Now make that with a continuous ink system.
I once worked at a place with a fancy printer that didn’t use cartridges and instead used blocks of wax that could be continually fed into it. It took 20 minutes to start up and warm the wax up to the point it could be used as ink, but I thought it was a nifty idea. They’re called solid ink printers if you’re interested in looking them up.
Yes, called sth like Tektronix Phaser, and I seem to remember some big company bought the patent…
So… Basically the same principle as 3d printing (melt the printing medium, send it through an extruding head that directs it onto a surface in a desired pattern), but just in 2 dimensions, and with a different printing medium? Neat.
“while a CUPS based driver…”
So you’re volunteering to re-write the entire Unix print stack?
No, I’m gonna complain about using CUPS on the internet like any sane user would.
What are the complaints?
In my nearly 15 years of using Linux, CUPS only ever worked consistently with one printer I’ve owned. That printer was hooked up via cable. For all the wireless printers I’ve ever owned CUPS would recognize the printer once, do a job, and then magically act like the printer either doesn’t exist or is currently doing a job in perpetuity.
The last time this happened CUPS literally locked my PC out of using my printer because it always recognized it as job pending. Restart the printer, restart the PC, reinstall CUPS, change configs, and nothing worked. This was in the middle of working for a client. So I gave up and ended up emailing the documents to myself to use my phone.
Printing had been the most unreliable part of my workflow and CUPS is to blame for that.
For all the wireless printers I’ve ever owned CUPS would recognize the printer once, do a job, and then magically act like the printer either doesn’t exist or is currently doing a job in perpetuity.
I mean odds are it’s the printer that’s acting up. I owned a brother laser printer that I had to connect via the network, and that thing only ever configured the network on boot. If anything happened to the network connection it’d go offline until you restarted it.
There’s a reason I’ve seen multiple people recommend using something like a Raspberry Pi connected via USB as a replacement print server rather than relying on the native printer support. And since the modern printer protocols are basically lifted from CUPS, compatibility would be improved.
Sure. It could be the printer, but 3 outta 4 Printers that had the same issue repeatedly is more likely to be the drivers fault. Then everybody and there mother suggests getting a Brother printer. That’s one brand out of how many? I don’t own a Brother printer. That’s the only brand you ever see Linux enthusiasts recommend. And then so top it off with Raspberry Pi. I shouldn’t have to hook up a Raspberry Pi to use my fucking printer. You can try to convince me that CUPS is improved, but my own experience is completely the opposite. I will shit on it every time.
This is so awesome!
Inkjet… I’m out.
Actually kind of a necessity.
With an inkjet, most of the ‘difficult’ engineering and manufacturing is in the print head. The rest is just a basic x/y bot to move the head and paper around- easy engineering and manufacturing. They use someone else’s print head so they get around all that. That makes this a fairly easy design- just figure out how to trigger the cartridge nozzles when the head is in the right spot, write some code for rasterizing the image into print strips, and you’re done.
With a laser, there’s a lot more work. You need an entire optical system (laser, spinning mirror, etc), you need high voltage stuff to charge the drum, you need a high wattage heating coil for the fuser, etc. There’s a lot more engineering and coding work involved and more manufacturing also.
yeah but they get dried out and waste ink if youre not a frequent printer
Everyone knows that, but the comment you replied to explains why anything else just isn’t feasible.
thank you, representative speaking on behalf of everyone, for telling me what the comment i replied to says
- I want to shit out gold instead of turds
- sure but that’s not possible
- yeah but I could be rich
- ok dumbass
- thank you, representative speaking on behalf of everyone …
what if you only shit a tiny amount of gold, and had to sift that out of daily turdage?
what percentage would still send you back into gold-panning the turds of what you ate yesterday?
these are the questions that keep me up at night.
Probably not that much I guess.
I mean if you could net $200 or so per hour of turd sifting I’d be game with the economy the way it is and all.
are they nuggets/flakes or is the gold homogeneously integrated into the poo such that we’d need a furnace or centrifuge to remove it?
furnaces that hot are expensive is all. cuts into my gold profits
An laser printer is just a reverse scanner. It’s basically a resin printer with toner. We’re well past those being diy doable. It’s a couple of wires to deliver a charge to the drum and paper, a laser to remove the charge from the drum for the image, and a reservoir for the toner for the drum to pick up. The most complex part is the laser and mirrors for alignment, which is well into hobby diy territory.
Ok, well… we’re all looking forward to you publishing the repo for an opensource laser printer then I guess.
