Jump to content

PRNU Help


ralphmalph
 Share

Recommended Posts

Just received my LCD after almost a year waiting.  For some reason the PRNU will not become uniform, I see little waves within the printed grey ( see attached)

I tried with the disk @ 9 pixels which did not work so I started from scratch going from 3/5/7/9 pixels. The grey looke good but I see little waves in it.

Any help appreciated.

Link to comment
Share on other sites

Good day sir. You may try doing prnu from scratch. Rename or save on a different folder the current prnub, prnug, prnur.bmp files under parametres folder, so you can easily revert back incase u messed up with the prnu procedure. Look for the PrnuBase24.bmp file under parameters folder ( this is the base image. Then copy and paste three times on the parametres folder rename the 3 bmp files as PrnuB.bmp, PrnuG.bmp and PrnuR.bmp. Then try to do the uniformity procedure again using the three base files. Hope thus helps

Link to comment
Share on other sites

Thanks Nyoman. I don't know if this is the same but i deleted the 3 prnu files that were in the parameters folder then exited out of the software. When i reopened the software 3 " virgin" prnu files were created. I then ran the prnu at 3/5/7/9 pixels. Is this basically the same procedure you are talking about?

I am using glossy paper does the paper make any difference?

Thanks

Link to comment
Share on other sites

Because you changed the LCD and now the lenses are probably not in perfect focus compared to the original LCD ;)

It is never a big difference, but sometimes when changing the LCD, the difference is bigger and lenses then to be refocused  :-/

One colleague technician solved the same problem only after he refocused the lens, and before that he tried everything else including starting from scratch  :o

I had a same issue one time, but starting PRNU from scratch solved that, but probably I was lucky and I did not need to refocus the lens  ;)

Looking now at the technical manual, the procedure for changing the LCD (X9 sub-assembly with LCD and EPSON LCD driver board) is mentioning the checking off the sharpness and adjusting it:

•Check the adjustment of the LCD turret Zero sensor:See chapter 10-•Adjust the flicker: See chapter 10-10

•Check the sharpness, adjust if necessary: see chapter 10-

•Adjust the mechanical center of the reference lens: see chapter 10-9

•Adjust the polarizer and carry out an LUT: see chapter 10-11 and 10-12

•Redo the PRNU: see chapter 10-15

•Adjust the X9 (piezos): see chapter 10-15

•Carry out a complete centering operation: see chapter 10-16

•Balance the filters: see chapter 10-17

•Print a final check order using the "SU/settings/check" key

Link to comment
Share on other sites

I had once experienced your problem and adjusted the lens. In my experience i did trial and error, first i tried rotating the lens approx 30degree to the right then do prnu. then 30deg to the left  then do prnu then another 30 deg and so on ...If i think i was able to get it on the third try. Just mark the initial position so you can easily get back to the original position. A very tiring procedure.

Btw you can  focus only the lens being used by when printing the prnu.

As PScaro said in most cases it can be solved by doing PRNU from scratch but if all else failed try the refocusing

Link to comment
Share on other sites

our machine here was 1500, as for me I dont turn the 3 black pins, I just rotated the lens clockwise or counterclockwise a few degrees just like focusing a manual camera lens.

You need to loosen the small plastic white screw holding the lens in place.

Link to comment
Share on other sites

If there is a problem just with a plane sharpness, then adjust only lens position (up or down by screwing or unscrewing the lens), but if there is also a problem with some corner sharpness, then there is a need to adjust also the X and Y screws, but do not touch pivot screw.

I am attaching the part off the service manual where this is explained, but also there were two Info Expresses for DKS3 which explain the sharpening off the lenses in more detail way, but this can be implemented on DKS2 also.

Long time ago I design 3 test images for each DKS2 format which helped me a lot when doing this sharpness adjustment on DKS2, and this 3 tests are design for each format/lens and they have LCD resolution 1:1 (aproximatly).

If the sharpness is not to far away the test with thin circles is very good, because visually you can judge the sharpness just by looking at the color off the circles (green or magenta) in what direction it is necessary to screw/unscrew the lens or which angle screw have to be adjusted. But be careful with angle screws as they are reversed to the image position ;)

My sharpness test designs: http://jumbo-old.iskon.hr/dl/190c6xdudnpjw6eg/

Link to comment
Share on other sites

Guys - don't fool him.

DON'T touch the lenses unless you really have to - but this will be possible to judge later one when you set all optic and print Piezo picture.

ralphmalph - from the early beginning you should REMOVE filters from the lamp. all preliminary settings u need to do without filters.

The basic - VERY BASIC - operation on the new LCD is to perform flicker at the early beginning As soon as U switch the machine.

than U can check zero LCD and cross centering.

After that U set polariser, than LUT (LCD response), LCD gain as well if needed, probably will be needed....

to perform PRNU from scratch you NEED work filters: RED 40+ NEUTRAL  40 (Red always on the top)

than you calibrate paper - ALWAYS GLOSSY IS USED FOR CALIBRATION.

then you can start PRNU form scratch.

