zav3nd's Trident 350 - Formbot Kit

[zav3nd]

Day 9 - Running more wires.

Yeah so I made a mess for future me. Also probably time to clear some un-necessary stuff, but again future me problem.

 

Think I saw someone with nicely twisted together wires. I think I need to go learn how to do that for the next one...

[claudermilk]

Ah yes, the initial wiring rat's nest. Keep plugging away, it gets neat in the end.

[TitusADuxass]

19 hours ago, claudermilk said:

Ah yes, the initial wiring rat's nest. Keep plugging away, it gets neat in the end.

Mine hasn't   

[atrushing]

19 minutes ago, TitusADuxass said:

Mine hasn't   

You must have forgotten to leave a coin under the bed for the rat's-nest fairy

[zav3nd]

Day 10 - Organizing gantry/cable chains.

Fished through both chains now.

Tape to help hold all the connectors at the right length. Removed it later on. Started sealing at toolhead and worked back from there.

It does clean up pretty quick. Still a mess at the toolhead that I haven't decided how to deal with yet, but that may be a bit off.

Might swap my PSU around to the other side, seems like maybe @mvdveer did that on a trident. Then again I have wires I might just leave it and make myself a couple longer wires...

Guess I should start looking into the klipper side of things pretty quick too.

[mvdveer]

4 hours ago, zav3nd said:

Might swap my PSU around to the other side, seems like maybe @mvdveer did that on a trident. Then again I have wires I might just leave it and make myself a couple longer wires...

Whatever works the best for you - there is no right or wrong

[zav3nd]

I wound up making longer wires to get the PSU scooted back over. Got everything else plugged in except z-probe (but tap should cover that, I believe).

 

Pretty sure I can get started on Klipper next. I'll probably get skirts put on and consider unwrapping the panels. Double check all my wiring a couple times then hopefully get it plugged in. Looks like I might get a bit of time this weekend to work on it so maybe further along than I think.

[claudermilk]

That wad of cables and connectors will cram under the cover, it just gets crowded. That's one of the nice things about the PCB options--it cleans that up & makes it easier to deal with when you have to get in there for modding and repairs (which you will, trust me).

[zav3nd]

Day 11  - Checked continuities. Burned Mainsail to SD. Plugged in, nothing caught on fire.

Probably just cliff notes.

-AngryIP scanner helps in finding IP, for some reason it shows up as octopi.local for hostname, and kicked my ender 3 to octopi-2.local.

-Make your way to the mainsail OS docs for the walkthrough it seems. Voron docs say install klipperscreen through putty. Seems mainsail is prepackaged with everything (including klipperscreen) via the pi imager now?

-I hit the update all button, feel free to post the gif, I was lost after I put the SD card in.

-At Update controller path i get "usb-STMicroelectronics_MARLIN_BIGTREE_OCT                                   OPUS_V1_CDC_in_FS_Mode_204332945553-if00"  Ok I skipped flashing somehow. Here.

- STM32F446 or STM32F429 not sure yet. STM32F446, duh just read the big chip. Keep following the flashing doc... I went with cyberduck and sd card. Not sure how to power off and on the octopus so I just flipped the main power *shrug*. And now controller path is correct.

-Updating printer.cfg keep getting temperature sensor error, not sure why attached my current config. printer.cfg

 

Need to:
Tap Config. Reconnect my heater wires. Probably something else I'm not thinking of? (Edit: make sure I attach the x endstop extender so it doesn't crash.)

 

Also is there any way to fix my octopi.local without reimaging? Not that it matters the .locals rarely work for me.

Edited March 25 by zav3nd
links for future me

[mvdveer]

2 hours ago, zav3nd said:

Also is there any way to fix my octopi.local without reimaging?

sudo raspi-config

Choose system options, then host name and change to what you want it to be 

 

2 hours ago, zav3nd said:

Reconnect my heater wires.

 

2 hours ago, zav3nd said:

keep getting temperature sensor error,

You will get an error if the heater wires aren't connected. 

[zav3nd]

Day 12 - Plugged Heater wires back in. Put the endstop doo-hicky on. Back to configuring.

 

-Still getting error: Unknown temperature sensor ' "Generic 3950" ' ... quick google leads me to believe the " are the problem.  So anyone googling (its sensor_type: Generic 3950 , not sensor_type: "Generic 3950" ).Also hotend thermister was still commented out, flipped ahead in guide but don't see any other time to enable the pin and sensor type.

-Didn't enable probe pin, good thing it tells me where to find the problems. z_tilt same issue (I see the guide shows v1 and v2, I skipped over it thinking I have a VT ).

