Advanced Search

tommylight wrote:

Can you post from a phone or a tablet?
 

Ich habe es heute getestet, denn das Forum war wie so oft nicht erreichbar. Es war auch nicht vom Handy oder Tablet-PC aus erreichbar! 

So i changed that and it still shows the same thing

Try just changing #<_inc> to 28.8 degrees (and see what the preview looks like)
That's 360 / 12.5 though it sounds a bit too large without testing, and I don't know what it will do at 120.
(Sorry, not at a LinuxCNC machine at the moment to test it)

I found some specs for my LC60HS112 motors in the catalog:
model: LC60HS112
step angle 1.8
holding torque 4.0NM
rated current 4.0A
phase resistance 1.56 ohms
phase inductance 3.9 mH
rotor inertia 1200 g.cm2
motor weight 1.807 Kg
shaft dia. 8mm

Here are details on my motors that don't match the above catalog specs to some degree:
motor weight 1.79 Kg
shaft dia. 10mm

I messaged Lichuan to see what they say.
 

The first surprise I see is in the INI file where all the P, I and D parameters are set to zero.

Try increasing P to see if that helps.
(You can do this interactively in Halshow, though that won't save to the INI file)

linuxcnc.org/docs/html/man/man1/emccalib.1.html will save to the INI and is a menu option in the Axis GUI. I don't know if there is a way to run it from the other GUIs.

i hve not used it before, but i think others have.. you will need to pay attention to the timer function in the postgui hal file and i think maybe something else in there but i don't remember. im sure someone using it can better directly assist with whats needed or a search in here may bring something up.

spindle.0.speed-out is clearly there in the screenshot.

Two possibilities spring to mind, one of them rather wierd.

1) You have the net commands for the spindle too early in the HAL file, ie before the motmod module has been loaded.

2) The - in speed-out is the wrong sort of dash. Though it looks right in this case. Maybe try re-typing it using the minus key. I think I typed out the example text on my Mac on a train, and it likes to be clever with typography.

spindle.0.speed-out is clearly there in the screenshot.

Two possibilities spring to mind, one of them rather wierd.

1) You have the net commands for the spindle too early in the HAL file, ie before the motmod module has been loaded.

2) The - in speed-out is the wrong sort of dash. Though it looks right in this case. Maybe try re-typing it using the minus key. I think I typed out the example text on my Mac on a train, and it likes to be clever with typography.

Hallo,
die gute Nachricht ist, es fängt an zu funktionieren.
Über die vermutlichen Ursachen werde ich demnächst noch etwas schreiben, ich habe momentan einen anderen PC verwendet, ich denke das das nicht das Problem ist. Die für das Projekt vorgesehene Hardware (MSI Windbox) hat scheinbar ein Bios Problem, Linux macht da Bootprobleme. Das ist aber eine andere Baustelle.
Der Reihe nach, geflasht wird der NVEM über einen STlink Adapter, ich verwendete hier eine Software unter MXlinux zum Flashen, ich grabe das jetzt mal nicht weiter aus. Interessant war das die Software das Flashen dieser Firmware hier zeitweilig verweigerte:
github.com/scottalford75/Remora-NVEM/tree/main/Firmware
Es gibt aber ein weiteres Verzeichnis hier:
github.com/scottalford75/Remora-NVEM/tre...-NVEM-STM32/Firmware
Diese ließ sich flashen, und die hatte ich bis gestern verwendet. Linuxcnc konnte mit dieser Firmware zumindest den NVEM finden, das Einschalten der "Maschine" fuktionierte, jeder Bewegungsversuch endete in obengenannten Fehlermeldungen.
Gestern hatte ich nun noch einmal den NVEM zerlegt, um den STlink anzuschließen und noch einen Versuch mit der erstgenannten Firmware zu machen. Das Ergebnis war dasselbe Flashen verweigert mit einer nichtssagenden Fehlermeldung. Noch ein Versuch, auf einmal Flashen erfolgreich. Keine Ahnung was da schiefgelaufen ist.
Heute den Testpc mit Linuxcnc und der Anpassung nach der Youtube Anleitung vom Scott aufgesetzt, Controller dran und siehe da: Alles gut.
Die Ausgänge liefern die Signale, keine Fehlermeldungen etc..
Es ist also so das der NVEM beim Flashen bereits eine Grundkonfiguration bekommt, die dann auch ansprechbar ist.
Ein Hinweis, bitte, das ist keine Kritik, vielmehr ein Hinweis auch die Akzeptanz dieser Software beträchtlich zu steigern, vielleicht kann Scott mal das Installationsvideo hier etwas anpassen:

Hier verwendet er alte Verzeichnisstrukturen und es fehlt ein wesentlicher Schritt, der halcompile Befehl beinhaltet hier nur das Compilieren der eth Komponente, es fehlt das Compilieren der Remora-nv Komponente, ohne diese läuft Linuxcnc ebenfalls nicht mit dem NVEM.
Ich bin also der neuen Steuerung und vor allem dem Handrad (GRBL hat das nicht) ein Stück näher. Ich denke das war noch nicht der letzte Post, danke auch an Luca bis hierhin ,
Gruß
Mario

 

Do you think the MPG XHC WHB04B 4 model will work well?

Wireless, wired, or does it not matter?

depending on the type f mpg it is possible. some usb mpg versions require some conflict resolution in the postgui hal but its not too bad.

Current mesaflash version is 3.5.16
 

use bookworm per the docs

I did that as well (think it ended with version .12 2-3 days ago) not sure. But not on both machine and at that point not systematic as there was no difference. But now I will do that again with the current setup... and epp...
I'll update... have a nice weekend...8)

NWE wrote:

 
I know the motor's phase inductance or voltage rating can make a huge difference in top speed attainable. These motors do not have either of those specs listed, making it harder to compare specs, but when I get a chance, I'll try to measure phase inductance.

Higher inductance motors need higher voltage to produce the same speed and torque. This is the first thing that comes to mind, seeing what these motors can do.
sed with steppers, the ones I previously used would never have got this speed on 24V, hardly even at 48V.


 

During COVID, I worked with Marshall, a US engineer as we were going to design a plasma table together. He (mostly) and I (the tester) developed an insanely comprehensive spreadsheet model that benchmarks a stepper drive for current consumption, voltage requirement ambiant heat generation. We wanted to utilize stepper motors insane low down torque for acceleration which we thought trunped a geared down servo (which is a lot easier to design for).
If you can find these specs, mainly inductance, I'll run it over your drive.
 
I ran countless NEMA 23,24 and 34 motors through this model.
I then validated his model by replacing motors on my plasma table with those that passed the spreadsheet test. 

After testing countless motors, For all that is said about low inductance, We busted that myth. Some of the better performing motors had quite high inductance. Rotor inertia seemed to be more imortant.
Some failed because they needed too much voltage (beyond what you could buy a power supply for,
Others failed because they drew to much current (above the motors rated amps),
And some failed because the estimated ambiant temperature eise would have melted the wiring (which is  usually rated at  about 130 deg C)
This particular Sanyo Denki has insanely high rotor inertia and weighs accordingly! I speculated that the reason this was limited to 2 amps was because there simply wasn't enough room left in the drive for heavier wire! On my plasma table X axis, I achieved 60000 mm/min rapids and 8000 mm/s/s acceleration with a 30mm dia pinion and 3:1 gearing. I paid about $20 AUD per drive on eBay and another $10 for the wiring harness. After this testing, I saw the guy had a few left and bought the lot. the RRP of the drives is still around AUD $250. They run silently as smooth as silk.

So I had a few left and thought I'd better use them. I have some DM542a stepper drives here I could have used but my pc doesn't have a parallel port so by the time I bought a card and breakouts or even a Chinese 7i92, I might as well buy a multiaxis Ethercat drive.

Anyway, that's why I am here. My next step is to check I have phase polarity correct. I have a feeling I might have changed something early in the piece trying to get things to work because I know these drives deliver.