Epson 7900 from the inside - out

[Farmer]

It's OK to be a little frustrated and upset when something breaks that we hope won't, however put it into perspective.

That is extremely well said.

[Alan Goldhammer]

Epson makes good heads, that isn't the issue. The issue is this retarded pressurized system they designed for some of the latest printers. Everybody knows that. Even the Epson techs know that. Look, using that many big carts in one unit where the carts are inserted horizontally surely doesn't help. Maybe if the 10K had used 8 carts instead of 6 it would have had issues too. Whatever the cause of it, it isn't a problem with poorly made heads, it is a bad ink flow system and it's driving people nuts.

It just may be that Epson need more pressure to deliver the ink for this class of printers.  Of course this is difficult to tell from the specifications but they do have double the nozzles compared to the x880 class of printers.

[Gemmtech]

It definitely doesn't seem to be a nozzle / head issue with the larger printers.  There's a fix just waiting to be found.....

[deanwork]

I think your right, and I think they will solve it eventually.

I don't want to come across knocking the 9890 printers. I'm producing some giant outstanding work with it and these heads are excellent. Most of what they are doing with this line of printers, including the price, is outstanding. I just think that if they WERE to improve anything the next time around, the pressure distribution situation would be first on my ( and many people's) list. They do that and they will stay in first place.

john

[Mark D Segal]

I would be truly surprised if there were a GENERIC problem of the type you are pinpointing. Firstly, one has to surmise that the company would have tested these systems THOROUGHLY before committing the huge amount of manufacturing and marketing resources needed to put a new model on the market. Not to say that companies haven't had these kinds of failures, but again - really surprising. Secondly, I'm not sure how much I would rely on the information from EVERY tech (how many have you discussed this with?) because their own knowledge of these systems is not necessarily all that deep - depends. I'm not saying there isn't a problem here, but how systemic it is remains an open question in my mind.

[Farmer]

I think your right, and I think they will solve it eventually.

I don't want to come across knocking the 9890 printers. I'm producing some giant outstanding work with it and these heads are excellent. Most of what they are doing with this line of printers, including the price, is outstanding. I just think that if they WERE to improve anything the next time around, the pressure distribution situation would be first on my ( and many people's) list. They do that and they will stay in first place.

I think that's a good angle to come from and a good suggestion.  May I suggest that, if you haven't already, feed that back through your dealer or direct to Epson?

[Gemmtech]

I would be truly surprised if there were a GENERIC problem of the type you are pinpointing. Firstly, one has to surmise that the company would have tested these systems THOROUGHLY before committing the huge amount of manufacturing and marketing resources needed to put a new model on the market. Not to say that companies haven't had these kinds of failures, but again - really surprising. Secondly, I'm not sure how much I would rely on the information from EVERY tech (how many have you discussed this with?) because their own knowledge of these systems is not necessarily all that deep - depends. I'm not saying there isn't a problem here, but how systemic it is remains an open question in my mind.

I wouldn't be surprised by any type of "generic" problem, cars have them and they put a lot more testing into cars than printers.  Obviously software has issues all the time, though more complicated than a printer.  I can recall (pun) Canon 10D having a back focus issue (I had to send mine in), wouldn't they have tested that camera "THOROUGHLY"?   Wasn't there an orange shift issue with the Epson 1270 or 1280?  Nothing surprises me, I believe these companies allow us to be guinea pigs to a certain extent.

[Mark D Segal]

Nothing surprises me, I believe these companies allow us to be guinea pigs to a certain extent.

Yes, "to a certain extent" - noting that companies need to make judgments about when products are ready for "prime time", and in haste to satisfy revenue objectives they may "jump the gun". BUT, whenever they contemplate doing so, they also need to consider the cost and reputational risks associated with fiascoes. Given how fundamental the ink supply system is to the overall functioning of the printer and the accumulated experience Epson has had designing these things, it is these considerations that make me wonder whether the problem John ("Deanwork") identified is indeed a generic issue. Again, I'm not saying its inconceivable - I'm just wondering how likely.

[Gemmtech]

Yes, "to a certain extent" - noting that companies need to make judgments about when products are ready for "prime time", and in haste to satisfy revenue objectives they may "jump the gun". BUT, whenever they contemplate doing so, they also need to consider the cost and reputational risks associated with fiascoes. Given how fundamental the ink supply system is to the overall functioning of the printer and the accumulated experience Epson has had designing these things, it is these considerations that make me wonder whether the problem John ("Deanwork") identified is indeed a generic issue. Again, I'm not saying its inconceivable - I'm just wondering how likely.

