Advanced Search

Brisbane here.

Personally, I would not use a wifi extender. Instead, I would install a Ubiquiti wireless link with a couple of these
www.computeralliance.com.au/ubiquiti-loc...nostation-cpe-airmax
I ran an 800m link with Ubiquiti Loco M5's which are about $10 more. It was a bulletproof 150 mb link for about 10 years. In the shed, attach to a switch and add a Wireless Access point. In the house, just attach it to your network. Run them on 5G where they can use frequencies still legal but outside of the congested? 5G space.

This would let you use a network port (even if a USB one) on your machine and would be much more robust.

Sorry, dead of night here.
Last test is that the VFD shear pin circuit is being tripped, oddly could coincide with the tensioned belt.
Need to sleep on it and investigate further.
Could be I need to introduce the limit2 and ramp these decels, assuming that limit2 does accels and decels?
Thanks for worrying.

What's the current status?

2.9.4
OK, thanks for your confirmation that this should work. I have another 7i93 in production, I'll check with that and come back with the results.

Sorry I meant mode 1 (there is no mode 0)

This is pretty odd. I can't duplicate this here. I tested by putting a couple LEDs on the
PWM/DIR lines and in output mode 2 the PWM output changes from the PWM pin to
the DIR pin when the PWM value is negative. In addition, the LED brightness was proportional
to the absolute PWM value as expected.

Its especially odd since the PWM module is unchanged since it s original creation (version 0)
and PWM output mode 2 is used on some of our interface cards (7I48,7I49)

What LinuxCNC version? 


 

github.com/multigcs/riocore/blob/dev/doc/RIOSUB.md

I think I need a different name for it
suggestions are welcome !

not with type 0, I couldn't set it but with type 1 works (I mean, logic level of the corresponding pin changes properly)
so I think there is no hardware issue.

Thanks Tommy,
Looks like my constant playing with speeds up and down the range has had something over as it won't now swap directions without the spindle enable relay dropping out.
Worked this morning.
Poor old thing. It, not me, or maybe me as well.

Not my forte, but probably missing "spindle at speed" signal in hal ???

So, armed with a working spindle encoder, I've tried my luck with a rigid tapping cycle.
Spindle runs, and the code gets to the G33 line, but after the spindle advances down a little, motion and the spindle stop without any error message.
Looking here  Docs/Examples/Spindle  section 6.1 I am assuming that I don't have to add yet another encoder to the hal, but can use this working one?
I did test the spindle running at around 500rpm, and simply changing direction, and the system manually happily does that.
So, not sure where to look.
Thanks

Complete programme below in case something is wrong.
%
(RIGID TAP M8X1.25)
(T1 D=8. CR=0. - ZMIN=-11. - RIGHT HAND TAP)
N10 G90 G94 G17 G91.1
N15 G21
N20 G53 G0 Z0.
(DRILL1)
N25 T1 M6
N30 S500 M3
N35 G17 G90 G94
N40 G54
N45 M8
N50 G0 X0. Y0.
N55 G43 Z15. H1
N60 G0 Z5.
N65 G33.1 Z-11. K1.25
N70 G0 Z5.
N75 X0. Y0. Z15.
N80 M5
N85 M9
N90 G53 G0 Z0.
N95 M30
%

rick-linuxcnc wrote:

I am trying to upgrade linuxcnc to 2.9.3 I was running 2.7.x I am using all the same hardware but when the parallel port tries to load it gets the I/O ports at e000 [disabled] =8error. On the old version is was running a port 0xe000 and irq 3


lspci -vv
05:00.0 Parallel controller: Nanjing Qinheng Microelectronics Co., Ltd. CH351 PCIe Parallel Port Adapter (rev 0f) (prog-if 01 [BiDir])
    Subsystem: Nanjing Qinheng Microelectronics Co., Ltd. CH351 PCIe Parallel Port Adapter
    Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
    Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-

 

Is this the same computer that was used before?  How about the parallel port card?
The [disabled] indication means that the port will not work, the kernel has done something to disable access to it.
I had this issue with a specific Dell desktop machine, and had to use a different PCIe parport card.  I recommend cards with the NetMos 9900 or 9901 chip for use on PCIe slots.  These have always worked well for me.  
Jon

That file is the Linuxcnc driver, if that is missing there is something wrong with installation.

Intel Pentium 4 CPU 3.00G
Grafica Radeon 9200 pro
512MG RAM

Ahhh thanks, that extra joints is what I need for sure.

Hoping I can make a custom M code to control them

Also looks like the number of joints can be increased if recompiled. Per the docs.

The limits for the following items are compile-time settings:
num_joints: Maximum number of joints is set by EMCMOT_MAX_JOINTS.

Will give this a shot, thanks

Invert the enable pin in StepConf.
You can edit the existing config, click on the right side where you have chosen that pin.