QY6-0067 diagnosis

[sneezer2]

First of all, yes, I am aware this printhead has been superseded by QY6-0075. Not sure what the
difference is but anyway this one was found in a used mp610 that I bought about a week ago.
First date of service (from the Eeprom dump) was Aug. 2013. Total pages printed 855. So,
except what you will see following, it should be a nice clean printer and it does appear to be so.

Of course the nozzle check on acquisition was awful but after a thorough cleaning all the CLI-8
checks are very good, so I don't show them here.

This is the PGBK check after cleaning. Exactly the same as before I brought the machine home.
Although I have never seen this pattern before, I suppose many of you have, so I submit it for
your diagnosis. My "conjecture" though is that somebody blew a few resistors by trying to
print some text off a dry cartridge.

Now, here is the next curiosity. Printing from Windows notepad (and I am assuming this is done
automatically from the PGBK) the print quality is not good but could be considered at least
acceptable under not very rigorous criteria:

All the lettters are at least fully formed but some (line by line) appear lighter than others. Well,
I never knew and never asked what Canon was doing with all those nozzles but this reveals a
little something.

So I guess I could keep this and use it occasionally as a backup but my final question for some of
the experts would be what you think my chances are if I put in another print head (maybe new,
maybe not).

 

[Ink stained Fingers]

this is an electrical failure of a chip in the printhead and cannot be cleaned away. I would recommend to replace this printhead soon to prevent further damage to the main board, or to consider an upgrade of the current printer.

 

[sneezer2]

@InkStainedFingers
Thanks for your quick answer. I never thought of it that way but I guess there have to be some chips in there. Obviously, the few connection pads on the
back of the printhead cannot serve individually so many thousands of nozzles. So there has to be a bunch of logic in there and where else would it be if
not on some chips. I tend to agree with you although I am not so clear. However, if someone were to blow out resistors, it might appear as more of
a random pattern than as the clean block of failures shown on the nozzle check. As would clogged nozzles which I didn't expect anyway.

As for cleaning, no I don't expect that to be cleaned away but, not having thought of the chip possibility, I cleaned it as a first step to see what it would do.
What it did do is exactly the same as what it was doing before being cleaned.

As for damage, that of course is my main concern, although I was focussed more on damage to a new head than on possible "further" damage to the logic
board. At this point I am wondering if there is any way to check the outputs from the board to see if they are as they should be. I have some good (but old)
electronic test equipment. It seems access would be the main drawback. What do you think?

 

[sneezer2]

By the way my intention (at least so far) is to stick with the mp610. It's built like a tank and does everything I want.
I have a couple of them now and I seriously don't trust many of the newer models as I do this one. Comments?

 

[Ink stained Fingers]

there is a decoder chip to decode binary addresses to the individual nozzles, and this or similar regular patterns are typical for a printhead reaching the end of its useful life, you may print with the glossy paper setting to utilize the other black ink instead but that prints very slow and only would be a short term work around.
In order to tinker around with the electronics you would need a scope, a circuit diagram, access to test points and sample timing diagrams, the firing of the nozzle resistors is a matter of pretty short pulses only. And there is not much repairable anyway.

 

[sneezer2]

Yes, I like your characterization of it as a "regular pattern".

"a printhead reaching the end of its useful life". Hmmm! This one didn't last very long. On the other hand, we don't know what has been done to it.

Are you saying that the decoder chip is known to be in the printhead rather than on the board? It would seem to be the better choice but I am not
informed on that.

I don't think I will bother with settings to use the other ink. Would rather do what I can to get it working as it should. In the event of complete failure
it could be junked without too much loss. Besides, I still have another one working just fine.

I do have a scope and some logic probes. I don't have a circuit diagram and don't expect to get one. I do have a long career behind me of repairing
"old stuff" without the benefit of diagrams, etc. (teleprinters, fuel separators, radio transmitters and receivers, radar and collision avoidance, fire sensor
systems, whistle controls, gyro, depth sounder, whatever electronics you could find on a merchant ship).

In this case access appears to be the main problem. Timing diagrams I guess would be nice but not necessary as I don't think I would be looking for a
perfect analysis. It seems to me that finding (and accessing) output points between the board and the printhead and comparing them to see if all
are in some kind of a range might be enough to determine what can be saved and what has to be thrown away.

 

[sneezer2]

Just out of curiosity, how many nozzles are likely to be involved in this one fault?
Does anyone have an idea?

 

[Ink stained Fingers]

the chip is not available as a stand alone component, it is a chip without case, glued to the flexible ciruit board and then encapsulated with a drop of epoxy as I have seen it - nothing to repair at this place. one quarter of the pigment black nozzles are not addressed anymore.

 

[sneezer2]

Glad you have seen it. Not repairable. Of course not.

But considering the numbers, 1/4 of 512 = 128 nozzles not addressed, (Do I read the specs correctly here?)
this could be a fault at one input to the decoder. Possibly on one input and still on the chip or on the flexible circuit
or on the ribbon cable or at the logic board output. So evidently not a result of any misuse at all (my first
conjecture, #1 just under the first image) but more likely just a component failure "somewhere".

So which side do I throw out and which side can I keep? Or where can I look next to develop a better idea.

Always remembering that this path could be completely misguided, it is my experience that "repairs"
occur, first and foremost, inside the head of the technician and before any hardware even needs to
be touched.

Thank you again for your help. Very enlightening.

 

[sneezer2]

According to pictures in the service manual, access to at least part of the logic board is
available fairly early in the disassembly process. It is unclear at this point what portion
of the board is exposed, what the layout of the board is or whether the machine would
operate at this stage. I suppose it would take some effort to determine these things
but it seems to me that if the result is satisfactory, then application of a logic probe
to the outputs would at least reveal whether or not a stream of pulses (5V? or what?)
is present at each output and therefore offer some indication of whether the board is
damaged or not.

Any comments?