Re: [Emc-users] Rifling machine

[Dave Caroline]

Mount the cutter bar on a rotary for controlled rotation, then a
A plain g1A(rotation degrees)x(barrellength) would be what I would do.

I have done some very coarse helix cuts that way

The taper wedge putting on the cut is interesting, that can be the
same/similar just adjusting the out stroke as it runs

Dave Caroline

On 05/09/2014, Gene Heskett  wrote:
> On Thursday 04 September 2014 22:04:46 Len Shelton did opine
> And Gene did reply:
>> We have a customer who is wanting to make a motor controlled rifling
>> machine. I spent a good bit of time researching it for him today and I
>> was surprised that it seems that no one is doing it with motors (or at
>> least I couldn't find anyone). I understand the need for the linear
>> relationship of the cutter to the position of the indexer in such a
>> machine. But I wanted to hear your guys thoughts on doing this with
>> motors, which will inherently have some degree of granular positioning.
>>
>> Here is the best machine design I could find that really demonstrates
>> the issue:
>> https://www.youtube.com/watch?v=L_yemjfgkE0
>>
>>  >Len
>
> I don't see a thing there that cannot be done with steppers, nema 23's
> driving the big pulley thru timing belts, with the rotation being a
> variation of the G33.1 function.  But I'd not use the pulley & crank at
> all, but just a bigger stepper and ball screw because then you would not
> need the math to translate the rotary motion into the linear position to
> drive the G33.1 inputs. The biggest problem would be in adjusting the
> number of grooves in software, but I don't believe that its a problem hal
> or gcode cannot solve by running a start point cycle back and forth with a
> mod function in the gcode math.
>
> Interesting problem, but one that I believe lcnc can do with creative
> coding.  But is cut rifling enough more accurate than the other two
> methods, both of which are at least 100x faster, to make it worthwhile to
> do it that way?  Thats the $64K question. IMO yes. :-)
>
> I wish I knew I had 10 years left, I'd be getting started on just such a
> machine.  Just for S&G.  And bug hole sized 10 shot groups of coarse.
>
> 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 
> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Rifling machine

[Gregg Eshelman]

On 9/4/2014 9:42 PM, Bruce Layne wrote:

> The manual rifling method requires a custom cutter to be made that is
> hammered/pulled through the bore.  That's how rifle barrels were rifled
> in the old days.

Here's a good site on rifling methods.

http://firearmshistory.blogspot.com/2010/05/rifling-manufacturing-hammer-forged.html

---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Rifling machine

[Gregg Eshelman]

On 9/4/2014 8:04 PM, Len Shelton wrote:
> We have a customer who is wanting to make a motor controlled rifling
> machine. I spent a good bit of time researching it for him today and I
> was surprised that it seems that no one is doing it with motors (or at
> least I couldn't find anyone). I understand the need for the linear
> relationship of the cutter to the position of the indexer in such a
> machine. But I wanted to hear your guys thoughts on doing this with
> motors, which will inherently have some degree of granular positioning.
>
> Here is the best machine design I could find that really demonstrates
> the issue:
> https://www.youtube.com/watch?v=L_yemjfgkE0

Amazing what was done with purely mechanical methods before being 
replaced with CNC. Neat to see how rifling cutting a grove at a time was 
automated before single pass 'button' cutting was invented.

I'd have the cutting slide just run straight and put the barrel in a 
computer controlled rotator to both index and control the twist of the 
rifling. That would make it a fairly simple task to cut any twist 
including gain twist which always has to be cut, can't be hammer forged 
on a mandrel.

The cutting bit advance control could also be operated by a stepper.

Those two changes, put under computer control, would drastically 
simplify the construction of the rest of the machine.

Instead of the crank to run the cutter back and forth you could use a 
motor driving a ballscrew. With a rotary encoder on that and the barrel 
indexer/rotator you'd have complete CNC rifling with infinitely 
adjustable twist and length of stroke.

Something else I'd try is backing the bottom of the cutter bar with a 
half round of UHMW. That would make it slide easy in the barrel and 
wouldn't make any lengthwise scratches.

As for the motor turning the barrel, to get fine, step-free positioning 
there are a few ways to do it. Got $ to spend? Try a direct drive with a 
hollow shaft servo motor. Not so much $? A double-enveloping, high ratio 
worm gear set driven by a fast motor, with the encoder mounted to 
directly read the barrel position. Well, that might not be so 
"inexpensive" but compared to a hollow shaft servo with a hole large 
enough to pass a rifle barrel...

