Epson 3800 with a clogged magenta channel

[Nigel Harris]

Thanks for the feedback everyone. Not had chance to move this forward this weekend, but a thought (and question) did occur to me, and it was regarding when I initially filled the cleaning cartridge.

Here are the instructions I received when I bought the cartridges (from Marrutt) they assume that the cartridge is being filled with ink, rather than cleaning solution:

Step 14 goes on to say put the cartridge in the printer etc etc.

Now, here’s my question. I’ve read elsewhere about having to “prime” the cartridge, by drawing some ink out before fitting. I didn’t do anything along those lines. Am just wondering if there is air in the line or damper, and perhaps that might be why I’ve not seen anything from the Magenta line.

Alas I don’t have a black light, and amusingly I then found myself learning all about how you can (maybe) rig up your smartphone led light to become one by taping over the led and using marker pens. But then it depends on the smartphone, and... another rabbit hole to descend down! So, I’ve not pursued that line of enquiry yet.

I’ll post back when I’ve done some more investigating.

 

[Nigel Harris]

Quick update, for your entertainment. Snatched a bit of time to add lots more more dye to the cleaning cart. Took the opportunity to empty it out completely and then refill, and to “prime” using the technique shown here: https://www.youtube.com/watch?v=OvrjjF-Gdf8

I definitely did *not* do the priming thing first time through, so I think I may have introduce air into the magenta line....possibly....

Anyhow, all of the above seems to not made a blind bit of difference. Am feeling fairly stupid at the moment: I’ve printed 27 a4 pages of a magenta purge file, am seeing the ink % drop on the printer display, am seeing the ink level drop in the cartridge (see photo), but I am not seeing any difference in the output. The purge image looks the same as before I added more dye, and the nozzle check is showing blank for the magenta channel. Yes, I am aware that sounds like a contradiction and I don’t understand it! At the moment, the score is: Printer 3, Nige 0.

The hypothesis about the dye just not showing up is still standing (not forgotten the suggestion about the black light).

The hypothesis that the ink just hasn’t come through yet doesn’t seem to stack, 27 x A4 pages? Surely that’s enough? As can be seen from photo, must be more than 15ml of ink used?

Wondering whether to dismantle the printer again, a job for the weekend, but this time look to see if there is a way of physically pulling the ink up to the print head.

I wanted to hold off buying non OEM ink until I had got the printer fully functional again, however ink levels have been steadily reducing as I have been printing purge files, so am now unsure how to proceed on that front...

Current levels as follows:
Matte black 58%
Photo black 55%
Light black 37%
Light light black 53%
Cyan 19%
Magenta 100% (cleaning cartridge. Did reduce to 92% three times during purges then reset)
Light cyan 78%
Light magenta 22%
Yellow 76%
Maint cart 57% (but now have a reset tool & spare cartridge)

Speak soon!
Nige

 

[Nigel Harris]

I'm back onto this after a long hiatus. I did end up buying a cheap UV torch, which showed nothing of interest on the previous test prints. It occurs to me that maybe next test should just be to put a new magenta cartridge in, and print the purge images off until it comes through (or not). This will tell me if the channel is same as before, completely blocked or fixed. It may be the dye in the cleaning fluid was too dilute, evaporating or doing something else I don't understand. Wish me luck Will post photos when I've done some more testing, hopefully tomorrow.

 

[Nigel Harris]

Well, having had the magenta line soaking in the cleaning solution for weeks, I finally changed the cleaning cart out for a full epson magenta cartridge. The ink came through after purging with the purge image, thereby proving that the nozzle wasn’t completely blocked. I conclude that the cleaning solution did not have enough dye for it to become visible on a nozzle check.

Anyway, I ran the nozzle check once the magenta was coming through nicely, and i was surprised to see the same gaps in the pattern as when i started. So, the cleaning cartridge and the head soaking seem to have not solved the problem.

I think my next step is to replace the damper for magenta. I have one I ordered a while back, and I have ordered a full set although that’s weeks away. I will have a go at taking photos as I go, and see whether replacing just the magenta damper is going to do the trick. Wish me luck

 

[Nigel Harris]

OK, well I guess if anyone is still following this they deserve a medal... welcome back Warning: long post, and lots of pictures.

If you recall, I started off with an incomplete nozzle check, which looked like this:

And having soaked the head, and line with cleaning solution for weeks(!), printed lots of purge images (draining my LM in the process) and having taken out the capping station and given that a good cleaning, I ended up sticking a magenta cartridge back in to see what would happen. And got this:

So, problem not solved. Consider this thread as a cautionary tale, I think, but with pictures! I decided I should replace the magenta damper, and I had a single new damper, so off I went.

Half disassembled printer. Remain impressed with the build, although you must watch those thin wires when taking off the top cover.

Finally got to the print head assembly.

Removed the connectors.

Removed the damper assembly from the print head, here's the damper assembly.

I didn't understand how the motor on the damper assembly worked. There are a small set of gears that move a "pincer" that presses on one side of the damper or the other.

I concluded this is how the switching mechanism works for MK / PK. Seems really complicated to me, and if anyone is reading this, I'd appreciate understanding the need for such complexity when (in my mind) it's just another colour...

Anyway, I took this off, as I needed to get to the magenta damper. If anyone does this in the future, a couple of key points. In order to get this "pincer" off the damper, I needed to manually turn the motor spindle until the pincer was in a neutral position. Lesson number 1 - make a note of the current position for when you're putting it back together!