-Query_Endstops, Z shows as triggered (with chaoticlab cnc tap), tried inverting the "probe:z_virtual_endstop" returned to errors, inverted [probe] pin, running the command again shows xyz open.

 

Well after Klipper/Mainsail held my hand, I don't have a red error anymore

 

Currently running PID tune for the bed. Will have to update the pi's name, seeing octopi next to the mainsail logo feels wrong. Hopefully something more exciting soon, just figure maybe where I have to think might help someone else.

[zav3nd]

Day 13 - Continuing setup. Attempted print.

Z_Tilt_Adjust

Getting "No trigger on probe after full movement", rehomed tried again, same thing, appears 3rd screw is too low and doesn't trigger tap soon enough. Changed position min to -5. Closer, didn't make it. Went -10 makes it through the first round then most of the way through the second and probe samples exceed samples_tolerance.

Down the rabbit hole, back to probe_accuracy, works now:

probe accuracy results:
maximum -0.000000,
minimum -0.005000,
range 0.005000,
average -0.002125,
median -0.001875,
standard deviation 0.001586

Also weird, Tilt adjust finishes now:

Retries: 0/5 Probed points range: 0.007500 tolerance: 0.007500

In the config code sample, under [probe] tolerance seems different but

samples: 3
samples_result: median
sample_retract_dist: 3.0
samples_tolerance: 0.006
samples_tolerance_retries: 3

Some of it may have been from coming off of Extruder PID tune I suppose, while now its back to ambient. Skip Z_Endstop_Calibrate doesn't apply to tap.

 

First printer poop from Extruder calibration:

Went ahead and sliced a cube in Cura just to see. Ran through everything and got extruder not hot enough, tap dropped to 150, nothing to bring it back up?

Have to look into macros and start gcodes I guess. I think I'm going to make the jump from Cura to OrcaSlicer at this point too. Also need to get wires stuffed in the toolhead and panels etc put on. It does everything it needs to individually, now it just needs to do them together.

[claudermilk]

If the bed is tilted too far, it throws that error. The quick fix is turn off the motors and manually get it closer, then re-run z_tilt_adjust to let the printer finalize the leveling.

[zav3nd]

Day 14 - Skirts, Playing with Macros, and Test Prints.

Oh and I turned the lights on on the screen finally, still need to get the ones on Stealthburner going... Electronics fans need wired in, and all the wires need some management. Also need to find a couple deck support prints to help keep the cover flat.

Messed around with start Gcode:

M104 S0 ; Stops OrcaSlicer from sending temp waits separately
M140 S0

print_start EXTRUDER=[first_layer_temperature] BED=[first_layer_bed_temperature]

and Start Macro:

[gcode_macro PRINT_START]
gcode:
  #--------------------------------------------------------------------------------------------------------
  #--- read values provided by slicer
  {% set temp_bed     = params.BED|int %}     #*** Required. BED value must be provided by slicer
  {% set temp_extruder  = params.EXTRUDER|int %}  #*** Required. EXTRUDER value must be provided by slicer
  #
  #--------------------------------------------------------------------------------------------------------
  #--- do the usual stuff
  M190 S{temp_bed}                            #*** Set bed temp and wait
  M109 S150                                   #*** Heat nozzle to 150 and wait
  M117 Homing...                 ; display message
  G28
  Z_TILT_ADJUST
  G28
  G0 X175 Y175 Z30 F3600
  ;G32                                         #*** Home all, probe bed, level the gantry and such
  G90                                         #*** Absolute positioning - don't remove
  ;BED_MESH_PROFILE LOAD=default               #*** Load mesh
  #--------------------------------------------------------------------------------------------------------
  #--- set hotend to the temp provided by the slicer
  M109 S{temp_extruder}                         #*** Set extruder temp and wait

Both of those got me past the "extruder not hot enough" print cancel. I notice a lot of macros run G32, wasn't sure how that would work on the trident so I went for 28 and tilt command.

 

Still some work to go on my offset, It started out too close, then I had it about perfect, then I moved it too far away. Will be messing with that more tomorrow, then maybe a "real" print.

 

Think it might also be good to get a brush for tap, sometimes it has to retry several times on the tilt, I suspect its a bit of leakage.

[zav3nd]

Found the documentation on adding bed mesh after chasing a bed level print around 3 times.

I don't think that's the ideal bed shape. Thats a cold mesh, maybe a gantry issue of the left side being too high/right side being too low?