Then there's ye olde cogged pulleys and toothed belts. Use two belts off 
both ends of a double shaft stepper just for added strength of doubling 
up on belts and balancing pressure on the motor bearings.

Whatever is used to rotate the barrel it must have enough torque at very 
slow speeds to keep the cutting pressure from moving it from the 
movement the control commands. No problem with the worm drive since 
those can't be back-driven unless specially designed for it.

Like so many things, how precise do you want it? How much money do you 
have or how good are you at scrounging for bargains on precision equipment?

Is this for competition grade tack driving guns or for mass produced 
everyday shooters?

---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Rifling machine

[Gene Heskett]

On Thursday 04 September 2014 22:04:46 Len Shelton did opine
And Gene did reply:
> We have a customer who is wanting to make a motor controlled rifling
> machine. I spent a good bit of time researching it for him today and I
> was surprised that it seems that no one is doing it with motors (or at
> least I couldn't find anyone). I understand the need for the linear
> relationship of the cutter to the position of the indexer in such a
> machine. But I wanted to hear your guys thoughts on doing this with
> motors, which will inherently have some degree of granular positioning.
> 
> Here is the best machine design I could find that really demonstrates
> the issue:
> https://www.youtube.com/watch?v=L_yemjfgkE0
> 
>  >Len

I don't see a thing there that cannot be done with steppers, nema 23's 
driving the big pulley thru timing belts, with the rotation being a 
variation of the G33.1 function.  But I'd not use the pulley & crank at 
all, but just a bigger stepper and ball screw because then you would not 
need the math to translate the rotary motion into the linear position to 
drive the G33.1 inputs. The biggest problem would be in adjusting the 
number of grooves in software, but I don't believe that its a problem hal 
or gcode cannot solve by running a start point cycle back and forth with a 
mod function in the gcode math.

Interesting problem, but one that I believe lcnc can do with creative 
coding.  But is cut rifling enough more accurate than the other two 
methods, both of which are at least 100x faster, to make it worthwhile to 
do it that way?  Thats the $64K question. IMO yes. :-)

I wish I knew I had 10 years left, I'd be getting started on just such a 
machine.  Just for S&G.  And bug hole sized 10 shot groups of coarse.

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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Rifling machine

[Bruce Layne]

I've thought that LinuxCNC might make a good basis for a universal 
barrel rifling machine, either using a small off-the-shelf carbide 
insert, or using custom carbide cutters.  There's a shop close to me 
that will make custom carbide cutters.

The very well done animated video from your link demonstrated some 
clever old school mechanisms (sine bar, rack and pinion, pawl, etc. 
Almost all of that could be replaced by LinuxCNC with some stepper 
motors or servo motors and encoders.

The manual rifling method requires a custom cutter to be made that is 
hammered/pulled through the bore.  That's how rifle barrels were rifled 
in the old days.  The angle on the end of the cutter determined the 
rifling rate (rate of twist), and it was a bit of an art with some trial 
and error.  A barrel smith might want one turn in nine inches and might 
get one turn in 10.3 inches.  Using LinuxCNC to essentially thread the 
inside of the bore should ensure a precise and repeatable rate of twist.

If you follow through with this project, I'd love to see what you create 
and the results you get.  Boring and rifling barrels is a specialized 
trade.  There aren't many companies that do it.  The equipment is quite 
specific to the task.  Most firearms manufacturers buy rifled barrel 
blanks and do the finish work - threading the muzzle, cutting the 
chamber, fluting, etc.  It's about time for CNC to make it much easier 
to cut barrel rifling.





On 09/04/2014 10:04 PM, Len Shelton wrote:
> We have a customer who is wanting to make a motor controlled rifling
> machine. I spent a good bit of time researching it for him today and I
> was surprised that it seems that no one is doing it with motors (or at
> least I couldn't find anyone). I understand the need for the linear
> relationship of the cutter to the position of the indexer in such a
> machine. But I wanted to hear your guys thoughts on doing this with
> motors, which will inherently have some degree of granular positioning.
>
> Here is the best machine design I could find that really demonstrates
> the issue:
> https://www.youtube.com/watch?v=L_yemjfgkE0
>
>   >Len
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Rifling machine

[Len Shelton]

We have a customer who is wanting to make a motor controlled rifling 
machine. I spent a good bit of time researching it for him today and I 
was surprised that it seems that no one is doing it with motors (or at 
least I couldn't find anyone). I understand the need for the linear 
relationship of the cutter to the position of the indexer in such a 
machine. But I wanted to hear your guys thoughts on doing this with 
motors, which will inherently have some degree of granular positioning.