Necessity or not, I don’t want one. I’ve done enough IT support to know Inkjets eat ink unless they are used every few days.
My thoughts exactly when reading the headline.
So close, yet so far…
Bring back dot matrix printers and continuous feed paper!
You can still buy them. They never went away.
First, I always appreciate the effort for creating open systems:
and it’s entirely open source — bar its off-the-shelf print heads and ink cartridges.
…but the cartridge is usually the worst offender of commercial implementations for a number of reasons.
…leading companies including Brother, Epson, and Hewlett-Packard to implement a range of restrictions in hardware and firmware in an effort to lock printers down to their specific first-party cartridges.
The Open Printer, its creators claim, won’t do that — although it’s based around off-the-shelf Hewlett-Packard color and black ink cartridges with built-in print heads, the tiny microfluidic nozzles of a high-resolution inkjet being a little beyond the realm of do-it-yourself hardware. These cartridges, which can be third-party compatibles or refilled originals, are installed in a cartridge board driven by an STMicroelectronics STM32 microcontroller — which is, in turn, connected to a Raspberry Pi Zero W single-board computer acting as the central brain of the system.
So they’ve built their own driver for the cartridge which is good as it would prevent the vendor from denying the use of third party or expired first party cartridges from operating. There’s still the expense of acquiring the first party cartridges, and the questionable quality of third party/refilled to contend with.
I believe the idea is you would refill the original cartridge. You can get refill ink and a type of syringe to put ink back into the original cartridge. Those have been available for a while.
I’ve always thought it was interesting we have open source 3D printers but with how often 2D printers break and how expensive ink is no one has made an open source 2D printer. It’s nice to see some progress in this field
Pretty sure you can find schematics to make a BT100 printer somewhere ( https://en.wikipedia.org/wiki/BT100 ). You wouldn’t want to, but you can.
More seriously, I think the main problem is simply achieving a usable resolution for a 3D printer is way easier than for a 2D one. Hence why the real hard part (the print head) is using an HP cartridge instead of rolling out their own here. Given my quick googling, it’s usually that or building an inaccurate pen plotter and calling it a printer…
I often wonder why people think they have to start from scratch and build an entire printer. Few people’s printer problems are printer problems. They’re usually problems engineered into the printer at the firmware level. The stuff that actually does the printing is dumb components that do whatever you tell them and mechanical engineering someone else has already done for you. Right down to the commercially available connectors that connect the dumb components to the broken-by-design control board.
Why remake the entire printer instead of just the control board?
Not to mention, you can add features that should be there on every printer but that no manufacturer has considered including. Like an emergency stop feature for when the printer gets a corrupted print job and starts printing out as many blank pages as it can with the occasional page with a single line of gibberish. Tell it to stop, and it actually stops and spits out whatever sheet of paper it’s working on at the time. No holding down the power button. No clearing the jam that results. No uselessly canceling the job at the source. No questions asked. Just stop printing, clear the paper path, ignore the rest of the job, and lie your ass off and say you finished the job so no software gets any funny ideas about resending it.
The last thing you mentioned is something that HP did Implement some years ago into their (at least one of their) printers.
Wow. I figured no one would ever do that.
Course, then you’re stuck with the rest of the problems HP engineered into their printer.
That printer is in fact quite solid. No bullshit like forcing you to buy their ink, no batshit proprietary shit, doesn’t break that often etc. Only problem it has is, that when you try to scan something it sometimes happens that quite a lot of colourful stripes appear on the document.
That’s a really good idea. Something like OpenWrt but for printers would be amazing.
It’s funny, they have their own hardware now. Maybe starting with a open source printer firmware would eventually lead to open source printer hardware.
See, I was thinking replacement boards on the grounds that printer manufacturers have a lot more financial incentive to make firmware flashing difficult than router manufacturers do.
now do laser. fuck inkjets
Inkjet
No thanks
I can see by down votes not a popular opinion, but I’d pay over $1k for an open source laser printer with Brother’s reliability.
I’m all for open source laser, and lasers are great for some things, but they’re trash for photo printing for instance. Inkjets are brilliant for lots of stuff.
Unless you’re a professional printing hundreds of photos and buy the proper industrial printer, it’s always cheaper to take your photos to a store and have them printed there than deal with your own inkjet. Budget inkjets are a waste, through and through.
Or sometimes it’s nice to have a good inkjet so you can print stuff on a whim, or tweak the colours if they don’t turn out right. You can pick up a good inkjet for well under £500, it’s not outrageous for a hobby. Regardless, inkjets have their place.
