Page 2 of 17

Re: Hacking cartridge HP 84/85

Posted: Sat Sep 01, 2018 7:23 pm
by MAsic12345
Here's how the cartridge divides the nozzles into print
I pulled out the cable from the outlet
the printer prints a black square

Re: Hacking cartridge HP 84/85

Posted: Mon Sep 03, 2018 4:21 pm
by dragonator
Judging from the waveform, I am pretty sure 9 and 13 are data pins. Perhaps each pin does one side of the printhead, or top and bottom, since the black square is one continuous burst of data (I suspect the gap in the 3 columns is bigger than 4mm, and with the black square it is smaller, is this correct?). I think 19 is the most likely clock. I then think 5 and 16 are trigger pins for either side. What 14 is doing is a bit of a mystery to me. It is helpful that you labeled the pins. Knowing what contact each channel is is really helpful.

The overlap is something I never really thought about, but I have seen it before. I wonder what the advantages are of printing this way.

What are the waveforms captured in? Are the files something other people can open? I would be curious to see how the channels look up close.What frequency are the channels? Is it something you could reproduce on a microcontroller?

Re: Hacking cartridge HP 84/85

Posted: Mon Sep 03, 2018 4:45 pm
by MAsic12345
perhaps the speed of the shield of the analyzer was small and therefore the clock did not catch

Re: Hacking cartridge HP 84/85

Posted: Mon Sep 03, 2018 5:12 pm
by MAsic12345
Is it possible that 9 and 13 are responsible for the right and left nozzles, and not to the top and bottom?
you said (I suspect that the gap in 3 columns is more than 4 mm) what did you mean
what do the trigger contacts do or where I can read about it?
but the photos of the logical analyzer of the number of lines of logic coincide with the numbers of the print head,I gave each channel the number that is responsible for the contact number on the cartridge.

break - I think that this way of printing increases dpi
Is it something you could reproduce on a microcontroller?
What exactly?

Re: Hacking cartridge HP 84/85

Posted: Tue Sep 04, 2018 10:47 am
by dragonator
MAsic12345 wrote: Mon Sep 03, 2018 5:12 pm Is it possible that 9 and 13 are responsible for the right and left nozzles, and not to the top and bottom?
Now that I look at it better I am not sure they do different things. 9 and 13 are at different voltages according to your previous posts. That can not be doing the exact same thing if that is true. What the difference is then though, I am not sure.
MAsic12345 wrote: Mon Sep 03, 2018 5:12 pm you said (I suspect that the gap in 3 columns is more than 4 mm) what did you mean
If the gaps is more than 4mm (the distance between the 2 rows of nozzles on the printhead(?)) then the nozzles will stop firing for a short while while the printhead is in the white gap. If the gaps is narrower than 4mm, the right row will start firing again before the left row reaches the white are, so there will always be data sent to the head.
MAsic12345 wrote: Mon Sep 03, 2018 5:12 pm Is it something you could reproduce on a microcontroller?
If you take a perfect copy of what the printer is sending to the printhead, at the right voltages, and send it to your printhead using a microcontroller, it should do exactly the same thing. You can then slightly change what you send to the printhead (removing and adding pulses) and see what happens.

There is not much more I can do from here right now sadly. I do not have the time nor the tools right now to properly help.

Re: Hacking cartridge HP 84/85

Posted: Wed Sep 05, 2018 10:41 pm
by Wonko
I would recommend that you look at all the signals with an analog scope as well. What looks like neat square waves in a logic analyzer may actually be a quite complex analog signal. It is actually rather unlikely that the ink cartridges use MCU level digital inputs. Those are throw-away cartridges. Why would HP put logic chips on them with power drivers and heat sensors? Those chips are expensive, even in lager numbers. More likely, every nozzle driver is a FET or whatever semiconductor that shorts out, heats up, and fires the drop. If that is so, then - like in the HP45 - temperature control is one huge issue to keep drop size consistent and to keep the FETs from burning up. Timing and a power stage is the next huge issue. Think about it: how many drops do you have to produce in the 0.7 seconds that 7 ink heads need to print a thick line across the paper. Calculate back und you find the MHz of transmission speed on those lines, and why your digital scope may not show you the truth about the signals at all.

Happy hacking.

Re: Hacking cartridge HP 84/85

Posted: Thu Sep 06, 2018 7:45 pm
by MAsic12345
Here is more detailed measured and checked
previous ignore there many mistakes made
I checked the file again and corrected errors

Re: Hacking cartridge HP 84/85

Posted: Thu Sep 06, 2018 7:57 pm
by MAsic12345
I made a more detailed digital sample and it covered the whole time of the whole printing
I'm sure that 9 and 13 are responsible for the left and right row of nozzles
One can see on the oscillograph the ninth terminates faster than the thirteenth
this is the same oscillogram with each photo scale increases

Re: Hacking cartridge HP 84/85

Posted: Thu Sep 06, 2018 7:58 pm
by MAsic12345
closer

Re: Hacking cartridge HP 84/85

Posted: Thu Sep 06, 2018 8:01 pm
by MAsic12345
I would recommend that you look at all the signals with an analog scope as well

why do you think that this can be analog?