It’s been a little while since I updated some of the tests I’ve done.
All of these still failed within a couple minutes of cutting
@bar would it be possible/helpful to send out a survey with the weekly email with questions such as
I know it’s a long shot, but for the people who can’t even get a few minutes into a cut, have you tested your outlets to make sure they are grounded instead of just expecting them to be?
How about another question? Of the people who experience this error, are any of you working in a humid environment (basement without dehumidifier, garage in a humid climate, etc.)?
Anyone not experiencing this error while working in a dry environment (desert Southwest for example)?
I am in Colorado (dry climate, low humidity) and experiencing the error. Vertical, cutting 1/2" MDF, only seeing the issue when cutting. Dry runs, even with router and vacuum on, have all worked for me. Had the router and maslow cords not near each other, have router ferrite core up by router. In process of trying the mount in another thread with copper tape under it which I will ground. I did get my long, bare, grounding wire in the mail today so once I’m off work, I can attach it to the shield, run it through the hose, etc. I’m time-limited during the week so probably won’t get to trying any cutting today… might be the weekend before I can. Will also try other material once I’m cutting, because first, I’m almost out of scrap MDF to use, and I don’t have unlimited funds to by sheets of material 
I do have some scrap plywood, melamine, and OSB I can try next.
Another possibility : the length of our DC power cord is L-O-N-G (especially considering its not shielded)
The length of a cable significantly influences Electromagnetic Interference (EMI) issues in several ways:
Because of these factors, careful consideration of cable routing, shielding, and other mitigation techniques becomes more crucial as the length of the cable increases
So I tried a couple more things today. First I checked my outlets to make sure they are properly grounded, then I tried putting a wire down the dust tube and grounded it on my machine (I noticed a significant reduction in static woodchips) and I also tried some of the Shielded Ethernet cables that Bar was going to try. None of these solved the issue but I feel like I was able to get a couple minutes longer than I typically do with the shielded ethernet cables not sure if it was related to the cables or not though.
So, would it be better to mount the power supply box to the sled and have a much shorter DC cable, and somewhat longer AC cable nearby? We added 20’ extensions to our DC and it’s not any worse…
I’d be interested in ideas to try! Ferrite beads? lead sheet over everything? Faraday cage around everything and try the USB? Any top theories? We could go anti-static on the dust collection, etc. but it seems some people have done that w/o significant impact. My shielded Cat6 cables not arriving until tomorrow.
If you were on a survivor style reality show “Maslow’s Pyramid”, what’s your first move to reduce potential electromagnetic interference, with each success getting you closer to the top and attaining self actualization?
I’m hoping its a software problem
I think so.
Also try to shield the short DC run with shielding sleeve. Ground the sleeve to your AC earth ground and add a ferrite bead on the cathode and anode wire right before the connector to the mainboard.
I think I have a related issue with my setup:
When cutting on the right half of the sheet, I get a lot of encoder errors:
[MSG:INFO: Center point deviation: TL: 0.441 TR: 0.321 BL: 0.343 BR: -2.530]
[MSG:INFO: Center point deviation: TL: 0.441 TR: 0.321 BL: 0.343 BR: -2.530]
[MSG:INFO: Center point deviation within 15.000mm, your coordinate system is accurate]
[G54:0.000,-500.000,48.500]
[G54:0.000,-500.000,48.500]
[GC:G1 G54 G17 G21 G90 G94 M5 M9 T0 F229 S0]
[GC:G0 G54 G17 G21 G90 G94 M5 M9 T0 F229 S0]
[GC:G1 G54 G17 G21 G90 G94 M5 M9 T0 F305 S0]
[GC:G1 G54 G17 G21 G90 G94 M5 M9 T0 F1016 S0]
[MSG:WARN: Encoder read failure on Top Left]
[MSG:WARN: Bad connection on Top Left encoder, failed to read 5 times in the last second]
[MSG:INFO: Reset during file job at line: 22]
[MSG:ERR: Emergency stop. Update function not being called enough.1002ms since last call]
[MSG:WARN: Bad connection on Top Left encoder, failed to read 20 times in the last second]
[MSG:ERR: Emergency stop. Update function not being called enough.1001ms since last call]
[MSG:WARN: Position error on Bottom Left axis exceeded 15mm while running. Error is 15.000mm Counter: 1]
[MSG:WARN: Previous error was 15.000mm]
[MSG:WARN: Position error on Bottom Left axis exceeded 15mm while running. Error is 15.013mm Counter: 2]
While cutting on the left half of the sheet everything went fine. I do suspect some kind of slippage between the belt and the encoder gear. The only effected belt is the top left one, all others work fine.
I will try to rebuild the arm and maybe replace the belt. All electrical connections are fine, as far as I can tell.
I might be wrong and this is a software bug, but why would this only occur for one of the arms?
I will report back sometime next week when I had a chance to rebuild the arm unless someone tells me not to bother
This to me sounds like an issue with the wire to the top left encoder. I would try unplugging and re-plugging that before taking it apart
Could try switching two arms before pulling it down to see if the fault follows the arm?
So I did this… connected the board to the (M4 Router Be Gone (installed the spindle))
and copper-taped the bottom of it, soldered a big bare ground wire to it, ran it down to and through my vacuum hose to earth ground. (I know… janky looking but its fine. I have a better clamp on order for this connection between the hose and the ground wire)
I then ran some gcode pockets for 100mm circles, 100mm squares all cutting from 3 to 6mm deep – about 30-40 minutes of cutting and didn’t get a single error. not a long enough session to claim victory, but I’m slightly encouraged by it.
I did notice that the sled does tip a bit on movement, which hurts accuracy on the corners / turns. I have over 20% slope on this so I’m thinking I’ll wax the bottom when I care more. Right now I am after consistently not ruining wood vs perfect accuracy
That is definitely encouraging news! I’ll be trying some more grounding in the future
Good to see potential progress. And damn thats a heavy ground wire! Looks like you attached it to both the vacume and the shielding plate. Good!
Another thing to try is to twist each insulated wire of the two 24+ power supply lines around the other wire all the way from the PSU to the M4. Twisted pairs are less prone to transmit or receive EMI.
Same - for my sled I mean
I had an issue with a red blinking light so I replaced one of the encoder wires with one I have laying around, since then I have completed 3 cuts, one was kind of long, maybe an hour all with no issues.
Maybe I’ll switch the wire back and see if I get errors again. It would be nice to narrow this down.
It would be great to get a conclusive result on that if you are able to test it
I tried some additional grounding under the computer and wrapped around the body of the router and that didn’t seem to help at all. My only other thought is that doing it horizontally might help? has anybody doing it horizontally had this issue?
do you need to recalibrate when you transition from vertical to horizontal?
