Ad
Ad
Ad
Pages: « 1 ... 50 51 [52] 53 54 ... 74 »   Bottom of Page
Print
Author Topic: Epson 7900 from the inside - out  (Read 262234 times)
kdphotography
Sr. Member
****
Online Online

Posts: 686


WWW
« Reply #1020 on: December 27, 2012, 02:07:06 PM »
ReplyReply

Hi Eric,

Nice handiwork!   Grin

Be aware that hygrometers can be "inaccurate" with regard to actual measurements, but as you've found, the key point here is to just find a level on the gauge at which your printer is "happy"---so although actual humidity may only be 55% and your gauge reads 80%----it really doesn't matter as long as you've found that happy printer humidity level to maintain.

Carmel humidity levels are constant, and only turning on the heat (which I don't) dries out the air.  I've got an extra humidifier if you'd like to give it a try; it's a lot less work than you think.

ken
Logged

Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1021 on: December 27, 2012, 03:24:32 PM »
ReplyReply

Discovery number 2:

About mid-way through this journey I sent our Epson Stylus Pro 7900 printhead to Canada, after me and my genius buddy decided we could do no more to help clear it on our own.  Once in Canada our head was subject to both an ultrasonic cleaning, and also a flushing.  When we got the head back into our 7900, all it produced were fatal error messages.  Our goose (printhead) was cooked.

In the many months since experiencing this disaster, I have done a lot more learning - thanks in great part to other (more intelligent/better educated/and vastly more experienced) users right here on Luminous Landscape.  I have also done a lot more thinking.  And, finally, I have done quite a lot of poking around INSIDE the very printheads which we have all grown to enjoy a love (beautiful prints) hate (ink-time-money-printer-destroying-clogs) relationship with.  While I do still have hope that we can find a cure for X900 clogs, for quite some time now I have quietly hidden my greatest fear - that terminal clogs are not clogs at all.  After all, the wonderful world of Piezoelectrics is not only amazingly tiny, but it is also amazingly fragile.  For sure with all of the aggressive Power Cleanings we are forced to resort to in order to combat the most tenacious clogs; the sucking, the smearing, the pressures of ink being forced through the printhead's internals - it is quite possible that we all walk a fine line between helping our printheads with cleanings, and hurting our printheads with cleanings.  So for quite some time my greatest fear has been a discovery that terminal clogs are actually damaged Piezoelectrics, rather than simply clogs.  

Since this very point is such a vital element of our journey, late last night I finally gathered the courage to act on finding our answer.  I performed an autopsy on our "Oh Canada" X900 Printhead.  Remember, before we shipped it to Canada it simply had un-clearable clogs.  It returned with "terminal errors".  While this is a tragedy, it is also an opportunity.  If the piezoelectrics in this Oh Canada head are damaged, well then here would be our answer - damaged piezos = fatal errors, dropouts = clogs.    Anyone else curious what I found?

It WAS too much fluid pressure that killed our Oh Canada head.  I feel confident in this assessment due to the fact that both channel pairs show damage exactly at the source of where the pressurized fluid would have entered from - which is the same place where ink enters from.  Take a look for yourself.  

The following images line up/stack:

1 - healthy piezoelectric printhead (minus it's missing face, plus some microscopic garbage here and there)



2 - Partial workings of an Epson Stylus Pro 7900 printhead identified







2 - Damaged "Oh Canada" printhead










« Last Edit: December 27, 2012, 03:39:28 PM by Eric Gulbransen » Logged

Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1022 on: December 27, 2012, 08:27:36 PM »
ReplyReply

Thanks for this tip Ken, I have little experience or knowledge about hygrometers (especially spelling).  How people can sell a product with such inaccurate tolerances/performance baffles me.  My world is so much simpler - you build it nice you get paid.  You build it crooked you don't. 

...by the way it's hygrometers I'm referring to here.


Hi Eric,

Nice handiwork!   Grin

Be aware that hygrometers can be "inaccurate" with regard to actual measurements, but as you've found, the key point here is to just find a level on the gauge at which your printer is "happy"---so although actual humidity may only be 55% and your gauge reads 80%----it really doesn't matter as long as you've found that happy printer humidity level to maintain.