PRNU setting according to the generic KIS procedure consist of:

a) 5 scans at pixel_size=3

B) 1 scan at pixel_size=7 (with bitmap correction and fault masking)

c) Filter choice and paper calibration with new filter (this could be a problem since machine can request filters that you probably won't have)

d) 1 scan at pixel_size=7

Instead 2 scans at p=3 Ucan do 1 scan at p=5 but need to correct the picture already (mask dirt and dust)

scanning with p=9 is a hardcore - rather used for real problems with uniformity.

Then you can check piezo - generally should be OK.

If you have magenta or green colors on the corners U can do up to 3 scans ant it will get back to normal.

I am really scared that you are proposing the calibration greenhorn to touch the lenses. NOONE proposed him to mark somehow initial position of the lens in order to set it back on the factory position when he screws everything off.

And this can be done very easy since tolerance on the optics is rather narrow. ??)(naughty)

Link to comment
Share on other sites

Hi Sir relaxia, we are just sharing our experience with this kind of problem which worked in our case. Btw i did advised him mark the initial position of the lens ( refer to july 14 2:33 post) as i know the effect it may cause if he messed up with the adjustment.

Link to comment
Share on other sites

I am not touching the lens. My tech is calling France to find a solution. I will send this LCD bac before I touch the lens. I have had 3 LCD's in the last year and this is the only one to do this. ( this is number 4)

Focusing the my 1670 was a nightmare no way I am touching the Lens on the 1550, They were in perfect focus until this LCD came

Link to comment
Share on other sites

nyohan I dont think you are following me on this post, first let me start off by saying I have run a PRNU 4x times before with 0 problems on 3 other LCD's

1. I have done the uniformity with the supplied PRNU files = "got the canvas effect"

2. I ran PRNU with Virgin files ( deleted the PRNU files exited out of software") ran the New files at 3/5/7/9 pixles = "got the canvas effect"

3. Virgin files again ran 3x @ 3 and 2x @9 as the help file said too when starting from scratch= "got the canvas effect"

4.Virgin files again 5 scans at pixel_size=3  2 scan at pixel_size=7 as posted above="got the canvas effect"

Started with flicker, set poliriser,LCD response and Gain.  I dont have the service filters so I can't do all the steps in the above post

"Maybe you could ask france to do the LCD uniformity for you, ask for a CD with the prnu files." Not going to help but thanks for the suggestion

Link to comment
Share on other sites

I did'nt know that kis provided you the prnu files, there are cases that kis dont provide those prnu files ( in cd) especially when you order just the lcd alone (without the piezo plate and 2 pcb). This is the reason i recommended you to ask for the prnu cd.  Kindly disregard my post if not applicable

Link to comment
Share on other sites

Latest from France, I will try this in the a.m.

To be clear, to get the PRNU base, you need to quit the application, to delete under Parameters folder the 3 files :

PRNU G.BMP

PRNU B.BMP

PRNU R.BMP

Then you restart the application, and it will create 3 new files with PRNU from base.

Then, the paper setup must be OK on your machine.

Then you go to super user mode, settings menu and PRNU tab.

You print the target and normally, you must get a gray (not pink) picture with a scale on the side.

Then, the 1st calculation parameters will be : 17 and 3.

Then you reprint, between each calculation and you scan carefully without dust on the patterns.

Parameters will be :

2nd calculation : 17 ; 3

3rd calculation : 13 ; 3

4th calculation : 13 ; 3

5th calculation : 13 ; 6

6th calculation : 13 ; 9

Link to comment
Share on other sites

This is from-scratch-procedure with modification - parameter called Nombre_pixel_reperage by default set at 13 can be increased when you have sharpness problems. max value 19.

This is used rarely - sometimes soft has problem to calculate that... but I do agree with that attempt and these parameters.

Reminder - before calculation at pixel size 6 and 9 scan must must be totally reviewed and all dust and scratches must be masked.

No info about filters, so I am repeating - according the factory procedure:

1. put R40+N40 filters in the head

2. calibrate the paper with these filters before U start PRNU - in order to get exposure times over 100-120 ms for each color.

3. Before last calculation U remove work filters and U check what machine wants in tab "Filters" - U put desired values, calibrate paper again.

4. Last calculation

5. Piezo calculations - recommended 3 calculations.

GO AHEAD ralph :) Good luck.

Link to comment
Share on other sites

www.leefilters.com  - LEE Technical filters. Originally used by KIS.

Should be available in good photographic stores. DOn't know how they are sold in shops but originally it should be a sheet 150x150mm allowing to cut out 9 rectangles for DKS needs.

I have quite nice range but you are too far to get this fast...and you can need others after PRNU so I really suggest you to take a tour around the city and find source.

According to the LEE map from their site you have 2 official dealers in your state.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...