Hacking the HP45

Powder and inkjet printing
ghvanhoosesr
Posts: 6
Joined: Tue Nov 15, 2022 2:59 pm

Re: Hacking the HP45

Post by ghvanhoosesr »

Greetings Dragonator!

I'm having a doozie of a time trying to get this HP45 printhead and controller to work with Inkjet Commander.
I only have the older hardware in front of me; specifically, the HP45 Standalone V3.01 Controller.
From going through the code for both HP45_Standalone V3 and Inkjet Commander, plus comparing HP45_Standalone V3 code with the HP45_Standalone V4 code, it looks as though the V3 software is missing some code to actually trigger printing. (priming still works though) The V3 software does not appear to be set up to interpret any "Trigger" commands from Inkjet Commander.

What signal from Inkjet Commander to the HP45_Standalone V3 software will make the printhead print?
or,
Is there an older version of Inkjet Commander that will work with HP45_Standalone V3 software?
or,
Are there any code segments I can add to the HP45_Standalone V3 software that will make the printhead print?
User avatar
dragonator
Site Admin
Posts: 575
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

There are basically 2 ways the controller currently can print. The first is through an encoder. This is how it was originally designed for Oasis. The controller keeps it's position and when it sees the to be printed image is passing under the printhead in the right direction, it prints the image.

The second way is through virtual motion. This takes an expected speed, and a trigger, software or hardware, to determine a position. After that it handles the actual printing of an image the same. Both ways tell the head that it is moving and printing should happen in the right locations.

The older versions have not been supported for a while, but the virtual motion should be hackable into the older V3.01 controller. From memory, the biggest difference is the lack of the 74HC123 chip, that was needed for DMA printing. Your head should print a bit slower, but should still be able to handle things.

If you have a quadrature encoder you can try hook that up. Inkjet commander might (no guarantees though) work, and this is designed to simply send images and let the user handle the movement of the printhead.

https://github.com/yvodehaas/Inkjet-commander
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

dragonator,

I have some questions regarding printing with HP45 using controller version V3.02. Recently I did some tests with regular image printing (not SVG) and I noticed some weird, incorrectly sprayed vertical lines of ink. Here are the images:
Screenshot 2022-11-26 130307.png
Screenshot 2022-11-26 130307.png (502.81 KiB) Viewed 427 times
Screenshot 2022-11-26 114149.png
Screenshot 2022-11-26 114149.png (450.34 KiB) Viewed 427 times
I marked the direction of swipe. It does happen at the beginning and also during the swipe. Sometimes there are multiple mis-jets sometimes there are only single vertical lines. That is really what I do not understand.

Also, perhaps you have experienced with the issue I have outlined in red, where initial sprays produce some foggy/cloudy like areas? It is weird as subsequent lines of swipes produce sharper starting edges of the image as you can see. The ending edges also tend to be relatively sharp. DPI is ~200. Interestingly, new black ink HP45 produce sharp images/edges as you can see, but head with my own ink tends to have this issue. It also seems that the longer pause between each swipe, the more chances of getting this cloudy start. It might be related to the binder liquid it self, perhaps it dries too quickly and that results in some kind of failure to start smoothly, but I do not know for sure now.

Thanks.
User avatar
dragonator
Site Admin
Posts: 575
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

I marked the direction of swipe. It does happen at the beginning and also during the swipe. Sometimes there are multiple mis-jets sometimes there are only single vertical lines. That is really what I do not understand.
There is a known issue that on one side the nozzle table has a few swaps in it. I have had this on my list for a while, but I have not had much time for HP45 related stuff in the last while. This shows itself as lines in the printing direction itself, that clearly are supposed to be printed on a different height.

What your images are showing are lines perpendicular to the printing directions. These are a bit more of a mystery to me. Can you share (either with an image or in text) the command line of the image being sent? I am curious to see if this artifact is sent, or if it comes from the printhead itself. I have not ever seen these lines in my prints.
Also, perhaps you have experienced with the issue I have outlined in red, where initial sprays produce some foggy/cloudy like areas? It is weird as subsequent lines of swipes produce sharper starting edges of the image as you can see. The ending edges also tend to be relatively sharp. DPI is ~200. Interestingly, new black ink HP45 produce sharp images/edges as you can see, but head with my own ink tends to have this issue. It also seems that the longer pause between each swipe, the more chances of getting this cloudy start. It might be related to the binder liquid it self, perhaps it dries too quickly and that results in some kind of failure to start smoothly, but I do not know for sure now.
This issue I have seen before. This to me happens with older heads and heads that have not printed for a while. This is the ink drying in the nozzles. What has worked for me is soaking a rag in isopropyl alcohol and properly cleaning the head with this. For your own ink I do not know what else might help, but for me the issue has mostly been ink drying in the nozzles.
majian
Posts: 1
Joined: Sun Dec 18, 2022 12:57 am

Re: Hacking the HP45

Post by majian »

Hello Dragonator

I have a project for custom printer with HP45 printhead,
I want to start with C8855 manuel, is it possible to share the C8855 manuel?