If you rarely print, your cartridges dry out and the cost of replacing those is more expensive than if you just went to the store for the few prints you do. But if you’re willing to pay for convenience, I understand, but that’s all your paying for, not quality or reliability or consistency, especially with a >£500 printer. That’s still in beginner/budget territory.
I just bought my parents a Brother HLL3280CDW color laser printer and printed some pretty colorful graphics and omg the color quality was straight up tra—fine actually, good actually?
I was pretty surprised because I tend to see posts talk about how horrible laser color printers are and my experience seems to be exactly the opposite. At least on this Brother printer, printing big color charts, graphics, with full color backgrounds and patterns, seems great. I can’t tell the difference between laser and inkjet.
they’re trash for photo printing
I haven’t tried this yet, but maybe one of these days I’ll buy some glossy photo paper and try printing photos. I’ll have to report back.
They’re fine for charts and graphics, but they’re going to lack depth and vibrancy on photos. It is what it is, I’m absolutely not dissing laser printers they’re great, inkjets are just better for that job.
“Open Printer will use the Creative Commons BY-NC-SA 4.0 [Attribution-NonCommercial-ShareAlike] license for all of its files,”
That isn’t an open source licence…
While the firmware could be using a classic software license, in the hardware world these licenses don’t mean much. Afaik Creative Commons licenses are what’s generally used for open hardware
Non commercial means nobody else can sell it, so even if you print one for a friend you can’t legally recoup the costs from them
So your friend buys all the components and a case of beer (or non alcoholic equivalent) and asks you to help assemble it?
I believe you can. I’m not a lawyer, but over the years the NC clause of Creative Commons has been interpreted to mean that you can’t make a profit from sales of the product; recovering your production costs has typically been allowed. I believe that there used to be an example to that end on the CC FAQ, but I’m not sure.
In any instance, the copyright holder is allowed to offer additional or alternate terms. They may very well say “you can sell a printer made from these plans, but the sale price may not exceed your cost of the materials used” when they release it.
This would be allowed from my understanding of the license. You are not gaining commercial advantage or monetary compensation from “selling” it at cost to your friend once
One of the (many) problems with non-commercial licences is that “commercial” is really hard to define, and case law varies between countries.
That’s true, but that’s why there is a wiki page explaining the intent of the NonCommercial CC license.
Definitely not a perfect machine, but a giant step in the right direction,
Exactly stop looking at it as a bunch of faults, look at it as a proof of concept. The bubblejet ink – I’m sorry, the inkjet bubble – will pop eventually!
Well I’ll be damned. I had just recently made a comment about how open source printers have been hard to make due to all of the challenges associated with aligning the paper. This is an absolutely genius solution to the problem! Gonna have to plan on getting one of these.
Tons of 2d printer challenges.
The ink jet head at a resolution of <50um (as opposed to 3d printing resolution of usually >200um)
Combined with printing on many different surfaces and such: cardstock, printing paper, glossy, weights, stickers, etc…
This project “cheated” (I mean that not in a negative way, tons of open source hardware projects use proprietary components, my own does too for now) by using a proprietary ready-mades printhead, which saves the most cost and effort of the whole machine, and is the component that causes the most issues, generally.
I definitely want to try this out too.
Can you elaborate on the paper alignment challenge? Why is it difficult to do and why this roll solve the problem?
Getting the printer to pick up a single sheet off a stack of .003” thick sheets at a variable depth, and then dealing with variable sized-sheets, rotation, paper jam detection to avoid burning out motors, just to name a few. Things get even more complicated with things like 2-sided printing.
If it was just one of these at once, it might be pretty approachable, but the classic and convenient 8 1/2 X 11” paper tray that modern printers have is a genuine challenge to manage without lots of careful communication with all the different parts.
Freaking awesome start! I will revisit this when it gets to laser printing.
0 chance for DIY laser from scratch. 0 Chance for LED without someone’s printhead.
Those printers contain a lot of high-precision custom stuff that you can’t replicate without a substantial lab.
0 chance in what time frame?
Until the death of humanity or the point at which we get molecular 3D printers.
You need a high enough intensity light at 300 LEDs per inch to knock some electrons around on a sheet of perfect plastic. Then you need a perfect source of toner, the moving the paper bit and the melting bit are the only parts we can do adequately.
To make it outside of a fabrication facility simply isn’t reasonable. It’s too small.
Its the same reason we’re not going to be making our own desktop monitors.
The only reason they’re able to pull this inkjet off is because the head is coming from a fab, we don’t have that kind of precision at home to make a inkjet head.
