Pages: [1] 2   Go Down

Author Topic: Please Help:Did my Cart explode? Can't get rid of banding. Multiple Nozzle check  (Read 3491 times)

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Greetings!

I'm on a Epson 9890. I have to get this order out for a customer by Monday. it is a single image they want to see on several different papers. I'm getting some banding in the sunset sky. I thought it was the light Magenta. Ran a ton of CL1 and accouple CL 2 and even 1 CL 3 cleaning. Did print head alignment. Replaced my wiper, cleaned the capping station etc. Am I correct from the picture it is most likely the Light Magenta?

This is a blurry image. Any tricks i may use to reduce the look of the banding? Use a draft or Max quality setting? Maybe use a Wide GAP. The banding is more obvious in person, the scan is a bit soft.

The test patterns are latest on top going back between cleaning/printing cycles. The cowboy photo the print head is traveling vertically in that print note the shadows around the clouds.

Thanks

Max
« Last Edit: October 05, 2024, 11:51:49 pm by disneytoy »
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Well Boys and girls, Tell me what you think? So I thought I'd remove the Light Magenta maybe give it a light shake and look what I found. Looked like a slasher movie. Anyone ever see this before?

It does not appear to be coming from the little port. I think it may becoming from the seam of the cartridge. Yes I've done several cleaning 1 and 2 over the last day or so but this is crazy.
Logged

dgberg

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2954
    • http://bergsprintstudio.com http://bergscustomfurniture.com

15+ years old so it looks like you got your money out of it. 10 is good.
Call Ink Owl and see if they still sell replacement refillables for that model.
Pull your ink out of that cart with a syringe and put it into the refillable.
Logged
Dan Berg
http://www.bergsprintstudio.com/workshops/
Our Studio is a Canson Certified Print Lab

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

huh?  I'm pretty sure that says "2018" so unless you're living in the future Dan....

I'd just get a new LM cartridge-- the unstable nature of those missing nozzles does suggest that there is some issue with the pressure system, and that one having exploded confirms it.  Did you check your other carts?  If that's the only one experiencing the issue then it's most likely just a failure of an O ring or something else in that cart itself-- if any of the others also have this happening then you could have a bigger problem, but your nozzles all look relatively good except for the randomness of the dropout pattern, which is almost always caused by a bad seal somewhere causing air bubbles to be introduced into the lines.

The whole pressurization system is in parallel, so one failure like this could easily cause problems with the other ink lines as well.  Hopefully just replacing that one will fix it.
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Thank you.

That was a relatively full cartridge. Just curious how it imploded. Like I said it is leaking out of the seam not the nozzle. I put in a 1% left cart I had, but I have a new sealed 700ml cart I guess I should open.

Can anyone advise on my cleaning cycles> I will just do the LM/LC pair. should I forgo a CL1 and just do a CL2?

Its frustrating because I'm known for my fast turn around, and I basically have to get this working or I'll lose this client.

Do you think it was a defect in the bladder of the cart or did the printer pop it under too much pressure?

I haven't checked the left bank but all right bank carts were fine other than the LM blow out.

I left the cartridge in the sink in the upright "installed position" and it continued to leak from the bottom seam.

Max
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

it's not terribly difficult to open up the plastic cartridge-- the ink is in a bladder inside of it, so there's really no way to know where it's leaking from unless you get inside the shell.  There's nothing wrong with your specific cleaning cycles, basically do the least that will clear the issue.

As far as being known for your fast turnaround... it seems like a fairly obvious point to make that if this is your main selling point, you should have a backup machine...
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Unfortunately, my printing is by no means very profitable. I have some repeat clients that pop up but I wouldn't particularly in this economy invest in more equipment. The printer has more than paid for itself even for my own printing needs, but I could never get this to a consistent business. Unfortunately, this is a good client, with a very large job coming up for me.

