Re: [Emc-users] Lathe saddle protection

[Les Newell]

I have a Denford Orac with a switch on the saddle and an adjustable stop 
that clamps on the bed. I also use another machine that has a fixed 
switch on the headstock and is an adjustable rod on the saddle to set 
the trip point. These has saved the machines from damage a couple of 
times and I have been meaning to do something similar with my Colchester 
Triumph.

This sort of setup only works well if all of your tools are about the 
same length. For instance I have access to another machine with a trip 
switch and an ATC. On many jobs you end up with the trip switch being 
unable to protect all of the tools. For instance you can't protect 
boring tools.

Now what would be really useful is to be able to set soft limits in work 
coordinates rather than machine coordinates. That way you can protect 
all tools, no matter what size they are. You would need to be able to 
set the limit position from code to allow for boring operations etc.

Les

On 20/11/15 11:00, andy pugh wrote:
> Has anyone ever tried embedding proximity sensors in their lathe
> saddle to protect against chuck-strikes or backing up against the
> tailstock?
>
> If so, any "lessons learned" about where to put the sensors?
>


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

Re: [Emc-users] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 06:29:18 Les Newell wrote:

> I have a Denford Orac with a switch on the saddle and an adjustable
> stop that clamps on the bed. I also use another machine that has a
> fixed switch on the headstock and is an adjustable rod on the saddle
> to set the trip point. These has saved the machines from damage a
> couple of times and I have been meaning to do something similar with
> my Colchester Triumph.
>
> This sort of setup only works well if all of your tools are about the
> same length. For instance I have access to another machine with a trip
> switch and an ATC. On many jobs you end up with the trip switch being
> unable to protect all of the tools. For instance you can't protect
> boring tools.

Or taps, same reason.

> Now what would be really useful is to be able to set soft limits in
> work coordinates rather than machine coordinates. That way you can
> protect all tools, no matter what size they are. You would need to be
> able to set the limit position from code to allow for boring
> operations etc.
>
> Les

I've thought about that Les, but haven't voiced it here on the list.  I'd 
sure put a plus vote in for that ability.  Another option in the touch 
off menu would seem to be the ideal place for it.  Run it to where it 
needs to stop and set the appropriate left/right/in/out limit point.  
That would have saved me several hundred $ in crushed carbide chips over 
the last decade+.

> On 20/11/15 11:00, andy pugh wrote:
> > Has anyone ever tried embedding proximity sensors in their lathe
> > saddle to protect against chuck-strikes or backing up against the
> > tailstock?
> >
> > If so, any "lessons learned" about where to put the sensors?
>
> --
> ___
> 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] Lathe saddle protection

[andy pugh]

On 20 November 2015 at 12:58, Les Newell  wrote:

> A mechanical switch is probably best in this application. Swarf is
> unlikely to provide the solid push needed to operate a switch

The danger there seems to be that the chuck jaws might simply shorten
the plunger by a small amount each time, without tripping it.

Possibly the simplest solution is a wire glued to the parts that need
to be protected, if the wire is broken, activate feed-hold.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

Re: [Emc-users] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 06:00:00 andy pugh wrote:

> Has anyone ever tried embedding proximity sensors in their lathe
> saddle to protect against chuck-strikes or backing up against the
> tailstock?
>
> If so, any "lessons learned" about where to put the sensors?

Chuck strikes would seem to be so "tool mounted" a variable as to be 
unsolvable. But I have considered hanging a straight simple button micro 
microswitch on the saddle such that if its button is compressed, it 
would stop the saddle as it touched the tailstock.  I have a similar 
setup on the front of the carriage that senses the cross-slide, stopping 
it about 10 thou from a crash stop at the front end of the slot in the 
carriage, that has been used as a homing switch for a couple years.  My 
switch positions are such that I can, unless I've a really huge work 
piece in the chuck, remove the QC holder, and home the machine without 
crashing into something.

