[not yet solved] Tevo Little Monster DuetWifi errors

boldnuts

Try this guide and join this TLM group, https://www.facebook.com/groups/TEVO.Monster.Delta.Owners/

0_1550049851528_TLM Ultimate Calibration Tutorial (1).pdf

RafB

I thank you, I will study with precision.

RafB

Hello,
I just control the angle of the towers and I do not have the same exactly the same value. This can be my geometry problem?
I will now control the length of my Delta Diagonal Rod

dc42

Auto calibration will sort out the tower angles for you.

RafB

I think it would be complementary to bring illustrations and or videos to the Duet manual on the essential control points of a perfect machine calibration. Especially for the illiterate technicians like me.

RafB

I still do not understand the procedure of calibrating a delta with Duet but I will not give up. how to disable auto calibration? I think I can have a 1st more homogeneous without the self calibration. I checked everything manually and my tray with borosilicate glass is really flat.

RafB

My geometry problem may come from this problem (see video)
When I make a home and I go back 5mm to put out the leds of the endstops I always have an endstop that stays on.
I change endstop, screwed to racemess and unscrewed to lengthen the head of the screw, set the ten micro to have exactly the same distance between the three cart and the problem is the same, the led remains lit while the trolley is perfectly in front of the optical fork. When I push the cart slightly, the led goes out.

Can you help me understand this problem?

https://www.icloud.com/iclouddrive/0od7VaaViYMMcV5_HiBiwm8dA#IMG_5224

synergy41

Hello,
I have exactly the same problem as you.
I have an identical printer is impossible to turn off the limit switches,

RafB

Tu es Français non?
Tu as une couche de hauteur homogène avec ta TLM?

you have a 1st layer of uniform height with your TLM?

Stephen6309

@rafb Paint the flags with something that IR doesn't go through. Many plastics are transparent to IR.

synergy41

@rafb
Oui je suis bien Français...
Pour la 1er couche c'est pas trop mal à condition que je n'imprime pas quelque chose qui est une trop grande surface

RafB

@stephen6309 Hi, thank you for your proposal but it is a screw without black head

Dougal1957

This post is deleted!

Dougal1957

@rafb sounds like you need to make it wider so that it definitely blocks the IR on the Endstops sitches

RafB

@synergy41 Ça m’a fait tilt quand tu as mis le lien les imprimantes 3D.
Je pense pas que le problème vienne Du réglage de la mécanique. J’ai tout réglé au pied à coulisse. Je vais l’installer micromètre À la place du hotend Et vérifier la planéité du plateau. Si tout est parfaitement plat, je ferai une vidéo et demanderai de l’aide à dc 42 pour comprendre d’où vient le problème. Qu’en penses-tu as-tu d’autres propositions?

synergy41

@rafb I think at home my tray is not really Top Top.
I have to find a glass to try to see if I can print a larger area.
What annoys me the most is the problem of limit stop.

RafB

yes the problem of endstop optical is a known problem with the duet. This is probably what leads to the problem of geometry and therefore the impossibility of having a suitable first layer. I have a borosilicate glass and I use a Thickness Gage to adjust the space between my nozzle and the boro glass bed. My bed is on 3 spring.
I really have a hard time understanding? Having a 350mm bed to print pieces of max 100mm drives me crazy !!

dc42

@rafb said in [solved] Tevo Little Monster + DuetWiFi + DuetTouch7" errors:

yes the problem of endstop optical is a known problem with the duet.

What problem are you referring to? The only problem I am aware of is that some optical endstops are designed to work with a 5V supply and don't work properly on 3.3V. The common ones can be fixed by changing one resistor (the one that feeds the IR emitter in the slotted opto switch). If it is an SMD resistor, soldering another resistor on top (to connect it in parallel) is easier because then you don't need to remove the old one.

boldnuts

The optical ones supplied with the TLM work just fine with the Duet as is (well mine do) and I have a few spare ones that also work fine, it's just that the led switches on and off so fast that could be confusing you.

giostark

Never mind... I reopen the 3d. Unlucky. After age of attempts I'm still here
As I have changed so many parts for reach more accuracy I still can't find a barely good calibration.

If I try to manually to calibrate the printer following this video :
https://www.youtube.com/watch?v=Xk7remEhUx8
I reach some good stuffs , but absolutely not perfect and still full of mistakes.