Here is the best machine design I could find that really demonstrates 
the issue:
https://www.youtube.com/watch?v=L_yemjfgkE0

 >Len

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Mixing BLDC and brushed motors

[Peter C. Wallace]

On Thu, 4 Sep 2014, John Prentice (FS) wrote:

> Date: Thu, 4 Sep 2014 22:44:13 +0100
> From: "John Prentice (FS)" 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "'Enhanced Machine Controller (EMC)'" 
> Subject: Re: [Emc-users] Mixing BLDC and brushed motors
> 
> Peter thank you
>
>> -Original Message-
>> From: Peter C. Wallace [mailto:p...@mesanet.com]
>
>> Yes you can look at the 7i39 as 6 individual 1/2 bridges and drive 3 DC
> motors from one 7i39. if a brushed motor was connected to half bridges A and
> B, I _think_ >what is needed is to drive A with the PID output and B with
> -PID output.
>
>> You can also dedicate one 1/2 bridge output as a 1/2 way up reference and
> then run up to 5 small dc  motors (1/2 voltage 1/5 current) from the other 5
> 1/2 bridges
>
> Ingenious - but it leaves one short on encoder interfacing. I will try the A
> and B with ++Pid-out and --Pid-out
>>>
>>>
>>> The ideal card for the brushed motor would be a 7i40. Would it be
>>> practicable to have Hostmot2 firmware for both 7i39 and 7i40 co-exist
>>> in the 5i20? If so does a BIT file exist with this combination.
>>>
>>>
>
>
>> Not sure if theres a standard bitfile for that but its easy copy-paste a
> pinout up and make one
>
> I have a spare 7i40 so I think I ought to learn how to do this - I see
> resources on the web that will start me going.
>
> Is ISE Webpack 9.2 still your preferred version for 5i20 and 5i22 (Spartan
> 2) boards? What OS would you run it on - I feel concerned that it is so old
> that it will give trouble on my Win 8.1 (64 bit) system. I can run some
> Linux flavour in a V-Box to provide any environment to best suit the app.

I think 10 is the latest that supports Spartan but 9.2 is my preferred
version (before they broke the editor)



>
> Do you have a link to a document showing the  "compile" workflow for your
> firmware?  A few years ago I have written standalone FPGA and CPLD code for
> Xilinx - albeit in Verilog so hope not to prove a total nooby.
>
> TIA
>
> John Prentice
>


Yes Tom Light has a nice tutorial here:

http://tom-itx.no-ip.biz:81/~webpage/emc/xilinx/xilinx14_install_index.php

>
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Gregg Eshelman]

On 9/4/2014 9:07 AM, Viesturs Lācis wrote:
> Hello!
>
> I have a strange situation:
> Stepper motor of Z axis on a cnc router is losing it's position -
> doing a series of numerous holes in material will end up with various
> depths of holes.

Just thought of this, have you tried homing the Z axis up before 
drilling each hole? That would reset the position so the control will 
always start from the same known position each time.


---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Gregg Eshelman]

On 9/4/2014 9:07 AM, Viesturs Lācis wrote:
> Hello!
>
> I have a strange situation:
> Stepper motor of Z axis on a cnc router is losing it's position -
> doing a series of numerous holes in material will end up with various
> depths of holes.
>
> The strange part is:
> 1) for each time, when machine is running, it will lose position only
> in one direction; if holes are getting deeper, then they only get
> deeper, not the other way;
> 2) the direction of "drift" will change randomly as machine is
> restarted and only then; simply running another g-code file will not
> change the direction of "drift";

First thing I'd look at is your spindle speed and feed rate.

If the bit is pulling itself into the work faster than the control is 
commanding, it can force the motor to 'run ahead' of the commanded steps.

If the control is trying to push the bit in faster than it can cut, the 
effect is a 'push back' where the step signals go to the motor but it 
doesn't actually move.

Combine either or both with a slight variance in density of the material 
and you'll get randomness.