Laser printers exist today. We can’t make FOSS ones because of patent encumbered technology. Some day that will not be the case.
The printer from the parent particle has “off-the-shelf print heads and ink cartridges”. Why would a laser printer equivalent need molecular level printing instead of using off the shelf components that exist today?
How do you plan to fab the print element there? What part are you going to source? Just give me the smallest bit of a plan for this and I’ll give you a pass. Because I honestly think you have no idea what you are getting yourself into.
re: laser printing - the tech has been functional since the 70s. Honestly I don’t see it as fundamentally more ‘high precision’ than inkjet heads that require similar tolerances. It’s not as easy as their path, sure, but it’s hardly impossible.
I think the trick to both will be using already proven components on the market and encouraging third party ink, toner, print head etc manufacturers to supply an open source project, and it’s kinda in their interests imho…
hopefully this project succeeds and other avenues open.
There’s nobody making open inkjet heads either.
Look, there’s very few processes available to us in the non-professional scope. The devices that fabricate these printheads and LED rays are ultra-high precision themselves. It’s giants standing on the backs of giants.
Yes, laser printers existed in the 70s. And an HP laser jet 2 would be about the best engine to try to replicate I expect. It had relatively few parts and they were all big enough that we could work on it by hand.
The software part we can do, rasterizing the data is straightforward and the tech has been there for years. We can turn a laser on and off with such precision that we could completely replicate how the commercial units do it.
Power supply, easy peasy. Chroma wire, no big deal. We’re going to electrostatic this shit out of that transfer roller.
Where are you going to get the materials to scratch make a transfer roller that can handle 300/dpi? It needs to be utterly perfect. It needs to be absolutely exact so in the light hits it, it knocks off the electric charge. You’re going to need a plastics lab in a clean room.
Let’s put a pass on that and let’s assume we’re going to use a toner cartridge with a built-in transfer roller manufactured for some printer that exists and they’ll keep making toner for it for some number of years.
Let’s talk about diy manufacturing the scanning mirror. A 3D print’s not gonna cut that. You need perfect segmented mirrors, beveled on the back to allow them to touch point to point. The positioning on them is going to be sub micrometer resolution. Even with the best 3D printing we have right now, you’re going to end up with a wobbly mess.
So, we’re going to move into a servo mirror. We put a fairly high-quality mirror between two servos and use them to make it go up and down. Teck here, it’s doable. Until you realize you’re going to need to have 300dpi. Now you need something capable of doing 2,500 steps over a few degrees of arc. You can’t get closer to the paper to fix it because you’ll end up with focusing problems.
Okay, so using a laser is probably a lost cause, but we’re not out of tech yet. We just need a really bright, fast, focused light source. Let’s move over to LED.
They put 2,500 LEDs and they strip 8 inches wide.
You’re not even going to find a pick and place service that can put 300 LEDs per linear inch.
DIY is getting better in leaps and bounds with 3D printing, software, and medium to small sized PCB manufacturing.
But we aren’t getting any closer anytime soon for true microprocesses. We’re not going to be making amoled screens, LCD screens, we’re probably not going to be making true processors. There’s just a lot of manufacturing techniques we need that aren’t going to happen outside of a lab or a clean room.
The project hasn’t even launched yet. So I’d say that it is not an “is a” situation but rather a “might be a” sort of thing.
You can follow the project here: https://hackaday.io/project/202990-open-printer
Unless this thing forgoes that stupid print code that puts yellow dots on each page to identify you when you print stuff like every other printer out there, it isn’t worth the hassle.
It’s open source. So even if it had it you could remove it.
actually that’s only for laser printers - inkjets don’t have tracking dots
This isn’t entirely correct.
https://office-watch.com/2017/secret-printer-tracking-dots/
Generally they are not found in lower end budget inkjets, but are found in higher end ones.
What’s this? I’ve never seen yellow dots on stuff I print
Edit: looked it up, intresting and definitely a bit disturbing from a privacy perspective. Particularly that it was kept secret until 2004.
theyre microscopic. theyre typically used to track printers doing illegal stuff (e.g counterfeiting) and leak tracking if someone when to scan some document and post it somewhere it shouldnt be. Of course, this is antithetical to privacy. The yellow dots are microscopic throughout the page.
You can’t. They’re microscopic.
It’s only common on color laser printers.
That rules. I wonder what the print speed it, and if they’ll be a way to use toner carts
Not a super complicated device, but toner would be a very different tech stack, probably not.
