Epson 9900 Printing Size Accuracy

[JasonHopkins]

Hi L-L Forum!

I have had an Epson 9900 for about 6 months.

Recently I noticed that when I print on different media at say 50cm wide by 100cm length, the print comes out at 50cm wide but 99cm length.

This happens on several media types, over say a metre in length I am losing between 5mm and 10mm. As if the paper is slipping.

Any ideas as to whether this is normal on printers?

Am I missing a setting such as paper feed adjustment or roll tension?

ANY help or suggestions would be appreciated.

Many thanks,

Jason


PS tried to join here several times with no luck since I got an Epson 9900 6 months ago. Had several problems with it that have been resolved.

[Doombrain]

what software do you use to print?

[JasonHopkins]

what software do you use to print?

I am using the latest version of Qimage.

It is a problem that I have just noticed as I print Seperations for someone and they have come out at different lengths.

I have not had a chance to print from a different program.

Surely it should print at the exact size?

[Ernst Dinkla]

I am using the latest version of Qimage.

It is a problem that I have just noticed as I print Seperations for someone and they have come out at different lengths.

I have not had a chance to print from a different program.

Surely it should print at the exact size?

It is very unlikely that Qimage is the problem. Sounds more like a paper transport issue if the lengths are changing for what should be exactly the same print run: 4 separations done with exactly the same Qimage and driver settings. Film used too slippery?

What happens: is the image scaled in one direction or is the image cropped to 99 cm, in other words missing 5 mm at both ends? Often if there is no interesting content at the ends one may not see the difference between the two possibilities. Qimage doesn't know a resampling to another aspect ratio, it only has a crop function feature. Is the maximum print size displayed above the Qimage preview 50 x 100 cm ?



met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[jule]

One way to check is to make a file with a big rectangle with the line about .5cm (quarter of an inch or so) thick. Print it directly through Photoshop, and then repeat through Qimage. Get an accurate steel rule or tape and measure exactly each print. Compare the actual measurements with the file dimensions. You would then be able to determine whether the problem lies in the paper feed or the programme itself.

To be even more accurate and specific, you could also compare a file with the actiual dimensions of the rectangle to be printed, with a smaller sized rectangle interpolated up through Qimage. If there is any difference in the size of the two printed rectangles you know the interpolation in Qimage is affecting the image size, and you may need to examine your settings more carefully.

You can also adjust the paper feed on the main control panel, but it does take a lot of tweeking, and I struggle with getting the adjustments right. You will sometimes notice a slight banding in certain light conditions if the paper feed is not correct. (which I did have) Have a good look at the prints, especially in really dark areas. If there is no banding, and your prints are looking stunning, I would be hesitant and not advise touching anything to alter the paper feed.  

I know a fair bit of paper is being used in these tests, but you can trim the rectangular line and cut sheets of various sizes from the inprinted paper in the middle so it is not all wasted.

hope this helps.
Julie

[JasonHopkins]

Hi Julie and Ernst,

thanks you for your replies, this forum is the bomb.

Having trawled through posts yesterday I found some info similar to the rectangle test.


2 things:

I think that there may be an operator error involved here

Since I got Qimage I thought that it handled ALL printer driver fucntions, therefore I stopped changing the paper type on the printer display - just left it at custom setting X. So I have been putting .2ml paper in and the printer hardware thinks it is .4ml etc.

Like I say I thought the driver, through Qimage, was overriding this as everything prints perfect (except dimensions!).


I think secondly the media is slipping as there was a difference of 5mm from the beginning of the roll (99.5cm) to the end of the roll (99cm), having used the same settings from beginning to end. - Not sure how I can compensate for this.

I have done some testing on cheap non-epson matte paper and I have got it back to within 1mm over 1m (99.9% accurate) - do you think this is an acceptable tolerance?

Will test on the other media today.


Thanks,

Jason

[Ernst Dinkla]

I have done some testing on cheap non-epson matte paper and I have got it back to within 1mm over 1m (99.9% accurate) - do you think this is an acceptable tolerance?

Will test on the other media today.


Thanks,

Jason

A 1/1000 error on thin paper is acceptable if the studio is not climate controlled. You will notice more errors + or - with different papers. Third party papers have to suit the Epson OEM media presets on coating qualities and transport aspects. It also depends on the period the rolls are in the studio to acclimatise, the humidity control, the paper thickness, 2" or 3" core, tightly wound. etc and the image content, a heavy image will influence the dimensions too. In an ideal world the rolls have been stored in the manufacturer/distributor warehouses at 55 degr humidity, undergo short transport trips and are kept at 55 in your shop. The plastic inside the box does something to keep humidity levels consistent but not for a long time.

PET film like used for separations should behave better as it will not change dimensions due to air humidity and the inks. It will change dimensions to temperature but less than vinyl qualities. Any sign of static electricity when unwinding the roll ? It may be an interesting question whether the 9900 is better than the 9600 for separations. The last had a good reputation on ink density etc.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[JasonHopkins]

Hey Peeps,

Thought I would report back.


