Dye inkset ok for PDN?

I am just calibrating for using the PDN system. Just wondering if its ok to use a dye colour inkset in a epson 1280/1290.

I have tried printing with agfa copyjet transparency on my epson 3800 but it seems to be pooling the ink so i will have to use the 1280 instead but wanted to know if their are any disadvantages in using dye ink over the latest K3 ultrachrome inks.

many thanks

Hi Dave,

You will have no problems using the OEM inks for the 1280/1290 printers. As a matter of fact you may wish to experiment with some of the inexpensive dye based inks for that printer as the may work just as well.

At anyrate the Epson dye inks for the 1280/1290 are ample enough for most if not all alt. processes as they have a high UV density when used with PDN or other colorized negative schemes.

Hope this helps,

Don

Thanks Don, I intend to spend the next few days trying to get to grips with PDN as i know people have produced some wonderful results with the system, Keith Taylor for example, his platinum prints look superb.

Don Bryant said:

Hi Dave,

You will have no problems using the OEM inks for the 1280/1290 printers. As a matter of fact you may wish to experiment with some of the inexpensive dye based inks for that printer as the may work just as well.

At anyrate the Epson dye inks for the 1280/1290 are ample enough for most if not all alt. processes as they have a high UV density when used with PDN or other colorized negative schemes.

Hope this helps,

Don

Click to expand...

Just one other question is it ink side down or up when placing the inkjet negative onto the sensitised paper?

For all my previous cyanotype inkjet negs before PDN it was ink side up and i seemed to be getting what i would call good results.

thanks

Dave,

For sharpest results reverse the image when printing and then print ink side down.

I also forgot to mention that performing an alignment check on the 1280 on the actual transparency material is important to get the sharpest results. It doesn't matter for calibration but it will matter when you begin to print your images, especially with images with a lot of continous tones such as scenes with large sky areas with subtle tonal transitions.

The 1280 is know to occasionaly have some microbanfing issues with some images like the one I just mentioned. Some users have found that taping the negative substrate to a sheet of paper helped to minimize or eliminate the effect.

Don

Thanks Don.

Am having a bit of trouble and need some more help if possible.

I have followed the calibration procedures mentioned in the PDF book and have got the stage where I have printed out the tonal palette for the first time with the alternative process i am using which is cyanotype.

However when evaluating the print after dry down. Squares 81 to 101 on the tonal palette have no tone and are paper white. According to the book there should be tone in these highlight areas.

I have looked in the book and it does not seem to mention a remidy for this. Any help would be much appriciated.

Davec said:

Thanks Don.

Am having a bit of trouble and need some more help if possible.

I have followed the calibration procedures mentioned in the PDF book and have got the stage where I have printed out the tonal palette for the first time with the alternative process i am using which is cyanotype.

However when evaluating the print after dry down. Squares 81 to 101 on the tonal palette have no tone and are paper white. According to the book there should be tone in these highlight areas.

I have looked in the book and it does not seem to mention a remidy for this. Any help would be much appriciated.

Click to expand...

Can you give us some more information about how the tonal palette was printed? Having 20 white squares is a pretty large error.

Don

Hi Don

My baseline printing time to I worked out to be 29mins (31mins x 0.7) I have a small Philips UV unit!

Then I worked out the standard colour density to be in square no 130 on the second from top set of grids, this I believe is green to yellow strand. (See attached pic where I marked it in green)

I then used this finding and fed it into the tonal palette by using the following combination Red=130, Green 255 and Blue=0 . I used the paint bucket and filled the tonal palette with that colour. The tonal palette went into various shades of that colour. Printed this out onto transparency


Printed it using cyanotype, with the settings all being constant 29mins etc..etc..

And I got the result as shown in the pic attached, i.e from square 81 to 101 paper white.

I hope this helps a little.

Cheers

Don Bryant said:

Can you give us some more information about how the tonal palette was printed? Having 20 white squares is a pretty large error.

Don

Click to expand...

attached is the tonal palette before printing

Update

I picked another colour that i thought also looked paper white for the standard colour density. It was on the same line, Grid 220. I have printed this out with the cyanotype process and it seems to have restored some tone now from grid 80-90, 90-101 its hard to say but they look quite similar in tone.