I have the new 700ml LM installed. Did a couple CL1s and a CL2 something looked like progress, but the main cowboy image I need to print on 4 different papers still has cringy banding in the sky. I will try again tonight.  I've really never had an issue with this printer. I think I was in a sweet spot. Not commercial quantity printing, but enough large prints on a regular basis to keep the printer happy.
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

the only constant issues you seem to have between the different nozzle checks you posted originally are the one deflected nozzle in the top row of the PK and the missing nozzle in the yellow towards the bottom.  You might be able to fix the yellow nozzle with some manual cleaning with cleaning solution or a head soak with cleaning solution in the cap, but it doesn't feel like this should be enough to cause such obvious banding.  It looks to me like it would have to be something with the LM-- like it's not firing at the same density on half the head or something...  Have you tried jacking up the print mode to the highest quality (1440x2880 it should be) and using the lowest possible head height?  You may be able to get something usable at least to close out this one print...

also try printing unidirectionally-- it may end up being something as simple as a bidirectional alignment being out of wack after all the cleanings, and if that ends up solving it you could definitely go back to your desired print speed to make the print, then recalibrate your bidirecitonal offsets once the job is complete.  IIRC the way to change to uni is to uncheck "high speed" in the epson driver.

Another method for trying to pinpoint the source of the issue would be to print on your LOWEST quality (like, set the printer for plain paper or bond paper or something, and use the lowest possible quality draft mode also with high speed unchecked-- this should make the printer do a single pass the full width of the head, then move the paper forward the full width, and as long as microweave and all that other jargon is off, you should then be able to see if there is a problem with a certain portion of the print head.  You could go even further by printing just magenta or just cyan etc. etc.  Without a RIP you won't be able to isolate single channels this way (except possibly for yellow if you really hit the color value of the ink itself on the head) you'd still be able to see majority of, say, M+LM and see if it is indeed one particular channel (still betting on LM here) causing the problem.

Unfortunately were you to isolate the issue to a specific channel, there wouldn't be anything else you could do without a RIP of some kind which would let you disable the offending channel and reprofile using the remaining good ones.  Gutenprint should be able to do this for that model printer, but the process is defintely not for the faint of heart, and if you don't have a spectrophotometer to make your own profiles you'd be SOL anyway...
« Last Edit: October 06, 2024, 07:38:21 pm by Lessbones »
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Wow Lessbones!

Great info. I will explore all your suggestions tonight. I did create in CMYK magenta gradient in PS and was printing that to check the nozzles. I'm now a little afraid to run the LM with too many CL 1 and CL 2 and perhaps blow out another Cartridge. I've overlaid my nozzle checks and do see a missing yellow lower left corner which hasn't come back with some CL 1 cleanings. Should I continue only doing pairs, or is it worth doing a full cleaning? Also I've run the alignment calibration a few times.

They printers History hs been excellent. I've always been able to get all nozzles printing perfectly. I always start the day with a nozzle check and have been quite pleased that everything has been perfect.   I will google manual head cleaning. I did clean the capping station and put in a new wiper.

Logic would tell me. the "banding" is in the LM. The only Cartridge that is blown out was the LM. So is it the chicken or the egg? Was it leaking and that effected the nozzle, or my attempts to clear the nozzle popped the cartridge. Crazy stuff.

Really appreciate your advice!

Max

« Last Edit: October 06, 2024, 09:36:38 pm by disneytoy »
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

I was reading at: https://myx900.com/

Does this apply to the 9890? Would anyone recommend this?

SS Cleaning: Maintenance Mode 1 (Hold the Pause button at power on)

SS Cleaning has a strong ultra sonic component. The Piezo Elements internal to the Print Head are used to attempt to break up pigment particles that may be clogging the interior of the Print Head.

Use SS Cleaning when a Nozzle can not be cleared through normal cleaning.
Logged

dgberg

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2954
    • http://bergsprintstudio.com http://bergscustomfurniture.com

huh?  I'm pretty sure that says "2018" so unless you're living in the future Dan....