Got the transparency film to 2-3mm/1000mm accurate - having to physically scale the image to compensate so not ideal!

Tried every setting under the sun and ended up with rectangle tests coming out of my ears!

The biggest effect was turning the suction right down to -4 and roll tension to standard. Still can't get it to print 1m properly.


Had no such luck with Hahnemuhle paper at the moment - 3mm/1000mm off. Cannot do better with any setting I try.

Epson have told me to contact Hahnemuhle before proceeding with them - great.


Does anybody use Hahn Photo Rag 308gsm with their Epson 9900? - Could you please check your next prints for dimension accuracy and let me know?

Thanks,

Jason

[Ernst Dinkla]

Epson have told me to contact Hahnemuhle before proceeding with them - great.

Thanks,

Jason

That answer stinks in my opinion.

Are all prints with any substrate too short ?

It should be possible to get the two dimensions correct. The width on the printer should be correct on the film right away and very acceptable with papers. If you have to fall back to image aspect ratio changes it isn't good.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[pleverington]

I have a 9900 --same problem. When printing canvas profile tagets off a roll, I loose almost 1/4 inch on the ten inch side. I have tried everyting as far as I know with settings. On a large print I'll loose an inch. Plus I get pretty wide, light banding, and the head passes are overlapping too much as indicated by color seen in the white divisional space lines between the color rows on a target. The only settings that help with the problem are to max out the paper feed and print at 2880.

But this is not a solution. I believe that the problem lies with the backtensioning. If I print the same target as mentioned before, but cut a piece off the roll and print in sheet mode --the problems go completely away. Yes, I went into service mode and turned down the tensioning--no help.

My personal workaround is to construct a media holder over the 9900 using my old 9600 spindle. I'll use this then for the media that is having the problem--canvas so far--until epson gets a fix. I have had no problems with paper media by the way, just that canvas. But honestly more testing with paper is needed for that statement to be verified as fact. I have not printed film or photo rag 308. Never had a problem of this type with the 9600 fwiw.

Paul

[jule]

Hey Peeps,



Does anybody use Hahn Photo Rag 308gsm with their Epson 9900? - Could you please check your next prints for dimension accuracy and let me know?

Thanks,

Jason

Hi Jason,
Just did a square, 100cm x 100cm printed through Qimage. I used a 110cm steel rule to measure. The width of the square ( width of the printer- parallell with the print head) was spot on at exactly 100cm, to the outside of my 30pixel wide line.

The length/height of the square (perpendicular to the print head) was 99.8cm therefore 2mm short.

I might go and try changing the colour of the line and reprint as a sheet instead of off the roll and see if there is any difference in size - before I go and change the paper feed on the control panel or cut the paper up into usable sheets.


Julie

[Desmond]

Hi Jason,
Just did a square, 100cm x 100cm printed through Qimage. I used a 110cm steel rule to measure. The width of the square ( width of the printer- parallell with the print head) was spot on at exactly 100cm, to the outside of my 30pixel wide line.

The length/height of the square (perpendicular to the print head) was 99.8cm therefore 2mm short.

I might go and try changing the colour of the line and reprint as a sheet instead of off the roll and see if there is any difference in size - before I go and change the paper feed on the control panel or cut the paper up into usable sheets.


Julie

This sounds absolutely reasonable. One direction is affected by the rate of print head travel which is nothing to do with the paper thickness. While the other is controlled by the rate of the roller turning and the diameter of the roller/ paper combo. I suggest to further experiment with same paper with different media settings, and also same media setting with different paper thickness. You can verify two things. 1) whether medal setting compensate the roller turning 2) whether the printer is running inconsistent manner.

[Ernst Dinkla]

You can verify two things. 1) whether medal setting compensate the roller turning 2) whether the printer is running inconsistent manner.

As I understand it the last happened with the separations on film. It is possible that the media presets in the Epson driver also dictate the clamping pressure of the paper transport pinch rolls and one should select another media in the driver to get that right. Or a fault in the auto loading mechanism, not activating the right roll pressure. Filthy transport roll is another possibility.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[ooblik]

hello,

I used a 9900 and a Caldera RIP (Vision +) mostly on Photo Rag 308 and noticed some strange size too...
I recently print 6 square print of 95x95cm, and all of them were print in fact about 95x94,7 cm...
I'm still asking Caldera an Epson why...and they just answer me : you have to test your "paper adjust" !

à suivre...

[Doombrain]

I am using the latest version of Qimage.

It is a problem that I have just noticed as I print Seperations for someone and they have come out at different lengths.

I have not had a chance to print from a different program.

Surely it should print at the exact size?

Any raster based program running through the normal driver is going to give you a problem.
Try printing from illustrator if you have it. I seriously doubt there’s a problem with the feed system on your printer.

