Hacking the HP45

Powder and inkjet printing
User avatar
dragonator
Site Admin
Posts: 595
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

Do please let us know when you have actual results. I myself am sadly extremely busy preparing for SHA2017 right now, but I can't wait to see how the L6452 works. I have testing that high on my list.
davidk
Posts: 77
Joined: Sun Mar 12, 2017 6:48 pm

Re: Hacking the HP45

Post by davidk »

Of course. This is a 50-50% hobby-professional project and I don't have much time, otherwise it would have been finished. But I can share it here so I'll let you know about my conclusions.
chanhnguyen28
Posts: 1
Joined: Sun Jul 23, 2017 10:33 am

Re: Hacking the HP45

Post by chanhnguyen28 »

Hi dragonator

I want to buy HP45 connector socket.Please help me how to order? link or store website... thanks
Attachments
connector socket.jpg
connector socket.jpg (180.32 KiB) Viewed 13587 times
Jurif
Posts: 6
Joined: Mon Aug 15, 2016 10:38 am

Re: Hacking the HP45

Post by Jurif »

I have a question for those who have gotten the printhead to function properly. I am having few problems with getting the printhead to function as it should. While debugging I noticed that the nozzles are getting really hot to the touch (probably over 50°C). Is this normal or did anyone else encounter this. I am using the first version of dragonators board. Also I have noticed that the printhead needs few seconds to start functioning (this happens in even about 1 minute after it was last triggered). I am triggering the 2 nozzles at a time as dragonator already pointed out.

Best regards, JP
User avatar
dragonator
Site Admin
Posts: 595
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

Sorry for the long delay, I was busy with the Turret (see props topic). Anyway, I am picking HP45 projects up again, and also have time for replies again.

To my knowledge the HP45 will heat up. I have heard that it can run up to 60 degrees when in use. There is a threshold (I think 60?) after which a normal HP45 printer will go into a cooldown. The original driver has a way to alter the voltage so the printhead will heat up as little as possible. It uses the temperature sensor and 1000 pulses on each of the 10 or so possible voltages to determine which voltage heats the head up the least. Apparently too low a voltage means that no droplets deposit and the head heats up with all the power applied to it. Too much voltage and the droplet takes a certain amount of energy and the rest stays in the head. There is a sweetspot where there is just enough to vaporize a droplet, but not so much that you have excess.

The startup is a bit more complicated. Sometimes I have had it, sometimes I haven't. I have noticed that if I clean the head with a damp paper towel first, it starts easier. Also the Pen driver documentation speaks of a warmup function, that pulses at around 700ns(?) for a few thousand pulses to warm up the head. The best I got for the startup was within a few tenths of a second. I have yet to test a warmup cycle.

You can trigger more than 2 nozzles. I trigger a full address in 3 passes. doing 5, 5 and 4 nozzles. Any more than that starts giving problems.

In other news. I finally got ink out of the new standalone HP45 driver. There were a few bugs in the firmware, but all hardware seems to be working just fine. More on that in the coming weeks. Also, the HP15 seems to be drop in compatible with the HP45.
davidk
Posts: 77
Joined: Sun Mar 12, 2017 6:48 pm

Re: Hacking the HP45

Post by davidk »

Hi,

I finally managed to fire this cartridge with my L6452 driver that I wrote above. The missing horizontal lines aren't clogged nozzles but a short on address 21/22 that I'll repair later.
If I will redesign the PCB, I will rearrange primitive outputs. Since odd and even ones are on different rows, I will put odd primitives first and then even. This way it would be much easier to store the data arrays. Now I have to rearrange each single byte and this takes cpu time.

This driver can fire all 14 primitives at once but I only worked with the odd row.
I didn't noticed any sound nor noticed the head heating during tests. Timing is: address on - 2us - primitives on - 1.7us- primitives off- 3us - address off and the firing voltage is about 11V. As you can see, the amount of ink is very low.