Is this normal and is it ok to go ahead to scan the tone palette and start adding the readings to the spreadsheet?

thanks

Davec said:

Update

I picked another colour that i thought also looked paper white for the standard colour density. It was on the same line, Grid 220. I have printed this out with the cyanotype process and it seems to have restored some tone now from grid 80-90, 90-101 its hard to say but they look quite similar in tone.

Is this normal and is it ok to go ahead to scan the tone palette and start adding the readings to the spreadsheet?

thanks

Click to expand...

Dave,

Thanks for providing all of the scans.

First I can't imagine why your Baseline exposure would be 29 mins. That is a really really long exposure for a print made from an inkjet negative. Looking at your scan of the printed CDRP it appears that you have tone reversal. The leftmost patches in the Green/Red path should not have density if you have paper white patches farther to the right. Tonal reversal is the result of gross over exposure.

Assuming that you maintain the same exposure time I would pick the G255/R235 color or G255/R230.

The one thing you didn't scan is your lapped 31 step tablet showing your Baseline Exposure. That could be informative. I would also join Mark's PDN Yahoo group if you haven't already done so and run this problem by him and the group. Your results are a new wrinkle for me. My tests of cyanotype with several different UV light sources typically produce at least two color density paths which can be considered for use and never with any patches with density after the first white patch.

Don

Thanks Don. I am using the 21 step wedge to calculate my baseline exposure. Please see the attached image, it is at the bottom of the scan where is says stuffer ( i overlapped the stuffer with my own transparency material)

Step 1 and 2 only started to merge at a time of 41 mins in my tests , so i divided that by 0.7 to get my baseline exposure of 29 mins.

I dont know if this might make things differant but it is the Cyanotype II process by Mike Ware.

Hi Don the attached scan is for G255/R220, the stuffer step wedge is on the right this time showing the baseline.

There seems to be more highlight tone now from 80-90 on the palette but from 90-101 it seems to be very similar. I will try your suggestion of G255/235 to see if that makes any difference and scan later.

Davec said:

Hi Don the attached scan is for G255/R220, the stuffer step wedge is on the right this time showing the baseline.

There seems to be more highlight tone now from 80-90 on the palette but from 90-101 it seems to be very similar. I will try your suggestion of G255/235 to see if that makes any difference and scan later.

Click to expand...

Dave,

What kind of light source are you using? Something isn't kosher here with those long printing times.

The Ware Cyano process does print faster with higher DMAX.

Don

Its a Philips desktop home solarium, quite small with 4 tubes, see attached pic. I have it around 20 inches away from the paper to be exposed.

Don Bryant said:

Dave,

What kind of light source are you using? Something isn't kosher here with those long printing times.

The Ware Cyano process does print faster with higher DMAX.

Don

Click to expand...

Davec said:

Its a Philips desktop home solarium, quite small with 4 tubes, see attached pic. I have it around 20 inches away from the paper to be exposed.

Click to expand...

I think this is the root of your problem. I would suggest considering acquiring or building a high intensity UV lightsource.

Don

Don Bryant said:

I think this is the root of your problem. I would suggest considering acquiring or building a high intensity UV lightsource.

Don

Click to expand...

Dave,

You might consider something like this:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=120154101589&ssPageName=STRK:MEWA:IT&ih=002

Don

Don Bryant said:

I think this is the root of your problem. I would suggest considering acquiring or building a high intensity UV lightsource.

Don

Click to expand...

For Cyanotype 2 I liked how the Philips unit printed however as you say I think its time to try something else , I will move the philips half the distance (i.e 10 inches) this supposedly should reduce my time by half and see what happens.

I do have a Parker vacuum unit that I should probably try that has built in UV ( A lot more powerful) like a Nuarc, however on larger prints I seem to have been getting UV exposure fall off at the edges where it seems to become lighter, rather annoying and I think it must be down to the UV lamps.

I am now also calibrating for Platinum/Palladium so I will give that a try in the Vacuum unit, exposures are usually around 5 mins. However this unit does not allow me to change the vacuum pressure which might cause more problems like mottling!

Anyway thanks Don for your continuing help

Dave,
In Photoshop, after you colour-filled your 101 step wedge with green, what are the RGB values for block 81? This is your (new) blocking colour. I'm basing this on the 101 print out you published on Aug 31 3:29 p.m.

~m

mkochsch said:

Dave,
In Photoshop, after you colour-filled you 101 step wedge with green, what are the RGB values for block 81? This is your (new) blocking colour. I'm basing this on the 101 print out you published on Aug 31 3:29 p.m.

