Advanced Search

Hi
I have the 400w version, should be the same. I'll post my comfig as soon as I have access, but it rather looks like an issue of drive side configuration.

-does the manual jog according to the manual work?

-can you readout any error codes? For example when toggling the arrow buttons on the drive

-did you do configure the drive with their software over usb? I did PID tuning there and also had first time power on the motors through this.

-there are some digital inputs that default to overtravel/undertravel with active low according to the manual page 212, make sure these are connected properly or disabled

Again, not steppers, those are permanent magnet brushed or brushless servo motors.
I would not change the drives and motors, that would introduce mechanical issues that might be easy to fix if you have access to a mill.
I would also not change any automation and e-stop wiring, just wire in the required stuff to Mesa and LinuxCNC.
-
Transformers do not blow, usually they just burn in place with some smoke and a distinct smell, but if left like this longer they do blow violently if any wire brakes....makes me ask: Where are the fuses? GFI?

the transformer for the drives is broken, and replacements are old we got one to try and fix it but it blew again so im looking at replacing as much of the old stuff as possible. Im keeping the motors im just not sure where to find new modern drives for steppers this big. I guess what im looking for is would it be easier to find a new drive boards to hook up to a mesa board or should i make a new power delivery for the old drive boards and try and use thoes with the mesa board

Perfect, ty so much.
I have a problem with the 7i84u.
In the datasheet mention W1 y W3
But apear W4.
What this do?

Do not change the drives and motors, if they work.
Those seem like analog servo drives, so you would need to wire in the analog input from drives to Mesa 7i77 or 7i97T analog outputs, drive enables, and encoder feedback. In general, that is it.
For detailed info on wiring and testing and tuning the drives (do not move the pots on the drives, ever, no matter what chatgpt tells you) have a look here, and when you start wiring, do not skip steps:
forum.linuxcnc.org/10-advanced-configura...to-example-mesa-7i77

phino wrote:

If a mod could fix the video links in my original post and remove this post, that would be great.

Nope, can not be fixed, something in the links is messing with the editor.
Use youtube if possible as that works fine.

This is the motor drive/power supply the greay wires on the bottom go to the old computer and the ones on the top, i belive the 11 pin connector is the encoder and the 3 is the servo motor. Would i need to use thees drives or another drive board with the 7i7 board? 

What Mesa cards do you have. To reuse the original drives you would probably need a 7i77 Mesa card or similar analog servo control card.

For some reason, the video links in my post above are not showing, and attempts to edit the post don't show me the full post to reach the video links.

I'll try to post the video links in this post.

Drilling holes:

i.imgur.com/mOQMjpb.mp4

i.imgur.com/CB8KpLx.mp4


Straight cuts:

i.imgur.com/YGVQZon.mp4


If a mod could fix the video links in my original post and remove this post, that would be great. A post preview function would help troubleshoot.

That machine uses servo motors not stepper motors. (Probably using analog commands between the control and the drives.)

Lukasz from Proma-Elektronika kindly sent us a unit to test, and we are quite happy with how it performs on our fiber laser machine. 





The height sensor is used to automatically set and/or maintain a fixed nozzle-to-material distance which has been determined appropriate according to the focal point of the laser beam, i.e. "autofocus".

Autofocus is enabled in the following clips, and the readout from the device shows it is maintaining the set height closely, including during/after indexing. The drilling speed was slowed down to make it easier to observe.

The C2V Laser Sense has not had issues with noise from operation or the sparks generated, which eventually became a problem with an optical height sensor used previously with this machine.

Drilling holes:

i.imgur.com/mOQMjpb.mp4
i.imgur.com/CB8KpLx.mp4

Entry holes: 


Exit holes:


Straight cuts. Note the visible autofocusing at the very beginning:

i.imgur.com/YGVQZon.mp4

Result:




The testing we did is so far largely qualitative but we can say it comfortably achieves its specified 10 um (0.01mm) resolution, which is important to us. At some point it would also be good to validate the stated sampling frequency (3.65 kHz) and response time (< 2 ms).

Aside from performance, one standout feature of this unit is the information feedback right on the unit. The LED and built-in display are tremendously helpful, both during calibration, and during operation. Really a very useful feature!

Setup was quite simple and the documentation of the device function, modes, wiring diagrams, and calibration routines were helpful. A cat6a cable (shielded, twisted pairs) was perfect for making the data/power cable. 

For integrating the C2V Laser Sense to our existing machine using an Aerotech A3200 controller, it was straightforward to just use the analog output from the device. It would be great to see others configure this device with LinuxCNC. Given the many modes of operation offered, it should be possible to get it set up with quite a variety of hardware.

Since precision is our priority, we went with Mode 0 which has a non-linear (exponential decay) relationship between voltage and distance.

Good grounding of the chassis was important to getting good calibration results, and luckily the calibration process is easy. Fast calibration for the reference position is just a button press. The precision calibration requires programming a small routine to get the motion speed and distance required. 

The best part is the colors of the LED indicator make it easy to see when calibration should begin, when it is in progress, and most importantly, whether it has failed or completed successfully. And the built-in display even shows the calibration curve; a very nice touch!

In one instance, upon powering up the machine, the device's reading was "frozen" at a fixed voltage regardless of distance, or even touching the nozzle tip. We never figured out what the cause was. Re-initializing the system and checking the grounding did not help. But resetting the unit and recalibrating it resolved the matter. We were not able to reproduce this issue, so I guess chalk it up to machine gremlins.

It goes to show how useful the readout from the built-in display is, as that was the first and most obvious clue that something was not as expected. 

Overall, very happy with the C2V Laser Sense. Thanks to Lukasz for sending us this unit. We'll be looking to purchase more units for future machines, hopefully running on LinuxCNC. If others have used this unit with LinuxCNC, please do post your results!

"I typed stepper motor when i ment servo motor, there are no stepper motors on the machine sorry for the confusion "

Im modernizing a Clausing Komdia fv-1 it has a dynapath cnc control. The dynapath computer is old and has issues, and the servo motor drivers/power supply are having issues as well. I wanna look into getting a modern driver for them that can work with a linuxcnc computer. The stepper motors are baldor motor big mho 4201990, its a 39amp peak 120v 40v/krpm 2500rpm and the encoder is 1000l im guessing thats 1000 ticks per revolution. Im also not sure what connection type i should go with i see the main 2 mesa baords have either ethernet or serial

OK that sounds.

Not sure why you are getting the watchdog bite

What is the host computer, and have all power saving option been disabled in the BIOS?
 

Hallo, da habe ich eine Null vergessen.

PING 10.10.10.10 (10.10.10.10) 56(84) bytes of data
--- 10.10.10.10 ping statistics ---
60000 packets transmitted, 60000 received, 0% packet loss, time 60110ms
rtt min/arg/max/mdev = 0.055/0.071/0.181/0.004ms

That looks OK, but I  would go longer, that's why I suggested 60000 (= 1 minute)