If tweaking the speed and feed doesn't completely cure it there are 
hardware solutions.

Were this my machine I'd find a way to put a scale (magnetic type, crud 
and fluids don't mess them up and can be cut to length) or rotary 
encoder on so the control always knows where the axis is.

Another possibility is more power. Put on a larger stepper with more 
torque so it can't be pulled ahead of or pushed behind the stream of 
step signals.

Another cause of it might be variability of chip loading in the bit. 
What do you have for keeping the bit clear? For material that can't take 
liquids use an air nozzle and vacuum.

---
This email is free from viruses and malware because avast! Antivirus protection 
is active.
http://www.avast.com


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Mixing BLDC and brushed motors

[John Prentice (FS)]

Peter thank you

>-Original Message-
>From: Peter C. Wallace [mailto:p...@mesanet.com] 

>Yes you can look at the 7i39 as 6 individual 1/2 bridges and drive 3 DC
motors from one 7i39. if a brushed motor was connected to half bridges A and
B, I _think_ >what is needed is to drive A with the PID output and B with
-PID output.

>You can also dedicate one 1/2 bridge output as a 1/2 way up reference and
then run up to 5 small dc  motors (1/2 voltage 1/5 current) from the other 5
1/2 bridges

Ingenious - but it leaves one short on encoder interfacing. I will try the A
and B with ++Pid-out and --Pid-out
>>
>>
>> The ideal card for the brushed motor would be a 7i40. Would it be 
>> practicable to have Hostmot2 firmware for both 7i39 and 7i40 co-exist 
>> in the 5i20? If so does a BIT file exist with this combination.
>>
>>


>Not sure if theres a standard bitfile for that but its easy copy-paste a
pinout up and make one

I have a spare 7i40 so I think I ought to learn how to do this - I see
resources on the web that will start me going. 

Is ISE Webpack 9.2 still your preferred version for 5i20 and 5i22 (Spartan
2) boards? What OS would you run it on - I feel concerned that it is so old
that it will give trouble on my Win 8.1 (64 bit) system. I can run some
Linux flavour in a V-Box to provide any environment to best suit the app.

Do you have a link to a document showing the  "compile" workflow for your
firmware?  A few years ago I have written standalone FPGA and CPLD code for
Xilinx - albeit in Verilog so hope not to prove a total nooby.

TIA 

John Prentice



--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[Gene Heskett]

On Thursday 04 September 2014 12:08:45 John Kasunich did opine
And Gene did reply:
> On Thu, Sep 4, 2014, at 11:15 AM, Gene Heskett wrote:
> > pieces of wood to hold a pair of huge 15k uf, 60 volt Sangamo
> > capacitors in a power supply I'm building for the lathe motor
> > 
> > I don't trust the 140
> > volt surge ratings of the smaller 100 volt caps
> > With this
> > transformer, those 100 volt caps have a no signal voltage of 113
> > volts.
> > 
> > I'll put 7 watt night light lamps
> > across them for charge equality and power down bleeders.
> 
> Be very carefull using light bulbs as balancing resistors.  They have a
> resistance vs temperature curve that can bite you.  You can get in a
> situation where one bulb has a bit less voltage across it than the
> other. The one with less voltage dissipates less power, it cools down
> and the other one heats up.  Its resistance drops and the other one
> increased. That results in even less voltage across the cool bulb, and
> more across the hot bulb.  The cool one gets even cooler.  Depending
> on the exact characteristics of the bulbs and the total applied
> voltage, you can wind up with a 2:1 or bigger difference in voltage.
> 
> I would dig in my scrap boxes for real resistors if I was you.

You're absolutely right John.  But 99.99% of my resistors are puny power, 
1/4 watt and less.  And because the time related failure of bulk carbon in 
that resistor range, related to folks commonly exceeding the rated 
voltage, regardless of the wattage being plumb safe, they will not be 
carbon.  So I'll have to find a pair of 10 watt wirewound's in the 5k-10k 
range.

I hate to admit it, but you headed me off at the pass on that one, and I 
needed it.  Might even have to go raid the old NTSC tx site. The shack of 
course has nothing remotely close, so I got a 10 pack of 5k 10 watters 
that will burn about 1.5 watts each from PartsExpress. Along with half a 
pound of the new 4% silver, lead free solder since my spools of 15 year 
old GC 3% silver lead/tin are just about ~30~.