I'd just get a new LM cartridge-- the unstable nature of those missing nozzles does suggest that there is some issue with the pressure system, and that one having exploded confirms it.  Did you check your other carts?  If that's the only one experiencing the issue then it's most likely just a failure of an O ring or something else in that cart itself-- if any of the others also have this happening then you could have a bigger problem, but your nozzles all look relatively good except for the randomness of the dropout pattern, which is almost always caused by a bad seal somewhere causing air bubbles to be introduced into the lines.

The whole pressurization system is in parallel, so one failure like this could easily cause problems with the other ink lines as well.  Hopefully just replacing that one will fix it.

I meant the printer. Introduced in Oct. 2010, that makes it 14 years old from introduction, although it certainly could be newer. It really doesn't matter.
I would start with replacing the cart. Anything beyond that mechanically not sure I would spend anything on it.
My 9890 was the best printer I ever had. Never clogged. The head lasted 10 years which is exceptional. Gave it away for parts when the head went south.
Logged
Dan Berg
http://www.bergsprintstudio.com/workshops/
Our Studio is a Canson Certified Print Lab

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Hello Everyone!

Thank you for sharing your knowledge with me.

just applying observational logic, though the printer is not new and heads do die. the correlation with the fact the Light Magenta cartridge is the only cartridge to blowout and leak, and having difficulty with my test pattern, lends me to believe they are connected. I have not found anyone online yet who has experienced a blowout of a relatively full 350ml cartridge? This is not out of the rubber port. Clearly the ink has escaped all withing the cartridge. And that ink has leaked through the parting line of the container. I had only done gentle CL 1s and a couple CL 2s.

I have a fresh 700ml Light Magenta in it now. I will work on it this morning. Just so I don't blow out another Cartridge. Any precautions on cleaning cycles? I will only do pairs.Should I just avoid CL 1 since it is too gentle. Do an 8x10 after each cleaning?

I've only ever done one CL 3. Never a CL 4. I read about how it creates a vacuum in the capping station to clear the head. I don't believe it is dry ink. I print regularly enough. This printer is going to be 9 years old in January. but it never has commercial lever use. I've run many 24" rolls of Cold Press, and several 44" of Premium luster and Matte Canvas. Its always been powered up, as recommend to me.

Is it okay to just run CL 2s?

Your thought on CL 3 or CL 4? Any concerns on those? Are they just more vacuum, or do they do some other task? Should I attempt a CL 4.

I'll replace my maintenance tank today just in case.

Its a shame, I just got an email this morning from my client to ask for updated quote on the other sizes they need for their interior project.

Thanks.
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

HThis is not out of the rubber port. Clearly the ink has escaped all withing the cartridge.

this really isn't clear until you've opened the cartridge up, but it's also not all that important.  In thinking about this more overnight, you could be having an issue with the LM damper-- there might be some kind of starvation happening because of dried ink in the damper that is causing bubbles-- there really are so many potential issues, but you're not gonna be able to drill down to them unless you try out the steps that I've given you earlier and pinpoint the issue.

I personally hate to think of these machines as consumable products, which is why I've spent so much of the last 15 years of my life tearing them down and rebuilding them.  Peizoelectric heads like those used by Epson and most of the rest of the printing industry (except for the notable exceptions of Canon and HP) require levels of maintenance that simply can't be done automatically-- they really do require manual removal and cleaning and purging to stay in good shape over long periods of time, and they most certainly ARE consumables.  Epson seems to finally have acknowledged this in it's industrial level printers but creating methods for the user to be able to replace the heads, but they still sell their fine-art printers at a much lower price and expect the entire unit to be replaced when theres a single problematic nozzle.  It really sucks as a business model for someone who understands basic mechanical appliances, and I've done my damndest to learn as much as I can about how their systems work, and I've replaced the heads/dampers/capping stations, main boards, and just about everything else possible in these machines in my time with them.

