Which Pigment Ink for Epson 1500W?

I continued some of the fading tests and got some pretty quick results in the bright spring sun at this time.
I used the EV6 inks by precisioncolors , the cyan and black, they fade faster than the Surelab/Drylab inks and just give quicker results than the Surelab/Drylab inks. I'm testing these cyan and black PC inks on various glossy papers to see their impact on the results - how much a different paper makes a difference on the fading of this ink. This was not a test of various inks against each other but a test of some papers.
I'm using

-------------------------------------- Cyan---- Delta ------- Black --- Delta

Tecco Photo Glossy 240 gr RC -----85----- no change ---- 41 - 50 --- 9

HP Premium Photo Glossy RC----- 84 ---- no change----- 38 - 46 --- 8

LS 180 ---CC ----------------------- 86 - 95 --- 9 ----------- 37 - 52 --- 15

Labelheaven 180 ---CC ----------- 88 - 94 --- 6 ----------- 34 - 50 --- 16

Sihl/Aldi ----------- RC ----------- 90 - 99 --- 9 ----------- 36 - 49 --- 13

Netbit/Aldi ------- RC ------------ 93 - 97 --- 4 ----------- 33 - 47 --- 14

CC - cast coated - cheaper glossy papers

RC - resin coated - the typical premium photo papers

I'm measuring the luminance and difference of the exposed and unexposed patches via the histogram function in PaintShop Pro, printed on the same sheet of paper, the patches are out in the sun since 3 days. I'm not measuring any color shifts this way but the gray equivalent lightness changes.
There are quite some differences between all these papers, the blacks either stay black up to this time or get lighter and turn brown which makes an unpleasant look overall and create a suntan look to the fading images.
The best performing papers in this test are the HP Premium Plus Photo Glossy and the Tecco Photo Glossy 240gr, the other papers - both CC and RC type - are quite similar in their performance, actually as the ink performs and fades on these papers.

The result of aldi/sihl and netbit is a surprise to me, more precisely their similarity to cast coated paper.
Because my eye-opener regarding the importance of the paper was a fading test with aldi/sihl vs. some obviously much worse cast coated paper:
http://www.printerknowledge.com/threads/refilling-canon-with-oem-ink.10712/page-2#post-90875
Your test is a better result for your two CC papers ... or a worse result for aldi/sihl - just as you like to see

LS180 is a paper with a trade name Logic-Seek by tintendienst.de, it feels and looks pretty much like the Labelheaven paperr, and is as well pretty low-cost. The different performance of the papers is most likely a question about the coating - the top most layer creating the gloss, and that's a detail no paper company is specifying in any detail, and I cannot assess whether the coating, the chemistry would have an inmpact on the pricing of such papers.
And I'm testing with two colors only - o.k. - black and cyan have the most visible impact on the image degradation by fading. I'll keep the samples out in the sun for a few more days to let the luminance changes grow further.

the summerlike weather let me continue the fading tests from above - the histograms of the (un)exposed patches spread further, the trend from above remains :

the cast coated papers Labelheaven and Logic-Seek let the same ink fade fastest compared
to the RC papers - Aldi/Sihl and Netbit/Aldi which perform slightly better, just slightly better than the
CC papers,
the RC papers Tecco Glossy and HP Premium Plus Glossy let the inks perform visibly better - the inks fade slower . Since there are so many different papers on the market it is impossible to transfer these findings to any other paper or brand, and I don't know either whether a higher price would give you better fade performance in all cases.

Some Inkthrift CL inks were as well in the sun, they perform equivalent to the Surelab/Drylab Epson/Fuji inks, this is a class of inks of its own with a much better fade performance than all other inks tested so far.
So I have as a user lots of options - a wide bandwidth of performance and price from the cheapest fast fade combination to the best in class.

Let me show you some details of the fading tests, the black patch, like the cyan patches are in two parts - one exposed to the sun, the other part covered up as a reference for the changes

The histogram looks like this

There a two spikes - for the luminance of the two parts of the patch, the luminance difference of the peaks allows a direct comparison to other patches which were exposed to the sun for the same time.

Each pixel in the scanned patch image has an RGB value - or a recalcuated Lab value.
Displaying these pixels in the Lab space looks quite different

There are two blobs of pixels - one of the darker unexposed section, and another blob somewhat higher and shifting away from the L-axis, the gray axis, this indicates that the color shifts from
neutral into the +b-Axis which makes the brown tone of the exposed part of the patch.

martin0reg said:

I don't understand, why they sell their best (and most expensive) dye ink set only for epson 6 channel printers - not for K3 printers like R2880, 3000, 3880, ... For the K3 models there is only the less stable "inkthrift pro", why?

Click to expand...

This is a very good question and the one I actually asked Jon Cone in January this year. He said something very interesting (which will be very useful to anyone who dilutes these stable dye inks out there to get lighter colors). He said that they had this idea of expanding their 6 ink set to 8 ink printers long time ago and the reason why they didn't go ahead with it was because diluted blacks were not fade resistant. Since K3 printers mix these LK and LLK inks into pretty much every color, using unstable light K's causes very quick visible fading. I asked him if this applies to diluting C and M as well, he said yes, to some degree, but LC and LM do not affect the overall visible fading as much as diluted blacks do. So, with present 'stable dye' tech 6 color ink sets is all we can have. It should also mean that if one excludes LC and LM completely he should get a more fade resistant result.

