Electrical failure? Pro 9000 Mark II

[Nirvash]

Well after a long day of testing, I can say that I am nowhere closer to figuring out what causes the issue.
Despite my best efforts I can't figure out a pattern of what causes it. Using all Canon carts seems to get a clear output, but that may just be due to having less of them.

Anyone of have suggestions?

Normal test page, showing variation in color i.imgur.com/uAxmDSL.jpg
Here is a high resolution version (cropped) of that, i.imgur.com/MHsEVyE.jpg The oddest thing I found looking at that is the bottom line of the magenta test, it looks nearly cyan.
Which seems to cause this on prints i.imgur.com/28R1qMI.jpg

 

[The Hat]

On reviewing your three JPEGS you do seem to have some cyan leakage rather than
a problem with the magenta, But and heres your problem.

You say in your post when using all Canon carts the output is clear, but what isnt clear is the next bit in the same line
but that may just be due to having less of them. Please explain what exactly do you mean by less of them ?

Now does that imply that youre using some cartridges that are not the original OEM inks and cartridges that came with the new printer ?

To test you printer correctly you should have one set of values that dont change for that test and then study the print output.

For your next test you change only one variable and test print again and so on till your satisfied
you have found the exact cause of your varied output, only then can you begin to fix the problem..

 

[Grandad35]

The 1st image was generated from your high resolution cropped image. The "levels" were adjusted to darken the entire image; it was then converted to the CMYK color space and split into individual C/M/Y/K channels. The yellow and black channels were clean and are not shown, but the cyan and magenta are interesting. As you stated, the bottom row of dots shows darker cyan, but this is not the only area that shows darker than expected cyan:

The 2nd image is the cyan channel cropped from the magenta test, following the same procedure:

This is beyond strange, especially how the pattern changes in the second image. I agree with The Hat that the print head appears to be leaking cyan into the magenta (aren't these 2 carts next to each other?), and that the volume of the leak and where it comes out changes as the ink usage changes.
If you set the print head on a folded paper towel, the various inks will wick into the towel and you may be able to see the cyan in the magenta.

 

[Nirvash]

On reviewing your three JPEGS you do seem to have some cyan leakage rather than
a problem with the magenta, But and heres your problem.

You say in your post when using all Canon carts the output is clear, but what isnt clear is the next bit in the same line
but that may just be due to having less of them. Please explain what exactly do you mean by less of them ?

Now does that imply that youre using some cartridges that are not the original OEM inks and cartridges that came with the new printer ?

To test you printer correctly you should have one set of values that dont change for that test and then study the print output.

For your next test you change only one variable and test print again and so on till your satisfied
you have found the exact cause of your varied output, only then can you begin to fix the problem..

Sorry should of explained better, what I mean by that is that I only have one set of Canon carts, so I can't run quite as many tests on them before they run out.

What has made this so frustrating is I have been trying to go through changing one thing at a time till I get the same output error but I have been unable to reliably reproduce it. I can get perfectly fine tests sometimes, then they just go bad.

The 1st image was generated from your high resolution cropped image. The "levels" were adjusted to darken the entire image; it was then converted to the CMYK color space and split into individual C/M/Y/K channels. The yellow and black channels were clean and are not shown, but the cyan and magenta are interesting. As you stated, the bottom row of dots shows darker cyan, but this is not the only area that shows darker than expected cyan:

The 2nd image is the cyan channel cropped from the magenta test, following the same procedure:

This is beyond strange, especially how the pattern changes in the second image. I agree with The Hat that the print head appears to be leaking cyan into the magenta (aren't these 2 carts next to each other?), and that the volume of the leak and where it comes out changes as the ink usage changes.
If you set the print head on a folded paper towel, the various inks will wick into the towel and you may be able to see the cyan in the magenta.

Separating the image into channels does make it a lot more visible and yes, I do believe that Cyan and Magenta are next to each other. And I will make sure to test the wicking when I get home

 

[Nirvash]

So I had a chance to test ink wicking through the printhead.
Just leaving the print head on a paper towel easily shows that the head matches the cart arrangement, G R PM BK PC C M Y
i.imgur.com/BK6YGW1.png
Though not much wicks out, not enough to tell anything for sure and it is all very close together. So what I did next was I held a Cyan cart on the input for the print head and tested to make sure which nozzle was Cyan, which was easy based on the massive amount that wicked through.
What I did for this next picture was to hold the paper towel up to the nozzles just next to the Cyan and see if anything bleed out of them.
i.imgur.com/TeDDJsC.jpg
Something did, what should be Magenta and Yellow, ended up being Purple and Green. Which would support the idea that Cyan was leaking into the other colors in the head.

 

[mikling]

Sounds to be an improperly assembled printhead. possibly one of the gasket inside is not aligned properly. A bad Apple. If it is under warranty, get it replaced first After it is replaced, then you might consider disassembly and reassembly. Grandad35 is good at that, there was once a thread on doing it. I don't have much luck in doing these.

 

[Smile]

Sounds to be an improperly assembled printhead. possibly one of the gasket inside is not aligned properly. A bad Apple. If it is under warranty, get it replaced first After it is replaced, then you might consider disassembly and reassembly. Grandad35 is good at that, there was once a thread on doing it. I don't have much luck in doing these.

Isnt the gasket glued?

 

[Grandad35]

Isnt the gasket glued?

There are several places where the ink could leak in the print head. The first is inside the print head body. This is a glued joint and is not accessible. The second is at a rubber gasket between the print head body and the ceramic plate. The gasket pilots over small nibs on the print head body to hold it in place for assembly. That same link also describes how to get to this gasket. Since your new print head is under warranty, I wouldn't attempt to take it apart. The third is where the thin nozzle plate is glued onto the ceramic plate, and is also not accessible.

Even if there was a leak at the rubber gasket, it would probably leak to the outside of the print head instead of into another channel.