[Ernst Dinkla]

Any raster based program running through the normal driver is going to give you a problem.
Try printing from illustrator if you have it. I seriously doubt there’s a problem with the feed system on your printer.

Whether one feeds pixels or vectors, any driver has to rasterise the input again and translate that to a distribution of droplets on an area.
Illustrator, Photoshop, Autocad, Qimage, it doesn't matter what program you print from, identical images through the same pipeline should deliver identical print sizes. A RIP should produce equally sized prints too.

An image setter will produce better register, a better defined halftone screen and a higher actinic density. I'm not an advocate for using inkjet separation films but mm's difference between separations printed after another is telling something about the transport mechanism of that particular printer.

Some in this thread mention the use of canvas and deviations of the expected size. That is normal in practice, most likely caused by prestretching of the canvas on the rolls and a related shrink after printing.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[pleverington]

Not like this Ernest. Never had that on my 9600 running same canvas--same settings.. Doesn't happen when I sheet feed this same canvas media(9900) --as I mentioned earlier. I know all about canvas shortening, but almost a 1/4 over ten inches is something else going on. I would love to be wrong but all clues point to the tensioner as far as I can tell.

Paul

[Ernst Dinkla]

Not like this Ernest. Never had that on my 9600 running same canvas--same settings.. Doesn't happen when I sheet feed this same canvas media(9900) --as I mentioned earlier. I know all about canvas shortening, but almost a 1/4 over ten inches is something else going on. I would love to be wrong but all clues point to the tensioner as far as I can tell.

Paul

I should have made the distinction, I agree. Usually it is mm's on a meter. There must be something wrong with transport. While you mention the 9600, that one still has a nice reputation with silkscreen printers for its separation film output: density and register.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

[JasonHopkins]

I have a 9900 --same problem. When printing canvas profile tagets off a roll, I loose almost 1/4 inch on the ten inch side. I have tried everyting as far as I know with settings. On a large print I'll loose an inch.

I don't print canvas at the moment but that sounds worrying!

Just did a square, 100cm x 100cm printed through Qimage. I used a 110cm steel rule to measure. The width of the square ( width of the printer- parallell with the print head) was spot on at exactly 100cm, to the outside of my 30pixel wide line.

The length/height of the square (perpendicular to the print head) was 99.8cm therefore 2mm short.

Hi Julie, thanks for doing that test, what paper did you use? Please don't waste any expensive paper on account of me.

This sounds absolutely reasonable. One direction is affected by the rate of print head travel which is nothing to do with the paper thickness. While the other is controlled by the rate of the roller turning and the diameter of the roller/ paper combo. I suggest to further experiment with same paper with different media settings, and also same media setting with different paper thickness. You can verify two things. 1) whether medal setting compensate the roller turning 2) whether the printer is running inconsistent manner.

I have tried all sorts of settings, the only thing I can get accurate is cheap thin matte paper - which is basically the only media I am never going to use.

Changing settings (tension and suction and paper type) on the thicker media does not seem to have any effect at all!

I used a 9900 and a Caldera RIP (Vision +) mostly on Photo Rag 308 and noticed some strange size too...
I recently print 6 square print of 95x95cm, and all of them were print in fact about 95x94,7 cm...
I'm still asking Caldera an Epson why...and they just answer me : you have to test your "paper adjust" !

Hi, Epson said that this is not to be changed if the print quality is perfect. You should only change these settings if you are getting light or dark banding.

Let me know what settings you had for that print.

Any raster based program running through the normal driver is going to give you a problem.
Try printing from illustrator if you have it. I seriously doubt there’s a problem with the feed system on your printer.

I thought you might of cracked it with that idea! But alas it is still 3mm short over 1m.

I have Illustrator CS3 (the only vector software I have). It seemed to bitmap the image quite badly even though I was printing an A1 vector file - so I think illustrator or the printer was converting it on the fly.


I talked to Hahnemuhle UK and suprisingly they don't think it is their media but something to do with Epson.

I am going to buy a roll of thick Epson Media so that I can do a quantified test.  


Let me know if you are printing with Hahnemuhle or Canson Infinity on an Epson 9900/7900

Thanks for all the comments!

[pleverington]

I don't print canvas at the moment but that sounds worrying!








I thought you might of cracked it with that idea! But alas it is still 3mm short over 1m.

I have Illustrator CS3 (the only vector software I have). It seemed to bitmap the image quite badly even though I was printing an A1 vector file - so I think illustrator or the printer was converting it on the fly.


I talked to Hahnemuhle UK and suprisingly they don't think it is their media but something to do with Epson.

I am going to buy a roll of thick Epson Media so that I can do a quantified test.  


Let me know if you are printing with Hahnemuhle or Canson Infinity on an Epson 9900/7900

Thanks for all the comments!

Jason --try printing it as a sheet instead of off the roll and see if you get the same problem.

Paul