Stop and Save System


#1

The WASP Resurrection System Provides an Open Source Stop and Save System for 3D Printers.

https://3dprint.com/28079/wasp-start-and-stop/

I had a power outage today near the completion of a 22-hour print. Battery Backups don’t power the device long enough to complete a large print. It would be valuable if Flux Delta included a feature to auto-save the last position of the printer and resume a print job when power comes back on. The WASP system is supposed to be Open Source.


#2

+1

It’s rainy season here in SE Asia too, and I am scared every time I click ‘Start’. Not printing anything over 2 hours for now.

WASP is releasing this as Open Source under Creative Commons license.

FLUX Team, it would be amazing if this could be incorporated into FLUX Studio! Please!


#3

On a printer with a heated bed, it’s likely that the machine cooling down would let the part contract and unstick from the bed. On a machine like the Flux, this probably wouldn’t matter. The printer would have to remember its calibration so that it didn’t have to probe the bed when it restarts. You’d also probably want to manually feed a little bit of filament so that you didn’t cause a gap in the extrusion of the print.

The biggest issue would probably be the hotend melting into the part when the power goes out. Unless you were there to catch the machine and raise the hotend, the part would probably be ruined anyway.


#4

Very good points Jim. For me, it’s more about being able to manually stop and resume, I can double tap the FLUX button now to pause (I never tried it long term though), this would be like that with a safety net. I would be able to stop the machine when I am going to be away and resume it when I return, so moving the hotend isn’t really a problem.

I don’t have any experience with heated beds, but did not realize there was that much of an issue with contraction, just thought it was more of an adhesion thing for the first layers. Definitely makes me more pleased with the FLUX!

The with the way they are planning to be able to redirect power via diodes, combined with a UPS system it wouldn’t really be a stretch to allow enough power to the motors to bring the hotend to zero point before cutting it and only supplying power to the arduino/pi.

Although any implementation of this kind of feature is probably further down the road and not so high on the priority list, it does not seem logical that with all the technological brilliance that is involved in 3D printing and the FLUX in particular, that it is hamstrung so badly by any slight power interruption. Just seems kind of silly and backwards.


#5

The way I understand it, the problem is that the axis stepper motors do not have encoders fitted to give position feedback. Just using home switches without encoders does not give the same home position twice. The machine has to do a calibration routine to find an accurate reference position for each axis, EVERY time the power has been removed from the motors.
If the machine was to “home” very very slowly each axis then the home position from the switches could be more accurate and the previous calibration data may still be usable but it would have to home very very slowly every time you turn on the motor power.
It would be nice to have maybe a “slow home” option and possible recovery mode enabled by such a feature.


#6

Actually, the homing sequence could be the usual fast upward motion then a very very slow downward motion off the switch again to get the most consistent home position possible and would not need a separate “slow home option”.
The machine just needs to keep a record of it’s position in program and it’s calibration data somehow.


#7

You’re correct, the stepper motors used on almost every consumer grade printer don’t have any means of positional feedback, it’s not closed loop positioning like you’d see with servos and encoders on higher end CNC equipment.

The homing process with the limit switches at the top of the FLUX should be very accurate and repeatable, at least more accurate than we’d ever be able to tell and more accurate than the thermal shift of the plastic that we’re printing with. On another one of my printers, I’ve calibrated it with a dial indicator and gotten homing repeatability about the width of the indicator needle on a gauge with thousandths (inch).

A bed that’s not heating up and cooling down wouldn’t have any shift either, so it’s just a matter of FLUX remembering its settings and logging which line of GCode it left off on (or at the very least, what layer height). I don’t know how the FLUX circuitry or firmware works, perhaps it’s RepRap based and these calues could be stored to EEPROM while the print is running?


#8

It would be nice to see a history of calibration values to see how much it varies.
There is the SD card to save data to but I suppose that could be corrupted by writing to it continuously then cutting the power. Maybe there is another bit of EEPROM on the pie but I don’t know. Maybe the info could be constantly sent to a server somewhere.


#9

Wow, you guys know way more about the ‘innards’ than I do! Thank you for educating me in these forums!

If I’m following along correctly, FLUX or any other consumer grade printer isn’t going to be capable of say, medical-grade precision on homing the hot-end consistently to the same point, but quite possibly to a base-level point where most consumers wouldn’t know otherwise anyway?

There are a lot of other factors involved regarding power diversion and the memory write issues that could be incurred as well, but this still seems like something that could eventually be possible?

With the concept and open source code coming from WASP, who had the need because of using clay to print housing structures, I’d guess the degree of precision wasn’t quite as high in that application.

Some seriously great information here, this community is just as impressive as the FLUX itself!


#10

It will be good to allow setting break points at specified height and pause there. Like I did in the picture manually. I was able to put some paper clay and to support the floating hand without having to add a long support from the bottom.