I do not remember exactly, but I am sure it is measured from touching the sheet to the rollers and as far as I remember (but I'm not sure if 3cm from the sensor tip selection, you need to check it) ~ 5mm from the edge. Make a hole of 0.5mm and if you do not hit it you can enlarge it to 0.8 or 1mm.
I will write how it works so that you know where to look for a possible error. You have three sensors in the printer, named in the tutorial mfps, pis, es and npis is an additional, because these original three sensors are removed from the printer. All three sensors must go through the specified time for all three sensors. The most important is pis. When the uC board starts, it switches mfps programmatically, after ~ 500-550ms (time between the sheet metal position on the drum and the hole position in the sheet on the photoresist), when the hole in the sheet is on the photoresist (npis), the UC switches on, counting from the writing ~ ms delay and printing starts. In other words, the most important thing is to write, since it starts printing, mfps and es are actually paper jam sensors.
3d printers are probably sprayed? And every inkjet works on this principle. BWT: I used to think about making tiles with a UV inkjet printer. There are those that can print on flat sheets, but this is a very expensive case .
I wonder if there is anything like a shower printer
The only thing that comes to mind is an inkjet printer converted I also have one. It works great too, but the problem is drying up the carcass. All the time you have to remember to print something from time to time, which was just my reason for changing to a laser.
I heard that the best pcb come from thermo transfer from printing on color newspapers
I have different experiences.
Perhaps you have used the wrong paper (newspapers) - it must be slippery and thin, then it goes really elegant - much better than on chalk paper.
It is also important what paints the printing house uses - some pages of newspapers with multicolored pictures are not good, while when there is only one - two colors (print, for example) there are no problems.
Okay. Now I have a real turn signal - once it prints, it does not print once.
I made a hole in the plate. 1 mm hole, about 5 cm below the NPIS sensor, 2 mm from the edge.
I am waiting for the buzzer to stop buzzing (I am buzzer for 5 seconds, not 3.5) and sometimes it will print and sometimes it will not.
So actually NPIS does not fit in the window.
I wonder if 5 cm is too far away and it is not better that the hole is at max 10 mm to the lower edge of NPIS ??
I think I've found a solution. The original tutorial says:
Remove the carrier. Now, a small hole in the carrier, located about 3mm. I used a .026 inch drill.
And according to it, I made a hole 5 cm below the sensor. For this reason, sometimes I was located, and sometimes exceeded the "time window". I made a second hole about 7 mm below the sensor and everything works now, but the tests will show.
Well, I'll praise you. After many attempts and something came out. I do not know why Eagle prints very poorly. The effects are just like in Eagle will print to PDF and then scream for the printer.
No paths. The thinner they are, the better the blackening, and the thick ones do not come out very well. In the picture, the path is 8 mils, and unfortunately [with these dimensions some are already drained, so 10 mils is probably the maximum of this technology.
When printing this second tile, I "killed" the drum.
I did 8 or even 7 mils between pads and it came out pretty. It is true that this "print wizard" in the Eagle is some kind of lip, especially with large tiles, strange things do, sometimes it also does not keep dimensions when moving away from the margins, etc., hence combines when printing with older versions of EAGL or PDF. The second thing is drums, and massaging the tiles. One of the drums, or toners, some "prefer" a perfectly smooth surface, other 400/500 paper. As for the amount of toner, try printing on a sheet, how it will print perfectly on a piece of paper, then print on a plate.