In the FWIW category, the tailstock clamping on a 7x12 is so poor that my 
1605 z screw can actually push it when you think its locked.

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] Lathe saddle protection

[Dave Caroline]

I have the mechanical version on one of the lathes too, it is job
specific I dont see how a proximity switch could know all possible
errors of eg hitting jaws, chuck body or work or even the headstock.
The little denford cnc lathe has the electrical stop movable on the
bed.

Dave

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

[Emc-users] Lathe saddle protection

[andy pugh]

Has anyone ever tried embedding proximity sensors in their lathe
saddle to protect against chuck-strikes or backing up against the
tailstock?

If so, any "lessons learned" about where to put the sensors?

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

Re: [Emc-users] Lathe saddle protection

[andy pugh]

On 20 November 2015 at 11:13, John Thornton  wrote:
> My Samson lathe has an adjustable kick out for the saddle feed. I can
> move it along the Z axis to where I need the feed to kick out. It is
> under the saddle, I don't have a photo of it atm. You could do the same
> with a prox or switch.

That assumes that I remember to check. I was looking for ways to save
myself from my own carelessness.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

Re: [Emc-users] Lathe saddle protection

[andy pugh]

On 20 November 2015 at 11:50, Gene Heskett  wrote:
> Chuck strikes would seem to be so "tool mounted" a variable as to be
> unsolvable.

I don't see why. The same area of the saddle is likely to be at risk
of hitting the chuck jaws in must setups.
I am not trying to protect lathe tools or the chuck jaws.

One concern with any such idea would be that swarf might trip the
sensor. Some super-clever frequency analysis might be needed (3x or 4x
spindle speed) to tell the difference.



-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

Re: [Emc-users] Lathe saddle protection

[Les Newell]

On 20/11/15 12:30, andy pugh wrote:
> One concern with any such idea would be that swarf might trip the 
> sensor. Some super-clever frequency analysis might be needed (3x or 4x 
> spindle speed) to tell the difference. 

A mechanical switch is probably best in this application. Swarf is 
unlikely to provide the solid push needed to operate a switch. It is 
possible that swarf can get caught between the switch and actuator 
making it trip early. Even then it is only a problem if you are running 
very close to the trip point in normal operation.

Les


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

Re: [Emc-users] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 07:58:20 Les Newell wrote:

> On 20/11/15 12:30, andy pugh wrote:
> > One concern with any such idea would be that swarf might trip the
> > sensor. Some super-clever frequency analysis might be needed (3x or
> > 4x spindle speed) to tell the difference.
>
> A mechanical switch is probably best in this application. Swarf is
> unlikely to provide the solid push needed to operate a switch. It is
> possible that swarf can get caught between the switch and actuator
> making it trip early. Even then it is only a problem if you are
> running very close to the trip point in normal operation.
>
> Les

I wonder how a hall effect might work, to sense the chucks jaws going by 
when it gets too close?  I think I'd boost the length of its output 
pulse with a one shot to make sure its triggering wasn't missed though 
when the servo thread is the only one, no base thread to read it every 
30 microseconds.

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] Lathe saddle protection

[andy pugh]

On 20 November 2015 at 15:30, Gene Heskett  wrote:
> I have probably worn off an eighth of an
> inch off the nose of the compound feed since it usually hangs out in
> front of the saddle.  Just one of the reasons I have considered removing
> it and replacing it with a block of steel

Right and below-centre in this photo
https://picasaweb.google.com/lh/photo/gQuPtDB0J_ZiqVVDHaf8vtMTjNZETYmyPJy0liipFm0?feat=directlink
Is the pattern that I am taking to the foundry tomorrow to make just
such a block for my Holbrook.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

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

Re: [Emc-users] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 07:30:49 andy pugh wrote:

> On 20 November 2015 at 11:50, Gene Heskett  wrote:
> > Chuck strikes would seem to be so "tool mounted" a variable as to be
> > unsolvable.
>
> I don't see why. The same area of the saddle is likely to be at risk
> of hitting the chuck jaws in must setups.
> I am not trying to protect lathe tools or the chuck jaws.
>
> One concern with any such idea would be that swarf might trip the
> sensor. Some super-clever frequency analysis might be needed (3x or 4x
> spindle speed) to tell the difference.

