Advanced Search

tommylight wrote:

The link i posted, explicitly mentions making the machine work without encoder feedback to LinuxCNC first, so enable the stepgen feedback in the hal file and omit the encoder feedback, test the machine, set the scaling correctly, and only after getting this done activate the encoder feedback and omit the stepgen feedback.
Closed loop stepers when loop is closed on the drives are viewed as normal open loop steppers from LinuxCNC.
So yes, it should work with your drives, just make sure they work properly in open loop mode to LinuxCNC.

Thanks tommy,
so i will test it today.
hopefully this will work.
This was a lot of work adapting these linear-encoders and make all work, but it would be an extreme optimization of my machine to get it to the level of accuracy I always wanted. Not least to finally get a handle on the occasional step losses on the X-axis.

But with you all, i am sure. At the end this will work! ;o)

So I've got my machine to the point of wiring up and testing motors. Made adapter plates and fitted everything off temporarily.

I'm running into an issue when I hit machine enable in LCNC.
One or more motors will jump, move that is, violently for a split second.
Only for a split second and only once, a few seconds after I hit enable.
Sometimes it'll cause LCNC to error out (servo amp error?) and other times not.
The machine continues to work as if nothing's happened though, if it throws an error i just hit enable again and it's good to go.
No jump after the first startup. I can turn the machine enable button on and off as many times as I like and it's fine after the first time.
Sometimes it doesn't do it at all!
I've checked all the encoder cables and motor cables.
I've swapped all the cat5 cables connecting the drives.
I've auto tuned all the drives in ASDA soft. Any jogging or tuning functions from the drive or the drive software work flawlessly, no jump.

I'm at a bit of a loss. I assume LCNC is sending something to the drive and maybe it's position dependent? I've messed around with how I close out of LCNC looking for patterns but I can't really narrow it down.

I've uploaded a video of the Z axis jumping. It's pretty violent, like max speed and acceleration for 1/10 of a second.

Youtube Video

Hal, ini and ethercat xml are attached. I worked from someone else's files, hence the EL8 name.

PCW, thanks again for your hints! 
Maybe you remember, that I have had that latency problem posted somewhere this year and had it solved. But, I will follow your advice and go through this process agein, tomorrow. I did never do that diagnosis with HALscope but I will do this, too. 
Yes, my thoughts up to yesterday was mechanical fault but the measurements today say no. These hundreds that I found are totally acceptable for me, but the tenths not. I really don't know where these deviations come from. Maybe I check the latency again, and take a  look at Halscope, and got to all screws that fix the Z axis.
BTW _ I made measurements at the milling spindle head with a spring scale: I got 1/10 in Y direction at 10 kg, 2/10th in X direction at 10kg, so this is way under my deviations at the milled part. 

I will report my findings, thanks for your hints again!
Hanno 
 

Thanks for your reply, Lcvette!The alarm comes at the motor, Lcnc does not even react on this, because my alarm loop does not work actually. I have to find out why, in the next steps. Yes, indeed, i have ballscrews and especially on Y, a C5 with double Nut. Yes, I know that paired bearings would be better, but i cannot afford them. The bearings are from NSK,and the dealer says I have to set a spacer in between (0.5mm) to set the backlash to exactly zero. Otherwise I would have a small amount of play within. So, I hope this information was ok. I know other bearings that are paired without backlash, but they are way more expensive. I took measurements today and found just 1/100mm backlash, measured at the end of the ballscrew. I found some more hundreds, measuring at high feeds, driven by my handwheel that makes stepping the axis, not smooth running, there were some more 2-3 hundreds, measured at the side part that holds the Bearing and the motor, I think it's a resonance of the whole system at that side. Maybe additionally triggered by the belt drive. Its a HTD15 mm belt, 2:1 reduction, same as at X axis. Belt tension is strong enough, I can move it just 5-6mm from above in the middle of the belt. Well, I dont have a feedback loop back to LCNC, the motor is a JMC 3.3 Nm Nema 24 CL Stepper with its own closed loop.
So, what FF do you mean, FF1 that is set to 1.0? Because somebody else said a tiny amount of maybe 0,003 should help too.  
 
 
(photo from whole machine is not accepted althoug it is under 1 MB)

Thanks a lot for thinking about my problem! 
Ahh just seeing your next posting - what exactly does seem too big for you!
 

I don't see an option anywhere in pnnconf to set pwmgens to offset mode. Is that something I would have to manually add to hal?

Also what are some tuning settings to use for the servos to get started?

Ive already tried it on 3 different ways now,

1. A youtube Video from Talla82 (looked awsome how easy it would be....)
2. with the Instructions from you (Forum) @rodw
3. ChatGPT whats kind of ... depressing and for sure wont work

I can hang you the current lcec.conf whitch is the setup for my EK1100 hopefully you can help me correct it

Is /etc/apt/sources.list broken?

hängt ab von means depends on
keine means none

unfortunatly i did do the sudo apt update already that would be to easy
just tried again and tried the build essentials afterwards it tells me it is already the newest version 12.9


should I do a reinstall with the english language? It seems od to me that this would depend on the language.

Added this to the link i posted here to better explain what can be used :Thank you for reminding me.
Danke schon.

The link i posted, explicitly mentions making the machine work without encoder feedback to LinuxCNC first, so enable the stepgen feedback in the hal file and omit the encoder feedback, test the machine, set the scaling correctly, and only after getting this done activate the encoder feedback and omit the stepgen feedback.
Closed loop stepers when loop is closed on the drives are viewed as normal open loop steppers from LinuxCNC.
So yes, it should work with your drives, just make sure they work properly in open loop mode to LinuxCNC.

Did you do
sudo apt update
before
sudo apt install build-essential
?

The GPIO pins/ parameters are not really useful except on the
7I95T's parallel expansion connector (P1)

raw inmux or inm inputs are unfiltered ( not debounced )

man hostmot2 for inmux details ( search for inmux )

Ah i see ok.
Pins of the FPGA Chip i asume.
So basicly this would be the name of one of the pins ( hm2_7i95.0.inmux.00.input-00  ....input-23)
is there something different to this pin (hm2_7i95.0.inmux.00.raw-input-00) ?


And those are basicly those that i can use as input or output
hm2_7i95.0.gpio.041.in
hm2_7i95.0.gpio.041.in_not (What does that mean)
hm2_7i95.0.gpio.041.out


Sorry for asking so many questions

Try


For me is says it can find libtool at 
 *** 2.5.4-4 500
        500 deb.debian.org/debian trixie/main amd64 Packages

But I'm on trixie so not identical.
 

Mesaflash shows only physical pins on the FPGA

The 7I95T uses multiplexing to reduce the number  of external connections,
so there are many more TB pins than FPGA pins. Also much of the I/O
is differential so have 2 TB pins per FPGA pin.

One example of this multiplexing is the isolated inputs. The 7I95T has
24 isolated inputs but these use only 6 FPGA pins to access.