So it will be 3 or 4 days, in the meantime I can finish the mountings for 
the big 15k uf x 60 WV Sangamo's.  Gotta figure out a way to put some 
pressure on the mount caps, s/b fun since they made from MDF.

Thanks John.

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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Setting of HAL pin from gcode

[Marius Liebenberg]

Thanks Andy, it works great.

On 2014-09-04 17:10, andy pugh wrote:
> On 4 September 2014 15:52, Marius Liebenberg  wrote:
>
>> motion.analog-out-00 => mycomp-float-in.
> Yes, exactly that.
>

-- 

Regards /Groete

Marius D. Liebenberg
+27 82 698 3251
+27 12 743 6064
QQ 1767394877


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[Gene Heskett]

On Thursday 04 September 2014 11:39:57 John Thornton did opine
And Gene did reply:
> I added it to the wiki.
> 
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Simple_LinuxCNC_G-Code_Generat
> ors
> 
> When you download arcbuddy15.py change the name to arcbuddy.py.
> 
> JT
> 
> On 9/4/2014 10:16 AM, Gene Heskett wrote:
> > On Thursday 04 September 2014 10:57:49 John Thornton did opine
> > 
> > And Gene did reply:
> >> Gene,
> >> 
> >> Try Arc Buddy 1.5, I've incorporated the ability to use in Gedit and
> >> instructions on how to do so.
> >> 
> >> JT
> > 

And they work!  The only stumbling block was that the 'star' near the help 
button was actually a missing icon icon.  But clicking on it worked as 
expected.

Thank you John.  A big improvement in its utility.

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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Mixing BLDC and brushed motors

[Peter C. Wallace]

On Thu, 4 Sep 2014, John Prentice (FS) wrote:

> Date: Thu, 4 Sep 2014 13:50:23 +0100
> From: "John Prentice (FS)" 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: [Emc-users] Mixing BLDC and brushed motors
> 
> Greetings
>
>
>
> This is probably a question for PCW but others might know or have dome it.
>
>
>
> I am looking at a small machine design where two axes will be BLDC motors
> and one (or perhaps two) brushed DC.
>
>
>
> The BLDC motors suit a 7i39 very well (interfaced by a 5i20). Are there any
> problems in using just two "phases" of another 7i39 to drive the brushed
> motor? I presume that one would drive, say A, with the desired level
> (positive or negative) holding  inputs B and C at zero - i.e. no bldc
> component is needed.
>


Yes you can look at the 7i39 as 6 individual 1/2 bridges and drive 3 DC motors 
from one 7i39. if a brushed motor was connected to half bridges A and B, I 
_think_ what is needed is to drive A with the PID output and B with -PID 
output.

You can also dedicate one 1/2 bridge output as a 1/2 way up reference
and then run up to 5 small dc  motors (1/2 voltage 1/5 current) from
the other 5 1/2 bridges


>
>
> The ideal card for the brushed motor would be a 7i40. Would it be
> practicable to have Hostmot2 firmware for both 7i39 and 7i40 co-exist in the
> 5i20? If so does a BIT file exist with this combination.
>
>


Not sure if theres a standard bitfile for that but its easy copy-paste a 
pinout up and make one

>
> TIA
>
>
>
> John Prentice
>
>
>
>
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[John Prentice (FS)]

Greetings

>-Original Message-
>From: Viesturs Lācis [mailto:viesturs.la...@gmail.com] 
>2014-09-04 18:16 GMT+03:00 alex chiosso :
>> If the holes are deeper doesn't it means that the motor is making "more"
>> steps and not "less" ?
>> Is it possible that you have extra steps due to noises or bad 
>> connections PC ---> BoB ---> Gecko Drive ?

>Yes, as I said, motor is "drifting" in particular direction, which may
change (not always), if system is restarted, but the direction of drift does
not change, once the >system is running - if it gradually goes up, then it
does so; if down, then it keeps getting deeper.

>There is no BoB, step/dir signals go directly from 5i23 to Gecko.

Two thoughts - sadly neither addressing the "direction of drift constant for
a given boot" symptom:

(a) It is not possible to infer if steps are being lost or gained from a
series of holes. Noise might give additional steps - but IMO this is unusual
-  but this might be when going down when holes get deeper or going up when
holes get shallower.

