Advanced Search

My crystal ball has stopped working....
More info, pictures, configs...

Hi all;

My question is simple. No changes have been made to wiring or configuration, yet Y2 stopped working. I verified connectivity between the motor and the breakout board, all is good. I next tried swapping the plug position between my A axis and the Y2 axis, modified the hal file to reflect the swap and restarted LCNC. Now the X axis has joined the problem. Z still responds correctly.
What is more likely, a failure of my MESA 7i76U driver board or the breakout board or both?

Kind Regards,

Ich habe einen Fehler
beim Anklicken von Hilfe bekomme ich diese Fehlermeldung
I have a mistake…..
When clicking on help I get this error message

king regarde
Juergen

Yes, exactly.

Yes it is the C axis. It's just that the A, B axes in the image are not in the referenced orientation.

metalkesh wrote:

C axis is for rotation of the welding torch and this torch has offset between torch tip and rotation center of the C 

This is not the C-axis.

I see. Yes, that would require a new kinematic model. Basically the 'xyzab-tdr' expanded with the tool side c rotation and the torch tip offset.

Hi,

i attached representive photo of the machine. 

C axis is for rotation of the welding torch and this torch has offset between torch tip and rotation center of the C axis This ofsett can modified by operator according to welding jig.

A and B axis are for rotation of the object that operator needs to weld.  

I'm afraid I don't have a ready made kinematic for a machine with 2 work side rotaries and one spindle rotary.

Do you have a picture of the machine geometry? I'm having trouble picturing a spindle with a C rotation only. C would rotate around the Z axis and usually the spindle itself is aligned with the Z axis.

Hi Aciera;

First of all, thanks for your great work.

with config "xyzbca-trsrn" it is very easy to make nutation-angle = 90 if machine doesn't have nutating head and it works properly with 90 degree nutation agle.

what about if A axis is on the plane like B axis, and only C axis is on spindle ? is there any easy way to do it with config  "xyzbca-trsrn" ?

i checked "xyzab-tdr" config too but this config doesnt have C axis on spindle. and also i checked "xybc-trt" .

what is the correct way for  configuration xyzbac- table rotay- table rotary  - spindle rotary ?   

 

6 SmartSerial interfaces with ZYNQ 7010 and SPI Ethernet , on first one ethernet port Ethercat is configured

I ordered new switches and replaced them and now the homing sequence is working and latching fully. Now homing works all the time with no errors and I did not change any settings in the configuration.
 
The old switches are at least two years old so it could have been contact bounce. Big thank you to everyone for your resourceful feedback and kind help.

So I have a Precix machine I picked up a while ago. This thread has been a HUGE help in finding some information that is kept secret. If you own a Precix machine, you may know what I am getting at.

Anyways, I wanted to join in on getting more information out there for it, mainly focusing on the encoder side and how I am managing it using some DC drivers. These drivers are a bit older, but using the Gecko G320 DC servo drive units along with the C46 Differential Single-Ended Converter. Taking the encoder that you have established for the wires coming from the machine, converting the singles to a DC signal for the Gecko Drivers to be able to understand. Again, this is a bit older stuff, but for anyone retrofitting the Precix machine, this will be a key factor, I promise you!

I did go a different route using a C11G breakout board instead of the Mesa Card using the LinuxCNC, but that's because I already had the board for another project and was just using what I had on hand. Check out the photo I included to help anyone else out there.

I will be doing a full wiring diagram on the machine once I sort out some of the other items suchs as spindle wiring, limit switches and such. I'd like to be able to use a RapidChangeATC (rapidchangeatc.com/) for this machine, but not sure the collet and nut would match up correctly. 

randypetersen wrote:

As always thanks for everyone's input!

I updated my wiring diagram per PCW's recommendation for the drive interface. I also took a guess on the encoder interface with the details I got from Magtronic. Hopefully someone can tell me if it make sense or not.

I took off the CAP of the encoder and would have to start to take it apart further to find details PCW mentioned.

I am thinking with the manual I found and the drawing Magtronic sent I can be fairly confident in the wiring, what do you think?

Anyone want to see more pictures (why I don't know) can see them here:
photos.app.goo.gl/wBgjA9fuxgXsEZX96

don't rely solely on the machine setup files, also look over the docs config section that is more kept updated and relevant, the machine setup files are a very very basic setup and may actually be out of date, i need to look at that when i have a moment to sit down at dev next.

use the docs those have been recently updated and contain the more current necessary items you may be missing.

probe basic itself has requirements beyond axis so again only rely on axis as a starting point, simply changing display is not going to work without the remaining items being present.

the post gui hal file CANNOT be removed the base items in the supplied postgui are requirements not optional if you do not use the stock items in the probe_basic_postgui.hal file, it will 100% have issues.

If you are using qtplasmac, I would upgrade to Linuxcnc 2.9.10 using the ISO. Seeing Python 2.7 errors raises alarm bells for me
Also, you may find with qtplasmac you will get better results with master branch so it would be good to learn how to build LinuxCNC from source. Its pretty simple these days. QTplasmac is still being pretty actively developed.