Lesson number 2 - there's a microswitch which rests against the white circular cam that is turned by the motor. The microswitch is the thing that tells the printer that the "pincers" have reached a certain position.

When dismantling the components I lost the position of where that microswitch should be sitting, and re-assembled it at the wrong angle. Then when trying to switch black inks later as a test, I got a nasty noise as the motor tried to keep turning the gears when they couldn't go any further. So, don't do that. For reference, it actually has a little lug on it, and Epson was good enough to drill a reference hole to locate it.

It is then secured by a single screw. Here it is in the correct position:

Anyway, that wasn't why I was taking the printer apart. I now had access to the dampers (and the print head!).
Next step I removed the ink lines from the dampers, so I could work on the component on the bench.
Taped up the ink lines so they wouldn't drain out.

Assembly on the bench.

And how the print head now looked:

An opportune time to gently push some cleaning solution through the magenta channel, having placed kitchen towel underneath the print head in readiness.

Foil covered ink tunnels need to be removed before you can get the damper out.

Carefully prise off.

Leaving two components.

Thought it would be worthwhile shoving a bit of cleaning solution through the magenta ink line.

Original magenta damper removed.

Here's the old and new damper side by side, not quite identical, but same function I guess.

I wanted to pull some cleaning solution through the line and through the damper, I had read an excellent guide on this on another website (inkjetmall), due credit to jkschreiber's post which was really helpful. This indicated that in order to make the valve open in the damper, the cleaning cart needed to be pressured with 20ml of air. Something along the lines of this:

Lesson number 3 - my 10ml syringes weren't enough!
Here's a photo of me trying to pull the fluid through the damper. It looks like I'm pushing, but I was pulling - I was using an ipad and I think I pressed the photo button with my nose, not enough hands!

I ended up buying some 80ml syringes off amazon, and using the same technique finally I managed to pull the solution through into the damper.

Reassembly was fairly non-eventful, but I did want to test it before closing everything up. The display can be reconnected via the ribbon cable, pile of tissue boxes and books is not essential, but helpful. The problem with that is there is a sensor in the top cover that detects if the ink cover is open or not, and the printer won't play nicely if it's not closed. The mechanism is the bit near my set of screwdrivers at the bottom right of this picture:

Closeup:

And here's how you fool it (I suggest a bit of paper folded over):

So, having re-assembled, then taken it all apart again to fix the MK/PK switching problem I created, I finally put it all back together and discovered I DIDN'T HAVE ENOUGH INK TO RUN A CLEAN. So I await a light magenta cartridge, after which I'll run a power clean, and hopefully things will be improved.

I'll post again once I've got some ink in the machine, along with my test results. Any comments about stuff I've done wrong, or things I've not done and should do, are all appreciated. Hopefully I'll get a clear nozzle check once I've replaced the empty ink cartridge.

Cheers
Nige

 

[The Hat]

Wow there’s enough Pics there to keep the POW going for months..
Best of luck with your nozzle check..

 

[Nigel Harris]

Wow there’s enough Pics there to keep the POW going for months..
Best of luck with your nozzle check..

Thank you. Well, finally sourced some light magenta. Ran a cleaning cycle, then a nozzle check.
So, after having done all of that maintenance, I appear to have improved the situation from 12 gaps in the nozzle check, to 10 gaps in the nozzle check.

I'm a fairly tenacious person, however I must say I'm now wondering whether to live with it. Unless anyone has any bright ideas, I'm going to move on from the maintenance, ignore the nozzle check and just run some real-world images and see if there are any visible indications of banding, gaps etc. Was intending to use this machine to do invites and other stationery for my daughter's wedding next year, am now unsure as I wanted it "right" before embarking on that project.

I have these nagging questions:-
1. Was the cleaning solution I mixed up strong enough? It was left in the cleaning cart and print line for weeks and weeks (not deliberately, but just driven by available time) - it appears to have done nothing;
2. The damper clearly wasn't the problem? So where's the problem?
3. If this is an electronic problem with the print head, would the nozzle check look like that, or would it fail completely? Can you get the equivalent of "dead pixels"?!

Any advice appreciated.
Nige

 

[The Hat]

What I would suggest now is... Stick in 30 or 40 scrap sheets of paper into the feeder and run them as test sheets, it can’t do any harm but it may clear up your problem by getting the printers juices going..

Try this as a test sheet and print on normal mode…

 

[Nigel Harris]

What I would suggest now is... Stick in 30 or 40 scrap sheets of paper into the feeder and run them as test sheets, it can’t do any harm but it may clear up your problem by getting the printers juices going..

Try this as a test sheet and print on normal mode…

Thanks again... I've run this off (just a single copy for now), on standard a4 inkjet paper. I've photographed the output, along with a close up of the magenta output. I am thinking that just using the printer will be the best course of action, I'm not an expert in this regard, but I cannot see any major defects in the magenta output compared to any of the others. Maybe I should move onto some better quality paper and see how I go with that.

 

[The Hat]

I've run this off (just a single copy for now), on standard a4 inkjet paper.

Just use a couple more rubbish sheets, but there’s no need to waste a whole page of the expensive stuff, cut a small piece and only print the red and magenta..
The sample test print is looking good..
Just remember 1 swallow doesn’t make a summer, so you may need to print a lot more to be sure you stay clog free..