~m

Click to expand...

If you set the white point with the color found on patch 81 you will compress the highlights. You will be loosing 3 to 4 steps of the digital step tablet (the 21 step tablet shown at the bottom of the tonal palette not the Stouffers step tablet.)

Don

Maybe you see something that I don't. I'm seeing one of two things (barring emulsion issues). The initial blocking colour was slightly off (too dense) or possibly the exposure changed (not as likely).
The colour should be about R154, G255, B0 by my estimation. If the whole thing gets re-filled that "digital" 21-step on the bottom should read correctly (as far as highlights go). Then the curve is created from the palette. Tho' I'm not at all familiar with finer points of his system's (ok I can say it "P.D.N.") work flow but it would appear from this that the curve is calculated and applied to the negative rather than applied to the positive. OK Fine. But I had to look twice. Does the "tonal" print out really call the square 1 black and square 101 white? This is backwards (and counter-intuitive) since most artists and programmers (read: graphic artists and Photoshop programmers) use scales that count up to black (100 per cent). Did Davec just forget to "invert". Please don't take this as an attack (it's debate I'm after) and I'm not asking you to defend anyone either, but I'm baffled (more than usual) ;^)

~m

mkochsch said:

Maybe you see something that I don't. I'm seeing one of two things (barring emulsion issues). The initial blocking colour was slightly off (too dense) or possibly the exposure changed (not as likely).
The colour should be about R154, G255, B0 by my estimation. If the whole thing gets re-filled that "digital" 21-step on the bottom should read correctly (as far as highlights go). Then the curve is created from the palette. Tho' I'm not at all familiar with finer points of his system's (ok I can say it "P.D.N.") work flow but it would appear from this that the curve is calculated and applied to the negative rather than applied to the positive. OK Fine. But I had to look twice. Does the "tonal" print out really call the square 1 black and square 101 white? This is backwards (and counter-intuitive) since most artists and programmers (read: graphic artists and Photoshop programmers) use scales that count up to black (100 per cent). Did Davec just forget to "invert". Please don't take this as an attack (it's debate I'm after) and I'm not asking you to defend anyone either, but I'm baffled (more than usual) ;^)

~m

Click to expand...

Using the approach that you suggest will definitely clip the highlights in the final image when one uses Curve Calculator.

Based on this test, choosing color G255R154 is the wrong choice. And choosing the color from the tonal palette printing is the wrong approach with PDN.

After looking at the test some more it appears that that the cyanotype is a bit longer scale than would probably be obtained (possibly due to the light source spectrum and long printing time). As a result I would suggest to Dave that since he is using the 2200 printer to increase the ink density by about 10% in the Epson driver and reprint the CDRP and then reprint it to obtain a new SCD (Standard Color Density or blocking color). Then reprint the Tonal Palette again with that fill color. If that doesn't work then increase the percentage some more.

I recently had to do something similar this summer when calibrating for Kallitype on a very "fast" BL UV printer. There was almost no white patches in any of the color paths shown in the CDRP print and one of the paths had some tonal inversion. In that instance I finally had to make a 20% increase in ink deposition. Once I did that everything 'clicked'
Actually I have several CDRPs already printed for use as work arounds for such problems and it's probably not a bad idea for other PDN users if their driver allows.

Perhaps Mark's documentation should be expanded for these kind of situations but on the other hand this might lead to confusion for beginners. That's why Mark has established a private Yahoo group for PDN users (much like Gene Laughter does with his Yahoo bromoil group - established for students of his workshops).

The first printing of the Tonal Palette is printed without a curve, it serves as the source of data for generating an adjustment curve in Curve Calculator, therefore it needs to be full scale. The tonal palette is not inverted so it is a 'negative'. Not counter-intuitive if you read the documentation or RTFM as they say. Not that I'm suggesting that Dave or anyone else didn't or doesn't, that's just the way Mark set up his system.

Hope this clears up some mysteries,

Don

Hi Don

Thanks again for your continued help, i only hope that once i grasp this i will be as helpful as you have to others on this forum.

Over the last two days i put my calibration problems with Cyanotype 2 and my epson 1290 on hold while i started calibrating for Platinum and Palladium with my new epson 3800. I encounted similar problems to the cyanotype calibration and have just sent this email to Mark at PDN as i have paid for a phone consulation with him in the next few days to sort this whole thing out, rather than reword it i will just cut and paste it :

' I am calibrating for Platinum and Palladium using my Epson 3800.