 

Also tried a cube last night, but canceled after it started rolling over waves (tap riding up and down on the walls). Not really sure whats causing it. Still working through Ellis Tuning guide when I have time, so maybe that will solve it. Just haven't really found any other examples of it. Also its hard to take a picture of...

[claudermilk]

Yes, that's absolutely your Y rails not being level. Re-measure those distances and get them absolutely as close to 110mm from the top extrusions on all four corners as you can. Take your time and get it as perfect as you can.

[zav3nd]

25 minutes ago, claudermilk said:

Yes, that's absolutely your Y rails not being level. Re-measure those distances and get them absolutely as close to 110mm from the top extrusions on all four corners as you can. Take your time and get it as perfect as you can.

Yeah thinking back I used the printed spacers, but never actually double checked that distance with calipers. Just checked right side comes in 109.7ish, left at 110.25         

 

(And mesh looks the same when heated too)

 

Edited March 30 by zav3nd

[Dirk]

4 hours ago, zav3nd said:

I don't think that's the ideal bed shape.

That's a normal bed shape for a tilted gantry (on a voron 2.4). I would run a Quad_Gantry_level on my 2.4. But I guess you would need to run a Z_Tilt_Adjust  and it will all be flat. 

[zav3nd]

I run G28, Z_Tilt_Adjust, G28 Z, Bed_Mesh_Calibrate in my print_start (and every time for these).

 

Got the Y extrusions all at 110.04 - 110.06

 

Moved the front right down to 110.12 for a slight loss.

 

I see the .2mm range and think that's the size of my layers, but its way better than the 1.2mm from earlier, and there's 350x350 to work with so I'm going to stop remeasuring and try and print something again.

[claudermilk]

That looks lots better.

[zav3nd]

On 3/29/2024 at 2:13 PM, zav3nd said:

rolling over waves

Well I think I've started to narrow it down. Noticed that everything beyond the first layer is heavy skinned/ridged looking like over extrusion. And letting it go on wound up letting TAP ride up and down over the ridges. I went with rotation_distance of 8 (I think it was the default). Tried looking up my z leads but oukeda motors don't come up with much, dug through and formbot does send a paper with information on all the motors, probably should've looked at it before going this far. Screws are infact TR8x4's.

 

 

Will report back. Think I'm going back to z offset prints again.

 

Edit: Bed Mesh, yep I'm an idiot.

Edit 2: Printed a cube not long after that. Been messing about with adding purge lines to start macro, and working on tuning since then.

 

Edited March 31 by zav3nd

[zav3nd]

Its "done". Well its been printing for the past few days.

Horseshoe Spool Holder is clean, but a little difficult to change filaments. Considering trying to cut the reverse bowden tube instead of running it all the way to the toolhead, but that might be a bad idea?

Printed some more panel clips on it and assembled.

 

I suppose next up is (not in order):

-Serial

-Chamber Thermistor - Thinking this one. And mounting on the rear z extrusion.

-Need to mess with Input Shaping, and probably more tuning.

-Print Spare Parts.

-Clicky-Clack Door? Considering more a top door and just a solid panel for the front, but so far its lived in the middle of the floor. Undecided...

 

[zav3nd]

Chamber Temperature

Printed this mount.

Firmware Code from here. Then just a matter of editing the sensor type and pin.

 

Mounted top right instead of the rear z drive. Figure even though it may heat faster there its a little further away from the bed.

I  taped off the white wires, then hid behind the rest of the hotend wires. And passed through the extrusion and the A Drive.

 

Happy with the panel clips, this would've been no fun to do with panels on.

[blaze toast215]

build is looks really good i also built a formbot trident kit i added tap and canbus to it have fun printing

[zav3nd]

Ok Time for Input Shaper, can only watch so many videos and still be clueless.

I went with LDO's Input Shaper Kit from west3d.

That board is staying put since it feeds power now. Kit is cheap enough I think I can manage one per printer...

Obviously ribbon cable will come off, does everyone take the board off the toolhead?

 

I also went with default config so my probe point was the front left of the bed instead of the center for these graphs. Is it worth doing more points or just generally the middle?

X Shaper type of zv and max accel 15700

Y shaper type of 2hump-El and max accel of 4900

 

Looks like this should be my printer.cfg addition? Based on the graphs.

[input_shaper]
shaper_freq_x: 63.4
shaper_type_x: zv
shaper_freq_y: 66.2
shaper_type_y: 2HUMP-El

[printer]
max_accel: 4800  # should not exceed the estimated max_accel for X and Y axes

 

But now its time to watch the "how to read input shaper graph" video again.