[Emc-users] head scratcher

[Gene Heskett]

Greetings all;

Trying to troubleshoot my failed G0704. It apparently took a hit while I 
was in the shop with all sorts of tubing etc stuck here and there 
watching a heart thats working at about 30%. They'll fix that with a new 
valve next month.

Got a pair of control boxes I've assembled. Power button refuses to turn 
green on the computer if a cable is connected from the 5i25's P3 to a 
7i76D doing most of the I/O on this machine. Its in a separate box with 
its own power supplies, which are all good and all the led's are saying 
its all ok. A mesaflash --device 5i25 --readhmid looks like it has 7i76 
drivers hitting both P3 and P2.
root@GO704:~# mesaflash --device 5i25 --readhmid
Configuration Name: HOSTMOT2

General configuration information:

  BoardName : MESA5I25
  FPGA Size: 9 KGates
  FPGA Pins: 144
  Number of IO Ports: 2
  Width of one I/O port: 17
  Clock Low frequency: 33. MHz
  Clock High frequency: 200. MHz
  IDROM Type: 3
  Instance Stride 0: 4
  Instance Stride 1: 64
  Register Stride 0: 256
  Register Stride 1: 256

Modules in configuration:

  Module: DPLL
  There are 1 of DPLL in configuration
  Version: 0
  Registers: 7
  BaseAddress: 7000
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: WatchDog
  There are 1 of WatchDog in configuration
  Version: 0
  Registers: 3
  BaseAddress: 0C00
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: IOPort
  There are 2 of IOPort in configuration
  Version: 0
  Registers: 5
  BaseAddress: 1000
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: PWM
  There are 2 of PWM in configuration
  Version: 0
  Registers: 5
  BaseAddress: 4100
  ClockFrequency: 200.000 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: QCount
  There are 2 of QCount in configuration
  Version: 2
  Registers: 5
  BaseAddress: 3000
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: SSerial
  There are 1 of SSerial in configuration
  Version: 0
  Registers: 6
  BaseAddress: 5B00
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 64 bytes

  Module: StepGen
  There are 8 of StepGen in configuration
  Version: 2
  Registers: 10
  BaseAddress: 2000
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

  Module: LED
  There are 1 of LED in configuration
  Version: 0
  Registers: 1
  BaseAddress: 0200
  ClockFrequency: 33.333 MHz
  Register Stride: 256 bytes
  Instance Stride: 4 bytes

Configuration pin-out:

IO Connections for P3
Pin#  I/O   Pri. funcSec. func   Chan  Pin funcPin 
Dir

 1  0   IOPort   StepGen  0Dir/Table2  (Out)
14  1   IOPort   StepGen  0Step/Table1 (Out)
 2  2   IOPort   StepGen  1Dir/Table2  (Out)
15  3   IOPort   StepGen  1Step/Table1 (Out)
 3  4   IOPort   StepGen  2Dir/Table2  (Out)
16  5   IOPort   StepGen  2Step/Table1 (Out)
 4  6   IOPort   StepGen  3Dir/Table2  (Out)
17  7   IOPort   StepGen  3Step/Table1 (Out)
 5  8   IOPort   PWM  0Dir (Out)
 6  9   IOPort   PWM  0PWM (Out)
 7 10   IOPort   SSerial  0TXData1 (Out)
 8 11   IOPort   SSerial  0RXData1 (In)
 9 12   IOPort   SSerial  0TXData2 (Out)
10 13   IOPort   SSerial  0RXData2 (In)
11 14   IOPort   QCount   0Quad-IDX(In)
12 15   IOPort   QCount   0Quad-B  (In)
13 16   IOPort   QCount   0Quad-A  (In)

IO Connections for P2
Pin#  I/O   Pri. funcSec. func   Chan  Pin funcPin 
Dir

 1 17   IOPort   StepGen  4Dir/Table2  (Out)
14 18   IOPort   StepGen  4Step/Table1 (Out)
 2 19   IOPort   StepGen  5Dir/Table2  (Out)
15 20   IOPort   StepGen  5Step/Table1 (Out)
 3 21   IOPort   StepGen  6Dir/Table2  (Out)
16 22   IOPort   StepGen  6Step/Table1 (Out)
 4 23   IOPort   StepGen  7Dir/Table2  (Out)
17 24   IOPort   StepGen  7Step/Table1 (Out)
 5 25   IOPort   PWM  1Dir (Out)
 6 26   IOPort   PWM  1PWM (Out)
 7 27   IOPort   SSerial  0TXData3 (Out)
 8 28   IOPort   SSerial  0RXData3 (In)
 9 29   IOPort   SSerial  0   

Re: [Emc-users] head scratcher, updated some more

[Gene Heskett]

On Monday 25 February 2019 08:11:39 MC Cason via Emc-users wrote:

> On 2/24/19 11:45 PM, Gene Heskett wrote:
> > To stay organized, I'd need a pdf reader that does the mouse slide
> > copy so I could paste it in. I've not seen one of them critters
> > lately, neither evince nor okular support the mouse copy.
> >
> > Give me a pdf reader that does, and a browser that just works, and I
> > might give it a shot, IF I can ever find whats made firefox so
> > darned allergic to anything that even smells like javascript.  One
> > of the reasons I've been trying to make konqueror the default
> > browser, but click on a link in konqueror, and it opens a copy of
> > firefox. Which of course throws up the javascript error like it
> > can't find a javascript interpretor anyplace on the system. When
> > probably 75% of the javascript stuff in the repo's IS INSTALLED.
> >
> > Cheers, Gene Heskett
>
>    Take the PDF and copy it.  If you do not have PDF-Shuffler
> installed, install it and use it to delete all but the remaining pages
> you need to translate.  Use PDF-Shuffler again to add them back into a
> new translated PDF.
>
>    I use it all the time to edit PDF files.
>
> ---Mark
>
You must be writing about a different pwd-shuffler than the one I just 
installed.  Its badly "out of focus" and cannot be magnified enough to 
read the individual page to determine if its engrish or chinese.

LCNC is sending it the correct signals. But its only responding to the 
non-zero input from the analog pot the 7i76 outpits its speed voltage 
over  While theres 10.3 volts present on both the fwd and rev labeled 
terminals, connecting them to the analog common is ignored. Such a 
connection is equ to manual pushbutton, but pushing it only has zero 
effect. LCNC's drive pulls it some below a TTL logic zero. Anything from 
LCNC is treated as a fwd request, even if the wires for fwd/ena and rev 
are physically disconnected!

I can use it this way, but it bugs the hell out of me that the control 
terminal strip has well labeled fwd and rev terminals that don't do 
squat.


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] head scratcher, updated some more

[Roland Jollivet]

On Sun, 24 Feb 2019 at 23:16, Gene Heskett  wrote:

> On Sunday 24 February 2019 15:07:59 Nicklas Karlsson wrote:
>
> > > Why did they plainly mark the control strip with fwd and rev if they
> > > have no effect?
> >
> > Configuration?
> >
> The best I have for that is in chinese, and google translate stops
> translating in the middle of page 40.  Its about a 70 page manual so it
> stops short of getting to the register descriptions that probably
> control that. And for some reason, mozilla refuses to retry the
> translation.  So now I'm asking how to make konqueror the default
> browser because every link I try to follow in konqueror open a new
> mozilla (firefox) to access the link.
>
> Since its a long document, is there some trick to making google translate
> the whole thing?
>

The simplest way is to 'print' pages 40 - 70 or whatever, but send it to a
virtual pdf printer, like PDF ReDirect. Of your pdf reader might support
'printing' to pdf.

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Gene Heskett]

On Monday 25 February 2019 01:48:26 jrmitchellj wrote:

> On the Yaskawa VFD I am using, there are registers that can be changed
> to alter the function of the run & fwd/rev inputs, so the names really
> are silly.
> There are a large number of those registers that change a bunch of
> things. It took me several days for me to find the ones I needed to
> tweek.
>
> Ray
>
> --J. Ray Mitchell Jr.
> jrmitche...@gmail.com
>
Generally, the lack of docs on most of these things is a major PIMA.

It took a while to find the ones I needed to play with when I bought one 
of the then $130 clones from ebay to run a 3 phase motor on my cnc'd 
Sheldon 11x54.

But I can, within reason, do rigid tapping on that Sheldon now. These vfd 
things can be made to do amazing things.

