Hacking cartridge HP 84/85

Powder and inkjet printing
User avatar
dragonator
Site Admin
Posts: 441
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking cartridge HP 84/85

Post by dragonator » Thu Sep 13, 2018 5:49 am

What file format are your logic analyzer files in? Can the files be opened in something that is fairly easily downloaded? The images of the waveforms worked so far, but now when you zoom in enough that I can see the details I need, I no longer see enough to see a few bursts of the head. Are the files sharable or are they too big for that?

MAsic12345
Posts: 96
Joined: Sat Aug 25, 2018 5:26 pm
Location: Poland

Re: Hacking cartridge HP 84/85

Post by MAsic12345 » Thu Sep 13, 2018 8:09 pm



MAsic12345
Posts: 96
Joined: Sat Aug 25, 2018 5:26 pm
Location: Poland

Re: Hacking cartridge HP 84/85

Post by MAsic12345 » Sun Sep 16, 2018 6:36 pm


User avatar
dragonator
Site Admin
Posts: 441
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking cartridge HP 84/85

Post by dragonator » Mon Sep 17, 2018 5:21 pm

I will take a better look at the scope file when a. I have found or made a reader, and b. have time. But so far all there seems to be little new, other than the fact that all data seems to be at 2.3V (2.5V logic level probably)

8 and 11 take 12.8V and are probably power for the addresses. 17 and 18 take 9.6V (which probably varies with temperature) and is I think for the primitive power side. I suspect that the pads 17 and 18 have a decent amount of current going through them, they have bigger contacts. 5 is a clock pin, and while I would expect a square wave, not triangle, but it is the highest frequency pin, and it does not matter much. While you said 1.3V (which struck me as odd) on the scope it shows peaks of 2.3V. 9 and 13 are data pins, operating at 2.3V. 19 is the trigger signal. While it is high, the nozzles are firing It too seems to be 2.3V, not 1.4V.

Only remaining pins that I do not know are: 14, Does have some sign of signals, but only when the head is not printing from what I can see so far. It probably does feedback. It too is 2.3V. 15 gives a low signal (2.3V) every 11 bursts. This seems odd to me because everything in the head should be in orders of 22, so what exactly happens here is a bit of a mystery, but it does seem to be related to the address. It also gives it during the first high of 19, adding a bit to the mystery. 16 only gives a signals of 0.04V, but gives a consistent square wave of 2.5us, same as the trigger signal. Whether this is noise or important I do not know.

Are you sure the remaining pins carry no signals (the small ground pads and 10?). If so I think this is what all pins do.

Now it becomes a bit of a question what you want to do. I do not have the time right now to pursue another head, and I will not have much time for quite a while. At some point you will have to take a microcontroller and try to recreate this signal. If the printhead prints with that signal, you know how to drive it and all you need to do is figure out some of the details like some pin functions and the nozzle mapping. If it doesn't you are missing a signal.

MAsic12345
Posts: 96
Joined: Sat Aug 25, 2018 5:26 pm
Location: Poland

Re: Hacking cartridge HP 84/85

Post by MAsic12345 » Tue Sep 18, 2018 8:13 am

to which microcontroller to pay attention
on what data does 9/13 output to SPI or I²C or another?

User avatar
dragonator
Site Admin
Posts: 441
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking cartridge HP 84/85

Post by dragonator » Tue Sep 18, 2018 6:09 pm

If you have never used anything I would not know. Most people pick what they are accustomed to. I like arduino based coding, so in this case I would pick a Teensy (3.2 or even 3.6, considering the speed required). However a lot of different fast micro controllers exist (or even FPGA's).

You will probably need a fast level shifter because most micro controllers do not operate on 2.5V (Teensy is mostly 3.3V).

While I would say it most closely resembles SPI, But I think the protocol is custom, close to what the signal on a shift register looks like.

MAsic12345
Posts: 96
Joined: Sat Aug 25, 2018 5:26 pm
Location: Poland

Re: Hacking cartridge HP 84/85

Post by MAsic12345 » Mon Sep 24, 2018 8:27 am

Corekt

MAsic12345
Posts: 96
Joined: Sat Aug 25, 2018 5:26 pm
Location: Poland

Re: Hacking cartridge HP 84/85

Post by MAsic12345 » Mon Sep 24, 2018 8:27 am

Corekt
Attachments
IMG-1d4091b585329a6cb16fb76cf63129e2-V.jpg
IMG-1d4091b585329a6cb16fb76cf63129e2-V.jpg (187.73 KiB) Viewed 2724 times

math
Posts: 21
Joined: Tue May 24, 2016 1:08 am

Re: Hacking cartridge HP 84/85

Post by math » Mon Oct 01, 2018 4:57 pm

Seem to me that signals are very compatible with hp45'structure: 22 address x 14 primitives. In 5 we have groups of 7 signals. 2 of this groups equal 14 primitives. Each signal in 15 embrace 11 groups of 5 = 11x7 = 77. Two groups of signals in 15 embraces 154 of signals in 5. Four groups of 15 results in 308 signals in 5, wich corrrespond all nozzles to be fired in a HP45 cartridge. I think that signals in 5 are clocks to obtain nozzles data to be fired from 9 and 13.

Post Reply