@dc42 said in Runoff 3.3V on expansion socket:

My impression is that you have purchased one of the few relay boards that doesn't have optically isolated inputs; but I can't be certain form the photos you have posted. Please provide a link to the datasheet or seller's page for that relay module.

Hello, I finally choosed a simple relay board optically isolated as written in other posts here.

BTW, this is the link of the object in aliexpress

https://it.aliexpress.com/item/32974992635.html?spm=a2g0s.9042311.0.0.27424c4dYNXVMk

@jrsphoto The ability to set negative numbers with M208 allows the firmware to know exactly how far off the bed the carriages can move. This is important if you have an offset probe, because you may need to probe a point close to the edge, which causes the nozzle (which is usually the reference point) to be off the bed. The firmware will know if it's possible for the probe to reach that point, because of the M208 settings. Perhaps it's best to think of M208 as showing the 'bounds' of the machine, while the numbers show where the bed actually is relative to the homing positions.

When printing, it's the job of the slicer to generate code that keeps the printer within the boundaries of the bed. There are some exceptions (eg delta printers have a 'print radius' in config.g), so setting up the slicer correctly also helps. There's also M564 S0, which will allow movement outside the boundaries set by M208.

Also, the machine can't know that it's not supposed to go beyond 0. Many people have (including myself, and everyone who uses a Delta printer) have the origin (X0 Y0) at the centre of the bed.

Ian

The huge drop shown in the temp graph indicates that the sensor has lost communication with the board entirely. Failing crimp or cable exacerbated by movement and flexing during a print as you've found.

@dc42 Thanks! I shall get some on order.

I knew the RepRapPro machines used 10k on the bed which made me wonder which would give the best resolution in respect to the duet circuitry and ADC.

Yeah, I'll consider moving the board...

@DIY-O-Sphere
thank you

Fixed the issue with upgrade.

Thank you DC42 and everyone for the heads up. Should have done an upgrade from the start..

separate bin

while the whole thing was pretty neat; that would definitively take it to 11!

You are right, thank you very much. With the correct parameters the readings are now correct.

👍 they're usually quite amenable indeed!

@droftarts I couldn't use the exact same calculation on the second line because it hit the limit for max length of a gcode in a single line, however "current position" is the position of the first probe, so returns the same value.
For the same reason I had to move to X and Y independently in homeall.g
Not ideal, but necessary for now.
Aside from that the first probe point moves to a different Y position to the homeall.g yet they use the same code.
There is quite likely something I'm not understanding in the way the two processes are calculated, but it's odd. 🤔

If anyone is interested, my friend and I developed a Cura post processing script.

You pass it a single parameter (mesh spacing), and it parses the first layer gcode for min/max X and Y coordinates, and then replaces the M557 line in your start gcode.

LevelingMeshOptimizer.py.txt

Just remove the .txt, and drop it here: C:\Program Files\Ultimaker Cura 4.5.0\plugins\PostProcessingPlugin\scripts\

You must have a static mesh leveling command in your start gcode, like: M557 X0:200 Y0:200 S20

@droftarts said in duex5 PWM/servo pins as parts cooling fan help:

@tekstyle So all the tool fans will be wired in parallel? Exciting!

Ian

all the parts fan in parallel. its a lot of wiring. hopefully it works. tool fans are all to their own fan output

If running RRF 3.01RC you could generate two pseudo-random numbers from the object model and those in a G10 L2 P1 command to adjust the X and Y workplace coordinate offsets.

Here are some ideas for generating pseudo-random numbers from the number of seconds that the Duet has been running:

sin(radians(state.upTime))
cos(radians(state.upTime*3))
mod(state.upTime,15)

yeah, i just didn't see your previous response while writing that question. all good!

@dc42 any way we can actually inform the firmware or DWC about the estimated finish time? something like, read data from M73 and send it to firmware? (if firmware is not supporting M73 on it's own)

@Tony-Owens basically I it comes down to different takes on what would cause artifacts. I do not think my frame will cause issues. It's all 2040, not particularly big, metal corners, strapped tightly to the table with ratchet straps, and gains some additional stability from the 6mm Plexi attaching each tower to both neighbors the whole length of the tower. My thought is I'd best minimize artifacts by minimizing the moving mass on the effector/arms/belts. I'm currently using a flying effector attached via elastic to the carriages and happy with the results but want to see how fast I can get quality prints, and that bouncing on mass isn't going to help at higher speeds. This plan takes keeps the Bowden very short, but moves the mass of the extruder to a completely separate motion system. I'd share 2040 but otherwise it's weight is not shared to the hotend, different belts, arms, carriages and rods. I don't think sharing a frame is going to impact the effector personally, but am always open to other views!

Think i resolved the problem, replaced the Z motor and seems to be working fine now. Recon its time for complete motor replacement from e3d.

Thanks for the help guys.

@KenD Could it be that whatever hobbed bolt arrangement is used to grip the filament is slipping on the motor shaft? It would explain why you can hear the motors after the filament had stopped extruding.
For info, the only Quad I saw running a demo print was soooo slow. Maybe 5mm/sec extruder speed is the max that those tiny motors can do.

@massynrt This is the official support channel.

There's only one person who can follow this up, and that's dc42, and he's replied to you already asking for additional information. Keep all the information in one place (your other thread https://forum.duet3d.com/topic/15227/duet-maestro-2-stops-mid-print ), and tag him when you post something new. He'll get to it when he can.

Ian