Mark, I'd hope very unlikely, but that would be an altruistic POV.  I suppose my naivety has waned since 1996 when my wife educated me on how big businesses are run, since then I have a much different outlook.  I'd love to believe that companies test products and make sure they are "perfect" before releasing them, but I believe they get them to the stage of "Good Enough" and then start to ship them.  They probably know what percentage will come back, how many repairs per unit (just like cars) and adjust accordingly.  Also, being surrounded by so many engineers some of whom do finite analysis and failure analysis work for some pretty large and well known corporations leads me to believe a lot of products are released before they should be.  You don't think Apple knew about the antenna issue before they released the IPhone 4?  Sure they did and to placate their customers they gave us all a "Free" case.... Was it worth it to them to release the product without fixing the problem?  HELL YEAH, it didn't hurt their reputation and they sold millions of them, no delays in shipping or cash flow.  It was the cheapest way to solve the problem and what's a public company's number one priority? 

[Mark D Segal]

I don't disagree with any of this in principle. Notwithstanding, it is still reasonable to be not totally convinced of its applicability to the case at hand.

[Gemmtech]

I don't disagree with any of this in principle. Notwithstanding, it is still reasonable to be not totally convinced of its applicability to the case at hand.

One can't be totally convinced in this case or any case until the problem has been fully diagnosed and documented.  My opinion is if clogging were a major issue with a mass produced printer there would be a lot more "squeaky wheels".  I don't know what the percentages are, but I would guess Epson only sells a handful of large format printers compared to their consumer / prosumer units.   Has Epson ever gone on record as saying "We have a clogging issue with our large format printers" ?  I've had nothing but success with Epson products, so I can't commiserate with the poor souls who are having time consuming problems.... I'm sure trying to get a print completed only to have a printer not work would be frustrating.   "Product Karma" there's something to be said...

[Mark D Segal]

Yes, I agree. I would add that we are unlikely to be given access to the kind of data that would definitively make these determinations, but it's likely that the total number of x900 models in use world-wide is probably quite large.

[enduser]

As a statistician might tell you, rather than seeking to validate the absolute importance of an issue from a single sample, it can be a lot more enlightening to take a comparative approach.  In this case I see that compared to other brands, printer forums with "clogging" threads combined with the Epson brand, are much, much more common than "clogging" threads combined with the Canon or HP brands.  (No "brand-wars" please)

As for "expectations", I think it's reasonable that a printer will work for a long time before becoming unusable without major expense.  This a reasonable user expectation.  I think that if auto makers can offer 5 year, unlimited mileage warranties, large format printer makers should be doing something similar.

[Mark D Segal]

As a statistician might tell you, rather than seeking to validate the absolute importance of an issue from a single sample, it can be a lot more enlightening to take a comparative approach.  In this case I see that compared to other brands, printer forums with "clogging" threads combined with the Epson brand, are much, much more common than "clogging" threads combined with the Canon or HP brands.  (No "brand-wars" please)

As for "expectations", I think it's reasonable that a printer will work for a long time before becoming unusable without major expense.  This a reasonable user expectation.  I think that if auto makers can offer 5 year, unlimited mileage warranties, large format printer makers should be doing something similar.

Much depends on the quality of the sample and how representative it is of the population. As for a comparative approach - sure it can help, but then you want to be sure about what you are comparing. There is much discussion here and elsewhere about "clogs", but the issue may not be clogs. Then there is the fact that even if it were clogs,  different printer technologies have different ways of dealing with them. One technology makes clogs apparent to the user to do something about; others do not. The difference between measuring what you see and not what you don't see can be very large and distort the value of the exercise.

I'd love to have a five year warranty on my printer (but then the price would have been correspondingly higher); however, I just don't know on what the basis to expect that just because I get a 5 yr./ 60,000km. limited warranty on my car, I should expect the same of a printer.

[Gemmtech]

As a statistician might tell you, rather than seeking to validate the absolute importance of an issue from a single sample, it can be a lot more enlightening to take a comparative approach.  In this case I see that compared to other brands, printer forums with "clogging" threads combined with the Epson brand, are much, much more common than "clogging" threads combined with the Canon or HP brands.  (No "brand-wars" please)

As for "expectations", I think it's reasonable that a printer will work for a long time before becoming unusable without major expense.  This a reasonable user expectation.  I think that if auto makers can offer 5 year, unlimited mileage warranties, large format printer makers should be doing something similar.