At 300 revs I have to compensate for the overtravel at the bottom of a 
blind hole else the tap will hit the bottom of the hole due to overshoot 
at the g33.1 reversal, with an 8" chuck mounted, thats quite a flywheel, 
so its around 3.5 turns. Much less of a problem at slower speeds.  And 
the drive belts are yelping.  So I wrote some hal code to measure the 
overshoot while cutting air, display it in both distance traveled and in 
chuck turns, then I subtract about 95% of it for the depth of the tap 
travel in the resultant g33.1 code.  Haven't broken a tap since. But 
trying to hold a tap in a common tool holder like for a boring bar, 
allows slippage. I need to design a tap holder that fits the qctp that 
grabs a tap by the square butt to stop any slippage, thats still being a 
problem on the Sheldon.

The tap "hats" for the milling machine have solved that problem there by 
keying the hat into the r8 holder with a notch for the head of a 4mm 
screw in the side of the hat.

So I need to design and make something that holds a tap mounted in a hat, 
right on the Sheldon's qctp. That will put torque into the qctp, and may 
make me put tapered gibs on its carriage.  Doing that to TLM  made a 
whole new machine out of it.

So if we had the docs, we can likely make it work. The docs we have, fed 
to google translate, are a usable translation up to page 40, but the 
translator stops translating and the other 19 pages are worthless, so we 
have descriptions up to only d050's for registers, and theres 200 of 
them. We do have a list of defaults, but zero explanations for the last 
140 registers.

Thanks J. Ray.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[MC Cason via Emc-users]

On 2/24/19 11:45 PM, Gene Heskett wrote:

To stay organized, I'd need a pdf reader that does the mouse slide copy
so I could paste it in. I've not seen one of them critters lately,
neither evince nor okular support the mouse copy.

Give me a pdf reader that does, and a browser that just works, and I
might give it a shot, IF I can ever find whats made firefox so darned
allergic to anything that even smells like javascript.  One of the
reasons I've been trying to make konqueror the default browser, but
click on a link in konqueror, and it opens a copy of firefox. Which of
course throws up the javascript error like it can't find a javascript
interpretor anyplace on the system. When probably 75% of the javascript
stuff in the repo's IS INSTALLED.

Cheers, Gene Heskett



  Take the PDF and copy it.  If you do not have PDF-Shuffler installed, 
install it and use it to delete all but the remaining pages you need to 
translate.  Use PDF-Shuffler again to add them back into a new 
translated PDF.


  I use it all the time to edit PDF files.

---Mark





___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[jrmitchellj]

On the Yaskawa VFD I am using, there are registers that can be changed to
alter the function of the run & fwd/rev inputs, so the names really are
silly.
There are a large number of those registers that change a bunch of things.
It took me several days for me to find the ones I needed to tweek.

Ray

--J. Ray Mitchell Jr.
jrmitche...@gmail.com



"No problem can be solved from the same level of consciousness that created
it"Albert Einstein


On Sun, Feb 24, 2019 at 9:48 PM Gene Heskett  wrote:

> On Sunday 24 February 2019 23:06:52 Przemek Klosowski wrote:
>
> > >  a long document, is there some trick to making google translate
> > > the whole thing?
> >
> > if you just need to translate a section about some particular register
> > you're interested in, just cut and paste into the direct translation
> > box. Otherwise, download and split the Chinese doc and then upload
> > individual pieces, using teh Documents button above the translation
> > box. I haven't tried translating my own Google Drive documents, but it
> > could work too, and save uploading them back.
> >
> To stay organized, I'd need a pdf reader that does the mouse slide copy
> so I could paste it in. I've not seen one of them critters lately,
> neither evince nor okular support the mouse copy.
>
> Give me a pdf reader that does, and a browser that just works, and I
> might give it a shot, IF I can ever find whats made firefox so darned
> allergic to anything that even smells like javascript.  One of the
> reasons I've been trying to make konqueror the default browser, but
> click on a link in konqueror, and it opens a copy of firefox. Which of
> course throws up the javascript error like it can't find a javascript
> interpretor anyplace on the system. When probably 75% of the javascript
> stuff in the repo's IS INSTALLED.
>
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
> Cheers, Gene Heskett
> --
> "There are four boxes to be used in defense of liberty:
>  soap, ballot, jury, and ammo. Please use in that order."
> -Ed Howdershelt (Author)
> Genes Web page 
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Gene Heskett]

On Sunday 24 February 2019 23:06:52 Przemek Klosowski wrote:

> >  a long document, is there some trick to making google translate
> > the whole thing?
>
> if you just need to translate a section about some particular register
> you're interested in, just cut and paste into the direct translation
> box. Otherwise, download and split the Chinese doc and then upload
> individual pieces, using teh Documents button above the translation
> box. I haven't tried translating my own Google Drive documents, but it
> could work too, and save uploading them back.
>
To stay organized, I'd need a pdf reader that does the mouse slide copy 
so I could paste it in. I've not seen one of them critters lately, 
neither evince nor okular support the mouse copy.

Give me a pdf reader that does, and a browser that just works, and I 
might give it a shot, IF I can ever find whats made firefox so darned 
allergic to anything that even smells like javascript.  One of the 
reasons I've been trying to make konqueror the default browser, but 
click on a link in konqueror, and it opens a copy of firefox. Which of 
course throws up the javascript error like it can't find a javascript 
interpretor anyplace on the system. When probably 75% of the javascript 
stuff in the repo's IS INSTALLED.

> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Przemek Klosowski]

>
>
>  a long document, is there some trick to making google translate
> the whole thing?
>
if you just need to translate a section about some particular register
you're interested in, just cut and paste into the direct translation box.
Otherwise, download and split the Chinese doc and then upload individual
pieces, using teh Documents button above the translation box. I haven't
tried translating my own Google Drive documents, but it could work too, and
save uploading them back.

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Gene Heskett]

On Sunday 24 February 2019 15:07:59 Nicklas Karlsson wrote:

> > Why did they plainly mark the control strip with fwd and rev if they
> > have no effect?
>
> Configuration?
>
The best I have for that is in chinese, and google translate stops 
translating in the middle of page 40.  Its about a 70 page manual so it 
stops short of getting to the register descriptions that probably 
control that. And for some reason, mozilla refuses to retry the 
translation.  So now I'm asking how to make konqueror the default 
browser because every link I try to follow in konqueror open a new 
mozilla (firefox) to access the link.

Since its a long document, is there some trick to making google translate 
the whole thing?

The answer may well be there, but I don't read Chinese.

Thanks Nic.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Nicklas Karlsson]

> Why did they plainly mark the control strip with fwd and rev if they have 
> no effect?

Configuration?


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated some more

[Gene Heskett]

On Saturday 23 February 2019 17:39:04 Gene Heskett wrote:

Trying to get control over this @#&%$#)* Chinese vfd.

Found a doc that I think applies to the clone vfd thats running my 
Sheldon, it shows all the control buttons as making ground contact to 
exert the control, so I rewire the 7i76 to do that. But this thing 
ignores the fwd and rev signals, responding only to the voltage sent to 
the spinout terminal. I've now done everything in the way of control 
combo's and the only way I get a response is if its stopped, but the fwd 
led on it panel is on, then it will run fwd only. at whatever speed lcnc 
asks for.  And thats disabled if the last button push on the local panel 
is stop, the last button pushed must be run.  If I switch it to MANual, 
it runs wide open, ignoring the speed knob setting.

Why did they plainly mark the control strip with fwd and rev if they have 
no effect?

Theres one register that supposedly controls how thats supposed to work, 
also has no effect.  d091.

Has anyone actually made the 1500 watt spindle in the bigger 6040 gantry 
mill actually work?with LCNC?
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> Cheers, Gene Heskett


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Saturday 23 February 2019 17:16:26 Jon Elson wrote:

> On 02/23/2019 02:34 PM, Gene Heskett wrote:
> > On Saturday 23 February 2019 12:22:54 Jon Elson wrote:
> >> On 02/23/2019 10:41 AM, Peter C. Wallace wrote:
> >>> Umm just loaded a software stepgen (lastet master) and it
> >>> has no such pins/parameters
> >>
> >> Right, MIN_LIMIT and MAX_LIMIT are not handled by stepgen or
> >> any motion interface hardware, they are handled much higher
> >> in LinuxCNC.  To reach a smooth deceleration right at the
> >> limit position requires it to be limited before the
> >> trajectory planner. So, these parameters are picked up out
> >> of the [JOINT_n] stanza by something in the motion section.
> >> Even the G-code interpreter knows about them, as you will
> >> get "program exceeds +X limit at line 1234" messages when
> >> loading a program.
> >>
> >> Jon
> >
> > I've seen that memo hundreds of times. Maddening when its only 1
> > thou outside of that limit.
>
> Much LESS maddening than if you didn't have homing and soft
> limits set up, and you started in cutting material when it
> hit a mechanical limit and wrecked the material or a cutter
> or whatever.
> I'M SO GLAD I finally set up home and limits on my system.
>
Well, now I'll have enough i/o to do all that, Jon.  And the switches are 
in a post bag someplace on the continent. I didn't realize just how much 
of a limit the parport i/o was until I started adding purties to the 
Sheldon. Haveing 72 bits, 60+ of them for gingerbread by useing the pi 
and a 7i90. Biggest problem there is the cost of the 7i42TA's, BOTOH, 
they sure make wiring things up a heck of a lot simpler.