(b) It might be worth checking that the 5i23 pullups are set to 5volt for
the connector you are using. The 5i20 manual says that, with 3v3 pullups,
optos connected to a 5 volt rail may not fully turn off.

John Prentice




--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[andy pugh]

On 4 September 2014 16:07, Viesturs Lācis  wrote:

> There is Nema23 stepper motor and Gecko stepper driver.

A vaguely recall hearing of one channel of the G540 going "bad" and
then also that inverting the step signals makes it good again.
It certainly seems that it might be worth at least attempting an
inversion of the step pin. (both the parport and Mesa drivers allow an
output pin to be inverted)

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

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[John Kasunich]

On Thu, Sep 4, 2014, at 11:15 AM, Gene Heskett wrote:

> pieces of wood to hold a pair of huge 15k uf, 60 volt Sangamo capacitors 
> in a power supply I'm building for the lathe motor 

> I don't trust the 140 
> volt surge ratings of the smaller 100 volt caps
> With this 
> transformer, those 100 volt caps have a no signal voltage of 113 volts.

> I'll put 7 watt night light lamps 
> across them for charge equality and power down bleeders. 

Be very carefull using light bulbs as balancing resistors.  They have a
resistance vs temperature curve that can bite you.  You can get in a
situation where one bulb has a bit less voltage across it than the other.
The one with less voltage dissipates less power, it cools down and the
other one heats up.  Its resistance drops and the other one increased.
That results in even less voltage across the cool bulb, and more across
the hot bulb.  The cool one gets even cooler.  Depending on the exact
characteristics of the bulbs and the total applied voltage, you can wind
up with a 2:1 or bigger difference in voltage.

I would dig in my scrap boxes for real resistors if I was you.

-- 
  John Kasunich
  jmkasun...@fastmail.fm

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[alex chiosso]

This seems to be an electrical problem .
Have you tried to perform the machine cycle (i.e the holes drilling) in
"dry mode" (no material) and with a reduced feed rate ?
Just to understand if is related to the step frequency and torque .

Alex


On Thu, Sep 4, 2014 at 5:37 PM, Viesturs Lācis 
wrote:

> 2014-09-04 18:32 GMT+03:00 alex chiosso :
> >
> > Means that the "drift" in is both directions. But the axis (motor) is
> > always making more space than expected , isn't it ?
>
> Yes, "drift" can be in any direction, but the change of direction
> happens only on system restart.
>
> Viesturs
>
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[John Thornton]

I added it to the wiki.

http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Simple_LinuxCNC_G-Code_Generators

When you download arcbuddy15.py change the name to arcbuddy.py.

JT

On 9/4/2014 10:16 AM, Gene Heskett wrote:
> On Thursday 04 September 2014 10:57:49 John Thornton did opine
> And Gene did reply:
>> Gene,
>>
>> Try Arc Buddy 1.5, I've incorporated the ability to use in Gedit and
>> instructions on how to do so.
>>
>> JT
> URL?
>   
>> On 9/4/2014 7:58 AM, Gene Heskett wrote:
>>> Greetings;
>>>
>>> Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in
>>> that menu.
>>>
>>> gedit is version 2.30.3, but the gedit tools is 2.30.0 according to
>>> synaptic.  Is there a repo I need to enable in order to get the
>>> uptodate tools package that does contain the 'new' option?
>>>
>>> Running it from the axis interface, the data it generates is not
>>> saved anyplace, outputting this error in the cli I started lcnc
>>> from:
>>>
>>> File ended with no percent sign
>>> can't open /tmp/tmp55fjrP/arcbuddy-sje.py
>>>
>>> Since it does not unfocus even when you click someplace else, the
>>> only way I can use it is to generate the code, write it down, and
>>> hand copy it to the mdi interface for testing after arcbuddy-sje.py
>>> has been quit so I can change focus and type.  That is doing it the
>>> hard way. When I found the instructions about how to use it with
>>> gedit, which would also be handy, only to find my 10.04.4 LTS
>>> version of gedit cannot be configured according to those
>>> instructions.
>>>
>>> Thanks.
>>>
>>> Cheers, Gene Heskett
>> ---
>> --- Slashdot TV.
>> Video for Nerds.  Stuff that matters.
>> http://tv.slashdot.org/
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>
> Cheers, Gene Heskett


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Viesturs Lācis]

2014-09-04 18:32 GMT+03:00 alex chiosso :
>
> Means that the "drift" in is both directions. But the axis (motor) is
> always making more space than expected , isn't it ?