I think Mark's statement is straightforward and I agree.  There's no doubt that there are a lot more online complaints regarding Epson printers regarding "clogging" than the other two...Is it because Epson sells a lot more printers or is there another reason?  It's been established that pigment printers from all of the Big 3 (sounds like the car industry) clog. but that HP and Canon deal with it in the background while with Epson it's a part of the user experience (I have never had a clogging problem with an Epson printer).  I do agree that one should be able to own a printer for several years without expense, but as I have mentioned before, warranties are used in marketing and not engineering.  Hyundai has perhaps the longest warranty in the car industry (as long as you own the car), why?  It was a way to gain market share (they have) Apple doesn't need to gain market share (they'll eliminate PCs someday) and therefore don't give much of a warranty. what are they now 90 days?  That brings us to Epson, do they need to give a longer warranty in order to sell more printers?  I highly doubt it and neither do the other 2.  Warranties are also a way of getting customers back who have left the brand, imagine Epson or another company saying, "We are better now and we'll give you a 5 year 10,000 print warranty" 

The question is still, what's the problem?  Are any of the inks in the 7900/9900 the same formula as the 3880?  If the answer is yes, than I'd say it's probably not the ink.  What about the heads?  It's possible it's not a clogging issue at all, it's possible it's a delivery problem.

[Alan Goldhammer]

The question is still, what's the problem?  Are any of the inks in the 7900/9900 the same formula as the 3880?  If the answer is yes, than I'd say it's probably not the ink.  What about the heads?  It's possible it's not a clogging issue at all, it's possible it's a delivery problem.

It is likely that the inks are the same.  They are covered by the same patents and Material Safety Data Sheets (which give you the composition of the various chemicals).  We don't know the exact compounds since they are covered by trade secrets and one could reverse engineer them (as has been done by some third parties).  We've been all over the clogging issue with regard to Epson printers and as has been said, Epson has chosen a different print head technology than either HP or Canon.  Only Epson know for sure the scope of the clogging problem with regard to number of units in the field vs. number of repairs for this issue.

With respect to the engineering issues and testing about when a particular product is ready to go into the market, this is a call by the manufacturer.  It is very difficult to identify every single problem prior to market entry since a relatively small issue cannot be identified prior to "large" scale introduction into the market.  In my field of pharmaceutical development we see this a lot with respect to rare adverse drug events.  Routine clinical trials of a new drug involve maybe 5000 patients in a clinical trial at the high end (not counting vaccines which are given to healthy people).  If you have a rare event that occurs in 1 out 10,000 patients, you cannot reliably measure that unless you have a clinical trial of 30,000 (which leads to much higher development costs).  "If" clogging is a rare event, the manufacturer may not know about the issue until the printer sees widespread use and the question then becomes should the manufacturer do lots more pre-market testing (and delay sales which could lead to a weaker competitive position if say Canon or HP enter with a comparable printer).  Complicated business decisions.

[gwhitf]

Here's a Nozzle Check from yesterday, on Epson Exhibition. For weeks now, the same exact identical missing bars, only in the LLK area. Power Cleanings; Pair Cleanings; nothing will budge it. Frustrating on just how few bars need to be missing in order to show very boldly, on an actual print. It shows up as "horizontal banding".

[Mark D Segal]

I seem to recall reading in my 4900 manual that in such situations they recommend calling Epson technical support for a solution. Have you done that yet?

[Mark D Segal]

Aww c'mon Mark,
 All 'tech help' (and I use the term lightly), is going to do, is tell him to do what he already did! We already know the routine... If the printer is under warrantee it will be a free service call, if out of warrantee, good luck!

Two points: (1) Don't under estimate them, and (2) if there is an established route for problem resolution, use it. If it doesn't work or proves unsatisfactory for whatever reason, web forums become an excellent resource for sharing experience and alerting the company in a public kind of way to what the issues are - but first give them a real chance to help you - that's why they make a service option available both during and post-warranty. The main point I'm making is first things first.

[Mark D Segal]

Time for a service call.
 
David

Indeed - we end up at the same place! :-)