Vibration detection? On mine, I have probably worn off an eighth of an 
inch off the nose of the compound feed since it usually hangs out in 
front of the saddle.  Just one of the reasons I have considered removing 
it and replacing it with a block of steel that holds the QC base about 
1/2" out and to the right just to put the cutting forces back on top of 
the saddle as opposed to hanging off the left of it enough to lift the 
right edge against the gib strip. When the machine gets that small, the 
saddle footprint on the bed gets too darned narrow, and cutting forces 
can make it do things the screw didn't tell it to.  That narrow a saddle 
would only be at home wearing a lantern post tool holder, which would 
put the cutting forces closer to the center of its footprint on the bed. 
I tried to make one, but the compound is too high, no room for the big 
woodruff key and dished washer base. Getting rid of the compound and 
replacing it with a thinner block would solve that problem too.

But ATM I'm trying to finish those 3 blanket chests, and seem to have 
miss-laid my round tuit to take a break from that and do that sort of 
thing.  I spent yesterday trying to get enough filler into and sanded 
back flat, a nearly 23.5" x 41" lid panel.  I left in laying on the saw 
table last night, sanded to 320 grit.  No clue how much it has warped 
overnight just laying there.  I know it will, so the next step is 
getting the breadboard ends fitted, which will control that quite a bit, 
hopefully w/o skinning it up so my moving it around doesn't show in the 
finish.

And this morning I am having trouble walking again, that pinched nerve in 
my back seems to be making me think I have a cramp in the calf of my 
left leg. I wrapped a heating pad around it about 5ish & managed to get 
a bit more sleep though. Loaded up with vitamin B12, which is reported 
to help counteract the metformin's tendency to flush it out, the 
shortage of which causes leg cramps. Sigh.

I got 2 more saddle blocks in my back 2 weeks back, on the left side this 
time, but they aren't working.  Getting old is not for wimps, but I seem 
to be turning into one way too fast.  Heck, I still have stuff on my 
bucket list!

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] Lathe saddle protection

[Les Newell]

> I wonder how a hall effect might work, to sense the chucks jaws going 
> by when it gets too close? I think I'd boost the length of its output 
> pulse with a one shot to make sure its triggering wasn't missed though 
> when the servo thread is the only one, no base thread to read it every 
> 30 microseconds. Cheers, Gene Heskett 

Oh, I see, you want to sense the chuck jaws themselves. That leaves the 
sensor in a very vulnerable place, right in the firing line for coolant, 
swarf etc. What is wrong with having a switch on the saddle that trips 
when it hits an adjustable stop? It works quite well and is a lot less 
vulnerable.

Les

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

Re: [Emc-users] Configuring New PCI Parport

[jeshua]

