Tracking Proofer Drift – How Accurate is an Epson Over Time?
Although we know that an Epson proofer can be tuned to be highly accurate – often to within a deltaE of 1 to the target – most of us are unaware of how proofers drift. The question isnÃ•t if inkjet proofers drift, but why do they drift, and what do we need to do to control the drift. Inkjet proofers are not alone in this regard. Even high-end proofers such as the Kodak Approval and Creo Spectrum are subject to drift. Conditions such as temperature and humidity can cause huge color shifts in these devices, and it is not uncommon to see a drift of a deltaE of 2 across several shifts. Keeping these proofers stable can require frequent calibration and attention to environmental conditions.
Inkjet proofers have several reasons for drift. Early dye based Epsons drifted because of the ink composition. With these dye based inkjet proofers you could watch the proof drift over the course of minutes and hours. HP inkjet proofers are notorious for drift because of their thermal head technology. As the heads heat up and degrade, the proofs change. Even the more recent Ultracrome inkjet Epsons could be wildly different from machine to machine. Every Epson was a different animal, and print quality could vary even with Epsons of the same model.
The current Ultracrome K3 generation of Epsons are more stable than previous inkjet inksets have been. Not only does the ink come off the printer dry, but each proofer is tuned in the factory to be within a deltaE +/-2 of the factory standard for that proofer. But even with great technology such as the Ultracrome K3 inkset these proofers still are subject to drift. There are several reasons that proofers still drift. The first reason is changes in the printer. Epson printers are very stable and reportedly can go months and sometimes even years maintaining color with just head cleanings being performed as maintenance procedure. Even so, over time the printheads degrade, and the printer becomes a different animal. I see this on occasion when I perform a relinearization and am unable to get the printer back to its original state. As the printheads changes so does the printer. At first it is slight but over time it can become dramatic. The second factor is paper. Although manufacturers claim that their paper is stable and controlled, paper is a bigger factor than people know. Some large paper brands have their paper manufactured by multiple vendors, which leads to quality problems. On a premium paper the shift may be as much as a deltaE of .5, and with an average paper deltaE differences between batches can be as much as 1.5. If we are now capable of performing color matches of less than a 1 deltaE then a shift of 1.5 deltaE is troubling. The third factor is inks, which seem to be less volatile than paper, however ink cannot be discounted as a contributing factor.
How much do printers drift? From my experience, after leaving an Epson tuned and calibrated they seem to be stable for a period of several months. Measurements vary depending on the printing, but over a period of several months I have measured changes ranging from almost no change to a difference of several DeltaE. Over a year the change can be more dramatic, and when I go back to sites that have not recalibrated over the course of the year the change ranges from a deltaE of 1 to as much as 10.
So we know that printers drift. The proof is supposed to be the most stable thing in the plant. So how do we keep the proof stable? How often do we have to relinearize to keep the proof stable?
There are several theories on how to keep the proof stable. The first theory says that relinearizing using a proofer relin tool will be sufficient way to keep the proofer stable. The idea of relinearizing a proofer is that by adjusting the basic linearization for the proofer everything will be held stable. In many cases relinearizing seems to work. Although not perfect it provides some proof stability. Relinearizing results are not flawless though. When a dramatic change occurs to the printer, such as a degradation of print head or a change in paper, simply relinearizing may not be enough. Because of this many color consultants prefer to reprofile rather than relinearize. They have been burned by the slight variations that often occur when relinearizing. Because of the new technique of iterating color (reading patches in and having the rip make automatic adjustments until the color is good) that is common in RIPs such as EFI XF Colorproof and Oris ColorTuner there is an alternative method to performing a relin in the rip. If the color is tuned to a deltaE of less than 1, and the paper itself can shift that much from batch to batch then it can make more sense to reiterate the color down to a deltaE match of less than 1 rather than just trust the linearization. When using the iteration to pull the color back down you are assured that you still have an accurate match, and there no question that the printer is still linear and stable. It is a decision that comes with a price – the basic relin is fast but less accurate. The reiterating takes longer to do but is more accurate.
Whether relinearizing or reiterating, the next logical concern is how often to perform this function. If color is really important then once a week is reasonable. Doing it often also makes it an easy function to perform. If done less operators may forget it or decline to do it because they do not feel comfortable and familiar with the procedure. If once a week it too much then once a month may be sufficient. Tracking the proofer will indicate when the proofer needs to be relinearized or retuned.
Tracking the proofer can be performed with quality control software. Many RIPs such as CGS Oris Colortuner, EFI Colorproof XF, and GMG Colorproof have proof verification options. These let you read in a small control bar on the proof and verify that the proof is within the tolerance required to be an accurate proof. If you have a copy of GretagMacbeth ProfileMaker the a target can be run and compared to the original using MeasureToolÃ•s compare function. Either of these methods will tell you if the proofer has drifted significantly. For those without a spectrophotometer or any quality control software a densitometer can be a quick way to check the proof. When the proofer is tuned read the control strip with a densitometer and write down the values. The same strip can be read at a later time (with the same amount of dry time as the original) and if the density has shifted significantly it is an indicator that a relin is required. While density is not as accurate of an indicator as LAB values, it can give an indication that the state of the printer has changed.
Everyone has a different tolerance for proofer drift. The important thing is to recognize that inkjet proofers do drift, and require frequent recalibration. Monitoring the proofer helps us realize when these recalibrations are required.