I thought a brim was to prevent edge curl - LOL!
I’m getting closer, but the 3 point electromagnetical-metal-detecting in a emi noisy environment is far from perfect. Observant readers will note the little piece of aluminum duct tape on the bottom left corner of the bed to help trick the probe into plotting a more accurate plane. It has helped, but I’m still plowing into the bed at the top left, and nearly air-printing at the bottom right.
I also moved the printer over to the left to be further from the power cord strapped down to the bench. That has helped as well.
I’d like to get a couple more level points in, but I understand that means compiling new firmware - I don’t quite have that time/brain energy yet.
Doesn’t help that I’m printing at 0.1mm layer height (w/ 0.2mm nozzle). @Printrbot Smalls with 0.2mm nozzle.
Seems to me you need to just level your bed properly, get rid of the Buildtak and go back to blue painters tape. The objects your trying to print would be better printed on a SLA machine.
Yup - leveling the bed.
Been doing this for a few years. Printrbot Maker (laser cut plywood, fishing line belts) - massively upgraded now, Frankenbot with 10x10" bed. Both with leveling screws and a mechanical Z-axis home switch.
Did the painters tape, direct onto acrylic and now heated aluminum beds with Buildtak.
Very satisfied with those two printers.
This one however, came without a mechanical bed leveling system. It assumes:
a) bed is flat (which I believe it is - no light visible under a straight edge in both X & Y directions)
b) no electrical/magnetic interference on the bed probe
c) 3 points are sufficient to produce the correct Z plane.
All good in theory - but something is off on this one. Or maybe I’m just pushing it too far with the small nozzle.
As for SLA, yup - maybe someday. But this is what I got, and the customers who paid me to print these for them are happy so far.
3 points is not sufficient. All 4 corners and the middle of the bed need to be in alignment. Even automatic leveling systems use 5 point. Clamping and removing the prints just doesn’t things in .01mm alignment for very long.
@MidnightVisions 3 points is fine if your probe is accurate and the bed is really and truly flat. Lots of people and firmwares use 3-point leveling. All four-point or five-point leveling tells you that 3-point doesn’t is whether your bed is warped… and if it’s warped, there’s nothing you can do about it with this printer.
@Ryan_Carlyle Well as you say, something is off and your bed is not flat or else the tip wouldn’t be digging into the bed.
A gentleman in the community compiled the Printrbot firmware to allow mesh leveling…
Thanks @Brook_Drumm for the link! What are the odds the heavy plate that comes with the extended bed option would be warped? (Like I mentioned, it looks fine to me when I put a straight edge on it). I’m willing to try a multi-point leveling, but I am wondering if my problem is more interference related.
If a straight edge tells you it’s flat, I would think it is indeed flat. Have you tried a .3mm first layer?
I recently ran into this video. Maybe it can be of help…at least with print quality after you have other stuff figured out.
Vid looks promising - will study closer later this week. Thanks @NathanielStenzel !
Are the rails straight? The bed may be flat but moving along a bent axis.
@Baldur_Norddahl That’s an interesting theory. Will be a bit of work to determine that - so I’ll tuck in a proverbial back pocket for now. But certainly thanks for the suggestion!
I would try a dial indicator mounted on a stand independent of the printer. Then measure the bed moved to the extremes and the middle point. The middle point should be exactly the average.
Independent? I’m thinking that would introduce errors as the printer (potentially) shifts. It’s a Printrbot Smalls - the stock model lacks a solid base.
But from your previous post, I’m envisioning using feeler gauges (or now I might go and by a dial indicator) and manually mapping the bed without the printer compensating (it - normal home the Z, not 3 point map). And then compare that to the values the 3 point provides.
(That is - I’ll set the hot end tip a known distance from a homed point - say 0.2mm and then head to the 4 corners and remeasure and do some basic math to determine if the bed is flat/running true or if it or the rods are warped).
A cheap chinese dial indicator is plenty good for this. Just turn off the printer and move the axis manually. Mount the indicator to the printer head. Possible using tape :-).