Yes, "drift" can be in any direction, but the change of direction
happens only on system restart.

Viesturs

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Viesturs Lācis]

2014-09-04 18:29 GMT+03:00 alex chiosso :
> A question for you ... the router is brand new (I mean that you have
> finished the setup) or is a "already working" machine ?

I built that machine 3 years ago.

Viesturs

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[alex chiosso]

As you sad :
"Yes, as I said, motor is "drifting" in particular direction, which may
change (not always), if system is restarted, but the direction of
drift does not change, once the system is running - if it gradually
goes up, then it does so; if down, then it keeps getting deeper."

Means that the "drift" in is both directions. But the axis (motor) is
always making more space than expected , isn't it ?

Alex


On Thu, Sep 4, 2014 at 5:29 PM, alex chiosso  wrote:

> As Gene is suggesting , here
> 
> is a discussion on the forum regarding problems with "steps" and Gecko
> Drive .
> A question for you ... the router is brand new (I mean that you have
> finished the setup) or is a "already working" machine ?
> This question is to understand if the machine was working fine before and
> later start with the problem you are mentioning.
>
> Alex
>
>
> On Thu, Sep 4, 2014 at 5:21 PM, Gene Heskett  wrote:
>
>> On Thursday 04 September 2014 11:07:07 Viesturs Lācis did opine
>> And Gene did reply:
>> > Hello!
>> >
>> > I have a strange situation:
>> > Stepper motor of Z axis on a cnc router is losing it's position -
>> > doing a series of numerous holes in material will end up with various
>> > depths of holes.
>> >
>> > The strange part is:
>> > 1) for each time, when machine is running, it will lose position only
>> > in one direction; if holes are getting deeper, then they only get
>> > deeper, not the other way;
>> > 2) the direction of "drift" will change randomly as machine is
>> > restarted and only then; simply running another g-code file will not
>> > change the direction of "drift";
>> >
>> > There is Nema23 stepper motor and Gecko stepper driver.
>> > Motor cable is checked for loose contact, step/dir signal lines are
>> > checked for loose contact, ballscrew and linear rails are checked for
>> > any damage to cause extra resistance in particular direction, the
>> > clutch between motor and ballscrew is checked not to slip.
>> >
>> > I am totally out of ideas, where to look. Has anyone ever experienced
>> > anything like that? What puzzles me most is the consistency to
>> > maintain direction of "drift" and yet randomly change, when whole
>> > system is restarted, so any suggestions will be appreciated.
>> >
>> > Viesturs
>>
>> Can we assume you have lengthened the dirsetup and dirhold settings?
>>
>> I don't have any gecko's, but this sure seems to be an ongoing problem
>> with them.
>>
>> From previous messages, it seems a trip to gecko for repairs fixes it, but
>> a random direction fixed at machine boot?  Thats weird, and I've no clue.
>>
>> 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 
>> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
>>
>>
>> --
>> Slashdot TV.
>> Video for Nerds.  Stuff that matters.
>> http://tv.slashdot.org/
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>
>
--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[alex chiosso]

As Gene is suggesting , here

is a discussion on the forum regarding problems with "steps" and Gecko
Drive .
A question for you ... the router is brand new (I mean that you have
finished the setup) or is a "already working" machine ?
This question is to understand if the machine was working fine before and
later start with the problem you are mentioning.

Alex


On Thu, Sep 4, 2014 at 5:21 PM, Gene Heskett  wrote:

