wierd_w

This is getting out of hand. Package says "Delivered, at front porch" NO PACKAGE IS THERE. Who's cheerios did I shit in at UPS?

Unhappy. When installing the new glass bed topper, I bumped the connector (tight space) on my heated bed, and lifted the solder traces. It's slightly smaller than a true i3 bed (213x213mm, not 220x220). SOOOO I ordered aftermarket Y axis sled for real i3 (which has slotted linear bearing holes, so should fit existing bearing sleds), and an aftermarket bed for a real i3. 37$. Parts arrive the 30th

I have a few old flatbeds sitting in storage. Dug out a microtek one not that long ago. It can be "convinced" to work even on windows 10 with some prodding.

Better than hearing a meow... ?

OK. Here it is at original resolution And Waifu2x 'd at 400%

OK Omega-- Do you want: 1) Super clean black on white 2) Super clean Black on (averaged color) Blue I have (1) nearly completed, but if you want (2), I will have to do some extra processing. In either case, the result will be a very clean, sharp image that can be taken to Kinkos if you want.

Suggest: (Photoshop) Select color range. Hold down the CTRL key, and click in all the "blue" area to get mass selection, Turn up tolerance to something absurd, like 50%. Delete the selection. Convert to greyscale. Manually adjust levels. Convert to black and white 1bpp Feed into Waifu2X. Upscale 200%. (inkscape) Trace bitmap Save as PDF Give me a bit, and I will do exactly this.

I want to see the juicy engineering blueprints. Juicy. give to me.

Home from work now. Part printed absolutely perfectly with no warping, nor any other print artifacts. (While I was in fault-finding mode, I lubed the shit out of all the linear bearings, verified good action of all belts, lubed the Z-axis screw gears, and- since I had to remove the bed to fully examine it-- did a full manual re-calibration after re-installation.) Clearly the malfunctioning bed power connector has been my major issue. Very pleased now. Will continue printing the series. The part was the second of the 8 part set, and is adjacent to the first one. Peg layout was sufficiently exact for the parts to mate up and fully set down on the metal component, and remain straight with the edge. Things are looking good so far. Once I have the whole series printed (and thus fully fit-tested), I will release the updated set.

(several hours of printing later) Yup. Bed still nice and warm at 60C, part has not detached, print quality is nice. That connector to my bed was clearly the cause of my sorrows. I will replaced it properly once said replacement arrives.

*grumbly noises* After a rather spectacularly failed print today, (like, totally let loose, with black blobs and spaghetti everywhere), I went into hardcore fault checking mode. Found that my heated bed was not reliably heating/staying heated, due to the connector having gotten too hot in the past, and the plastic melting. I have ghetto-ized a "really, this is a terrible idea" fix, and also ordered a replacement cable from amazon. (The ghetto "fix" is only until the new cable arrives.) Also ordered a suitable borosilicate glass top for the bed. (Meant for a REAL i3, but the dimensions are identical. I fail to see why it wont work.) I have been fighting a slight warp in my aluminum bed that should be resolved nicely by the glass topper. I am running the part that failed again, with the ghetto fix in place. See if that really was the cause of all my sorrow.

Walmart has the Element TVs, IIRC. https://www.walmart.com/ip/Element-19-inch-Class-HD-LED-TV-Recertified/528957208?wmlspartner=wlpa&selectedSellerId=1631&adid=22222222227319453355&wl0=&wl1=g&wl2=c&wl3=397879488927&wl4=pla-840819067851&wl5=9024386&wl6=&wl7=&wl8=&wl9=pla&wl10=113509192&wl11=online&wl12=528957208&veh=sem&gclid=Cj0KCQiAyKrxBRDHARIsAKCzn8xPf0VnMbvhGygNHY8zPW2hV7N0sqAZHM7V3JekMV7wKEHJx99rVCoaArhJEALw_wcB (~60$)

It's been my understanding that a big portion of the system's slowness is not really the CPU's fault, but the way they designed the bus. Hobbling the 16bit processor with a multiplexed 8bit bus architecture hits it hard straight out of the gate. Further, the added waitstates for many things means it spends a good deal of time just sitting there waiting. Sadly, the only way to address that is to redesign the computer. One possibility might be a second CPU placed inside the multiplexed bus side. That way only the CPU-CPU communication gets the muliplex, and the now 'local' CPU can manage the memory and IO in the bus at 16 bits.

