The arms are the same length but one of the belts is a bit loose. When I have my table for the machine I’m going to print the arms again and find a better belt. Then it should do perfect straight lines.
I still need to wait for the right magnets to arrive in the mail.
I wanted to test the homing sequence yesterday but as soon as I clicked homing the robot accelerated to such a speed in seconds that it nearly fell from the table. 
When I have the table I’m going test the hall sensors as my end stops.
Thanks for your help.
Just one question concerning homing: is the homing only enable during homing cycle? What will be happen if a homing switch is trigged during the job? Will the firmware stop the job?
The hall sensors are only enabled during the homing sequence.
If one of the sensors is triggered while printing a pattern nothing will happen.
Ok. This is a special configuration or is it in the standard configuration?
That is the default in Marlin.
Yes I saw on the configuration file. Do you now if it the same form grbl?
I don’t know for sure. I believe you can change the behavior from $ commands though. Homing is definitely different in grbl. If you have endstops enabled, the machine won’t move until you send a home command. It also won’t let you put the home command in the gcode file.
So if someone want to use grbl I just have done the test for the homing. It work perfectly!!! After the homing , the endstop are not anymore working! Without any command!
2 Likes
That’s good news. Grbl on this machine is a good choice. The boards are a little cheaper, a little smaller, and there’s no demand for 2 drivers for either axis. There are even some serial control screens for grbl that should work (although they aren’t as snazzy as a raspberry pi).
1 Like
yes you can find some laser board compatible grbl for few $
Currently, I use cncjs for test and at the end, I will use a PI with sandypi on it connected to the grbl board. do you think it could works?
2 Likes
Yeah. I think it will be great. I haven’t used sandypi myself, but I want to. I imagine there may be a few issues to find, but it should work in theory.
1 Like
Today I just quickly tested the hall sensors as end stops and they worked perfectly.
(Sorry for the funny structure I build to test them)
But there’s something strange going on while homing the axis.
When the sensor detects the magnet it stops as it should but then travels a long distance back around and hits it again.
My 3d printers do the same but they are not moving back this long distance, only some centimeters.
I think there’s something to change in marlin that it does not move back that long. Any ideas?
Also, I bought 2 kilos of baking soda today and the cashier asked what cake I want to bake with this much baking soda. 
Hi. So nice I start my move also today and work well!
For the distance after the homing you can change with the following parametre…
2 Likes
Yeah. Haha. The home bump defaults to 5, which would be 5/6 of a rotation in our goofy units.
1 Like
@jeffeb3 hi. My first test is successful but I have to change the homing position. With you transfirmation code, the home position is top with both arms extended. But, in the case we have a radius table smaller than the sum of both arms, it’s does work.
For me it’s the big advantage of the scara. It can fit to any form with the same system.
What I need to do if I want a homing on the centre? offset -1,5 to X and + 1.5 to Y can work?
- You can home the elbow motor first, so it folds all the way in. Then home the shoulder until it reaches the 1.5 position. Then just tell Marlin where it is with G92 X1.5 Y1.5:
G28 Y
G28 X
G92 X1.5 Y1.5
- You can set the max radius for a pattern in the thr export dialog in sandify. If you set it to 0.9, it won’t go higher than 0.9 radius.
ok thank you.
For 1 you would say
G92 X1.5 Y-1.5 ?
For 2 ok I already use this function and work well during drawing
But what do you think about my proposition? putting an offset?
Sure -1.5 would be fine.