I wouldn't be able to operate properly at this stage in my printing life without a RIP that allows disabling/remapping of a channel.  Unfortunately there aren't many of them that can do this, and they can be expensive.  Almost all of them can turn off a single color and reprofile without it.  It's when you get into the issues of needing to swap one color with another that most of them start to balk.  Ergosoft is great for this purpose-- Onyx can do it, Flexi can do it, Caldera, and Colorgate (to a degree) but the only RIP I'd really recommend to anybody who isn't an absolute professional or at least a completely obsessed nerd would be Mirage, and last time I checked I don't think you can swap channels.

Anyway, rant aside, I would stop doing cleanings.  You need to figure out where the issue is, and these cleanings are doing nothing but potentially exacerbating it.  If you indeed have some dried ink in the LM damper or ink system leading to the head you're not gonna be able to push it thru those tiny nozzles to clear it if it was strong enough to cause the ink to blow back out of the port into your cartridge (which is what I think happened, I dont think the cart itself leaked, but no way to know for sure without opening it.  You CAN put it back together if that's what you're worried about.

By the way, I'm not even sure what CL1 CL2 correspond to these days, I would assume that they are just more passes of the wiper.

Basic theory of operation is this:

Regular cleanings are a process in which the head uses the piezo elements to fire a bunch of ink into the waste ink pad, and then the wiper physically wipes across the surface trying to clean off any excess ink on the face of the nozzle.  This is in theory useful for ink that could be dried up on the nozzle plate surface, but as it actually fires the nozzles there is a chance that you could cause more damage to the head if there's a really bad clog in a specific spot, because the peizo crystals in the nozzles that are responsible use the ink as coolant, and if there is no ink flow to that particular nozzle, it could burn out.

Power cleanings on the other hand, position the head over the sealed portion of the cap (in the 9890/9900 there is only a single width cap, so it can only do 1 pair of channels at a time, hence why it takes so long.  Other printers, and the older 11880 would seal the entire head and do all nozzles at once)  This procedure builds up negative pressure using the pump inside the capping station, then releases a stopper valve which causes it to PULL ink through the head-- therefore not firing the nozzles at all.  This can be a much safer option to clear certain types of clogs as it requires no nozzle firing, and therefore no strain on the head itself.  It uses quite a bit more ink, but it also can actually force through stubborn particles without having to actually "shoot" them out of the nozzles.  If you're going to continue cleaning, I would only use this type on that channel right now, but I still think you should try my troubleshooting steps and post pictures back first so we can evaluate further.
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Thanks Less!

I began your recommendations last night. I went to the lowest level quality 1 720x720, turned off high speed. and the result did not look better. I'll continue your other recommendations. I loaded a 44" roll of premium glossy, did a bunch of 6"x42" Magenta gradients to make sure all Magentas were being used. Still those lines.

You brought up an interesting point, I have always only done paired cleanings in Maintenance mode which gives me 4 options CL 1-4. I only do a CL 2 which I thought was a stronger cleaning than CL 1. Only once have I done a CL 3. I will have to search and see if the Power cleaning is a different process I have not tried. (Update I did in standard mode a paired "Execute Powerful")

Funny with the Gloss loaded I printed a strip of black and whites and some colors, all with a solid 1" black border and they came out perfect, but one with a dark blue sky I could perceive the missing lines.

This morning I ran the regular clean, then the Powerful, and in between print a large gradient. Here are the three test patterns. The missing nozzles do all move around in the same center zone.  I made an animated GIf that cycles through the three as well. Any Analysis?

Should I try Windex?