> On Nov 19, 2015, at 10:00 PM, Jon Elson  wrote:
> 
> On 11/19/2015 10:20 PM, Jeshua Lacock wrote:
>> On Nov 19, 2015, at 2:05 PM, Todd Zuercher  wrote:
>> 
>>> Are you sure that is actually the right listing for your pci card?
>> 
>> Hi Todd,
>> 
>> Yes, I am sure, if the card isn’t in the machine the listing isn’t there.
>> 
>>> The listing for my card (albeit a different card) is.
>>> 
>>> Try the command
>>> $ cat /proc/ioports | grep parport
>>> 
>>> 
>>> My system show this for the above command.
>>>  0378-037a : parport0
>>>e030-e032 : parport2
>>>e050-e052 : parport1
>> 
>> I have some ports shown for the address expected (from lspci) but they 
>> aren’t labeled as yours:
>> 
>>   d000-dfff : PCI Bus :04
>> d000-d0ff : :04:00.0
>> d100-d103 : :04:00.0
>> 
>> It is interesting to me that your lspci listing indicates "Kernel driver in 
>> use: parport_p” whereas my card does not. I guess Ubuntu must not even 
>> recognize it as a parport. I didn’t have any luck with trying to use the 
>> address ranges shown to connect or set it to epp with the showport utility.
>> 
> Oh, oh!  If the Linux kernel does not recognize it as a 
> parallel port, (probably because the card's BIOS extension 
> ROM code is buggy) then the extended control registers won't 
> be declared in the BIOS area at boot time.  Thus, the 
> various LinuxCNC drivers that try to intelligently handle 
> the parport (such as select EPP mode) will not work
> properly.

It might be nice to add some basic parport trouble shooting instructions 
somewhere. E.g.

1. Identify the parport card with:
lspci -vv
2. Check that the parport card is registered with the parport_p kernel. 
Potentially return the card if it fails
3. If it passes, locate the card with:
cat /proc/ioports | grep parport
4. If you need to set to epp mode, try:
sudo ./showport rangeBegin rangeEnd e
(replace rangeBegin and rangeEnd with values discovered with the 
command above
5. If it doesn’t work in epp mode still - consider returning the card


Just my 2¢ - anyone care to add to it?

> You might still be able to make this card work with various 
> tricks like my pcisetup program (but you have to know the 
> address of the ECR register to do that).  This card may just 
> be totally oddball, and only the specific Windows driver for 
> that card knows what to do with it.  There ARE standards, 
> but a lot of outfits choose to not follow them.

Might be worth the trouble if the card wasn’t $12 and I had another card coming 
today.

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

Re: [Emc-users] Configuring New PCI Parport

[Todd Zuercher]

I am using two Netmos cards on two installs (one is a 9815 and the other is a 
9865), and haven't had any trouble with either one.  But I am not using either 
one in EPP mode.  I am using the ports with X-mode in Linuxcnc witch requires 
the ports to be set to SPP mode. 

- Original Message -
From: "Jon Elson" 
To: "Enhanced Machine Controller (EMC)" 
Sent: Friday, November 20, 2015 12:06:23 PM
Subject: Re: [Emc-users] Configuring New PCI Parport

On 11/19/2015 11:24 PM, Gene Heskett wrote:
> Did I see the number 9815 in the chipset someplace back in 
> the beginning of this thread? IIRC thats a netmos number, 
> and is known to not work at this location from several 
> years ago. I then bought a StarTech card, which did work. 
> But that whole machine was retired when the D525MW atom 
> machines became available. Cheers,
The NetMos 9805 definitely did NOT work in EPP mode at all.  
The 9815 is SUPPOSED to work, and I think at least two 
people have confirmed that this revised chip does work in 
EPP mode.  I've never checked one here so I can't say for sure.

Jon

--
___
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] Configuring New PCI Parport

[Jon Elson]

On 11/19/2015 11:24 PM, Gene Heskett wrote:
> Did I see the number 9815 in the chipset someplace back in 
> the beginning of this thread? IIRC thats a netmos number, 
> and is known to not work at this location from several 
> years ago. I then bought a StarTech card, which did work. 
> But that whole machine was retired when the D525MW atom 
> machines became available. Cheers,
The NetMos 9805 definitely did NOT work in EPP mode at all.  
The 9815 is SUPPOSED to work, and I think at least two 
people have confirmed that this revised chip does work in 
EPP mode.  I've never checked one here so I can't say for sure.

