Maslow arrived Thursday, built it yesterday, have a frame with 8’x4’ spoilsheet, total dimensions 10’ by 8’ 3" as pe rthe website. Use a Mac M1 with no difficulty. Use .YAML files with my 3D printers and octoprint. Updated the software with ease. Set the parameters of x=3000, y=2000, frame vertical=true. Belts wouldn’t fully retract, upped power to 1800, belts now retract fully. Pulled belts out as per instructions, Heartbeat ceased, couldn’t communicate. Rebooted, repeated initial setup. Belts didn’t come out far enough, Did this several times pulling firmly on the belts, no change. Then released tension in setup screen. This allowed me to anchor the bottom corners. Tried calibration but got an error message that the belts weren’t fully extended. Hit the extend button, response was that the I needed to retract all again. Up on the buckets, redid it all, heartbeat ceased again. Repeated the same agian, belts too short to reach the corners. I a fitter now, but am approaching the definition of insanity. Any advice? I am open to ideas. I wire wrapped my first computer (cosmac elf) and it worked!
Hans
are you on the latest firmware version? (released every wednesday)
make sure your system doesn’t try going to sleep or allow the wifi to go to
sleep.
David Lang
Johann Cunningham wrote:
Updated all files this morning, no problem, used the secret codes to get more extension, but that doesn’t seem to count. used the set and new button to update the .yaml file, still no success. I use octaprint with my prusa printers, they never drop a connection. Thanks for the fast response.
Johann Cunningham wrote:
Updated all files this morning, no problem, used the secret codes to get more
extension, but that doesn’t seem to count. used the set and new button to
update the .yaml file, still no success. I use octaprint with my prusa
printers, they never drop a connection. Thanks for the fast response.
can you try a different browser or a different computer (including a
phone/tablet)?
I’m thinking that the first problem to deal with is the connection dropping.
David Lang
Changed the screen timeout on the Mac not really dropping out the connection anymore. Using the standard settings, set up a 4x5 grid, calibration started fine, then the sled dropped below the sacrificial spill board. Had to catch it, hold it steady. Next at either left or right points the sled was again in the gap between the frame and the board. Board is 4’ x 8’ just as in the diagram. Finally on occasion either the bottom left or right belt didn’t fully tension! However when at the edges, tension exceeded 4000! It finished the calibration, but provided less than useful numbers. How can I keep the sled on the spill board? Help please. What parameters must i enter to keep it there?Either I get this working, or my wife will have it carted to the dump as it fills a large part of my garage. Of note, the lower belts became so slack that they would catch under the bracing at the bottom. Need to watch them closely.
The settings for this are confusing right now. I am working on making those more intuitive and easy to edit in tomorrow’s weekly firmware update.
Far better day today.
1 updated the firmware, very nice to enter the data.
2 Set Spoilboard to 1000 x 2000.
3 Set Frame to 2000 x 3000.
4 Retracted and released belts; OK
5 Belts not reaching, so I slacked off the belts; no problem
6 Reapplied tension; OK
7 Ran 4 x 5 calibration. The sled didn’t drop off the bottom like two days ago. Belts came out way more controlled, thank you. However, at the top, the sled teetered at the top edge. Held it in place, flat on the board with my hand.
8 Calibration completed with low fitness. Used the new numbers, saved them.
9 Then I worked the sled back to the center of the spoil board with a couple of tension releases and left shifts; arrived safely, though was moving like an impaired driver, swimming towards the center, approximating rather poorly a straight line (may be the poor calibration).
10 Set home at center, wanted to move to top left DISASTER STRIKES, the top right belt snaps, fortunately right at the anchor point. I turn it off and decide it is time to go play with the laser cutter.
11 Should have lots of length to spare on the belt for the fix as there was lots on the reel. I didn’t think that the motors could overpower the belts. Correct me if I am wrong.
H
All in all happy
what area did the 4x5 calibration cover?
David Lang
200m horizontals 1000mm vertical.
H
Did it break right where it enters the plastic end piece? Do you have some close-up photos?
I’m wondering whether a sharp corner or even some sharp flash at the entrance damaged the belt, making it weaker and more likely to break. If there were a sharp edge against the belt then when/if you ‘rock’ the belt end to get the extend function to work, then the high forces involved might damage the belt.
I think that this is the problem that we need to solve. The low fitness means that the machine doesn’t have a good understanding of its anchor points so using those numbers is likely to cause trouble.
That makes sense, as it is a really hard rock to release the belt.
H
May I suggest that one set the sled reasonably close to the Center point on the spoil board, then have the belts tension to that point. That should keep the sled from riding too high or dropping to low. As others have noticed, there is a slight amount of torque with the current sled to anchor point design I will add extra triangles, especially top left to see if that will give a straighter run to the cable.
H
Yes, something is not right with the setup numbers. When you input the exact anchor point numbers the machine should know the amount of slack to let out and know exactly how much to retract so that the sled is centered on the board prior to the beginning of calibration.