Advanced Search

Hi - The 7i77D has been out of stock for a spell - anybody know if they are going to continue this version. Im tempted to use a 7i77 with relays but if the D sinking outputs is going to become available Ill hang on.
Thanks all for a great service on this forum. Good bedtime reading

As far as I know it shouldn't matter if you install it during compilation, otherwise it would be fixed with "apt reinstall linuxcnc-uspace".
What do you see with: uname -a ?
What do you see in the terminal when you start with: latency-histogram ?

royka wrote:

Without seeing (uart) logs it's hard to tell what's going wrong.

With the 6.6 kernel likely you didn't remove the localversion part from the patch like I mentioned before. Probably you can also remove it from the kernel config menu.

If it's something else, you should provide logs
 

Yes, for kernel 6.6 it was the localversion indeed, I forgot about it
but for bot kernels 6.12 and 6.6, when I tested the latency-test I got 0 for all of it parameters!!
(kernel 6.6 with rt patch from kernel.org and 6.12 with rt patch that you gave its link)
btw I installed linuxcnc using custompatch.sh file to be installed during the compilation, can it be the problem for this?

Yeah quite out of the original topic. Maybe there could be a generic consumable thread?

Might sound far fetched but it is true. You will have a molten puddle, bead or what ever you want to call it, in the “root” of the arc. You try to constrain and keep it as steady as you can. The copper ( or silver) around the hafnium/tungsten gives means to conduct heat and electricity.

Of course in the old stuff you didn’t have anything “fancy”. You had plasma with electrodes made of tig rods. When the electrode wore, you could pull more from the torch.

I think it was in the -90’s where hypertherm came out with venting. So you don’t only ramp down the current but you match it to the flow rate of the plasma gas at the end of the cut. So you have a separate valve (close to the torch) that lets out the plasma gas. You then see how fast it gets rid of the plasma gas and you ramp down at the same time and rate. This improves the consumable life significantly.

There is quite a lot of cool stuff behind the “scenes” the more complex the systems get.

I am way to far, but if someone comes from there, i might just do something as i am sure i have 2 of 7i92M and 2 of i have no idea 7i92H or 7i92TH.
Mailing would cost an arm and a leg as i am in Kosovo, still.
Free of charge as much as humanly possible as i can use the T ones easily.
I will check with my brothers and friends if i can arrange something, but nothing fast, that is for sure.

It's a ratio. units of skew in Y  per unit of travel in X. If you set 'skew-y' to 0.1 then moving the y-axis in the positive direction will have the x-axis move 0.1 units in the negative direction for every unit of y-travel.

 

It can not find axis number 3, but it does continue loading...

Hope this helps.

Full error from the popup error

JT

ok but wouldn't it be possible for you to leave a copy of the extracted firmware so that I can study something and implement it in the project I'm doing? I'm thinking about creating something but not using the original handwhee, I want to do something just using buttons, etc.

Hi Aciera
Thank you,  Now look its working  when jog   I will  caribrate my cnc new again.
this unit is angle ?

Thank you,
Asis

 

Well, I had it right here on the table.
Add encoder_config() to setup() like this


Also, in the call to the constructor, change the Index_Z pin to PA3 according to the schematics. I didn't check that, leave it to you to check that the callback is triggered when the encoder passes the index slot

Thank you cakeslob for the response,

I was checking in Linuxcnc system, and the observation was there is no /dev/i2c-1 even if i create it couldn't be read and shows "no such devices or address".

Is it possible to get it ?

What are all the process ?

What am i doing wrong ?

Without seeing (uart) logs it's hard to tell what's going wrong.

With the 6.6 kernel likely you didn't remove the localversion part from the patch like I mentioned before. Probably you can also remove it from the kernel config menu.

If it's something else, you should provide logs

royka wrote:

With 6.12 the kernel needs a small patch to be recognized as realtime for linuxcnc and needs a specific config otherwise it gives an error during compiling:
forum.linuxcnc.org/38-general-linuxcnc-q...6-12?start=10#315437

And see the next post for the necessary config.

LinuxCNC needs X to start, you can try something like:
raspberrypi.stackexchange.com/questions/...hout-showing-desktop
 
 

I've done based on what you said about installing X, and desktop got enabled using start X and linuxcnc showed up. hwoever its latency was really bad, something like 500 us. which it was probably because of not being real-time (without adding the rt patch that you mentioned)

After this, I used that patch that you said and I copied it to /patch/kernel/archive/rockchip64-6.12. However during this build I compiled it with XFCE Desktop. Now after compilation again it doesnt boot at all!!!
I'm trying to find out why it doesnt boot.
Or I can compile it without desktop (minimal) and install X to find out if the latency gets better or not.

Another thing that I did was, I added patch for 6.6.63 rt46 to my 6.6 kernel. but during the compilation I get some errors however it compiles without rt patch

@COFHAL: You didn't get it to work? I'll try to connect up one as I haven't tested it on this board. It may take a few days though.

@SOLD: Where do you need that extra speed in the Ethercat client? Anyways, it should work with the existing soes library. Check SPI settings, reset line pin, and the SYNC0 pin.