Dro magnétique pour schaublin 102 ou fraiseuse

P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
La résolution est divisée par deux :-D.
1-1 : 512 pas par mm
2-2 : 256 pas par mm
Il faudrait que je refasse des mesures pour voir la precision
 
D

Doctor_itchy

Compagnon
20 Avr 2007
7 771
Morlanwelz
ha oui :/ dans le premier cas on arrive a presque 2micron de résolution , dans l'autre a 4 micron , tout dépend du besoin :)

bon je vais continuer a chercher de la 1mm :) renisaw en avais mais le prix .... j'ai 4 ou 5 ttl chinoise pour le meme tarif +- ^^
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
have you some ideas to start... if you say its not difficult i would try but i dont know where.

a little bit help is great:)
Hello,

The idea is good but the menu of the dro need to be change.
I work with a new menu library for my new project but i want to know if you can change the navigation button, it's more easy to change a value and for the navigation ( need 6 buttons) : 5 way joystick and a cancel button

Up
Left OK Right Cancel
Down
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
After, i will be easy to add a tool menu for choose the tool and entrer the X,Y for each tool.
These settings will be save after a modification in flash memory
 
H

hitchhiker

Nouveau
4 Août 2020
16
its no problem! i can etch a new mainboard.
if you can code it, i would send you my pcb layout.

my layout is for 3 axis,i2c display,6buttons and hold a bluepill board.
its singlelayer!
its very easy ,because you dont need to solder smd parts.

the mainboard is conected to a second board which holds the display and the button shield.

where can i contact you direct?
i would send you layouts (eagle), some pictures.

for me its no option to buy big dro displays.thats why i hope you can help me to upgrade your simple dro with a tool lib.

thanks
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
Bonjour,

J'ai passé tout le code avec le nouveau menu et la nouvelle lib d'affichage u8g2, cette base va aussi me servir pour le filetage avec le moteur pas à pas.
Je réfléchis à comment implémenter le choix des outils et je cale un peu sur comment ça doit marcher.
Le choix de l'outil avec des paramètres x, y propre c'est bon mais comment ça doit marcher :???: .

Pour l'instant, chaque codeur à sa valeur absolue qui est prise à la mise sous tension et qui suit le déplacement de la règle. J'ai aussi 2 offset ( absolu ou relatif suivant ce qui est sélectionné).
Donc : Valeur_Affichée = Valeur_Absolue - offset
Comment doit fonctionner le correcteur d'outils ?

SI vous avez des idées.
 
H

hitchhiker

Nouveau
4 Août 2020
16
hi

my idea was:

tool1 is the master tool.
with a fine dial mounted to the chuck set x and z.

now save tool 1 and zero x and z
then put in tool 2 and set x and z
now put tool 1 again in, change to tool 1 and test if tool 1 is x/z 0 when its rouch the dail
then put in tool2, change to tool 2 and look if tool 2 is x/z 0 when its touch the dial..

if this work its only necessary to put tool 1 in, activate tool 1 and zero x/z via dial in the chuck... every time the dro is shutdown.

hope you understand my idea?
what do you think?

standart dro you can buy work same? i dont know... but if think the master key is to save the difference between tool 1 and tool 2
if this all work is the next step to measure in tool 3.

if the toolchange between 3 tools possible.. then it should work.

a china dro works same i think... we have no reference switches or something like this.

it calculate the difference between all the tools. find the zero points is another thing.

i have done it in the last days like this :

i put a tool in. crank it in Z to my backplate with thin paper between cutter and backplate. if the paper cant pull out i zero Z.
same with X. i crank x to the outside of the backplate and make the paper method.

after this i measure the diameter of the backplate and crank the radius in X in and zero.

for Z i crank to the chuck and zero it.

for me it works in a tolerance of 0.02mm.
if i need perfect tolerance i test cut a piece of brass to exactly 10mm in a collet.
then i crank in x the radius in zero.. crank back to 5mm radius and make a testcut.. 0.007mm tolerance. and this is my way to get returnable results.



hope you could understand me

thanks

what do you think?
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
OK, I understand, the tool 1 is the reference and i save the difference for the other tool. I use a select box for choose the tool ( tool1,...tool n) in the menu.
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
Hello,

I work on the code with the new lib. I just add the select box for the tool. Not yet code the behaviour...
The code : https://github.com/pailpoe/MiniDro2

FBDED093-875F-4474-9EEE-00740B4F70E6.jpeg


18A1DCEE-777B-42A8-B0E0-D1C498CEFA89.jpeg


2A02E602-FD76-4731-93AC-042E46640B38.jpeg
 
Dernière édition:
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
yes, a zip if you want...
 