Carmel humidity levels are constant, and only turning on the heat (which I don't) dries out the air.  I've got an extra humidifier if you'd like to give it a try; it's a lot less work than you think.

ken
Logged

Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1023 on: December 28, 2012, 01:31:28 AM »
ReplyReply

I call my buddy Steve a genius.  If you were at dinner tonight you would too.  I dumped all the new found data I could on him tonight.  Nothing intimidated him.  In fact each new piece I shared only drew him closer to better understanding the hidden science behind these X900 printheads.  Turns out as complex as these things are, they're also pretty simple.  I can't say we know exactly how every element of them work, but I can say we are getting pretty dam close.  I will explain this all in time, in a language I am comfortable with - images.  But this will take some work.  For now I'd like to share our most recent discovery - the third dimension of piezoelectric nozzle boards.

Up until this point (or point of view) one major unsolved mystery has been tracking the path of ink through the face of the piezo nozzle board.  As you can see from my previous posts, ink has a bit of traveling to do in order to get to the actual piezo nozzles just under each hole in the face of the printhead.  The reality behind these nozzles is actually very different than any illustrations that I have seen.  These "nozzles" are not actually nozzles at all.  Instead they are more like chambers.  Each chamber is surrounded by it's walls, which flex when electrically charged.  When these walls flex, the ink that is between them gets forced down (or across actually - since our heads sit sideways on the printer) the path of least resistance.  In the case of these X900 printheads, that path is out the face of the printhead and on to your print.  For sure the ink won't travel back into the head instead, because of the pressure backing up the ink in the supply lines - and throughout all the chambers (and tunnels) in the piezoboard. 

Hold on there I just used a new word.  Two in fact - chambers and tunnels.  Like I said "up until this point - of view - one major mystery has been tracking the path of ink through the face of the piezo nozzle board."  I shot video of ink entering a piezoboard and filling all the chambers.  It makes no sense though, I can't track it.  Somehow the chambers just get filled.  Remember there is only one ink supply port per channel - it is pretty small relative to the size of the channel board that it supplies with ink.  The bottom of a channel board is solid, no ink enters or exits out the bottom.  The top of a channel board sits smack up against the thin sheet metal printhead face.  No ink flows between the sheet metal face and the channel board - if it did the different inks in each channel pair would flow together.  So it is confirmed then, the only way for ink to travel from the supply port, to the nozzles, is THROUGH the channel board.

...feel like blowing your mind right now?  Imagine this - a channel board measures only about 1/32", or .8000mm thin.  That's pretty lean actually, yet somehow it is perforated not only vertically (seen in various pics I have recently uploaded here), but also in fact horizontally - on multiple levels (only seen in pics I will share now) - in what is a fascinating complex three dimensional maze of impressive intricacy. 


Here is a side-view pic of a channel UNIT.  This is NOT a pic of the channel board.  A channel unit measures about 3/16", or just under 5mm.




Here is the channel board, which sits at the very top of the channel unit.




I tried to remove the channel board from the channel unit but no dice, it's glued on pretty good and the board is fragile as you can imagine.  So I broke a piece of the channel board off to examine more than just the top.  Here we can see the bottom as well.











While these images are cool I learned nothing from them, other than we are dealing with some tiny stuff here.

...and then it happened.  I took multiple frame grabs from some video I had shot and put them together to reveal what to me is a gateway to understanding the paths (or tunnels) that carry ink through and to our piezoelectric nozzles.  This is by far the most revealing image yet.  Those square holes you can see in the side-face of the channel board are actually ink passages.  These are the paths, or tunnels, which lead ink to nozzle chambers.





Today I learned that "fatal error messages" report broken piezoelectric boards.  Your head is truly dead.  I also learned that "clogs" are most likely actually that  - simply clogs.  Tonight however I learned that "nozzle" clogs may not actually be limited to just nozzle clogs at all.  Could be clogs are anywhere inside the intricate maze that makes up piezoelectric nozzle boards.  Perhaps a clog in a specific tunnel, or a chamber, could end up starving the nozzle that chamber supplies.  I don't know yet.  ...But I will I promise you.















Logged

Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1024 on: December 28, 2012, 05:17:08 PM »
ReplyReply

I've got two more hours in me then it's back to planet earth.  

