Advanced Search

I attach the configuration files to handle the turret. The only difference with yours is that the one I have did not have the encoder and I had to design one, but without the use of parity pins, you can add it by modifying the turret.hal file. My turret only has a 6-position tool changer while yours has 12, that's not a problem, what I did was design my encoder to only detect 6 positions and only use 3 sense pins to get a maximum of 8 combinations, with your encoder you can do all 12, adding another sense pin to the turret.hal file. As I explained in this thread, the turret originally works bidirectionally, but I could not get it to work that way because the carrousel component does not allow it. In this thread they also discuss this type of turret forum.linuxcnc.org/10-advanced-configura...ute-encoder?start=30. I have to search my notes because I found in another thread someone who designed a component for a turret that works the same as the one we have, but is manufactured by another company. This component does allow the turret to work in a bidirectional manner.

tommylight wrote:

CandCNC ?

I was to polite to say anything but a lot of users have bought a THCAD and converted to QTplasmac.

I then looked at your diagram about io. Just check the docs to see what it says for internal homing and limits as if used, it will consume some pins.
I'm keen to get internal homing working so I can do torque homing to avoid adding limit and homing switches

Great move!
The IO is reasonable from Rtelligent and Deiwu. I have xml configs for both
github.com/rodw-au/linuxcnc-cia402
Rtelligent has PWM
I have both but used Rtelligent on my mill.
There is another one on Aliexpress with a few more features including Encoder but you may need to write the XML which is not that hard.
Note the IO's on the drives are more for homing etc but do work for general IO

I do not use internal homing. I intend to one day but I have had my little Ethercat homing lab on my desk for 12 months now untouched...

For spindle I used RS485. It actually interaces nicely with QTdragon GUI and is very stable speed wise when tested with a laser tacho

I did not end up adding an encoder due to cost. I ended up using a thread mill that gives perfect results. I now wonder what all the fuss is about with encoders. Its only ever needed for threading which is rarely done and thread mills are much easier on the machine and don't need an encoder!

+- velocity values are not possible in counter mode unless you have noise on the "B" input

Try jumpering the "B" input for single ended mode
(and make sure nothing is connected to either "B" input at TB2)

 

No and not planing to, the "influencer's" keep forgetting to tell us just what headache that is. It fails, a lot.
As it is, using the USB is enough liability, adding another set of comms is really asking for trouble.

40VDC is enough
57VAC is enough
Start wiring stuff, build the machine, make chips,...

yes! built with canbus. saves a lot of cables and is easy to implement.

For time like this, I generally make a now config  with pncconf with some rough settings. Once you see movement, then cut and paste stepper settings and pins from your old config.

>Halshow has the reading fluctuating from about -3650 -> +3650 with no plasma source having power.
you have a 0 volt frequency of 3637 in you system so 3650 is expected. The swings are not
There is something not right in hal, like you have a double ended encoder or something.

my1987toyota wrote:

There is rumor of Big Tree Tech making one, but that is still just a rumor.

Thats the one!
biqu.equipment/products/bigtreetech-pand...fP0vAWLH50KrWga94hgp

tommylight is this just more of the test?

If you don't want to stick to Ethercat drives for all axes, really you should not have gone down the Ethercat route. Rtelligent and others have closed loop Ethercat stepper drivers. I would get a couple of them. Leadshine and Rtelligent also have open loop drives that control 2 x steppers in a single drive.

Rtelligent and Deiwu both have more affordable Ethercat I/O options

Thanks for the answer but I wasn't referring to how to use and2, hahaha.

I was referring to the limit3 component when the limit3.enable hal pin is set to false the output goes to zero, in case of connecting the and2 output to the limit3 enable if you write a measurement to the limit3.in hal pin and press the buttons you may need to release the buttons during the movement, at that moment I need it to stop as quickly as possible but not to go back to zero, or better yet you can press the buttons and reach the measurement and then when you release them you go back to zero, hahaha.
It may be that some applications require going back to zero but not all, imagine a measuring stop on a saw, or a double head miter saw, no interpolation in this case it is necessary that when you give a measurement you move the joint to go to the measurement and stay there or if you release the buttons in the middle of the movement it stops and when you press them again the movement resumes, this is what I am referring to.

I have now achieved this behavior, I have had to make changes in the limit3 component, in how the enable is interpreted with respect to the measurement, later I will upload the modified component in case anyone wants to take a look.

As always, greetings and thanks to everyone.

The DM556 will be fine at 57 volts AC
I have drivers rated or 65v AC and have run them at 64V and had to pay attention to other parameters as that was so close to the max.
So next time I ran them at 60v AC and then finally found the manufacturers specs which I think said 55 volts was recommended!

As far as DM556 goes, 40 volts is probably OK too. It is possible to calculate the maximum voltage to achieve specific performance. A friend developed a spreadsheet that does this. In a recent design for  2 amp Sanyo denki steppers, it only required 38 volts to achieve 60m/min rapids and 5 m/s/s accelleration on a 29kg gantry. But that was after some trial and error with gearbox ratios and pinion diameter to get to the sweet spot.