we can use it to control both the capstan and takeup torque:
- switch off - capstan not used (for units without capstan) - takeup torque weak
- switch on - capstan active - torque strong for 20 feet then weak

If switch is on and then mid reel we turn the unit off then start later. The switch can be left off so that the torque is weak which lower the jitter. But the capstan will be off so after a while (after 20 feet of film has gone through) the switch can be turned on. It is a compromise.
Hope this helps.
Regards
Stan