Jon

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

Re: [Emc-users] Configuring New PCI Parport

[Jon Elson]

On 11/20/2015 11:19 AM, Sarah Armstrong wrote:
> i have quite a collection of parallel port cards that don't work with
> Linuxcnc  both pci & pcie
> the only ones so far i have found to work are the startech ( after patching
> the epp ) & one's with Oxford semi chipsets
> are the only ones so far iv'e found to work just about out the box .
>
>
I don't know if any other companies use the Oxford chips, 
but that is what is in all SIIG boards.
And, yes, they have been flawless, I have several here in 
various test stations.

Jon

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

Re: [Emc-users] Lathe saddle protection

[rayj]

Just thinking out loud here.

I have seen table saws that have emergency stop systems that use 
capacitance (I believe) to sense when something (a finger) touches the 
blade.  The touch sensor trips a brake that stops the blade virtually 
instantly.

I was thinking something similar could be used to sense contact.  It 
wouldn't prevent a strike, but it would minimize the consequences.

Raymond Julian
Kettle River, MN

The things we admire in men, kindness and generosity, openness, honesty, 
understanding and feeling are the concomitants of failure in our system. 
And those traits we detest, sharpness, greed, acquisitiveness, meanness, 
egotism and self-interest are the traits of success. And while men 
admire the quality of the first they love the produce of the second. 
-John Steinbeck, novelist, Nobel laureate (1902-1968)

On 11/20/2015 05:50 AM, Gene Heskett wrote:
> On Friday 20 November 2015 06:00:00 andy pugh wrote:
>
>> Has anyone ever tried embedding proximity sensors in their lathe
>> saddle to protect against chuck-strikes or backing up against the
>> tailstock?
>>
>> If so, any "lessons learned" about where to put the sensors?
>
> Chuck strikes would seem to be so "tool mounted" a variable as to be
> unsolvable. But I have considered hanging a straight simple button micro
> microswitch on the saddle such that if its button is compressed, it
> would stop the saddle as it touched the tailstock.  I have a similar
> setup on the front of the carriage that senses the cross-slide, stopping
> it about 10 thou from a crash stop at the front end of the slot in the
> carriage, that has been used as a homing switch for a couple years.  My
> switch positions are such that I can, unless I've a really huge work
> piece in the chuck, remove the QC holder, and home the machine without
> crashing into something.
>
> In the FWIW category, the tailstock clamping on a 7x12 is so poor that my
> 1605 z screw can actually push it when you think its locked.
>
> Cheers, Gene Heskett
>

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

Re: [Emc-users] Configuring New PCI Parport

[Gene Heskett]

On Friday 20 November 2015 12:06:23 Jon Elson wrote:

> On 11/19/2015 11:24 PM, Gene Heskett wrote:
> > Did I see the number 9815 in the chipset someplace back in
> > the beginning of this thread? IIRC thats a netmos number,
> > and is known to not work at this location from several
> > years ago. I then bought a StarTech card, which did work.
> > But that whole machine was retired when the D525MW atom
> > machines became available. Cheers,
>
> The NetMos 9805 definitely did NOT work in EPP mode at all.
> The 9815 is SUPPOSED to work, and I think at least two
> people have confirmed that this revised chip does work in
> EPP mode.  I've never checked one here so I can't say for sure.
>
> Jon

Well, that instance was been probably 6 years back up the log Jon, and 
you know what happens to short term memory loss at my year count.  I've 
displayed it several times right here.  Darnit...
However, the 9805 does NOT ring any bells at 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] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 13:10:05 Les Newell wrote:

> > I wonder how a hall effect might work, to sense the chucks jaws
> > going by when it gets too close? I think I'd boost the length of its
> > output pulse with a one shot to make sure its triggering wasn't
> > missed though when the servo thread is the only one, no base thread
> > to read it every 30 microseconds. Cheers, Gene Heskett
>
> Oh, I see, you want to sense the chuck jaws themselves. That leaves
> the sensor in a very vulnerable place, right in the firing line for
> coolant, swarf etc. What is wrong with having a switch on the saddle
> that trips when it hits an adjustable stop? It works quite well and is
> a lot less vulnerable.
>
> Les

True, but to be effective, they would all have to be adjusted during the 
setup for the job at hand.  Thats a bit of a PITA, so I listen for the 
first ticks.