Thanks!
rofl
Posts: 56
Joined: Mon Aug 26, 2019 9:13 pm

Re: Hacking the HP45

Post by rofl »

dragonator,

I am sending you text file with terminal output/log. Also here is the image of what was the typical output. Log consists of two prints of this image, but image is only one as they are similar, but not identical. This image always fails to print during the second swipe, so it prints just the first swipe and half of the second one, but that is expected. DPI is 300.
image_2022-12-18_224516247.png
image_2022-12-18_224516247.png (184.55 KiB) Viewed 296 times
oasis3.zip
(10.54 KiB) Downloaded 17 times
Just looking the the output, I can hardly see those vertical lines, am I wrong?

Thanks.
User avatar
dragonator
Site Admin
Posts: 575
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

I also do not see the lines. What I do see however is a lot of weird errors popping up.

Code: Select all

Dec 17 20:56:56 raspberrypi oasis: Qt: XKEYBOARD extension not present on the X server.
Dec 17 20:56:56 raspberrypi oasis: QStandardPaths: XDG_RUNTIME_DIR not set, defaulting to '/tmp/runtime-pi'
Dec 17 20:56:56 raspberrypi oasis: qt5ct: using qt5ct plugin
Dec 17 20:56:56 raspberrypi oasis: Xlib:  extension "RANDR" missing on display ":1.0".
Dec 17 20:57:07 raspberrypi oasis: QObject::connect: Cannot queue arguments of type 'QTextCursor'
Dec 17 20:57:07 raspberrypi oasis: (Make sure 'QTextCursor' is registered using qRegisterMetaType().)
Dec 17 20:57:09 raspberrypi oasis: QObject::connect: Cannot queue arguments of type 'QTextCursor'
Dec 17 20:57:09 raspberrypi oasis: (Make sure 'QTextCursor' is registered using qRegisterMetaType().)
These errors also pop up when you are sending some lines. For instance:

Code: Select all

Dec 17 20:56:36 raspberrypi oasis: SBR qry //////wAP///////////8AAAA
This is the last line of a sweep. A good sweep always ends with a series of A's, this is all nozzles off, like in this line:

Code: Select all

Dec 17 20:56:15 raspberrypi oasis: SBR viq AAAAAAAAAAAAAAAAAAAAAAAAA
If a line is not terminated at the end of a sweep, weir behavior can happen whenever the printhead tries to start again, with an existing line in the buffer. I think the errors are the cause of these errors. Can you verify the code works on a PC, where hopefully these errors do not occur? If that is the case, we can look into removing these errors.
amirslz74
Posts: 3
Joined: Tue Aug 23, 2022 2:00 pm

Re: Hacking the HP45

Post by amirslz74 »

Hi
I followed the Arduino board you made but have a problem.
After a few print, the color gradually fades until it becomes completely colorless, but it is quite good at the first use
After one day, when I want to use the cartridge, the same thing happens again, and it is well at first, but gradually it fades.
Thank you for telling me if you have any experience.
User avatar
dragonator
Site Admin
Posts: 575
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

I have not had this issue myself. It does sound heat related though. The slow buildup and the resetting behavior makes me thing that the printhead is overdriven. Do you have a way to lower the voltage on the primitives without lowering the voltage on the addresses?
amirslz74
Posts: 3
Joined: Tue Aug 23, 2022 2:00 pm

Re: Hacking the HP45

Post by amirslz74 »

dragonator wrote: Sun Oct 16, 2016 11:57 am Got a bit of work done today. I wrote a bit of software in my least favorite used program, processing. This because it is the best environment that requires no extra skill for me. What the program does is it imports an image, scales it to 150 by 150 pixels and converts it to bilevel. After that it can convert the file into an array that can be loaded into an arduino program as a constant.

I have attached the program (that needs to run in processing) and the current V2 firmware with the code. Especially the processing code is a mess, but for now it is intended for one time use. The firmware still has some issues getting all the pixels to fire in the right moment, but for now I am happy with it.

Converter user guide.jpg

And now a video of the code printing in order: the Ytec logo, the jolly wrencher and a banana.



161016 files.zip
I use that file for code and arduino shematic with 4017 for addreses and tcl50213 for primiritive
Post Reply