Thank you for this info, what Jon Cone says seems to be confirmed by my own B&W ink set.
I made six shades of grey, according to Paul Roark, with dilution ratios going down to 2% for the Y channel.
Normally this is intended for pigment ink - I made a set out of fuji drylab ink, which is one of most stable dye inks... Result: these prints fade to sepia, much more than a B&W print would fade, if printed with the whole color set of fuji DL
http://www.printerknowledge.com/threads/dye-or-pigment-ink-for-3880-question.9896/page-4#post-86168
My explanation: the full color set contains 2 light color dilutions of 30%... the six shades of grey contain 3 much higher dilutions of 9%, 6% and 2%...
Anyway I thought that a manufacturer of ink could do something to avoid this effect

Thanks to Martin0reg I could get some of this fantastic DL ink for my Epson R2000, which does not need LM or LC to obtain a smooth visibly grainless print quality, due to smaller ink droplets (1.5 pl), like Canon's approach to mimic LM and LC with much smaller droplets from the concentrated M and C. Not only this method saves ink, but because the concentrated colours are used, you will get better fade resistance. Now there are Epson printers without the LM and LC cartridges, but most of them don't have the necessary photo black to obtain the maximum contrast for visually superior photo printing. Only the Epson R800/R1800/R1900/R2000 and P-400 do have the PB and these printers are real photo printers using the CMYK or CMYKOR or CMYKRB printing method. Also the usage of the gloss optimizer will probably enhance the fade resistance by covering the dye droplets with a clear coating sealing them from the influence of gases.

I used aftermarket pigment inks in my Epson R2000 and after several month just cleaning the print head every week, because of the tenacious clogs (especially orange and red), I gave up these pigment inks and turned to dye ink and last year I already had a test with the DL ink in my Epson R2000 and was amazed by the fade resistance. The gloss of those pigment inks in my Epson R2000 was suboptimal, even my Epson Pro 3880 using Inktec Powerchrome K3 ink will yield much better gloss on glossy and pearl papers. Nothing can beat the original Epson R2000 pigment Ultrachrome Hi-Gloss 2 inks in glossiness, unless you use dye ink, which still beat any dye ink, including Epson's own Ultrachrome Hi-Gloss inks.

Martin asked my to try the mix the optimal Blue ink for the Epson R800/R1800 printers, so I will post the results later on to make your own superior fade resistant dye ink mix for these Epson photo printers, which are more economical to run compared to the CcMmYK printers, because there is no LM and LC ink in them, which are heavily used in the lighter area's.

During my fade tests last year I tested as well the effect of a gloss optimizer and can confirm that the Glop does indeed provides a level of protection - probably against UV and ozone in combination - test patches were fading visibly slower with a Glop overprint. I did a Glop overprint of dye ink patches with a WF2010W in a second print pass which I decommissioned shortly after. The Epson Glop, or the Canon Chroma optimizer are otherwise only running on pigment ink printers.
I did some tests recently - I ran the L800 as a genuine 6 color photo printer without the light colors but a clear cleaner in these ink channels only. You can get such ink set profiled, but the gamut is somewhat smaller - about 10% volumewise which is barely visible on printouts, and only in direct comparison. You would not get any decent printout when printing without a icm-profile, it looks pretty pale. The driver internal transition from regular to light inks is not accessible from the outside so there is nothing in the parameters which can be changed.
I ran as well the regular cyan and magenta inks in the light ink channels, that combination can be as well
profiled, the overall volume of the gamut is slightly bigger compared to the normal operating mode. Running this ink set without a profile gives totally oversaturated printouts and is not usable as such. I cannot see a difference in lighter and less saturated color areas compared to prints with regular light inks - the L800 droplets are apparently small enough. Since this mode is pretty far off from the regular operating mode the small numbers of color patches with ColorMunki for profiling may not be sufficient to catch the excessive ink in the lighter areas, I'm running a sheet with 968 patches with Profile maker which creates working profiles wihtout error message or other dropouts.

I will try the DL set in a r800 as soon as I get your formulation for blue, pharmacist..
But while I agree with your opinion that original "highgloss" ink from epson is the best pigment ink for gloss - I have some 3rd party ink in use which seems to be quite a good alternative, it's "photochrome r8.1".
Sadly it is not available anymore from the german seller "farbenwerk".
Of the other set "photochrome r20" there seem to be a rest of full sets, not cheap (8x 100ml 140€ actually.."while stocks last") but IMHO worth the price ...if it prints out as well as the set for my r800...

PS: anyway I have to underline, for really shiny gloss:

pharmacist said:

...Nothing can beat the original Epson R2000 pigment Ultrachrome Hi-Gloss 2 inks in glossiness, unless you use dye ink, which still beat any dye pigment ink, including Epson's own Ultrachrome Hi-Gloss inks....

Click to expand...