The hall effects lifetime can be shortened drastically by the wrong 
petroleum coming in contact with its epoxyB encapsulating materiel too.  
Been there, done that, paid the repair bill to prove it. WD-40 in 
particular is about a 3 days to dead device poison.

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] Configuring New PCI Parport

[jeshua]

> On Nov 20, 2015, at 11:46 AM, andy pugh  wrote:
> 
> n 18 November 2015 at 17:49, Jeshua Lacock  wrote:
>> Long story short; I had to install a new PCI Parport card and I am trying to 
>> figure out what address to use for my hal configuration file for Axis. The 
>> card has the same CH382L chip listed in the wiki:
>> 
>> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?WCH
> 
> It isn't clear to me what that Wiki page is for. It doesn't say that
> it works. It doesn't say that it doesn't work.

Well I found that section of the wiki from folks on the list discussing PCI 
cards that “just work”.

To me, it doesn’t make sense to me to list cards in the wiki that aren’t 
supported unless specifically state so (eg don’t buy this card), otherwise I 
interpret them as being supported by the project. Again; just my interpretation.


Best,

Jeshua Lacock
Founder/Engineer
<3DTOPO.com>
GlassPrinted.com
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Lathe saddle protection

[Gene Heskett]

On Friday 20 November 2015 10:47:15 andy pugh wrote:

> On 20 November 2015 at 15:30, Gene Heskett  wrote:
> > I have probably worn off an eighth of an
> > inch off the nose of the compound feed since it usually hangs out in
> > front of the saddle.  Just one of the reasons I have considered
> > removing it and replacing it with a block of steel
>
> Right and below-centre in this photo
> https://picasaweb.google.com/lh/photo/gQuPtDB0J_ZiqVVDHaf8vtMTjNZETYmy
>PJy0liipFm0?feat=directlink Is the pattern that I am taking to the
> foundry tomorrow to make just such a block for my Holbrook.

Looks quite a bit more complex than what I had in mind.  I assume you are 
doing a ball screw conversion from the saddle up, using a much smaller, 
geared down motor that will be hidden behind the apron.  I thought of 
using a 52 oz single stack thats not quite a nema 23, but bigger than a 
17.  It could do the job with enough gear down, but that might limit the 
G76 entry/exit ramp rate. It did not have the cojones to drive the screw 
direct.  Currently that 525 oz seems to work well, with a triple stack 
23 direct on the screw, but hanging off the rear of the saddle & wanting 
to lift the front way from the V-Way as its too darned heavy and too far 
back.  But if I shorten that, then there really is not room for the back 
of the compound, and I've crushed the swarf covers on the cross-feed 
several times with the rear of the compound.

Shuffling motors around to free up one of the double stack 252 oz motors 
would probably be a better deal as that would lose half a pound or more 
of overhanging weight.  That alone would set it more firmly on its only 
V-Way.  But there's that missing round tuit again.

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] Lathe saddle protection

[Dave Cole]

Do what the big companies do.   Make up a big sign or list of things you 
need to do and put it over your lathe..
You need to get tougher on yourself!   ;-)

Actually a procedural checklist might make sense.