I noticed: i could accomplished to make a perfect equilibrium of calibration in front of the 3 tower using the screws on the cartridge.
So now if I run this 3 commands.
G28
G1 X-121.24 Y-70.00 Z0 F5000 - tower X
G28
G1 X121.24 Y-70.00 Z0 F5000 - tower Y
G28
G1 X0 Y140 Z0 F5000 - tower Z

I can reach all the points at the same distance from the bed.
Hurra!!! I though... WTF , absolutely not.
If from the center G1 X0Y0Z0 , I go to X-100 the nozzle press toooo much against the bed. Fine , maybe the radius culprit ? Is settled as the original parameter as the rods as all the rest.

After i found the "right" offset for the 3 towers I left to zero the M666.

Now the drama, if I move the nozzle in the points in front of the tower all is good , but 5cm on the left or right all is wrong. No matter I try to change slightly in radius or rods DAMN !!!!!

So the last resort:
The last resort was the AUTO calibration. BUT HELL if that was able to adjust a thing ! All the time produce uneven results.
I tried all the permutation on the Esher bed generator. Tried all the factor from 3 to 9. NOTHING make that damn bed flat and perpendicular to the towers.
Few point or all the possible to use.
The auto cal change the end-stop offset despite is the only thing that seems to manually work. Dramatically declinate the bed. (this was the FIRST problem I face of when I start to calibrate this devil machine)
The H parameters are not useful . They change the height JUST and exactly in the points mentioned and dont declinate all the bed.
I hoped that setting up the offset manually I could help the auto calibration to make a better result for the bed inclination . BUT NO. NOTHING.

So, I'm lost (not in space but in my house looking this metal trap) .

I mounted the nema 17 , 09 degree and properly settled in :
; Motors 0.9 degree
M569 P0 S1 ; Drive 0 goes forwards
M569 P1 S1 ; Drive 1 goes forwards
M569 P2 S1 ; Drive 2 goes forwards
M569 P3 S0 ; Drive 3 goes forwards
M350 X16 Y16 Z16 E16 I1 ; Configure microstepping with interpolation
M92 X160 Y160 Z160 E920 ; Set steps per mm
M566 X1200 Y1200 Z1200 E1200 ; Set maximum instantaneous speed changes (mm/min)
M203 X18000 Y18000 Z18000 E1200 ; Set maximum speeds (mm/min)
M201 X1000 Y1000 Z1000 E1000 ; Set accelerations (mm/s^2)
M906 X1800 Y1800 Z1800 E1800 I60 ;

I changed the belt with the ones with iron cores for less stretch and more accuracy.
They are well tied . All sound pretty the same (they sound more or less "like" a"Mi" of a guitar)
The carriages are shimmed with Teflon for avoid torsion ad it work damn fine.
All the structure is pretty sturdy.
The glassed custom build made by a professional industry have tolerance in order of cents.

I have even tried to rise the bed using springs. I thought : if the automatic calibration see my bed declinate I'll correct manually the inclination of the bed. SOOOOOOOOOO ?!?! The auto calibration no matter what end declinate the bed !!!!!

NOW: I noticed something that some good souls could interpret for give me a light spot.

1)If I go to G1 Z0.2 (just for dont touch the bed) and then I move the head to X-100 , when turn to X0 the nozzle dig on the glass. As it was unable to turn in the same position. (belt tied much more or much less dont make difference - motor are at full power and very well cooled)
2)Without the automatic calibration the movements stay consistent inside the 100 radius (more or less). But between 100 and 140 the nozzle rise also of 0.2 (near). I suppose this is the effector shift . but it is really tiny till 100 and rise much more from 100 to 140. I'm avoiding 150/155 because the rising became much more evident.
3)When manually i Try to rise 0,05 sometime those commands are not received. I have to push twice the command for make one result.
When I press the button I can ear different sound form the motor. As first a whistle and then the movement. (the belt are tied , but I suppose that for a so tiny movement maybe tolerances of reverting the turn of the belts are the culprit? ) The motor are set to 1800mA and 160 micro steps.

Any suggestion apart throw everything from the windows?

ps: the Akismet spam filter is REALLY annoying (even if I just want correct a typo), no other alternatives ?