I think the pegs will work great. Slight revision on peg locations as I print parts. Test part released from bed slightly while printing (lifted corner), but was good test for new features otherwise. With the foam strips on the inside I am gonna have to be "clever" in how I make the actual retainer clip things though. Some will have to be multi-peg.

(nevermind, atari-age was pointing me to an ancient page, and I was responding to old discussion.)

Nothing some additional paramagnetic shielding tape could not overcome, but that would just increase installation complexity and unit cost. Nope, not suitable for neodym.

But now that I think about it.. Magnet is a pretty slick idea for holding. I could print little round recesses for standard disc size neodymium magnets. Say, these little monsters. 4 per section in the corners. Held in with epoxy. Wait... Floppies + neodym == bad idea.

Maybe.. PEB top is ~ 11" x 17.375" Would need trimming at the closest mat size at 12" x 18"

Cafe Press? Nope. Not big enough. I will see if I can get magnet sheet by the roll. (That, plus rolled shiny sticker paper, would let me make such things, since I happen to have *dun tish* a large format printer)

I need to make revisions. (this is kinda why I typically hold off on posting quick and dirty models.) Issues: 1) The retention spring tabs in the back have no allowance for operation in existing model(s). 2) the holes, while in the correct locations, lack sufficient depth to give good grip in any place except the thick ledge in the back. Proposed solutions: 1) Make cutout notches in impacted areas 2) Replace hole pattern with "Peg" pattern, use half-moon clips on interior side to retain mechanically, except the holes in the thick ledge, which work fine. The holes in the metal portion of the PEB top are 7mm across. I will produce 6mm diameter pegs, which should be plenty stout at the slicer settings I am using. I will make them tall enough to slip a 3D printed half-moon clip (2mm thick minimum) through a notch in their diameter, and still have enough "beef" to have good mechanical holding. The metal is .063" thick. I can model in a radiused reduction in the diameter to 5mm, that is 2mm tall to facilitate this. I will then reprint the corner piece I printed today with the changes, and test fitting.

I know the feeling, but my workplace would probably be real bad to try that.

If I am not locked in to lower resolutions by the limitations of the VDP, I have a preference for 80col mode. I find the 40 and 64 col modes available through the TIPI to be... barely functional. (Not bagging on Matt at all here, just that my eyes really hate those modes.)

I will have to modify the data a bit to make a back section like that. (For starters, the "step" in it needs to be removed-- AND it has to be mirrored after that so it matches...) Velcro is not my first choice. Neither is double-stick tape. The 8 thru holes that hold down the card slot backplane (and alignment grid) could have longer machine screws put through, that could then reach into the sections on the back. A modified plastic skid-plate for the floppy bay that has recesses (for screws to go the OTHER way through) could give a few more. I need to think about it more to come up with other clever holding that could be done

My ghetto chinese knockoff i3 is hardly the fastest hunk of junk in the galaxy you know. Still printing the first of the 8 sections. Since I intend to use the attachment point screw holes to hold it down (and each section conveniently has at least 3 holes already!) I did not add any tabs or anything. It's just 8 rectangular(ish) slabs that sit next to each other, that then form a complete PEB top. Like I said, quick and dirty. The bottom will be... Hmm... You know some of the screws that hold the PEB together are on the bottom, right? The only real way to accomplish that is gonna be to abuse some of the thru holes with longer screws, which I don't think I have. OK, you want STL files for the quick and dirty model? Here they are. First up-- If somebody happens to have a great big 3D printer that can do this monolithically... monolithic.7z and here it is divided into the 8 tab-less sections (For people with more consumer grade printers) PEB TOP sectional.7z AND-- If somebody wants the CAD data... PebTop.stp

I really have no idea if it would work, honestly. It was intended as a joke, but people kept asking. O.o