I dove deeper this morning in my quest to understand inkflow through these piezoboards.  I clamped a glass slide on top of a channel, hooked a syringe filled with Cyan ink to it's nipple, and filled the channel.  In my best representation, of my latest understanding, of how ink fills our X900 printheads - I need to report here it is both simpler and more complex than I thought last night.  Not that such a thing is even possible, but whatever.  I'm doing my best to reverse engineer these intricate machines in an effort to make my father proud, "Before you can fix it you need to understand it"

The following are a series of video frame grabs taken as the channel filled:


























So the channel long, deep hull-like groove that exists inside each channel is indeed a reservoir.  The reservoir fills first.  From here ink does definitely travel THROUGH these intricate piezoboards, but it also travels beneath certain levels of them, deeper in the reservoirs.  "Certain levels" of them you might ask?  Yes I know, it's sickening how complex these piezoboards are - they actually do have different levels.  Think of piezoboards like very wide three story buildings built on concrete slabs (ink never goes below the first floor), and capped with a thick sheet metal roof, with holes in them called "nozzle openings" (ink never goes above the roof, unless it's through a nozzle opening).  Starting to make sense yet?  Some of the areas of these buildings (piezoboards) are wide open, floor to ceiling.  These areas are the ink reservoirs.  They are contained within the dimensions of the piezoboard itself.  I can't tell yet what goes on on the first two floors of the piezoboard.  So far they seem made up of honeycomb-like tunnels.  I can however tell what goes on on the top floor.  This is where the nozzles fill, and this is where the nozzles fire.  At this point my confidence is growing surer by the minute that our struggles happen on this top floor.  This makes the most sense, being that this top floor is the one so vulnerably close to the open air - through the nozzle openings in the sheet metal roof just above.  For sure this floor is also vulnerable to half-dried sludge being forced back down into the nozzle openings in the tin face of the printhead, by our dirty rotten wiper blades...


Finally, after my having said all this, and you having read all this, here is your reward - THIS IS AN UN-CLEARABLE CLOG:




« Last Edit: December 28, 2012, 06:02:45 PM by Eric Gulbransen » Logged

Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1025 on: December 28, 2012, 06:01:20 PM »
ReplyReply

The top level of the piezoboards do indeed house the nozzles.  Just keep in mind what I said previously - they don't at all look like nozzles.  This is a piezoboard, top level, series of (clogged) nozzle chambers.  



Best I can tell ink enters these top level chambers (which DO have micro-membrane floors in them that seem flexible/vulnerable), from the reservoir side.  Ink then travels past what could be a one-way valve (looks like a speed bump) then sits, pressurized, and hopefully fluid, awaiting it's firing out into the great new world of dried up ink on photo paper.  




So the problem is obvious once you look at your head through a stereo microscope for two weeks, take hours of video, think about it forever, write about it for days, receive great feedback on LuLa, and have a genius buddy.  Our HDR ink is drying inbetween the chamber walls, on the top floor of the piezoboard.  Once dry in the chambers ( I know because I've dug it out with the tip of a razor knife) this ink takes on a semi-hard, rubber like consistency which for sure is what keeps the chamber walls (piezo nozzles) from flexing (firing), which is what keeps ink from exiting the nozzle opening in the tin printhead face, which is ultimately what results in drop outs, clogs, broken nozzle patterns, banding, and ultimately impotency.  


Last image I swear - here is the placement of the nozzle openings as they sit over the nozzle chambers.  There is exactly one nozzle for every one chamber, and here is exactly where each one sits.





Next step - fix it

« Last Edit: December 28, 2012, 06:06:22 PM by Eric Gulbransen » Logged

BrianWJH
Full Member
***
Offline Offline

Posts: 161


« Reply #1026 on: December 28, 2012, 06:12:32 PM »
ReplyReply

Hi Eric,

Been following your forensic analysis with interest even thought I don't have a 7900 and I have to say I admire your adventure and tenacity. Your last image of the un-clearable clog looks like a concentration of pigment particles gone hard. Is it possible that too many concurrent power cleanings 'bake the pigment' into a cement like blockage using the heat generated in the process?

Good luck in your quest.

Brian.
Logged
Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1027 on: December 29, 2012, 01:18:50 AM »
ReplyReply

Good question Brian.  Heat generated by the friction of multiple powerclean-firings, accelerating the drying of ink stuck inside the head is something we've talked about before.  It's like the potential solution to clogs could be causing more clogs, you are exactly right.  And then there is frying your head all together with powercleanings, which I think sfblue did.  It's definitely food for thought but either way they got here, once your head has an unclearable clog you're pretty well SOL  ...which is why this thread. 

So far I have limited my exploration to just the printheads.  I feel I understand them well enough now to know how they work, and why they don't anymore once clogged.  You are exactly right, now it's time to explore the world of drying inks.  I wish I were a chemist.  While I knew nothing about piezoelectrics six months ago, at least I am inherently mechanical.  Chemistry though, that's a world I need someone else's education to understand.  I did buy a compound microscope, and inks are my next target.  I plan to look at them wet, look at them dry, heat them up and look at them, and then ultimately I plan to try to loosen them up again. 