Thanks

Max

PS, if i had one of the RIPs you mention, I could exclude light magenta and be able to print without the lines. So i understand if someone has a 9890 and the RIP could they just make a profile for me Sans Light Magenta so I can complete this Job. Or is that not how it works.
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

unless there's a scratch in that area of the LM nozzle plate, then it looks like a damper issue to me.  There may be a hardened piece of ink in there that is causing the nozzles not to prime properly and its moving around slightly.  This is a way more in line looking nozzle check for having that bad of a banding pattern.  If it were me I would take the head out, suck some cleaning solution through it in reverse, flush out the damper, then put it back in and see if there was any improvement.

Do NOT use windex.  The other thing that this looks like unfortunately is the beginning signs of delamination-- I don't know if the LM is close to one of the edges of the print head, but it doesn't necessarily matter, the adhesives can fail in the middle too.  If that's the case, theres really nothing you can do (other than buying a RIP or a new printhead).  And no, you couldn't have someone with a RIP make you a profile, it needs to be driving the printer to control the individual channels.  Gutenprint can do this for you, but it's not gonna be easy to set up if you don't have a good understanding of ink limiting, dot size usage, color channel crossovers, and the ability to make your own ICC profiles...  Plus I heard they stopped pushing builds of the mac version due to lack of interest anyway.

I'd do another power cleaning on that pair and if it still looks like that I think you're gonna have to start researching a new machine...
Logged

disneytoy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 394

Any instructions online as you say to remove the head for cleaning?

You think they is dried in in front of the nozzle facing paper or within the head before the nozzles?

How hard is it to get a look at the nozzles? I have a digital microscope?

Clearly the nozzles gaps are moving around but staying in their lanes for the most part.

I have a older Mac, So Qimage may be able to do what I need even if it is difficult?

Thanks
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

Im talking about an actual disassembly-- there's almost definitely not anything you could accomplish by wiping the nozzle plate, you've done so much of that already I'm pretty sure it's completely clean.  I remember there being a way you could see it if you manually moved it a little bit to the right of the pump cap with the right side cover removed and looked with a flashlight through the middle of some wiper assembly...

anyway there is some information to wade through here:  https://myx900.com/  also you can find the service manual online without much difficulty if you want to go that route
Logged

Richard.Wills

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 373
    • Photofusion Photography Centre

I'd get a bottle of Piezo flush, and follow their capping station / head clean process, before getting to close to the screwdriver set. https://www.youtube.com/watch?v=m8NkgKw7mDw

Years ago, I brought a very long parked Epson 4800 back from the dead (mainly because I was bored, and thought it might be fun)
Logged

JCD

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 63

Yea, the head is not dead YET. If you keep doing nozzle cleans it will be.

Replace the damper assembly before you do anything. It’s the first thing to do after cleaning the cap station and the bottom of the head ( as in the good video he posted. You can use distilled water or better piezo flush and bounty paper towels. Keep cleaning until you see no more ink there.

 If you’ve done all this replace the dampers, 90% chance that’s the culprit. Been there several times. Most people replace heads unnecessarily or blow them with excessive cleaning of blocked dampers. We all get impatient when in the middle of work.

John


quote author=Richard.Wills link=topic=143976.msg1261395#msg1261395 date=1728407408]
I'd get a bottle of Piezo flush, and follow their capping station / head clean process, before getting to close to the screwdriver set. https://www.youtube.com/watch?v=m8NkgKw7mDw

Years ago, I brought a very long parked Epson 4800 back from the dead (mainly because I was bored, and thought it might be fun)
[/quote]
Logged

Lessbones

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 291
    • Color Center

Eh, personally I disagree.  I'd be willing to bet that yeah while the whole head isn't dead, the LM channel is likely hosed from delamination.  I've brought plenty of totally caked over heads back from the dead, and cleared out many a damper issue, but i've never seen anything target a specific LINE of nozzles like that except for an adhesive breakdown.  If they were all deflected 100% of the time i'd say it was a scratch, but that kind of weird random blipping in and out usually means the seal is broken at that point IME
Logged
Pages: [1] 2   Go Up