Now I have to test different the firing times and calculate precise firing time based on the head heating and voltages.
Attachments
2017-08-11 17.54.27.jpg
2017-08-11 17.54.27.jpg (405.86 KiB) Viewed 13390 times
midosi
Posts: 1
Joined: Thu Aug 17, 2017 8:54 pm

Re: Hacking the HP45

Post by midosi »

First of congrats to all you hacking this cartridge, it is amazing. I have been struggling to use C6602 for my grad studies. Easy to use cartridge but droplet volume is gigantic. I am printing conductive material on non-absorbent surfaces. Large drop volume is a major headache. So, I have been looking for something better and that brought me here.

David,

I am wondering if you are willing to share your schematic, pcb, firmware etc.. files? I like your concept as it is much more simple. I need a controller that would allow me to jet from a single nozzle at a frequency I can choose.
This will replace the C6602 setup for my XY plotter. Basically, I have a grbl setup on Arduino Uno. I send a single ON/OFF signal from it to another Arduino which turns on a single nozzle on C6602. So I really need to be able to enable single nozzle at a time for my setup to work properly.
If you are willing and have spares, I would be willing to pay for the whole thing (including cartridge carrier). Or I could get the pcb and everything done from universitites electronics shop. Let me know.

Dogan

davidk wrote:Hi,

I finally managed to fire this cartridge with my L6452 driver that I wrote above. The missing horizontal lines aren't clogged nozzles but a short on address 21/22 that I'll repair later.
If I will redesign the PCB, I will rearrange primitive outputs. Since odd and even ones are on different rows, I will put odd primitives first and then even. This way it would be much easier to store the data arrays. Now I have to rearrange each single byte and this takes cpu time.

This driver can fire all 14 primitives at once but I only worked with the odd row.
I didn't noticed any sound nor noticed the head heating during tests. Timing is: address on - 2us - primitives on - 1.7us- primitives off- 3us - address off and the firing voltage is about 11V. As you can see, the amount of ink is very low.

Now I have to test different the firing times and calculate precise firing time based on the head heating and voltages.
davidk
Posts: 77
Joined: Sun Mar 12, 2017 6:48 pm

Re: Hacking the HP45

Post by davidk »

Hi,
Please look back in my previous posts for the schematic and my observations.
Another thing, I bought an old HP printer and I took the printhead connector and the rubber capping system to seal the printhead.

The microcontroller is slow, because I loose so many time rearranging bits for firing cycle. The whole time for serial reading 16 bytes, rearranging data and firing at least 128 nozzles needs to be less than 200us. You do the math...
So, in other words, I have only a "demo" structure I'm not sure I can make it work properly.
Anyway Ill try to optimize the firmware and maybe I can make it work.

good luck,
David
User avatar
dragonator
Site Admin
Posts: 595
Joined: Fri Aug 14, 2015 4:48 pm
Location: The Nethelands
Contact:

Re: Hacking the HP45

Post by dragonator »

I am deeply unhappy with the current standalone controller I am working on. After having had the benefit of hindsight, the 3 layers and many features lead me to conclude that I will never want to sell this controller, nor really use it myself. It is simply to complicated to make and use. I will still do some final tests on it, but in the meantime I will try something a bit different. A new standalone controller that is a lot more barebones and simple. The V2.01 standalone controller will be on a single (though slightly bigger) PCB. It will have a Teensy 3.2 that sits at the center. This Teensy can be controlled using either SPI or Serial. Also there are a few I/O's connecting directly to the Teensy for either encoders, trigger signals or feedback signals. The variable voltage will still have a variable switching voltage regulator, but the 12V will now be on a linear regulator (since there is less draw).

I will still use the older TLC circuit for now, since it has proven to work for me and I first want something usable for my own and others' experiments. I do still want to play with the L6452 myself, maybe for a V3.

At this point I am not entirely sure whether I want the primitive pins on shift registers and the Teensy in a breadboard friendly format, or want all primitives directly on the Teensy, and have the Teensy fully broken out. I still have some time to make a decision, but right now I have it in the breadboard friendly format.

In the attachment roughly what I am thinking of making.
Attachments
HP45 Standalone V2.01.pdf.zip
(162.39 KiB) Downloaded 880 times
davidk
Posts: 77
Joined: Sun Mar 12, 2017 6:48 pm

Re: Hacking the HP45

Post by davidk »

What's the maximum fire rate you can achieve?
Post Reply