As for the clogs in the head pictured above, I soaked that head in distilled water - nothing happened.  I soaked it in rubbing alcohol - nothing happened.  I soaked it in Acetone - plenty happened.  The ink, ever so slowly, did loosen up, and did shrink away into tiny passable particles again.  This is very good news.  The flip side of this good news is Acetone is a lot more aggressive than I am comfortable with using just yet.  I need to explore it's affect on all the different materials in these X900 printheads.  Some plastics melt under the influence of Acetone.  Others aren't affected.  But it's not just plastics I am concerned with - there are many other semi-soft materials in these heads.  There is glue that sticks the channel pairs to the printhead framework.  There is a clear plastic silicone-like substance that insulates all the piezo wires.  There is the piezo board itself, made out of I have no idea what.  And then there is the glue used to stick the tin roof onto the piezoboards.  I need to be confident that none of these materials will be compromised due to soaking in whatever chemical we decide to use.  So there is plenty of work yet to do. 

Any chemists in the house willing to lend some thought..?
Logged

iladi
Newbie
*
Offline Offline

Posts: 49


« Reply #1028 on: December 29, 2012, 02:24:57 AM »
ReplyReply

Not a chemist, but, as far ar i know, the difference from watrebased head to ecosolvent head is the nipple. At least for the older generation heads you can buy a waterbased head, replace the nipple and you have a ecosolvent head. More, since epson heads are used for both water and solvent inks, i think is possible that epson don't have 2 distinct production lines, one for water and one for solvent (yes dx7 heads are also used for solvent printers). The problem here is that i understand that solvents and water don't mix well together, they precipitate.
Logged
Oldfox
Jr. Member
**
Offline Offline

Posts: 76


« Reply #1029 on: December 29, 2012, 02:36:06 AM »
ReplyReply

I am not a chemist either. I remember reading somewhere that ammonia or household ammonia could be used to clean the head.
« Last Edit: December 29, 2012, 02:56:25 AM by Oldfox » Logged
snsandrze
Newbie
*
Offline Offline

Posts: 3


« Reply #1030 on: December 29, 2012, 09:48:26 AM »
ReplyReply

Eric and all who have contributed,

I spent the better part of Christmas reading from page one of this thread, glued to the computer screen anticipating every twist and turn. I admire your staying power to determine the cause and solution! Awesome!!! What brought me to this place was my desire to purchase a 4900. I wanted to know as much about them as possible as I will be dipping my toes into the printing pool for the first time. I got to tell you that the water looks pretty cold from reading all of the issues with clogged print heads! I'm a masochist, so that is probably not going to stop me, however I will be mindful of how important maintenance is to the longevity of these printheads, thanks to you and all who have contributed there time and stories.

The one way valve (speed bump) would preclude you from drawing cleaning fluid backwards through the printhead? Or maybe I'm not completely following along with the path of ink flow.

Once again thank you for the education,

Steve
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6767


WWW
« Reply #1031 on: December 29, 2012, 10:13:24 AM »
ReplyReply

What brought me to this place was my desire to purchase a 4900.

Steve

The practical fact of the matter is that unless you use the printer every few days, you can expect it to require cleaning - a little or a lot depending, before using it after it has been unused for more than a few days. My experience confirms what Epson said about them (somewhere, sometime, don't ask me for specifics) from the get-go: these are production machines, and my experience confirms this. If you plan to have gaps between printing sessions exceeding more than several days and you don't need a roll-holder and you want the highest quality, most maintenance-free printing experience achievable with an Epson professional series printer, do yourself a favour and buy a 3880. If you will be printing at least every three days, the 4900 is a fabulous performer.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Sal Baker
Full Member
***
Offline Offline

Posts: 140


« Reply #1032 on: December 29, 2012, 10:44:51 AM »
ReplyReply

I agree.  I don't know what special magic Epson put in the 3880, but unlike all of my other Epsons it's been a clog-free machine.  It sometimes sits (covered) for a month at a time.  It required one standard nozzle cleaning about two years ago but that's it.  I still do nozzle checks before running an expensive sheet of 17x22 paper through it, but it continues to be clog-free.  A Google search reveals this is a common experience with most 3880 users.