Fichiers joints

  • MiniDro2-master.zip
    26.7 KB · Affichages: 62
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
Bonjour,
que pensez-vous de cette bande magnétique Bande ?
Je vendeur a confirmé ce jour l'auto-adhésif.
Merci.
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
Salut,

Comme je l'ai dit au dessus, il faut normalement selon la doc de la bande 1+1.
J'avais fait des essais rapide avec de la 2+2 et j'avais pu vu de différences à part la résolution qui passe à 256pas/mm.
 
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
Merci,
mais la 1+1 est introuvable à part 'Paulimo't qui ne livre pas en France en ce moment (cause Covid19 !).
Je ferai des essais avec la 2+2 mm.
Je comprends bien le dernier 2 mm qui indique le pas, mais le premier '2+' quelle est sa signification ?
Vous reste-t-il des 'pcb' avec le capteur ?
 
C

coredump

Compagnon
8 Jan 2007
5 242
FR-06
Pour info on trouve des capteurs tout fait pour environ 50€/60€ sur ali, je viens d'en prendre un pour le transversal de mon tour.
C'est par contre de la bande 5+5 qu'il faut, mais les vendeurs l'ont.

 
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
@coredump,
merci, j'avais déjà vu ce capteur : la bande est vraiment peu chère 5+5mm => résolution 1/100 mm
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
J'hesite aussi a acheter pour le Z de ma fraiseuse mais que vaut la précision de l'encodage de la bande ?
J'ai un peu peur sur de la bande chinoise.
Dans le lien ebay : "System accuracy: + (0.03+0.01*L) mm" C'est pas top.
Sur la bande que j'utilise, je suis sous le 1/100 de mm en réel.
 
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
Cette information est pour la 5+5 mm, mais pas d'indication sur la 2+2 mm.
Je viens de poser la question au vendeur de la 2+2 mm.
 
  • Réagir
Reactions: pailpoe
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
Réponse pour la précision de cette bande magnétique Bande
"2 + 2 mm is 1um"
Je vais commander au moins 1m.
 
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
C'est impossible ! il parle peut être de la résolution avec le capteur. En magnétique, j'ai n'ai pour l'instant rien vu qui tienne une précision inferieure au 1/100 en prenant le couple capteur + bande.
 
M

moufy55

Modérateur
23 Nov 2010
13 222
FR-55 Sud-Meusien
En magnétique, j'ai n'ai pour l'instant rien vu qui tienne une précision inferieure au 1/100 en prenant le couple capteur + bande.
Très intéressant, puisqu’il y en a qui montent (ou pensent monter), des résolutions annoncées commercialement de 5, 1microns...
 
  • Réagir
Reactions: pailpoe
P

pailpoe

Compagnon
29 Jan 2011
1 496
Haute-Saône
Résolution Oui ! mais précision Non !
Avec les capteurs AS5311, j'ai une résolution inférieure à 2µm mais un précision d'environ 1/100 de mm.
La courbe présente dans la datasheet du AS5311 est d'ailleurs claire et j'utilise la même bande qu'eux de marque allemande (MS10-10) :
as5311.JPG

Chez RLS, le mieux qu'ils annoncent est pour la série HiLin avec +-5µm. Le reste de leurs capteurs sont donnés à +-1/100 à +-4/100

Pour de la précision, le mieux reste de l'optique avec règle en verre !
 
M

moufy55

Modérateur
23 Nov 2010
13 222
FR-55 Sud-Meusien
J’en étais convaincu mais l’argument théorique est le meilleur :-D
 
C

coredump

Compagnon
8 Jan 2007
5 242
FR-06
Et en plus j'ai de gros doutes sur le comportement en température, la ou avec le verre c'est relativement constant.
 
L

LETARTARE

Compagnon
27 Sept 2010
553
Haute-Savoie
Cela étant, ce n'est qu'un revendeur qui ne l’annonçait pas directement et qui réponds à une question : cela a l'avantage de l'engager !
 

Sujets similaires

Pervenche
Réponses
2
Affichages
989
Pervenche
philou du 25
Réponses
21
Affichages
1 148
Tours
philou du 25
philou du 25
C
Réponses
5
Affichages
660
Tours
rabotnuc
R
C
Réponses
2
Affichages
642
Tours
cascador
C
@
Réponses
18
Affichages
1 609
Tours
thierry74
thierry74
thierry74
Réponses
53
Affichages
2 460
jon
@
Résolu SCHAUBLIN 50
Réponses
13
Affichages
1 189
Tours
@YakumO4444
@
thierry74
Réponses
18
Affichages
1 325
Tours
thierry74
thierry74
O
Réponses
13
Affichages
725
Tours
Comex007
C
M
Réponses
9
Affichages
5 548
Tours
MARC_GTHR
M