Dave


On 11/20/2015 1:48 PM, andy pugh wrote:
> On 20 November 2015 at 18:10, Les Newell  wrote:
>
>> What is wrong with having a switch on the saddle that trips
>> when it hits an adjustable stop?
> That requires me to remember to set the adjustable stop. If I can
> remember to do that, then I can remember to check the soft limits.
>

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

Re: [Emc-users] Configuring New PCI Parport

[jeshua]

> On Nov 20, 2015, at 11:41 AM, Jon Elson  wrote:
> 
> On 11/20/2015 11:19 AM, Sarah Armstrong wrote:
>> i have quite a collection of parallel port cards that don't work with
>> Linuxcnc  both pci & pcie
>> the only ones so far i have found to work are the startech ( after patching
>> the epp ) & one's with Oxford semi chipsets
>> are the only ones so far iv'e found to work just about out the box .
>> 
>> 
> I don't know if any other companies use the Oxford chips, 
> but that is what is in all SIIG boards.
> And, yes, they have been flawless, I have several here in 
> various test stations.

So I have installed the card, see that it has the parport_pc kernel registered, 
but cannot get it to go to epp. 

08:00.0 Parallel controller: Oxford Semiconductor Ltd OX9162 Mode 0 (parallel 
port) (prog-if 03 [IEEE1284])
Subsystem: Siig Inc Device 2020
Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- 
Stepping- SERR- FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- 
SERR- 
Kernel driver in use: parport_pc


0d00- : PCI Bus :00
  c000-cfff : PCI Bus :07
c000-cfff : PCI Bus :08
  c000-c01f : :08:00.0
  c020-c023 : :08:00.0
  c030-c037 : :08:00.0
c030-c032 : parport0
c033-c037 : parport0

I tried the c030 and c020 addresses. Not sure what ranges I would use for the 
showport command and/or the pcisetup command - tried what seemed likely and no 
dice. Also tried starting Axis with c020 and c030 addresses in hopes it would 
set the board to epp for me - no dice. 


Thanks,

Jeshua Lacock
Founder/Engineer
<3DTOPO.com>
GlassPrinted.com
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Configuring New PCI Parport

[jeshua]

> On Nov 20, 2015, at 9:14 PM, Jon Elson  wrote:
> 
> Here's the report from my older Siig card :
> 04:02.0 Parallel controller: Siig Inc CyberParallel (2-port) 
> (prog-if 02)
> Subsystem: Siig Inc CyberParallel (2-port)
> Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- 
> VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
> Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium 
>> TAbort- SERR-  Interrupt: pin A routed to IRQ 18
> Region 0: I/O ports at dc70 [size=8]
> Region 2: I/O ports at dc78 [size=8]
> Capabilities: [a0] Power Management version 1
> Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA 
> PME(D0-,D1-,D2-,D3hot-,D3cold-)
> Status: D0 PME-Enable- DSel=0 DScale=0 PME-
> Kernel driver in use: parport_pc
> Kernel modules: parport_pc
> 
> DC70 and DC78 are the two data ports addresses for the two 
> parallel ports it provides.
> Now, when I boot this system, I have to run the pcisetup 
> program before the diags will work, and I think I use the 
> dc70 address and it figures out the correct ECR address from 
> that.  LinuxCNC handles the EPP mode selection 
> automatically, I don't have to do anything.

Hi Jon,

Hmm - yours says " Kernel modules: parport_pc” whereas my only states: "Kernel 
driver in use: parport_pc”.

So I guess you are saying it has to be one of the following then:

Region 0: I/O ports at c030 [size=8]
Region 1: I/O ports at c020 [size=4]
Region 2: I/O ports at c000 [size=32]

But I get nothing on any of them.

jeshua@cncBot:~$ sudo ./pcisetup c030
io addr = c030
jeshua@cncBot:~$ sudo univstepdiag/univstepdiags c030 bus
io addr = c030
parport addr 0xc030 
Bus Map 
  Board Addr  Type Ver.
   0No Boardff
   1No Boardff


jeshua@cncBot:~$ sudo univstepdiag/univstepdiags c020 bus
io addr = c020
parport addr 0xc020 
Bus Map 
  Board Addr  Type Ver.
   0No Boardff
   1No Boardff


jeshua@cncBot:~$ sudo ./pcisetup c000
io addr = c000
jeshua@cncBot:~$ sudo univstepdiag/univstepdiags c000 bus
io addr = c000
parport addr 0xc000 
Bus Map 
  Board Addr  Type Ver.
   0Unknown  0
   1Unknown  0