What did Epson do different in the 3880?

Sal
Logged
Alto
Jr. Member
**
Offline Offline

Posts: 50


« Reply #1033 on: December 29, 2012, 11:14:55 AM »
ReplyReply

Hi Eric

Today our vintage tractor ran out of diesel so I had to bleed the injector system which got me thinking as I was standing out in the rain. In an injector system even the smallest and I mean smallest amount of air will cause the pump to malfunction . Are the nozzles sucking air back into the system ? if the piezoelectric chamber is trying to squeeze air the air will just compress and we see no output.?  If the chamber has an incompressible fluid i.e. ink it will be expelled from the chamber.

Jon

I have been following from the start .
Logged
Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1034 on: December 29, 2012, 11:47:10 AM »
ReplyReply

Mark D Segal, I've been wanting to bring this up for a while.  You own a 4900 don't you?  You seem to have your finger on the pulse of "X900" clogging issues so I ask you to think this over for a bit - in all of the reports we have heard about these X900 printhead killing terminal clogs, how many can you remember were 4900s?  I'm doing my best to remember, which has one of my dog ears cocked to one side.  Sure people mention clogging with 4900s, but how many dead heads...  ...any?  If there is a difference, and 4900s are indeed insulated from the most horrid of clogs - terminal clogs - then perhaps it's the very different angle that the 4900 head sits on it's rails.  I don't know, thoughts like these are my sheep to count late night.

iladi, that's an interesting point you bring up.  I know I've talked to HAL about some of his service calls on these X900s pushing out ecosolvent ink.  The stories are disgusting actually - how the lines get gunked up with muck and tar, it really sounds like a mess.  So perhaps this suggests there are solvents which could be used that won't do damage.  From what I could tell from the one channel that I did test with Acetone, none of the piezoelectrics seemed adversely affected, neither did the clear silicone (rubber-like) material surrounding all the microscopic wires on the underside of the channel.  The plastic framework of the head was not affected, the nipples were not affected, and neither was the rubber grommet that fits the nipples to the back of the head.  All clear there.  But there ARE two materials left which I am suspicious of, they may have been compromised.  The glue which adheres the channel to the plastic framework of the printhead, and the glue which fastens the tin roof to the face of the piezoboard.  It is indeed still difficult to remove an Acetone soaked channel from the printhead, but it doesn't seem AS difficult.  As far as the glue adhering the tin roof to the piezoboard, I have not tested that yet but I will.  I am paranoid about all glues now.. 
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6767


WWW
« Reply #1035 on: December 29, 2012, 12:00:58 PM »
ReplyReply

I agree.  I don't know what special magic Epson put in the 3880, but unlike all of my other Epsons it's been a clog-free machine.  It sometimes sits (covered) for a month at a time.  It required one standard nozzle cleaning about two years ago but that's it.  I still do nozzle checks before running an expensive sheet of 17x22 paper through it, but it continues to be clog-free.  A Google search reveals this is a common experience with most 3880 users.

What did Epson do different in the 3880?

Sal

Same experience with its predecessor the 3800, which was a breath of fresh-air from the previous 4800 and 4000 models both of which had clogging issues. So the question is what happened in the progression from the 3800/3880 series to the 4900. The new head technology is different - many more nozzles, hence smaller ones per square inch. The inks are named differently so the recipe may not be the same. The lines are longer between the cartridges and the print head. More than that I don't know. Epson did say - I believe in the initial promos - that the 4900 head was coated with a clog repellant material, but I have no idea what that would do for me if the clogs originate from drying ink within - or indeed if that's really what happens. This stuff is nano-technology, combining at least leading edge chemistry, materials technology, fluid mechanics and the relevant electronics; to really understand it I think requires a high level of very specialized education and experience.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
raydee
Newbie
*
Offline Offline

Posts: 13


« Reply #1036 on: December 29, 2012, 12:10:28 PM »
ReplyReply

So maybe the piezo heads are just as much a consumeable as the thermo heads are. They just aren't labeled as such, are hard to aquire for the end user, need a trained $$$  technician to replace and cost so much that it means the printers end of life in most cases.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6767


WWW
« Reply #1037 on: December 29, 2012, 12:11:25 PM »
ReplyReply

Mark D Segal, I've been wanting to bring this up for a while.  You own a 4900 don't you?  You seem to have your finger on the pulse of "X900" clogging issues so I ask you to think this over for a bit - in all of the reports we have heard about these X900 printhead killing terminal clogs, how many can you remember were 4900s?  I'm doing my best to remember, which has one of my dog ears cocked to one side.  Sure people mention clogging with 4900s, but how many dead heads...  ...any?  If there is a difference, and 4900s are indeed insulated from the most horrid of clogs - terminal clogs - then perhaps it's the very different angle that the 4900 head sits on it's rails.  I don't know, thoughts like these are my sheep to count late night.