I have established my standard printing time (10mins) however I seem to be getting stuck at the same problem and its regarding determination of the Standard colour density. I have attached a scan of the results from printing the color density range palette. I feel as though I have identified the right grid that ‘Just about prints paper white’ being R0 G255. My other option is R150 G255.

However when I printed out the tonal palette using that color RO G255, from grids 91-101 are all paper white (see attached scan.) I tried R15 G255 and grids 95-101 are paper white.

I then tried my other option of R150 G255 as a standard color density, when I printed using this color in the tonal palette it goes paper white at 94. I moved right onto grid R195 G255 and it goes paper white at 96-101. Is it just a case of moving on to the right I.e R225-G255.

Having done some density readings using my transmission densitometer on the negative of the tonal palette,R195-G255 being the chosen color, grid 101 is giving the same reading as the corresponding grid on the color density palette. ( a reading of .25) that’s fine, however what concerns me is grid 97,98,99, on the tonal palette transparency are all giving the same reading of 0.23 furthermore ,92,93 are also the same reading of 0.22. Going down the grids there seems to be others grids that are exhibiting this 47,48,49 are all giving a reading of 0.10.

Is this normal, should this be happening?

I am using Agfa Copyjet as the transparency, Epson 3800 (Premium Glossy Settings@2400dpi) Parker Vacuum UV exposure Unit with mercury vapor lamps. '


Any ideas Don? At the moment for me at least the calibration process has been quite challenging and rather a hit and miss affair. AlthoughI have been having a bit of fun using keriks method of using the 3 blacks in the epson 3800 and printing the negs in rgb and applying a rudementary curve to the image. This is not as 'precise' as i would like the whole process to be as i really want to understand the whole concept of PDN as i know some people have had great results.

Don Bryant said:

Using the approach that you suggest will definitely clip the highlights in the final image when one uses Curve Calculator.

Based on this test, choosing color G255R154 is the wrong choice. And choosing the color from the tonal palette printing is the wrong approach with PDN.

After looking at the test some more it appears that that the cyanotype is a bit longer scale than would probably be obtained (possibly due to the light source spectrum and long printing time). As a result I would suggest to Dave that since he is using the 2200 printer to increase the ink density by about 10% in the Epson driver and reprint the CDRP and then reprint it to obtain a new SCD (Standard Color Density or blocking color). Then reprint the Tonal Palette again with that fill color. If that doesn't work then increase the percentage some more.

I recently had to do something similar this summer when calibrating for Kallitype on a very "fast" BL UV printer. There was almost no white patches in any of the color paths shown in the CDRP print and one of the paths had some tonal inversion. In that instance I finally had to make a 20% increase in ink deposition. Once I did that everything 'clicked'
Actually I have several CDRPs already printed for use as work arounds for such problems and it's probably not a bad idea for other PDN users if their driver allows.

Perhaps Mark's documentation should be expanded for these kind of situations but on the other hand this might lead to confusion for beginners. That's why Mark has established a private Yahoo group for PDN users (much like Gene Laughter does with his Yahoo bromoil group - established for students of his workshops).

The first printing of the Tonal Palette is printed without a curve, it serves as the source of data for generating an adjustment curve in Curve Calculator, therefore it needs to be full scale. The tonal palette is not inverted so it is a 'negative'. Not counter-intuitive if you read the documentation or RTFM as they say. Not that I'm suggesting that Dave or anyone else didn't or doesn't, that's just the way Mark set up his system.

Hope this clears up some mysteries,

Don

Click to expand...

Dave,
The problem with the CDRP is it only shows you a narrow band of available colour and it doesn't show you what's going on "around" the colour that you initially choose. Try downloading and printing the RNP-HSL Array or look at some of the examples post here at hybrid in the digital negatives forum. The HSL-Array shows you the big picture and gives you many more options for choosing a colour -- not a PDN colour though. Often the printer driver is laying down ink in an irregular and unpredictable pattern you may "think" you have a square that's "just white" but in reality you're in a "density pool" and the path from your chosen starting colour to zero density is a mirage of sorts.

~m

My bad

I need to correct my second last post where I wrote 154,255,0 as Dave's new blocking colour....it's actually 154,255,48 I forgot to read the blue value...my bad. But I agree Don that this colour may not work in the Curve Calc. spreadsheet.