> Jon
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Jon Elson]

On 02/23/2019 02:34 PM, Gene Heskett wrote:

On Saturday 23 February 2019 12:22:54 Jon Elson wrote:


On 02/23/2019 10:41 AM, Peter C. Wallace wrote:

Umm just loaded a software stepgen (lastet master) and it
has no such pins/parameters

Right, MIN_LIMIT and MAX_LIMIT are not handled by stepgen or
any motion interface hardware, they are handled much higher
in LinuxCNC.  To reach a smooth deceleration right at the
limit position requires it to be limited before the
trajectory planner. So, these parameters are picked up out
of the [JOINT_n] stanza by something in the motion section.
Even the G-code interpreter knows about them, as you will
get "program exceeds +X limit at line 1234" messages when
loading a program.

Jon


I've seen that memo hundreds of times. Maddening when its only 1 thou
outside of that limit.

Much LESS maddening than if you didn't have homing and soft 
limits set up, and you started in cutting material when it 
hit a mechanical limit and wrecked the material or a cutter 
or whatever.

I'M SO GLAD I finally set up home and limits on my system.

Jon


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Saturday 23 February 2019 12:22:54 Jon Elson wrote:

> On 02/23/2019 10:41 AM, Peter C. Wallace wrote:
> > Umm just loaded a software stepgen (lastet master) and it
> > has no such pins/parameters
>
> Right, MIN_LIMIT and MAX_LIMIT are not handled by stepgen or
> any motion interface hardware, they are handled much higher
> in LinuxCNC.  To reach a smooth deceleration right at the
> limit position requires it to be limited before the
> trajectory planner. So, these parameters are picked up out
> of the [JOINT_n] stanza by something in the motion section.
> Even the G-code interpreter knows about them, as you will
> get "program exceeds +X limit at line 1234" messages when
> loading a program.
>
> Jon
>
I've seen that memo hundreds of times. Maddening when its only 1 thou 
outside of that limit.
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Jon Elson]

On 02/23/2019 10:41 AM, Peter C. Wallace wrote:



Umm just loaded a software stepgen (lastet master) and it 
has no such pins/parameters



Right, MIN_LIMIT and MAX_LIMIT are not handled by stepgen or 
any motion interface hardware, they are handled much higher 
in LinuxCNC.  To reach a smooth deceleration right at the 
limit position requires it to be limited before the 
trajectory planner. So, these parameters are picked up out 
of the [JOINT_n] stanza by something in the motion section.  
Even the G-code interpreter knows about them, as you will 
get "program exceeds +X limit at line 1234" messages when 
loading a program.


Jon


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Peter C. Wallace]

On Sat, 23 Feb 2019, Gene Heskett wrote:


Date: Sat, 23 Feb 2019 11:19:42 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Saturday 23 February 2019 10:16:34 Peter C. Wallace wrote:


On Fri, 22 Feb 2019, Gene Heskett wrote:

Date: Fri, 22 Feb 2019 22:35:21 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Friday 22 February 2019 19:20:36 Peter C. Wallace wrote:

On Fri, 22 Feb 2019, Gene Heskett wrote:

Date: Fri, 22 Feb 2019 17:10:16 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:

On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett



wrote:

memtest86 still has me scratching my head. If its so badly
broken, then how in hell is it running lcnc without a hiccup
while I go thru the copies of the original .ini and .hal files,
fixing the syntax errors that changing the hardware its running
on makes? Heck of a good question that. With some luck and a
bigger supply of giddyup I might even have it rigged to drive
some of the machine good enough to carve its own back panel by
the time the new memory gets here late Monday.

Well, maybe it's the memtest installation that's
broken---uninstall and


reinstall? It' should rewrite the boot areas memtest uses, in
case they got corrupted


I gave that a fleeting thought, so fleeting I had  forgotten about
it by the time I'd refilled my coffee cup.

Now, for PCW;
I have it running an admittedly stripped to the bones config.
Nothing is actually hooked up to the machine. But I can home it
w/o motion, and I can jug the dros to the iii file set limits. In
my old hal file I had:
setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
I put that in the new file and its a no-no, pin
hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or
max_limits show except as ini.data visible to a halmeter.  Yet the
dro's stop at the limits set in the ini file under
[AXIS_(xyz)]MIN_LIMIT or MAX_LIMIT.


I dont think there has ever been a stepgen limit
Is it possible you went a bit overboard with find+replace?


These are the axis "soft" limits. You normally set them relative to
the absolute $home of an axis (g53 measurements) to prevent its
running into the mechanical limits, usually I set them about 25 thou
off the crash points for each axis. When software stepping, they are
stepgen.N.min_limit and stepgen.N.max_limit. But there are no such
critters in the 5i25 stepgen's.  It apparently depends on axis to
set and enforce those, and I hadn't noticed they were missing in the
hal files I've hand carved previously.


Are you sure? I cant find these in the stepgen manual page...


See page 306 in the current master copy of Documentation.pdf, where it
lists them, but doesn't say where these limits are actually imposed.
They are also use to draw the machines operating envelope in the
backplot



Umm just loaded a software stepgen (lastet master) and it has no such 
pins/parameters





Without these, or set essentially to infinity, the machines can move at
any commanded speed until they crash into the end of travel.  Hard on
screws etc.



These are set with the ini file machine limit settings, not related to the 
stepgen at all



But as I also said later, this operating envelope, displayed in the
backplot if switched on, is also apparently a function of the gui, in
this case axis. Setting those limits in the ini file seems to be
sufficient to enforce them. Bear in mind that its possible they have
been moved in the last 15 years as the original file I am updating a
copy of, was for a machine that had no home switches and  only 2 limit
switches intended to keep me from unscrewing the little ball screws from
their nuts. That hal file was originally composed for a brain dead
install in the late 90's IIRC.

Unfortunately, the override limit switch logic let me do it anyway, and I
was unable to source another bag of .0635" balls to rebuild the x screw.
So I bought this 6040, and found its designed to work only with mach 3
electronics, was sorely lacking, hence the current linuxcnc build in
progress. But using your interface hardware instead of a single bob on a
parport. Motor drivers are the ones removed from the old mill, and are
moving this machine 10x faster from a parport drive than the old HF mill
ever moved from that same parport without stalls.

So that question is at this point, moot. I've one quite a bit more
important out there now. About spindle parameters for scaling and rev
limits using the spindle interface in the 7i76d. I've set those, but
they seem to have no effect even though the halmeter shows they are set.

Than

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Saturday 23 February 2019 10:16:34 Peter C. Wallace wrote:

> On Fri, 22 Feb 2019, Gene Heskett wrote:
> > Date: Fri, 22 Feb 2019 22:35:21 -0500
> > From: Gene Heskett 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] head scratcher, updated
> >
> > On Friday 22 February 2019 19:20:36 Peter C. Wallace wrote:
> >> On Fri, 22 Feb 2019, Gene Heskett wrote:
> >>> Date: Fri, 22 Feb 2019 17:10:16 -0500
> >>> From: Gene Heskett 
> >>> Reply-To: "Enhanced Machine Controller (EMC)"
> >>> 
> >>> To: emc-users@lists.sourceforge.net
> >>> Subject: Re: [Emc-users] head scratcher, updated
> >>>
> >>> On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:
> >>>> On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett
> >>>> 
> >>>
> >>> wrote:
> >>>>> memtest86 still has me scratching my head. If its so badly
> >>>>> broken, then how in hell is it running lcnc without a hiccup
> >>>>> while I go thru the copies of the original .ini and .hal files,
> >>>>> fixing the syntax errors that changing the hardware its running
> >>>>> on makes? Heck of a good question that. With some luck and a
> >>>>> bigger supply of giddyup I might even have it rigged to drive
> >>>>> some of the machine good enough to carve its own back panel by
> >>>>> the time the new memory gets here late Monday.
> >>>>>
> >>>>> Well, maybe it's the memtest installation that's
> >>>>> broken---uninstall and
> >>>>
> >>>> reinstall? It' should rewrite the boot areas memtest uses, in
> >>>> case they got corrupted
> >>>
> >>> I gave that a fleeting thought, so fleeting I had  forgotten about
> >>> it by the time I'd refilled my coffee cup.
> >>>
> >>> Now, for PCW;
> >>> I have it running an admittedly stripped to the bones config.
> >>> Nothing is actually hooked up to the machine. But I can home it
> >>> w/o motion, and I can jug the dros to the iii file set limits. In
> >>> my old hal file I had:
> >>> setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
> >>> I put that in the new file and its a no-no, pin
> >>> hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or
> >>> max_limits show except as ini.data visible to a halmeter.  Yet the
> >>> dro's stop at the limits set in the ini file under
> >>> [AXIS_(xyz)]MIN_LIMIT or MAX_LIMIT.
> >>
> >> I dont think there has ever been a stepgen limit
> >> Is it possible you went a bit overboard with find+replace?
> >
> > These are the axis "soft" limits. You normally set them relative to
> > the absolute $home of an axis (g53 measurements) to prevent its
> > running into the mechanical limits, usually I set them about 25 thou
> > off the crash points for each axis. When software stepping, they are
> > stepgen.N.min_limit and stepgen.N.max_limit. But there are no such
> > critters in the 5i25 stepgen's.  It apparently depends on axis to
> > set and enforce those, and I hadn't noticed they were missing in the
> > hal files I've hand carved previously.
>
> Are you sure? I cant find these in the stepgen manual page...
>
See page 306 in the current master copy of Documentation.pdf, where it 
lists them, but doesn't say where these limits are actually imposed. 
They are also use to draw the machines operating envelope in the 
backplot