................

Eric, yes, I own a 4900. I wouldn't say I have my finger on the pulse of "X900" clogging issues, but I do have a very clear perception of how MY 4900 behaves, and it behaves pretty much as predicted: use it regularly and there are seldom clogs; leave it alone for weeks on-end (as I need to do periodically) and cleaning will be required - sometimes repeatedly using a routine Epson Pro-Graphics advised me about some years ago (running prints between cleanings) and it always - eventually - recovers every nozzle. A nuisance to be sure, but no disasters - yet. The fact that there is so little forum noise about the 4900 indicates that few people are likely to be having fatal issues with them. I wouldn't have a clue why it *seems* to have fewer fatalities than the 7900, but these are not identical machines apart from size. The 4900 is a much more recent model - there was a gap of several years between the 7900 and the 4900, and the head technology is not simply a smaller sized clone of the 7900 head. They changed some stuff, but exactly what I don't know and it wouldn't mean much to me, not being a printer technologist.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Eric Gulbransen
Sr. Member
****
Offline Offline

Posts: 298


never surrender


WWW
« Reply #1038 on: December 29, 2012, 12:14:12 PM »
ReplyReply

The one way valve (speed bump) would preclude you from drawing cleaning fluid backwards through the printhead? Or maybe I'm not completely following along with the path of ink flow...
Steve

Good question Steve.  My genius brought up the same ideas..  It's easy to visualize the piezo-walls flexing, then pushing ink out of the nozzle opening in the printhead face (tin roof).  But it can't be that simple.  The reason being, each piezo chamber has two walls - a left wall and a right wall.  OK so fine, the walls flex inward and ink gets shot out the nozzle opening.  But what happens to the ink on the other side of these two walls?  After all if they flex inward from the perspective of the firing nozzle, they flex outward from the perspective of the two chambers on either side.  This would create suction in those two chambers wouldn't it?  Seems to me that it would..  I don't know that the "speedbumps" are actually two way valves.  I am assuming this.  It is possible that since these bumps are indeed on the chamber walls, that they do make contact with the wall on the other side during a piezo firing.  If this is true, then it is also possible that this one way valve is only a one way valve during a firing.  When not firing the passage is clear for ink to pass.  Keep in mind that these chamber walls have both a roof, and a floor.  The roof is the tin printhead face, the floor is a very thin plastic-like flexible membrane.  

Sorry for getting this technical, but you asked for it...   Smiley

One theory the genius and I came up with, which "might" explain how two neighboring piezo chambers can fire at once, is that they can't.  Perhaps, and please I am open to anyone's opinions here, the neighboring piezo chamber fires when the head comes back on it's return run.  If you've ever watched your criss-crossing printhead produce a print, and I am sure we all have, it DOES seem to print only partial areas of the full footprint of the printhead face with each pass.  I measured our X900 nozzle lineup, it is indeed one inch full of nozzles.  So why then don't our printers print complete one inch wide bands per pass?  It is possible that this is the reason why, two (or three actually) neighboring piezo chambers cannot fire at once.

We know that there is indeed one nozzle for every chamber.  We know that the chamber walls flex when electrically charged.  We know how the electric charge gets to each chamber wall.  We know the flow of the ink.  The rest is up to physics (which I dropped out of in college), and common sense (the only sense I have).  

The reason I post the pics that I do is that people more educated than me can share their theories as well.  We're all in this together.  So thank you, for both past and future contributions to this X900 clog clearing journey of ours.


« Last Edit: December 29, 2012, 12:20:07 PM by Eric Gulbransen » Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6767


WWW
« Reply #1039 on: December 29, 2012, 12:15:23 PM »
ReplyReply

So maybe the piezo heads are just as much a consumeable as the thermo heads are. They just aren't labeled as such, are hard to aquire for the end user, need a trained $$$  technician to replace and cost so much that it means the printers end of life in most cases.


No. Those heads are not meant to be a "consumable". They are essentially the printer, and they are meant to last a very long time.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Pages: « 1 ... 50 51 [52] 53 54 ... 74 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad