Just zip it up or put it on a file sharing like google drive, dropbox etc
give me another 10 minutes because putting it on my sandboxed web page is at least a 3 step process.
Done,
go to the link in my sig which this is a paste of:
Cheers, Gene Heskett.
“There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order.”
-Ed Howdershelt (Author, 1940)
If we desire respect for the law, we must first make the law respectable.
- Louis D. Brandeis
Genes Web page http://geneslinuxbox.net:6309/gene
And you’ll see an mp4movies directory, its in there. Just one g28 from the middle of the bed.you see x doing it about normally, but x still moves to the right while Y is moving back to the max_pos y switch. Z works normally I think. Or does it, it leaves the switch on, not dropping 2mm to assure clearance.
Take care and stay well, Sineos
you might have to remove the “gene” from the end of that link, let me know and I’ll edit the sig. Accessing this machine
from its home is at times odd.
Ok, lets try the following:
Make sure your steppers are configured as follows:
[stepper_x]
...
dir_pin: !PE2
[stepper_y]
...
dir_pin: !PB9
then:
- Power down the board completely!
- Swap the motor connectors on the board. The stepper plugged in X goes into Y and vice versa
If this does not help, then make a video of G28 X and G28 Y separately.
from the middle of the bed, g28 x comes forward an inch or 2, sorta spastically then finds the zero switch.
g28 y, from the middle of the bed, heads toward the left front till it hits the stopper at the left front and hammers the stop another 2 or 3 seconds. The switch is at +300mm.
So. I’ll nuke that 50 meg mp4 and make two more, but I’ll rename them g28x.mp4 and g28y.mp4
This may take a few. I’m finally hungry. and haven’t eaten in about 11 hours I’ve also been at the lathe, making a brass nut installer for the iron that came in the trident kit.
Had a bowl of beef terriyaki, and the 2 mp4 are ready for your viewing pleasure. Sorry I couldn’t round up some pretty girls on such short notice. ;o)>
Take care and stay well, Sineos
Ok, then lets try:
[stepper_y]
...
dir_pin: PB9
In addition check / make sure:
- Pulleys: Check them, make sure they are firmly mounted on their axis (stepper side) and free spinning on their idler sides
- Belt tension:
- Extremely important, especially on coreXY
- Check with the head in the middle, i.e. both sides of the belt have the same length
- Make sure they are evenly tight but not overly tight. Musically speaking somewhere around a G2 or 100 Hz when plucked (only rough indication. Would need more work to determine precisely)
- After tuning, move the head around, position back in the middle, then recheck
- After tuning the one belt, recheck the other. They influence each other
The pulley and belt stuff really is extremely important on corexy machines. What you learn now, will serve you well for your Voron project.
I agree, the first corexy I bought was an eyeopener. A Tronxy-400-plus. But it came with a broken marlin that insisted on starting the print about 4mm up in the air, so while it moved right, there was zero support from tronxy. Its going to be like the half finished Ender 5 Plus, with a lot of carbon fiber in place of the X axis, the Y drive has been converted to a nema 23 stepper/servo so it can throw the x axis around better. But I ran out of working printers before I got the linear bearings on a carbon fiber beam for the X axis completed, so its dead, this printer died resoldering the X switch, a biqu bx died while replacing the whole head. That highly touted H2 head is a bad design, puling the lever to change filaments crushes the pressure spring so it never feeds right again. A $1900 3d45 Dremel could not do PETG so I gave it to my stepson whose having a ball feeding it PLA, an Ender 3 failed the PETG test, a CR-10-pro Creality also died from PETG’s added heat, and if you modify a creality produnct, you are on your own, the plastic bowden feeder broke, I replaced it with a metal exact copy from creality then wanted a cable to make the filament switch reach an orbiter on top of the head, and got told to goto to hell by creality,
And somplace along the line a Newereal clone of the cr-10 died by blowing the top off the bed heater switching transistor about 5 minutes into its first print. Its on the back porch, a spare parts warehouse. So is the ender 3.
Unless I totally miss-understand corexy, an X motion only should occur by moving both motors in the same direction absolutely equal amounts. To achieve a Y only motion, you still have equal distances but one moves cw while the other moves that equal amount ccw, And this is something I have never seen it do since I resoldered the wire on the x switch. x commands also move y spasticly, Y commands move X a long ways with no Y motion until it hits the X stops.
The board has been changed, from the oem version 1.2, to the new version 3.1. And nothing has changed by changing the board. Two trees said they were sending me a new board. At least 2 weeks ago. I’ve queried them twice since asking for track numbers or a schedule and got zero response. It now Jan 13th here, and I’ve been screwing with this since late November. So I bought the voron since the prusa has also commited suicide, a $450 mosquito kit is leaking after about 3 weeks of PETG with a diamondback nozzle because quite a bit the the PETG has been +CF too. So if my ancient legs will do it, I’ll go finish the iron tip that will install 75 brass inserts into the holes in the plastic and hopefully I can resume construction of it. While you cogitate on this. And I’ll go pester two trees again. But first go make some breakfast & start the dishwasher.
Thank you, Sineos, take care and stay well.
Honestly, Gene, I’d really like to get this printer up and running with you but in order to also keep my invested time under control, I need your commitment to:
- Focus on this printer and nothing else (at least here in this thread)
- Answer my questions
- Follow my directions
- Whatever you do / is occurring attach a unmodified klippy.log
Doing such a diagnose remotely is already hard enough, there I really do not need wrapping my head around additional things or Gen(ph)enomenons each time.
If we cannot agree to this, we better stop here and now.
So would I, love to get it running, but we aren’t getting anywhere.
With my scope watching the Y en pin sitting at 4.9 volts it never wavered, It shoulld have dropped to less than .1 volts had it ever enabled the y motor
kilppy.og from this morning attached, probably 10+ restarts as I tried the menu.
klippy.log (181.4 KB)
I also seem to be having a t-bird problem, I saw a nitice of this incoming well before 11AM my time.
But I couldn’t find it, either here or in the inbox, gave up figure it was lost so I went to the grage and finished up the tool I need for the trident, then worked on it for about 2 hours. Fixed a sandwich and came in here to hide it and found it at 14:15.
The printer.cfg file is back to yesterday. The by-id/ hasn’t changed That should be in klippy.log.
Cheers Sineos. Take care & stay well
Lets see, where we are:
Got EOF when reading from device
Timeout with MCU 'mcu' (eventtime=384515.628496)
Transition to shutdown state: Lost communication with MCU 'mcu'
Looks like an USB communication problem:
- Check USB cable
- Reboot the BPi
What I currently believe:
- The EN-Pin story is red herring. The settings are correct as is. If they were not, we had no movement at all and just errors in the log
- The erratic movement is NO Klipper setting issue and NO board / driver issue but something else and we need to identify this “something else”.
- “Something else” could very well be a mechanical issue like
- Wrong / modified belt paths → Did you change anything here? → Check and compare against the build instructions if available
- Slipping pulley on the motor side → Check it
- Stuck / slipping pulleys on the idler side → Check it
- Octoprint - Do you have anything in the GCODE script settings?
Sorry to say but your understanding is completely wrong:
As you can clearly see, you have two independent belts. One could be slack like hell and one having a good tension and the head would still stay in its very same position → Check it
Way forward
- Systematically check / work through the bold marked items above
- Make sure we are on the initial configuration, i.e. both axes inverted and the motor plugs in their original position
- Provide a video of
G28 XandG28 Yseparately - Provide a video of moving the head 10mm in -X direction and 10mm in -Y direction separately (after homing and using Octoprints controls)
Lets see, where we are:
Got EOF when reading from device
Timeout with MCU 'mcu' (eventtime=384515.628496)
Transition to shutdown state: Lost communication with MCU 'mcu'
Looks like an USB communication problem:
- Check USB cable
- Reboot the BPi
cable, have 2 of them, both work the same.
bpi rebooted just now
all this has been done many times. except I’d guess a little lower frequency, 75hz perhaps. this is a factory assembled head the is just bolted to the top of the frame. And it has no visible method of adjusting the tensions. Not mentioned in the printed manual either.
What I currently believe:
- The EN-Pin story is red herring. The settings are correct as is. If they were not, we had no movement at all and just errors in the log
- The erratic movement is NO Klipper setting issue and NO board / driver issue but something else and we need to identify this “something else”.
- “Something else” could very well be a mechanical issue like
- Wrong / modified belt paths → Did you change anything here? → Check and compare against the build instructions if available
build instructions other than putting it together from the shipping box not available.
- Slipping pulley on the motor side → Check it
many times, I did get a wrench into the grubs screws and found then not as tight as I would have nade them, but none were loose enough to slip.
- Stuck / slipping pulleys on the idler side → Check it
powered down, can be moved easily, all idlers appear to turn normally.
- Octoprint - Do you have anything in the GCODE script settings?
There is some junk in the cura prelude, some of which can be edited, and some that is ahead that cannot be edited.
Let me see if a head on one of the files can be pasted;
;FLAVOR:Marlin
;TIME:16176
;Filament used: 13.0285m
;Layer height: 0.2
;MINX:104.412
;MINY:99.412
;MINZ:0.2
;MAXX:195.587
;MAXY:200.587
;MAXZ:5
;Generated with Cura_SteamEngine 5.2.1
M140 S90
M105
M190 S90
M104 S245
M105
M109 S245
M82 ;absolute extrusion mode
G4 P600; kill 10 minutes for bed temp to stabilize
G28 ;Home
;M420 S1 ;Enable ABL using saved Mesh and Fade Height
G92 E0 ;Reset Extruder
G1 Z2.0 F3000 ;Move bed down
G1 X10.1 Y20 Z0.28 F5000.0 ;Move to start position
G1 X10.1 Y200.0 Z0.28 F1500.0 E15 ;Draw the first line
G1 X10.4 Y200.0 Z0.28 F5000.0 ;Move to side a little
G1 X10.4 Y20 Z0.28 F1500.0 E30 ;Draw the second line
G92 E0 ;Reset Extruder
G1 Z2.0 F3000 ;Move Bed up
looks kosher to me
Sorry to say but your understanding is completely wrong:
Neither corexy printer here has the crossover shown at the rear in this image, the two belts are independent, yes
with one having its ends attached low on the carriage, the other high on the carriage. high or low path determined by turning the motor pulley upside down on one motor.
As you can clearly see, you have two independent belts. One could be slack like hell and one having a good tension and the head would still stay in its very same position → Check it
Have, many times, the tension are nearly equal and more than enough to prevent the belts from slipping a cog.
Way forward
- Systematically check / work through the bold marked items above
- Make sure we are on the initial configuration, i.e. both axes inverted and the motor plugs in their original position
Done, bpi5 rebooted, octoprint restarted in its own named venv. chromium watching localhost:5000/term atm.
- Provide a video of
G28 XandG28 Yseparately
New g28_x.mp4 on my web page, x homed normally, no y movement. left x motor locked, en low. will see if motor toggle works to release it. Yes, confirmed by scope. and , I can now reposition to middle of bed by han, will take movie of g28 y now.
New g28_y.mp4 on my web page.
I turned on the scope and watched both enables, x stays low after home, y does not, I’ve left it watching x home to see if it times out, I think it is supposed to to keep from burning up the motors.
- Provide a video of moving the head 10mm in -X direction and 10mm in -Y direction separately (after homing and using Octoprints controls)
I’ll go get those now.
No I won’t, the camera and the mouse to click are too far apart.
I had to use octoprints homing on all axis’s including z before it would let me just move it. Before it would move in any direction. Then I gave it 10mm of Z clearance. That homes it at the right rear but it homes X like its supposed to, then moves x to x_max before it starts any movement to the rear, so its homed at x_min but sitting at x & y max, if I then move y -10 it moves both x & y at a 45 degree angle, fwd and left. For every click on the y- button.
Doing that 10 or so times get me room enough from the stops to move it. from there I can move x plus only because it thinks its still on the home switch which I’ve checked many times, its normal. and once moved to the right, can be moved left only as far as it thinks is xhome.
Both enables are stuck down, but the right hand motor can be moved while the left hand motor is locked tight like it should be, the directon it moves x is correct when it thinks it has room to move, Y always moves at a 45 degree angle because the y motor is never actually enabled. And interchanging the tmc2209’s does NOT move the problem. Either the board is bad, or the corexy kinematics is busted. I vote for the board or you would be up to your neck in corexy complaints. New videos of g28_x.mp4 and g28_y.mp4 are on my web page.
It is sitting high on a dining room table so I took a stool to stand on to get a better angle. And you can see my pacemaker driven heartbeat in the videos. Sorry about that. I’m out of room for more printers so I’m working on the trident in the kitchen.
Take care and stay well, Sineos.
Are following observations correct?
- The X motor spins correctly
- The Y motor actually never spins, regardless if G28, G28 X or G28 Y?
- If you swap the TMC2209 between X and Y slot on the board, the situation stays the same as stated in 1. and 2.?
If yes then:
-
Issue a
STEPPER_BUZZ STEPPER=stepper_ywhile touching the y-stepper’s axis / pulley. Do you feel any movement / reaction? -
Post the output of
DUMP_TMC STEPPER=stepper_y -
Check the wiring of the y-stepper, especially the board connector. Often the wire breaks IN the connector or a pin is slightly pushed out when inserting the connector into the board
-
Move the y-stepper connector to E1:
and modify the cfg as follows:
[stepper_y] step_pin = PD15 dir_pin = !PA1 enable_pin = !PA3 microsteps = 16 rotation_distance = 40 endstop_pin = !PD2 position_endstop = 300 position_max = 300 homing_speed = 50 [tmc2209 stepper_y] uart_pin = PD8 run_current = .8
its caught on something in the bottom of ther printer and won’t reach the e1 socket so I’ll have to clear off enough room on the table to turn it up and take the bottom off again. Back later when I’d done that and edited the printer,cfg to match.
this is slightly different. X still moves like it did, but and y found home on the 2nd try.
Feeling the right hand belt while doing a third g28, its still homes y with the head about 50mm from xmax.
And the y motor, is locked but only when stopped, while running the left motor, the right one is free to be dragged along.
The Y cable is seated properly on both ends. Just for S&G, I’ll change the sign of y enable now.
Did, restarted klipper, exactly the same results except y is now free while its running, and after its homed. But I need to restart my locl network to get you a new klippy.log.
klippy.log (717.3 KB)
Take care and stay well.
Cheers, Gene
Would you go through the points above, please.
Do you have a spare stepper to exchange the y-stepper?
The videos of what is actually happening are a real help. Please continue to provide them.
I would be pleased as can be if, when you send me a msg, and I see the notice that its here on a popup, but it then takes at least half an hour to be able to see that msg, and now even the web link is not showing it to me in firefox.
WTF?
no output, no motion. G28 still moves x, y fails, leaving a at about 285 ,doesn’t try Z
I started a new kkippy.log:
klippy.log (138.9 KB)
Cheers Sineos.
Gene
I do have more motors, but powering down, I am measuring between 2.7 and 2.8 ohms for each coil, and if the motor is energized. its locked up tight. But this card is not energizing it when it should.
klippy.log (232.9 KB)
DUMP_TMC stepper_y
========== Write-only registers ==========
SLAVECONF: 00000200 senddelay=2
IHOLD_IRUN: 00081919 ihold=25 irun=25 iholddelay=8
TPOWERDOWN: 00000014 tpowerdown=20
SGTHRS: 00000000
========== Queried registers ==========
GCONF: 000001c4 en_spreadcycle=1 pdn_disable=1 mstep_reg_select=1 multistep_filt=1
GSTAT: 00000000
IFCNT: 0000004e ifcnt=78
OTP_READ: 00000010 otp_fclktrim=16
IOIN: 21000040 pdn_uart=1 version=0x21
FACTORY_CONF: 00000010 fclktrim=16
TSTEP: 000fffff tstep=1048575
MSCNT: 00000008 mscnt=8
MSCURACT: 00f7000c cur_a=12 cur_b=247
CHOPCONF: 34030053 toff=3 hstrt=5 tbl=2 vsense=1 mres=4(16usteps) intpol=1 dedge=1
DRV_STATUS: 80190080 olb=1(OpenLoad_B!) cs_actual=25 stst=1
PWMCONF: c80d0e24 pwm_ofs=36 pwm_grad=14 pwm_freq=1 pwm_autoscale=1 pwm_autograd=1 pwm_reg=8 pwm_lim=12
PWM_SCALE: 0000001d pwm_scale_sum=29
PWM_AUTO: 000e0024 pwm_ofs_auto=36 pwm_grad_auto=14
SG_RESULT: 00000002 sg_result=2
This is your output of DUMP_TMC STEPPER=stepper_y
DRV_STATUS: 80190080 olb=1(OpenLoad_B!) cs_actual=25 stst=1 shows that the TMC is not happy.
Since you verified that it is not the TMC itself, then we likely have an issue with the stepper. In my 3D printing career I have not had a dead stepper but quite some wiring issue.
Check the wiring again, especially on the boards side. Use a fine pliers to gently pull on each cable individually. Inspect the entire harness etc. If it does not help, try exchanging the stepper
Rather than pulling on each lead with a pair of pliers, could I recommend that you check each wire at each end with a digital multi-meter to make sure there is continuity?
Pulling on the cables won’t necessarily find a problem and could actually damage them down the road.
As I understand Gene’s problem, the Y stepper is locked solid - if there were a cable problem then you would see the motor vibrating and skipping.
@gene1934, I’m not completely up on the latest with your problems, but before you disconnect the Y Stepper cable from the controller to check continuity with the digital multi-meter, could you connect it to another stepper motor and tell us what happens? If it locks up solid then you most likely have a software problem. If it turns, then you need to take a look at your belt paths as the problem is mechanical and not electrical/software.
Sorry for budding in.