Without these, or set essentially to infinity, the machines can move at 
any commanded speed until they crash into the end of travel.  Hard on 
screws etc. 

But as I also said later, this operating envelope, displayed in the 
backplot if switched on, is also apparently a function of the gui, in 
this case axis. Setting those limits in the ini file seems to be 
sufficient to enforce them. Bear in mind that its possible they have 
been moved in the last 15 years as the original file I am updating a 
copy of, was for a machine that had no home switches and  only 2 limit 
switches intended to keep me from unscrewing the little ball screws from 
their nuts. That hal file was originally composed for a brain dead 
install in the late 90's IIRC.

Unfortunately, the override limit switch logic let me do it anyway, and I 
was unable to source another bag of .0635" balls to rebuild the x screw. 
So I bought this 6040, and found its designed to work only with mach 3 

Re: [Emc-users] head scratcher, updated

[Peter C. Wallace]

On Fri, 22 Feb 2019, Gene Heskett wrote:


Date: Fri, 22 Feb 2019 22:35:21 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Friday 22 February 2019 19:20:36 Peter C. Wallace wrote:


On Fri, 22 Feb 2019, Gene Heskett wrote:

Date: Fri, 22 Feb 2019 17:10:16 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:

On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett 


wrote:

memtest86 still has me scratching my head. If its so badly broken,
then how in hell is it running lcnc without a hiccup while I go
thru the copies of the original .ini and .hal files, fixing the
syntax errors that changing the hardware its running on makes?
Heck of a good question that. With some luck and a bigger supply
of giddyup I might even have it rigged to drive some of the
machine good enough to carve its own back panel by the time the
new memory gets here late Monday.

Well, maybe it's the memtest installation that's
broken---uninstall and


reinstall? It' should rewrite the boot areas memtest uses, in case
they got corrupted


I gave that a fleeting thought, so fleeting I had  forgotten about
it by the time I'd refilled my coffee cup.

Now, for PCW;
I have it running an admittedly stripped to the bones config.
Nothing is actually hooked up to the machine. But I can home it w/o
motion, and I can jug the dros to the iii file set limits.
In my old hal file I had:
setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
I put that in the new file and its a no-no, pin
hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or
max_limits show except as ini.data visible to a halmeter.  Yet the
dro's stop at the limits set in the ini file under
[AXIS_(xyz)]MIN_LIMIT or MAX_LIMIT.


I dont think there has ever been a stepgen limit
Is it possible you went a bit overboard with find+replace?



These are the axis "soft" limits. You normally set them relative to the
absolute $home of an axis (g53 measurements) to prevent its running into
the mechanical limits, usually I set them about 25 thou off the crash
points for each axis. When software stepping, they are
stepgen.N.min_limit and stepgen.N.max_limit. But there are no such
critters in the 5i25 stepgen's.  It apparently depends on axis to set
and enforce those, and I hadn't noticed they were missing in the hal
files I've hand carved previously.


Are you sure? I cant find these in the stepgen manual page...



Peter Wallace
Mesa Electronics


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Friday 22 February 2019 19:20:36 Peter C. Wallace wrote:

> On Fri, 22 Feb 2019, Gene Heskett wrote:
> > Date: Fri, 22 Feb 2019 17:10:16 -0500
> > From: Gene Heskett 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] head scratcher, updated
> >
> > On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:
> >> On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett 
> >
> > wrote:
> >>> memtest86 still has me scratching my head. If its so badly broken,
> >>> then how in hell is it running lcnc without a hiccup while I go
> >>> thru the copies of the original .ini and .hal files, fixing the
> >>> syntax errors that changing the hardware its running on makes?
> >>> Heck of a good question that. With some luck and a bigger supply
> >>> of giddyup I might even have it rigged to drive some of the
> >>> machine good enough to carve its own back panel by the time the
> >>> new memory gets here late Monday.
> >>>
> >>> Well, maybe it's the memtest installation that's
> >>> broken---uninstall and
> >>
> >> reinstall? It' should rewrite the boot areas memtest uses, in case
> >> they got corrupted
> >
> > I gave that a fleeting thought, so fleeting I had  forgotten about
> > it by the time I'd refilled my coffee cup.
> >
> > Now, for PCW;
> > I have it running an admittedly stripped to the bones config.
> > Nothing is actually hooked up to the machine. But I can home it w/o
> > motion, and I can jug the dros to the iii file set limits.
> > In my old hal file I had:
> > setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
> > I put that in the new file and its a no-no, pin
> > hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or
> > max_limits show except as ini.data visible to a halmeter.  Yet the
> > dro's stop at the limits set in the ini file under
> > [AXIS_(xyz)]MIN_LIMIT or MAX_LIMIT.
>
> I dont think there has ever been a stepgen limit
> Is it possible you went a bit overboard with find+replace?
>

These are the axis "soft" limits. You normally set them relative to the 
absolute $home of an axis (g53 measurements) to prevent its running into 
the mechanical limits, usually I set them about 25 thou off the crash 
points for each axis. When software stepping, they are 
stepgen.N.min_limit and stepgen.N.max_limit. But there are no such 
critters in the 5i25 stepgen's.  It apparently depends on axis to set 
and enforce those, and I hadn't noticed they were missing in the hal 
files I've hand carved previously.

So now I think I understand how they work, but it was a surprise when 
setting it, that hal said it was no longer present, so I couldn't setp 
it. Panic time till I figured it out. :)

Tomorrow, since I'm waiting yet on a slow junk from China, I'll cut up 
some cable and see if I can get it to run the vfd with the 7i76 spindle 
controls. I think I'll need 5 conductors so I should be able to do it in 
one 5 pin connector, if the pot common is also the ground reference for 
spinena and spindir. Most of the vfd's I've had, (a sample of one, which 
I'm running with a spinx1) share that. That I haven't verified with a 
meter yet on this vfd...

I did get the imitation Renishaw yesterday, and I certainly can't wax 
poetic about it. Very poorly made IMO.

Thank you Peter.

> Peter Wallace
> Mesa Electronics
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Peter C. Wallace]

On Fri, 22 Feb 2019, Gene Heskett wrote:


Date: Fri, 22 Feb 2019 17:10:16 -0500
From: Gene Heskett 
Reply-To: "Enhanced Machine Controller (EMC)"

To: emc-users@lists.sourceforge.net
Subject: Re: [Emc-users] head scratcher, updated

On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:


On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett 

wrote:

memtest86 still has me scratching my head. If its so badly broken,
then how in hell is it running lcnc without a hiccup while I go thru
the copies of the original .ini and .hal files, fixing the syntax
errors that changing the hardware its running on makes? Heck of a
good question that. With some luck and a bigger supply of giddyup I
might even have it rigged to drive some of the machine good enough
to carve its own back panel by the time the new memory gets here
late Monday.

Well, maybe it's the memtest installation that's broken---uninstall
and


reinstall? It' should rewrite the boot areas memtest uses, in case
they got corrupted


I gave that a fleeting thought, so fleeting I had  forgotten about it by
the time I'd refilled my coffee cup.

Now, for PCW;
I have it running an admittedly stripped to the bones config. Nothing is
actually hooked up to the machine. But I can home it w/o motion, and I
can jug the dros to the iii file set limits.
In my old hal file I had:
setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
I put that in the new file and its a no-no, pin
hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or
max_limits show except as ini.data visible to a halmeter.  Yet the dro's
stop at the limits set in the ini file under [AXIS_(xyz)]MIN_LIMIT or
MAX_LIMIT.


I dont think there has ever been a stepgen limit
Is it possible you went a bit overboard with find+replace?


Peter Wallace
Mesa Electronics


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Friday 22 February 2019 16:06:46 Przemek Klosowski wrote:

> On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett  
wrote:
> > memtest86 still has me scratching my head. If its so badly broken,
> > then how in hell is it running lcnc without a hiccup while I go thru
> > the copies of the original .ini and .hal files, fixing the syntax
> > errors that changing the hardware its running on makes? Heck of a
> > good question that. With some luck and a bigger supply of giddyup I
> > might even have it rigged to drive some of the machine good enough
> > to carve its own back panel by the time the new memory gets here
> > late Monday.
> >
> > Well, maybe it's the memtest installation that's broken---uninstall
> > and
>
> reinstall? It' should rewrite the boot areas memtest uses, in case
> they got corrupted
>
I gave that a fleeting thought, so fleeting I had  forgotten about it by 
the time I'd refilled my coffee cup.

Now, for PCW;
I have it running an admittedly stripped to the bones config. Nothing is 
actually hooked up to the machine. But I can home it w/o motion, and I 
can jug the dros to the iii file set limits.
In my old hal file I had:
setp stepgen.0.min_limit [AXIS_X]MIN_LIMIT
I put that in the new file and its a no-no, pin 
hm2_5i25.0.stepgen.0.0.min_limit does not exist. In fact no min or 
max_limits show except as ini.data visible to a halmeter.  Yet the dro's 
stop at the limits set in the ini file under [AXIS_(xyz)]MIN_LIMIT or 
MAX_LIMIT.

Weird, but since it works, I guess I can continue hooking it up to move 
the machine. But are there any gotchas to watch for? 
The switches for active home setters will be a few says arriving yet. So 
my next move will be to cable up the VFD and see if I can get it 
running.

So far, I have no PID's loaded or in a signal path. Somebody should tell 
me, when there are no encoders for feedback, why I need them.  The old 
now broken HF never had any and seemed to work ok.

And remove/reinstall memtest86 just for S

Progress! For some definition of the word...

> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Przemek Klosowski]

On Fri, Feb 22, 2019 at 1:07 PM Gene Heskett  wrote:

>
> memtest86 still has me scratching my head. If its so badly broken, then
> how in hell is it running lcnc without a hiccup while I go thru the
> copies of the original .ini and .hal files, fixing the syntax errors
> that changing the hardware its running on makes? Heck of a good question
> that. With some luck and a bigger supply of giddyup I might even have it
> rigged to drive some of the machine good enough to carve its own back
> panel by the time the new memory gets here late Monday.
>
> Well, maybe it's the memtest installation that's broken---uninstall and
reinstall? It' should rewrite the boot areas memtest uses, in case they got
corrupted

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher, updated

[Gene Heskett]

On Friday 22 February 2019 09:08:32 Gene Heskett wrote:

> On Friday 22 February 2019 00:58:16 Gene Heskett wrote:
> > On Friday 22 February 2019 00:04:10 Rafael Skodlar wrote:
> > > On 2/21/19 6:27 PM, Gene Heskett wrote:
> > > > On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:
> > > >> On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:
> > > >
> > > > Started another terminal, logged into it now I can copy/paste
> > > > gene@shop:~$ linuxcnc -l
> > > > LINUXCNC - 2.8.0-pre1-4590-g584314c
> > > > Machine configuration directory
> > > > is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
> > > > Machine configuration file is '6040-5i25-7i76.ini'
> > > > Starting LinuxCNC...
> > > > .
> > > > Found file(REL): ./6040-5i25-7i76.hal
> > > > ./6040-5i25-7i76.hal:107: parameter or pin
> > > > 'stepgen.0.position-scale' not found
> > > > Shutting down and cleaning up LinuxCNC...
> > > > line 107 of that hal:
> > > > setp stepgen.0.position-scale [JOINT_0]SCALE
> > > >
> > > > It exists in the ini file.
> > > >
> > > > So something is fubar. memtest86 won't run from grub, claims low
> > > > memory is too small, crashes and reboots to grub in 4 secs or
> > > > so. dmesg looks pretty normal. Except no hint of the 7i76's
> > > > presence can be detected.
> > >
> > > BIOS may have an option to test the memory. It checks certain
> > > addresses to make sure memory is there in the first place.
>
> I'll see if I can find that option in the bios and turn it on. The
> quick boot option I believe skips that, and TBT I had forgotten about
> it.
>
> Thanks.
>
> In the  meantime, since I'm going to transplant the existing bob to p2
> on the 5i25, and build it into the new interface box alongside the
> 7i76, I'll see if I have to make standoffs for that and maybe get it
> moved today. That should give me enough i/o AND the ability to run the
> spindle, if I ever get a response out of the 7i76. sseriel isn't
> starting from the clues I'm getting. dmesg is reporting those 4 pins
> as gpio.
>
The sseriel problem went away when I moved jumper 7i76 jumper w1 back to 
the left.  Now, if I'm getting its logic power from the 12 volt Field 
Power, what am I powering with the 5 volts I've fed the terminals 
immediately left of the field power input?

And editing the .ini and .hal files is slowly getting rid of the can't 
find stuff msgs quoted above.

The bios doesn't appear to have a scan memory function, at least not that 
I can find.

memtest86 still has me scratching my head. If its so badly broken, then 
how in hell is it running lcnc without a hiccup while I go thru the 
copies of the original .ini and .hal files, fixing the syntax errors 
that changing the hardware its running on makes? Heck of a good question 
that. With some luck and a bigger supply of giddyup I might even have it 
rigged to drive some of the machine good enough to carve its own back 
panel by the time the new memory gets here late Monday.

Thanks for reading my mental meanderings everybody. But I suspect the 
reason for lack of replies is that I have this unique ability to find 
problems no one else has seen before...

So if you can't be good, be good at it!

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Gene Heskett]

On Friday 22 February 2019 00:58:16 Gene Heskett wrote:

> On Friday 22 February 2019 00:04:10 Rafael Skodlar wrote:
> > On 2/21/19 6:27 PM, Gene Heskett wrote:
> > > On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:
> > >> On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:

> > > Started another terminal, logged into it now I can copy/paste
> > > gene@shop:~$ linuxcnc -l
> > > LINUXCNC - 2.8.0-pre1-4590-g584314c
> > > Machine configuration directory
> > > is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
> > > Machine configuration file is '6040-5i25-7i76.ini'
> > > Starting LinuxCNC...
> > > .
> > > Found file(REL): ./6040-5i25-7i76.hal
> > > ./6040-5i25-7i76.hal:107: parameter or pin
> > > 'stepgen.0.position-scale' not found
> > > Shutting down and cleaning up LinuxCNC...
> > > line 107 of that hal:
> > > setp stepgen.0.position-scale [JOINT_0]SCALE
> > >
> > > It exists in the ini file.
> > >
> > > So something is fubar. memtest86 won't run from grub, claims low
> > > memory is too small, crashes and reboots to grub in 4 secs or so.
> > > dmesg looks pretty normal. Except no hint of the 7i76's presence
> > > can be detected.
> >
> > BIOS may have an option to test the memory. It checks certain
> > addresses to make sure memory is there in the first place.

I'll see if I can find that option in the bios and turn it on. The quick 
boot option I believe skips that, and TBT I had forgotten about it.

Thanks.

In the  meantime, since I'm going to transplant the existing bob to p2 on 
the 5i25, and build it into the new interface box alongside the 7i76, 
I'll see if I have to make standoffs for that and maybe get it moved 
today. That should give me enough i/o AND the ability to run the 
spindle, if I ever get a response out of the 7i76. sseriel isn't 
starting from the clues I'm getting. dmesg is reporting those 4 pins as 
gpio.



Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Gene Heskett]

On Friday 22 February 2019 00:04:10 Rafael Skodlar wrote:

> On 2/21/19 6:27 PM, Gene Heskett wrote:
> > On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:
> >> On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:
> >>> On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett
> >>> 
> >>
> >> wrote:
>  Thought maybe I'd see if my memory (in the machine, I know /mine/
>  is shot)
>  was getting finicky. And the uptimes on this D525MW board haven't
>  been that great recently, but I reboot and choose memtest86 from
>  the grub menu, and it advises the mem is wrong, and reverts to
>  the grub menu in about 5 seconds.
> 
>  Run synaptic, see there is a slightly newer version available
>  under a slightly different name, but installed, it obviously
>  needs someone 100% familiar with the map available just to run
>  it.
> 
>  The msg at the top of the screen is wrong address, 0x99100,
>  0x8F000
> 
>  And it reverts to the grub menu in about 5 seconds.
> >>>
> >>> Memtest boots and runs as the only program on the system; 'revert
> >>> to grub' is probably memtest crashing and rebooting the system.
> >>
> >> Makes sense.
> >
> > After giving it a good D, even wiped the contacts with contact
> > cleaner,
>
> Do not use contact cleaner for electronics. It's not good in my
> experience and prohibited in Computer History Museum for good reason.
>
> > reset the bios to defaults and rekilled the hypertheading, no change
> > in anything. Still boots 3.4.9 like it should. Give up, wondering if
> > the low memory msgs is telling me the cpu cache is dead?  So I give
> > up there and continue with the new build. write a 5i25-7i76-x2.bit
> > to the 5i25.powerdown reboot,  power up and verify it, good.  The
> > 7i76 with the same dble male cable I've used for yonks, even try 2
> > more off the wall cables, both leds on the 7i76 are lit. w1, w2
> > right hand position because I've a 5 volt hooked up for the logic
> > stuffs, and an isolated 15 volt supply on the field power hookups.
> > But the leds are lit normally. But nothing in dmesg indicates
> > sserial is running, and the exit message when I try to run linuxcnc
> > -l makes no sense.
> >
> >
> >   Its as if the hal file has no linkage back to the ini file to get
> > setp stuff of the [JOINT_0]SCALE values for a setp
> > stepgen.0.position-scale.
> >
> > And my logins are crippled, can't highlight slide a copy/paste,
> > mouse clicks comeout as data echo's on the command line.
> >
> > Started another terminal, logged into it now I can copy/paste
> > gene@shop:~$ linuxcnc -l
> > LINUXCNC - 2.8.0-pre1-4590-g584314c
> > Machine configuration directory
> > is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
> > Machine configuration file is '6040-5i25-7i76.ini'
> > Starting LinuxCNC...
> > .
> > Found file(REL): ./6040-5i25-7i76.hal
> > ./6040-5i25-7i76.hal:107: parameter or pin
> > 'stepgen.0.position-scale' not found
> > Shutting down and cleaning up LinuxCNC...
> > line 107 of that hal:
> > setp stepgen.0.position-scale [JOINT_0]SCALE
> >
> > It exists in the ini file.
> >
> > So something is fubar. memtest86 won't run from grub, claims low
> > memory is too small, crashes and reboots to grub in 4 secs or so.
> > dmesg looks pretty normal. Except no hint of the 7i76's presence can
> > be detected.
>
> BIOS may have an option to test the memory. It checks certain
> addresses to make sure memory is there in the first place.
>
> > Educated guesses?
> >
> >>> I hmave never seen it behaving this way: even with bad memory I
> >>> have only seen it report errors while still running.
> >>> This may be actually good news: it's possible that your memory is
> >>> so bad that even the tiny region from which memtest runs is
> >>> hopelessly corrupt. Did you try re-seating the DIMMs?
> >
> > Yes, but just 1, 2gig dimm.  Several times. I haven't retrieved the
> > other box for test parts yet.
> >
> >> Thats this evenings project after i get the missus fed.
> >
> > Turned into the next days project. Legs/feet played out last night.
> > And I haven't really learned much. I jacked the cards up a half inch
> > so the power supplies would fit under the cards with 1/4" to spare &
> > hooked up the power supplies. Not much use doing anything else till
> > the cards talk to me.  And the 7i76 is muted.
> >
> > Just prowling the net. looking for someplace to src a couple sticks
> > of memory for one of these intel D525MW atom motherboards, and all
> > the usual businesses have dried up, not willing to talk to a
> > customer unless you have a business account with them.
> >
> > What I want is a couple 2Gb sticks of dimm, only 800 mhz so both
> > slow and old, lappy style, to fit the above motherboard. Where are
> > you folks getting your stuff these days?
> >
> > Thanks all.
> >
> > Cheers, Gene Heskett
>
> You still don't know for sure if the memory is bad. Have you swapped
> DIMMs among themselves to see if the problem moves to 

Re: [Emc-users] head scratcher

[Gene Heskett]

On Thursday 21 February 2019 21:27:56 Gene Heskett wrote:

> On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:
> > On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:
> > > On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett
> > > 
> >
> > wrote:
> > > > Thought maybe I'd see if my memory (in the machine, I know
> > > > /mine/ is shot)
> > > > was getting finicky. And the uptimes on this D525MW board
> > > > haven't been that great recently, but I reboot and choose
> > > > memtest86 from the grub menu, and it advises the mem is wrong,
> > > > and reverts to the grub menu in about 5 seconds.
> > > >
> > > > Run synaptic, see there is a slightly newer version available
> > > > under a slightly different name, but installed, it obviously
> > > > needs someone 100% familiar with the map available just to run
> > > > it.
> > > >
> > > > The msg at the top of the screen is wrong address, 0x99100,
> > > > 0x8F000
> > > >
> > > > And it reverts to the grub menu in about 5 seconds.
> > >
> > > Memtest boots and runs as the only program on the system; 'revert
> > > to grub' is probably memtest crashing and rebooting the system.
> >
> > Makes sense.
>
> After giving it a good D, even wiped the contacts with contact
> cleaner, reset the bios to defaults and rekilled the hypertheading, no
> change in anything. Still boots 3.4.9 like it should. Give up,
> wondering if the low memory msgs is telling me the cpu cache is dead? 
> So I give up there and continue with the new build. write a
> 5i25-7i76-x2.bit to the 5i25.powerdown reboot,  power up and verify
> it, good.  The 7i76 with the same dble male cable I've used for yonks,
> even try 2 more off the wall cables, both leds on the 7i76 are lit.
> w1, w2 right hand position because I've a 5 volt hooked up for the
> logic stuffs, and an isolated 15 volt supply on the field power
> hookups. But the leds are lit normally. But nothing in dmesg indicates
> sserial is running, and the exit message when I try to run linuxcnc -l
> makes no sense.
>
>
>  Its as if the hal file has no linkage back to the ini file to get
> setp stuff of the [JOINT_0]SCALE values for a setp
> stepgen.0.position-scale.
>
> And my logins are crippled, can't highlight slide a copy/paste, mouse
> clicks comeout as data echo's on the command line.
>
> Started another terminal, logged into it now I can copy/paste
> gene@shop:~$ linuxcnc -l
> LINUXCNC - 2.8.0-pre1-4590-g584314c
> Machine configuration directory
> is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
> Machine configuration file is '6040-5i25-7i76.ini'
> Starting LinuxCNC...
> .
> Found file(REL): ./6040-5i25-7i76.hal
> ./6040-5i25-7i76.hal:107: parameter or pin 'stepgen.0.position-scale'
> not found
> Shutting down and cleaning up LinuxCNC...
> line 107 of that hal:
> setp stepgen.0.position-scale [JOINT_0]SCALE
>
> It exists in the ini file.
>
> So something is fubar. memtest86 won't run from grub, claims low
> memory is too small, crashes and reboots to grub in 4 secs or so.
> dmesg looks pretty normal. Except no hint of the 7i76's presence can
> be detected.
>
> Educated guesses?
>
> > > I hmave never seen it behaving this way: even with bad memory I
> > > have only seen it report errors while still running.
> > > This may be actually good news: it's possible that your memory is
> > > so bad that even the tiny region from which memtest runs is
> > > hopelessly corrupt. Did you try re-seating the DIMMs?
>
> Yes, but just 1, 2gig dimm.  Several times. I haven't retrieved the
> other box for test parts yet.
>
> > Thats this evenings project after i get the missus fed.
>
> Turned into the next days project. Legs/feet played out last night.
> And I haven't really learned much. I jacked the cards up a half inch
> so the power supplies would fit under the cards with 1/4" to spare &
> hooked up the power supplies. Not much use doing anything else till
> the cards talk to me.  And the 7i76 is muted.
>
> Just prowling the net. looking for someplace to src a couple sticks of
> memory for one of these intel D525MW atom motherboards, and all the
> usual businesses have dried up, not willing to talk to a customer
> unless you have a business account with them.
>
> What I want is a couple 2Gb sticks of dimm, only 800 mhz so both slow
> and old, lappy style, to fit the above motherboard. Where are you
> folks getting your stuff these days?
>
> Thanks all.
>
> Cheers, Gene Heskett

I can't believe it, a simple search on ebay turned up a bunch of them at 
$13.51/copy, 1066 mhz instead of 800 mhz but should work. I bought 4 for 
insurance.  Be here Monday. I hope they are good, claimed to be brand 
new.

Not only that, there seems to be a plentiful supply of D525MW 
motherboards, lots of used ones, and quite a few that claim to be new.

So all is not lost on one of the best lcnc motherboards ever.
Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed 

Re: [Emc-users] head scratcher

[Rafael Skodlar]

On 2/21/19 6:27 PM, Gene Heskett wrote:

On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:


On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:

On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett 


wrote:

Thought maybe I'd see if my memory (in the machine, I know /mine/
is shot)
was getting finicky. And the uptimes on this D525MW board haven't
been that great recently, but I reboot and choose memtest86 from
the grub menu, and it advises the mem is wrong, and reverts to the
grub menu in about 5 seconds.

Run synaptic, see there is a slightly newer version available
under a slightly different name, but installed, it obviously needs
someone 100% familiar with the map available just to run it.

The msg at the top of the screen is wrong address, 0x99100,
0x8F000

And it reverts to the grub menu in about 5 seconds.


Memtest boots and runs as the only program on the system; 'revert to
grub' is probably memtest crashing and rebooting the system.


Makes sense.


After giving it a good D, even wiped the contacts with contact cleaner,


Do not use contact cleaner for electronics. It's not good in my 
experience and prohibited in Computer History Museum for good reason.



reset the bios to defaults and rekilled the hypertheading, no change in
anything. Still boots 3.4.9 like it should. Give up, wondering if the
low memory msgs is telling me the cpu cache is dead?  So I give up there
and continue with the new build. write a 5i25-7i76-x2.bit to the
5i25.powerdown reboot,  power up and verify it, good.  The 7i76 with the
same dble male cable I've used for yonks, even try 2 more off the wall
cables, both leds on the 7i76 are lit. w1, w2 right hand position
because I've a 5 volt hooked up for the logic stuffs, and an isolated 15
volt supply on the field power hookups. But the leds are lit normally.
But nothing in dmesg indicates sserial is running, and the exit message
when I try to run linuxcnc -l makes no sense.


  Its as if the hal file has no linkage back to the ini file to get setp
stuff of the [JOINT_0]SCALE values for a setp stepgen.0.position-scale.

And my logins are crippled, can't highlight slide a copy/paste, mouse
clicks comeout as data echo's on the command line.

Started another terminal, logged into it now I can copy/paste
gene@shop:~$ linuxcnc -l
LINUXCNC - 2.8.0-pre1-4590-g584314c
Machine configuration directory
is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
Machine configuration file is '6040-5i25-7i76.ini'
Starting LinuxCNC...
.
Found file(REL): ./6040-5i25-7i76.hal
./6040-5i25-7i76.hal:107: parameter or pin 'stepgen.0.position-scale' not
found
Shutting down and cleaning up LinuxCNC...
line 107 of that hal:
setp stepgen.0.position-scale [JOINT_0]SCALE

It exists in the ini file.

So something is fubar. memtest86 won't run from grub, claims low memory
is too small, crashes and reboots to grub in 4 secs or so. dmesg looks
pretty normal. Except no hint of the 7i76's presence can be detected.


BIOS may have an option to test the memory. It checks certain addresses 
to make sure memory is there in the first place.




Educated guesses?


I hmave never seen it behaving this way: even with bad memory I have
only seen it report errors while still running.
This may be actually good news: it's possible that your memory is so
bad that even the tiny region from which memtest runs is hopelessly
corrupt. Did you try re-seating the DIMMs?



Yes, but just 1, 2gig dimm.  Several times. I haven't retrieved the other
box for test parts yet.


Thats this evenings project after i get the missus fed.


Turned into the next days project. Legs/feet played out last night. And I
haven't really learned much. I jacked the cards up a half inch so the
power supplies would fit under the cards with 1/4" to spare & hooked up
the power supplies. Not much use doing anything else till the cards talk
to me.  And the 7i76 is muted.

Just prowling the net. looking for someplace to src a couple sticks of
memory for one of these intel D525MW atom motherboards, and all the
usual businesses have dried up, not willing to talk to a customer unless
you have a business account with them.

What I want is a couple 2Gb sticks of dimm, only 800 mhz so both slow and
old, lappy style, to fit the above motherboard. Where are you folks
getting your stuff these days?

Thanks all.

Cheers, Gene Heskett

You still don't know for sure if the memory is bad. Have you swapped 
DIMMs among themselves to see if the problem moves to different address?


Our local store in San Jose, CA seem to have what you are looking for:

https://www.frys.com/search?search_type=regular=1=_string=DIMM+memory=false

Surplus stores are disappearing in Silicon Valley so it's not easy to 
find old components these days. Best other place for electronics is 
https://anchor-electronics.com they still carry S-100 bus cards and 
components to build them, so there!


In any case, Gene try to start with the minimal system to test basics: 
motherboard, PSU, keyboard, 

Re: [Emc-users] head scratcher

[Gene Heskett]

On Wednesday 20 February 2019 17:00:19 Gene Heskett wrote:

> On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:
> > On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett 
>
> wrote:
> > > Thought maybe I'd see if my memory (in the machine, I know /mine/
> > > is shot)
> > > was getting finicky. And the uptimes on this D525MW board haven't
> > > been that great recently, but I reboot and choose memtest86 from
> > > the grub menu, and it advises the mem is wrong, and reverts to the
> > > grub menu in about 5 seconds.
> > >
> > > Run synaptic, see there is a slightly newer version available
> > > under a slightly different name, but installed, it obviously needs
> > > someone 100% familiar with the map available just to run it.
> > >
> > > The msg at the top of the screen is wrong address, 0x99100,
> > > 0x8F000
> > >
> > > And it reverts to the grub menu in about 5 seconds.
> >
> > Memtest boots and runs as the only program on the system; 'revert to
> > grub' is probably memtest crashing and rebooting the system.
>
> Makes sense.
>
After giving it a good D, even wiped the contacts with contact cleaner, 
reset the bios to defaults and rekilled the hypertheading, no change in 
anything. Still boots 3.4.9 like it should. Give up, wondering if the 
low memory msgs is telling me the cpu cache is dead?  So I give up there 
and continue with the new build. write a 5i25-7i76-x2.bit to the 
5i25.powerdown reboot,  power up and verify it, good.  The 7i76 with the 
same dble male cable I've used for yonks, even try 2 more off the wall 
cables, both leds on the 7i76 are lit. w1, w2 right hand position 
because I've a 5 volt hooked up for the logic stuffs, and an isolated 15 
volt supply on the field power hookups. But the leds are lit normally. 
But nothing in dmesg indicates sserial is running, and the exit message 
when I try to run linuxcnc -l makes no sense.


 Its as if the hal file has no linkage back to the ini file to get setp 
stuff of the [JOINT_0]SCALE values for a setp stepgen.0.position-scale.

And my logins are crippled, can't highlight slide a copy/paste, mouse 
clicks comeout as data echo's on the command line.

Started another terminal, logged into it now I can copy/paste
gene@shop:~$ linuxcnc -l
LINUXCNC - 2.8.0-pre1-4590-g584314c
Machine configuration directory 
is '/home/gene/linuxcnc/configs/6040-5i25-7i76'
Machine configuration file is '6040-5i25-7i76.ini'
Starting LinuxCNC...
.
Found file(REL): ./6040-5i25-7i76.hal
./6040-5i25-7i76.hal:107: parameter or pin 'stepgen.0.position-scale' not 
found
Shutting down and cleaning up LinuxCNC...
line 107 of that hal:
setp stepgen.0.position-scale [JOINT_0]SCALE

It exists in the ini file.

So something is fubar. memtest86 won't run from grub, claims low memory 
is too small, crashes and reboots to grub in 4 secs or so. dmesg looks 
pretty normal. Except no hint of the 7i76's presence can be detected.

Educated guesses?

> > I hmave never seen it behaving this way: even with bad memory I have
> > only seen it report errors while still running.
> > This may be actually good news: it's possible that your memory is so
> > bad that even the tiny region from which memtest runs is hopelessly
> > corrupt. Did you try re-seating the DIMMs?
>
Yes, but just 1, 2gig dimm.  Several times. I haven't retrieved the other 
box for test parts yet.

> Thats this evenings project after i get the missus fed.

Turned into the next days project. Legs/feet played out last night. And I 
haven't really learned much. I jacked the cards up a half inch so the 
power supplies would fit under the cards with 1/4" to spare & hooked up 
the power supplies. Not much use doing anything else till the cards talk 
to me.  And the 7i76 is muted.

Just prowling the net. looking for someplace to src a couple sticks of 
memory for one of these intel D525MW atom motherboards, and all the 
usual businesses have dried up, not willing to talk to a customer unless 
you have a business account with them.

What I want is a couple 2Gb sticks of dimm, only 800 mhz so both slow and 
old, lappy style, to fit the above motherboard. Where are you folks 
getting your stuff these days?

Thanks all.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Rafael Skodlar]

Hi Gene,

On 2/20/19 10:50 AM, Gene Heskett wrote:

Thought maybe I'd see if my memory (in the machine, I know /mine/ is
shot)
was getting finicky. And the uptimes on this D525MW board haven't been
that great recently, but I reboot and choose memtest86 from the grub
menu, and it advises the mem is wrong, and reverts to the grub menu in
about 5 seconds.



One way to troubleshoot is to swap DIMs or whatever type of memory you 
have and you should see if the memory error moves. Older motherboards 
might have electrolytic cap problems. One time I bought a bag of them 
and replace all on one motherboard.


Series of caps made in China was as fake as CNN news. They stole part of 
manufacturing process from Japan.



Run synaptic, see there is a slightly newer version available under a
slightly different name, but installed, it obviously needs someone 100%
familiar with the map available just to run it.

The msg at the top of the screen is wrong address, 0x99100, 0x8F000

And it reverts to the grub menu in about 5 seconds.


I would remove HDD and test booting from USB memory stick. Eliminate or 
narrow down components needed to see anthing on the screen to find the 
problem.




Anybody have a clue whats going on?
  
Thanks all;


Cheers, Gene Heskett



Good luck,

--
Rafael Skodlar


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Gene Heskett]

On Wednesday 20 February 2019 16:37:06 Przemek Klosowski wrote:

> On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett  
wrote:
> > Thought maybe I'd see if my memory (in the machine, I know /mine/ is
> > shot)
> > was getting finicky. And the uptimes on this D525MW board haven't
> > been that great recently, but I reboot and choose memtest86 from the
> > grub menu, and it advises the mem is wrong, and reverts to the grub
> > menu in about 5 seconds.
> >
> > Run synaptic, see there is a slightly newer version available under
> > a slightly different name, but installed, it obviously needs someone
> > 100% familiar with the map available just to run it.
> >
> > The msg at the top of the screen is wrong address, 0x99100, 0x8F000
> >
> > And it reverts to the grub menu in about 5 seconds.
>
> Memtest boots and runs as the only program on the system; 'revert to
> grub' is probably memtest crashing and rebooting the system.

Makes sense.

> I have never seen it behaving this way: even with bad memory I have
> only seen it report errors while still running.
> This may be actually good news: it's possible that your memory is so
> bad that even the tiny region from which memtest runs is hopelessly
> corrupt. Did you try re-seating the DIMMs?

Thats this evenings project after i get the missus fed.

> If that doesn't help, get 
> new ones and swap them out. It's possible that the memory controller
> went bad, too.. What was the history of the system---it worked fine
> and started miisbehaving recently without anything else contributing
> (like brownouts/lightning/fat-fingered nephews/BIOS battery dying and
> scrambling the memory controller settings)
>
Brownouts and power failures are endemic in bad weather here, so bad I 
had a 20kw NAT gas fired installed 4 years back. And although it starts 
and picks up the load in 10 seconds or so, the damage to a computer may 
well be done before an incandescent lamp is dark.

So I'll take the 2gig dimm out and reseat it a couple times, and reset 
the bios to defaults and turn the hyperthreading back off, then see if 
memtest86 will run.  If not, I'll check the other supposedly identical 
box as a src of supposedly good memory. But its out in the shop building 
at about 40F. So that won't get done this evening. This is the short, 
fat sticks that lay well down like used in lappy's. 2G's worth in both 
boxes.

Thanks Przemek Klosowski.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Przemek Klosowski]

On Wed, Feb 20, 2019 at 1:53 PM Gene Heskett  wrote:

> Thought maybe I'd see if my memory (in the machine, I know /mine/ is
> shot)
> was getting finicky. And the uptimes on this D525MW board haven't been
> that great recently, but I reboot and choose memtest86 from the grub
> menu, and it advises the mem is wrong, and reverts to the grub menu in
> about 5 seconds.
>
> Run synaptic, see there is a slightly newer version available under a
> slightly different name, but installed, it obviously needs someone 100%
> familiar with the map available just to run it.
>
> The msg at the top of the screen is wrong address, 0x99100, 0x8F000
>
> And it reverts to the grub menu in about 5 seconds.
>

Memtest boots and runs as the only program on the system; 'revert to grub'
is probably memtest crashing and rebooting the system.
I have never seen it behaving this way: even with bad memory I have only
seen it report errors while still running.
This may be actually good news: it's possible that your memory is so bad
that even the tiny region from which memtest runs is hopelessly corrupt.
Did you try re-seating the DIMMs? If that doesn't help, get new ones and
swap them out. It's possible that the memory controller went bad, too..
What was the history of the system---it worked fine and started
miisbehaving recently without anything else contributing (like
brownouts/lightning/fat-fingered nephews/BIOS battery dying and scrambling
the memory controller settings)

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] head scratcher

[Gene Heskett]

Thought maybe I'd see if my memory (in the machine, I know /mine/ is 
shot)
was getting finicky. And the uptimes on this D525MW board haven't been 
that great recently, but I reboot and choose memtest86 from the grub 
menu, and it advises the mem is wrong, and reverts to the grub menu in 
about 5 seconds.

Run synaptic, see there is a slightly newer version available under a 
slightly different name, but installed, it obviously needs someone 100% 
familiar with the map available just to run it.

The msg at the top of the screen is wrong address, 0x99100, 0x8F000

And it reverts to the grub menu in about 5 seconds.

Anybody have a clue whats going on?
 
Thanks all;

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[Gene Heskett]

On Sunday 08 April 2018 17:44:16 andy pugh wrote:

> On 8 April 2018 at 22:40, Gene Heskett  wrote:
> > Putting the final touches on a routine to tap a 4mm.70 hole, and
> > geany says it is saving the changes, by removing the red text for
> > normal text in its header bar, but I'm leaving the file open in
> > geany, and going back to lcnc, which also goes thru the motions of
> > reloading the file.
> >
> > But neither less from a shell from here ssh logged into that
> > machine, nor LCNC on that machine loads the new code.
>
> Are you sure that you are saving to the right place?
>
I opened the file via that name, and the saves (ctrl+s) had been working 
all day. This machine has a 60 GB SSD for a drive. And from here, there 
is only one "tap4mmhole.ngc" in the nc_files directory. I did an 
updatedb, then a "locate tap4mmhole.ngc", and theres a copy in the 
linuxcnc dir too. But its nowhere near the same file, and has a creation 
date 2 years back. Next I guess is to close geany w/o any saves, and 
rerun it. Except I, nor root can kill it. Go back out there and try. And 
out there, geany is gone from the workspace it was on, but still shows 
in a root session of htop, which cannot kill it. Shut down lcnc so it 
was a gracefull shut down, and reboot it I guess. I'll go back out and 
see if I can edit that file in a bit, helping Dee at the moment.

Weird stuff for sure.

Thanks Andy.


> I only ask because I managed to save an old config over my working
> config today when Geany lost the network drive and went back to a
> local drive containing a file of the same name.



-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] head scratcher

[andy pugh]

On 8 April 2018 at 22:40, Gene Heskett  wrote:

> Putting the final touches on a routine to tap a 4mm.70 hole, and geany
> says it is saving the changes, by removing the red text for normal text
> in its header bar, but I'm leaving the file open in geany, and going
> back to lcnc, which also goes thru the motions of reloading the file.
>
> But neither less from a shell from here ssh logged into that machine, nor
> LCNC on that machine loads the new code.

Are you sure that you are saving to the right place?

I only ask because I managed to save an old config over my working
config today when Geany lost the network drive and went back to a
local drive containing a file of the same name.


-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] head scratcher

[Gene Heskett]

Greetings all;

Putting the final touches on a routine to tap a 4mm.70 hole, and geany 
says it is saving the changes, by removing the red text for normal text 
in its header bar, but I'm leaving the file open in geany, and going 
back to lcnc, which also goes thru the motions of reloading the file.

But neither less from a shell from here ssh logged into that machine, nor 
LCNC on that machine loads the new code. Am I violating a file locking 
rule?, or is it just plain time to reboot?  Or a file system bug?

It has been working as expected most of the day. So I'm scratching whats 
left of my hair. I hate to just arbitrarily reboot, I have at times lost 
track of the homes doing that.

Probably unrelated, but one of the location points, x=-7.2 will show 
as -7.197 or 8 in the dro display when done from the mdi. No motion, and 
no errors are seen when that occurs.

Ideas?

-- 
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users