For my last SIIG card (also showed c030, that is what I used in fact), but I 
didn’t use the pcisetup command, I used this utility someone on this list 
pointed me to and the magic command was:

sudo ./showport C020 C030 e

Based on my shell history, it looks like I used trial and error to come up with 
those addresses. I am not sure who wrote it - they don’t have their name in the 
source but it is here:

http://glassprinted.com/showport.c.txt

The nice thing about the showport command is it flashed parport to stay in 
extended mode even on reboot. Well that is, until the Big Zap, then I had to 
run that showport command every reboot - which lead me to suspect the parport 
over the cable even more.

> Are you really sure your USC board is working?  Possibly, 
> something happened to either it or your IEEE-1284 cable.  It 
> WAS working a few days ago, and then SOMETHING quit.  You 
> assumed it was the parallel port, but maybe it was some 
> other unit.  When you power it on, does the Load Fail LED 
> flash for about 0.1 second and then go out?

The USC is brand new that you just tested. When I got it, it was working 
intermittently so I thought we figured it either had to have been the old PCI 
Parport card or the cable.

The cable I ordered from you, but do you really think the whole cable could 
have got fried? I will test all the pins if you think so.

But - if the cable is bad - it was at least letting me talk to the USC - albeit 
unreliably - however now with the new card nothing. I can swap the old card 
back in to see if it still can see it (I suspect that it will). Let me know if 
you think that would be helpful.

When I power it on, yes I get a red flash then go out to a solid green beside 
it; only when the power is also connected.  When just the parport cable is 
plugged in (no 12v to the USC) I get no lights on the board, which IIRC, is how 
it was in the past when it was woking until I set the card to epp then I would 
get dim green (and maybe dim red) lights...


Thanks,

Jeshua Lacock
Founder/Engineer
<3DTOPO.com>
GlassPrinted.com


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

Re: [Emc-users] Configuring New PCI Parport

[Jon Elson]

On 11/20/2015 06:56 PM, jeshua wrote:
>>
> So I have installed the card, see that it has the parport_pc kernel 
> registered, but cannot get it to go to epp.
>
> 08:00.0 Parallel controller: Oxford Semiconductor Ltd OX9162 Mode 0 (parallel 
> port) (prog-if 03 [IEEE1284])
>   Subsystem: Siig Inc Device 2020
>   Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- 
> Stepping- SERR- FastB2B- DisINTx-
>   Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort- 
> SERR-Interrupt: pin A routed to IRQ 18
>   Region 0: I/O ports at c030 [size=8]
>   Region 1: I/O ports at c020 [size=4]
>   Region 2: I/O ports at c000 [size=32]
>   Region 3: Memory at fb10 (32-bit, non-prefetchable) [size=4K]
>   Capabilities: 
>   Kernel driver in use: parport_pc
>
>
> 0d00- : PCI Bus :00
>c000-cfff : PCI Bus :07
>  c000-cfff : PCI Bus :08
>c000-c01f : :08:00.0
>c020-c023 : :08:00.0
>c030-c037 : :08:00.0
>  c030-c032 : parport0
>  c033-c037 : parport0
>
> I tried the c030 and c020 addresses. Not sure what ranges I would use for the 
> showport command and/or the pcisetup command - tried what seemed likely and 
> no dice. Also tried starting Axis with c020 and c030 addresses in hopes it 
> would set the board to epp for me - no dice.
>
>
Here's the report from my older Siig card :
04:02.0 Parallel controller: Siig Inc CyberParallel (2-port) 
(prog-if 02)
 Subsystem: Siig Inc CyberParallel (2-port)
 Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- 
VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx-
 Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium 
 >TAbort- SERR- https://lists.sourceforge.net/lists/listinfo/emc-users