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 »

I has PCB's
2016-01-19 18.19.51.jpg
2016-01-19 18.19.51.jpg (301.17 KiB) Viewed 70722 times
2016-01-19 18.56.56.jpg
2016-01-19 18.56.56.jpg (426.1 KiB) Viewed 70722 times
I will attach the pogo pins this weekend and print a new carrier. I mirrored the holes by accident. Maybe I will even try to eject ink this weekend, but no promises.
ezrec
Posts: 112
Joined: Sat Aug 15, 2015 1:31 pm

Re: Hacking the HP45

Post by ezrec »

Best of luck!

I'm looking into the HP10/11 printheads (as used on my ZCorp 310+) myself.
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 actually having trouble finding the components to source 300mA at 12V without resorting to custom huge arrays of fets and transistors. Mitsubishi had the M54564P darlington array, but it has a switch off time of 4us. Way too long for a 2us pulse.

Got any ideas to source 12V at relatively high currents at speed?
ezrec
Posts: 112
Joined: Sat Aug 15, 2015 1:31 pm

Re: Hacking the HP45

Post by ezrec »

Linksys Wifi router power bricks?
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 »

My apologies for being vague.

The power supply side is fairly straight forward, with 5-7A currents at 12V for the inkjet printhead.

The problem I have right now is finding an IC that can switch these currents fast enough for the printhead. My microcontroller cannot switch 12V or 300mA. The problem is that all parts of the printhead are source, not sink. most darlinton array (and all switching components for that matter) are sinking or common anode. I need a common cathode driver (or sourcing driver) that switches 12V, at 300mA continuously, that can switch on and of withing 0.5-1 microseconds. The M54564P is a rare sourcing darlinton array, but it has a switch off time of 4us.

I would rather not make my own arrays on a PCB, that would take a lot of space and I am convinced that the components do exist for this. I am just not looking the right way.
DigitalShadow
Posts: 20
Joined: Sat Jan 30, 2016 1:49 am

Re: Hacking the HP45

Post by DigitalShadow »

Awesome work! I found this forum/thread by first finding the thread in the old forum regarding the reverse engineering of the cn642a print head (As I have a spare one around collecting dust)

It looks like you have switched your efforts to the HP45 since, and I am considering picking myself one to experiment with. My intention was not originally a powder bed system, but to upgrade my 3d printer (RepRap Mendel, original not prusa) to have full color printing capability, by printing a layer of appropriately colored base plastic, and then use the print heads to colorize it. (Rinse and repeat per layer). I have a mixing nozzle and am intending to use that to control the balance of white to transparent for the base material.

If I do purchase one of these units, what might I have to do to obtain a set of those PCB's you designed to interface with it? I can print my own carrier and source my own parts if that is a possibility
DigitalShadow
Posts: 20
Joined: Sat Jan 30, 2016 1:49 am

Re: Hacking the HP45

Post by DigitalShadow »

And just out of curiosity, regarding your above question of the power switching; must they be able to supply 350 mA for the primitives? You did mention that that was the requirement of the gate, but wouldn't that be divided by 14 for the sources? (Considering they can all fire at once)

Do we know the resistances of the firing resistors?

Also, from what I understand, most switches are N-Type Mosfets due to their superior cost and current leakage properties. They should still be able to drive your positive signal in the configuration shown in the far right image

Image
DigitalShadow
Posts: 20
Joined: Sat Jan 30, 2016 1:49 am

Re: Hacking the HP45

Post by DigitalShadow »

Hmmm, this component may be just what we are looking for. The TLC59213 (Datasheet). It looks like it meets the voltage, current, and timing requirements. I couldn't help but notice, it even cites driving inkjet printer heads as one of it's intended uses :)
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 »

Hello. I did start with the CN642A, and I still believe it is a better printhead, but it also uses a way more complicated set of signals than the HP45. The C6602 was one contact per nozzle, the HP45 uses a multiplexed array. The CN642A is controller using only a few communication lines. All others are for supporting functions. Adding color to an FDM print is very interesting.

I have seen a few attempts with hacked inkjet printers, and the results were very quite good already. You could also print more functional thing on the part, like release agents, but that is an idea for the future.

On the HP45, every firing resistor uses 300mA at 12V according to the patent. The resistors are arranged in a multiplexed array with the primitives and the gates. The cathodes are common. First a line of gates is opened. Then the primitives that need to be fired in that line are made high. Every primitive will then draw 300mA. Then the primitive is closed, followed by the gate. This process is repeated with every gate line. The gates should take a lot less current, but require 12V as well. Maybe for that a normal level shifter would be enough.

I can send you a PCB if you want to. The 3D print files are outdated and require some modification, but the PCB seems to connect (though I haven't had time to run a full test yet). Flat rate shipping is not much, so if you vouch to help hacking, I will ship it for free. The HP45 also has a full color brother, so after the 45 is hacked, that is the next step.

At this moment I have no information on the resistance of the firing resistors, but given the 300mA, I assume that it is in the neighborhood of 40 ohm.

I will look into the fets. I can get SMD P-channels that should be capable fairly cheap, but I will see if N-channels can also do the trick. I will probably do a breadboard test with a dummy load to test it. To test the speed I will have to see what I can come up with.
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 »

And I did not see your last post regarding the tlc59213. That seems just what I needed. I will investigate more when it is not evening here. Thanks a million.
Post Reply