Advanced Search

If it weren't for you it wouldn't have been solved so quickly. Updating Ethercat may not be under Linuxcnc's control, but I think that such important files, versions and packages that were running in the past should be preserved or backed up. I appreciate your support and you serve the purpose of the Forum's existence.

Mecanix wrote:

Cant do this anymore bye all wrote:

If you have a look at the figures on the wiz net wiki you’ll see as packet size drops so does the overall throughput.
 
All I could find is the implementation and proposed data integrity stuff. I know you're past your daylight but when you have a chance tomorrow, see if you can find that bit of info you ref'ed and we'll talk to their engineer about it. See if we can flip a Magic Register to disable that supposedly/suspected throttling thingy...

Impl. Ref: docs.wiznet.io/Product/iEthernet/W5500/Application/udp

 

docs.wiznet.io/Product/iEthernet/W5500/A...tion/spi-performance

你不行不代表别人不行

Mesa 7i96S can do 10 axis just by adding a simple BOB (it can do without BOB, but wiring must be careful).
If you do not need much input/output pins, go for that one, later you can add whatever you might need, like 7i85S or 7i84 or even analog servo drives with 7i83.

The 8168-dkms driver requires the installation of DKMS and linux-headers for your particular kernel
These should be installed by the Linux ISO

There's always this:

forum.linuxcnc.org/27-driver-boards/5160...sa-card-should-i-buy

A Mesa 7i96S, plus a 7i85S, plus a 7i84 (or 7i84D) should cover just about everything you've described.

If you want a physical user control station with lots of buttons & handwheels, a 7i73 can be added.

What's got me interested is the 'auto' part of 'auto-lathe.'  Do you mean something with a parts loader or bar feeder?

I've got a 4-axis lathe with a subspindle and I still don't need 9 axes of positioning.

Cant do this anymore bye all wrote:

He’s made quite a few small machines. I like his channel.

After seeing this post, I wasted a lot of time watching all his 55 axis videos.....

Aciera wrote:

Have a look at the mesa boards:
www.mesanet.com/index.html
store.mesanet.com/

It won't get more flexible and expandable then that.

Well, the OP could consider Ethercat which is flexible and expandable but adds some complexity, reduces wiring  and probably adds  more cost than mesa cards

Moutomation wrote:

I buy and use professional devices, I pay money for these devices. I only shared this to prevent anyone from getting hurt. Because I can't do something I did a week ago a week later. For example, who can guess that something that worked a week ago won't work a week later. And since I worked it before, I could start a new job thinking it would work again and I could be disappointed. My aim is to prevent others from falling into these situations.

You really have an unrealistic view of Open source.
This has been fixed today 5 days after I became aware of the issue, during that time,  a lot happened.
1. Issue tested by me.
2. Bug report filed by me with etherlabmaster (5 days ago)
3. Bug was marked as fixed  4 days ago  by etherlabmaster and they kindly pushed changes to the Linuxcnc git repo
4. I asked Linuxcnc devs to  accept the changes 
5. Changes accepted into our git repo
6. I cloned and rebuilt ISO
7. New 1.6 Debs from etherlabmaster refused to install
8. I updated my earlier bug report
9. Bug marked resolved by etherlabmaster today
10 I tested the fix today, confirmed now working
11. I asked devs to deploy a new ISO to the downloads page.

Given what has happened to fix this, I think this is a very reasonable outage.
I pay for software that takes at least 2 weeks before they respond to a bug report so 5 days with open source is way faster than many commercial software services. You have to remember we are scattered all around the world so time zones become significant.

If you don't like the support you received, perhaps Linuxcnc is not for you.

 

Maybe I wasn't clear but my test was with the new Ethercat version 1.6 which now installs from repositories (if you have the new keys)

scottlaird wrote:

IIRC, the Ethercat people just did a major version upgrade after literally years of leading up to it. We should probably either pin LCEC to v 1.5.x or see what it takes to fix whatever upgrade problems are happening. I'll take a look at it later today if possible.
 

Several things happened at the Ethercat people's palace.
1. They rolled out the new 1.6 version of Ethercat
2. Their repository keys expired preventing us from installing Ethercat
3. They had a bug in their code that builds the DKMS (kernel part of) Ethercat

This is now all resolved. Their keys will expire in another 2 years.

Moutomation wrote:

rodw is very helpful to everyone regarding ethercat. Those who use ethercat should pray that nothing happens to him, otherwise you may not be able to install ethercat one day.

People were using Ethercat with Linuxcnc well before I got involved. I just made it easier by joining a few dots so they don't need to build from source and incorporating the great work by Scott Laird into one place.

Some bugs take time to resolve and in this instance they were not in Linuxcnc's control.
You can download a new ISO with the changes from my Google Drive
drive.google.com/file/d/1gEQzgHsj_LjHiGB...k1Z/view?usp=sharing
I have asked the devs to update the ISO on the downloads page with this.

I have tested this ISO and the installation process this morning and I can see my slaves

Now fixed. They had a bug in the DKMS deb files. Grab this Linuxcnc ISO which has the new keys in it.
ISO:  drive.google.com/file/d/1gEQzgHsj_LjHiGB...k1Z/view?usp=sharing
Folder with all generated files:  drive.google.com/drive/folders/1pwDh_jn8...QcQwcLv7?usp=sharing

This  ISO also corrects installing a pre-release version of Linuxcnc and no installs 2.93 as intended. 
I have asked the devs to copy this ISO to the downloads page.

The only thing I can think of at the moment is that your x axis is not homed which would inhibit axis jogging.

Aciera wrote:

You seem to have a lot going on there.
First thing I would suggest is to simplify things by choosing which jog method you actually want to use and then test this method in the 'Halshow' tool without all the connections in your hal.
For example this situation should jog the x-axis:

 

Alternatively this should also jog your x axis:
 

Also, posting your hal and ini files might give us a better idea of what you have going on.
 

 

I commented out the 3 lines related to the Arduino stuff in io.hal and postgui.hal to make sure that they are not doing something strange. I then tried both your suggestions and assuming that "setting" all three in the first example or just halui.axis.x.plus (in HalShow -watch), should cause a jog, it does not in either case. So I think there is something missing between those parameters in halshow and the real time thread causing then to be ineffective. I have attached my configuration files.
 

For anyone copying and pasting - typo in absolute
Should be