> On Thursday 04 September 2014 11:07:07 Viesturs Lācis did opine
> And Gene did reply:
> > Hello!
> >
> > I have a strange situation:
> > Stepper motor of Z axis on a cnc router is losing it's position -
> > doing a series of numerous holes in material will end up with various
> > depths of holes.
> >
> > The strange part is:
> > 1) for each time, when machine is running, it will lose position only
> > in one direction; if holes are getting deeper, then they only get
> > deeper, not the other way;
> > 2) the direction of "drift" will change randomly as machine is
> > restarted and only then; simply running another g-code file will not
> > change the direction of "drift";
> >
> > There is Nema23 stepper motor and Gecko stepper driver.
> > Motor cable is checked for loose contact, step/dir signal lines are
> > checked for loose contact, ballscrew and linear rails are checked for
> > any damage to cause extra resistance in particular direction, the
> > clutch between motor and ballscrew is checked not to slip.
> >
> > I am totally out of ideas, where to look. Has anyone ever experienced
> > anything like that? What puzzles me most is the consistency to
> > maintain direction of "drift" and yet randomly change, when whole
> > system is restarted, so any suggestions will be appreciated.
> >
> > Viesturs
>
> Can we assume you have lengthened the dirsetup and dirhold settings?
>
> I don't have any gecko's, but this sure seems to be an ongoing problem
> with them.
>
> From previous messages, it seems a trip to gecko for repairs fixes it, but
> a random direction fixed at machine boot?  Thats weird, and I've no clue.
>
> 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 
> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
>
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Viesturs Lācis]

2014-09-04 18:16 GMT+03:00 alex chiosso :
> If the holes are deeper doesn't it means that the motor is making "more"
> steps and not "less" ?
> Is it possible that you have extra steps due to noises or bad connections
> PC ---> BoB ---> Gecko Drive ?

Yes, as I said, motor is "drifting" in particular direction, which may
change (not always), if system is restarted, but the direction of
drift does not change, once the system is running - if it gradually
goes up, then it does so; if down, then it keeps getting deeper.

There is no BoB, step/dir signals go directly from 5i23 to Gecko.

Viesturs

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[Gene Heskett]

On Thursday 04 September 2014 11:07:07 Viesturs Lācis did opine
And Gene did reply:
> Hello!
> 
> I have a strange situation:
> Stepper motor of Z axis on a cnc router is losing it's position -
> doing a series of numerous holes in material will end up with various
> depths of holes.
> 
> The strange part is:
> 1) for each time, when machine is running, it will lose position only
> in one direction; if holes are getting deeper, then they only get
> deeper, not the other way;
> 2) the direction of "drift" will change randomly as machine is
> restarted and only then; simply running another g-code file will not
> change the direction of "drift";
> 
> There is Nema23 stepper motor and Gecko stepper driver.
> Motor cable is checked for loose contact, step/dir signal lines are
> checked for loose contact, ballscrew and linear rails are checked for
> any damage to cause extra resistance in particular direction, the
> clutch between motor and ballscrew is checked not to slip.
> 
> I am totally out of ideas, where to look. Has anyone ever experienced
> anything like that? What puzzles me most is the consistency to
> maintain direction of "drift" and yet randomly change, when whole
> system is restarted, so any suggestions will be appreciated.
> 
> Viesturs

Can we assume you have lengthened the dirsetup and dirhold settings?

I don't have any gecko's, but this sure seems to be an ongoing problem 
with them.

From previous messages, it seems a trip to gecko for repairs fixes it, but 
a random direction fixed at machine boot?  Thats weird, and I've no clue.

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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Steppers losing position

[alex chiosso]

If the holes are deeper doesn't it means that the motor is making "more"
steps and not "less" ?
Is it possible that you have extra steps due to noises or bad connections
PC ---> BoB ---> Gecko Drive ?

Alex


On Thu, Sep 4, 2014 at 5:07 PM, Viesturs Lācis 
wrote:

> Hello!
>
> I have a strange situation:
> Stepper motor of Z axis on a cnc router is losing it's position -
> doing a series of numerous holes in material will end up with various
> depths of holes.
>
> The strange part is:
> 1) for each time, when machine is running, it will lose position only
> in one direction; if holes are getting deeper, then they only get
> deeper, not the other way;
> 2) the direction of "drift" will change randomly as machine is
> restarted and only then; simply running another g-code file will not
> change the direction of "drift";
>
> There is Nema23 stepper motor and Gecko stepper driver.
> Motor cable is checked for loose contact, step/dir signal lines are
> checked for loose contact, ballscrew and linear rails are checked for
> any damage to cause extra resistance in particular direction, the
> clutch between motor and ballscrew is checked not to slip.
>
> I am totally out of ideas, where to look. Has anyone ever experienced
> anything like that? What puzzles me most is the consistency to
> maintain direction of "drift" and yet randomly change, when whole
> system is restarted, so any suggestions will be appreciated.
>
> Viesturs
>
>
> --
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[Gene Heskett]

On Thursday 04 September 2014 10:57:49 John Thornton did opine
And Gene did reply:
> Gene,
> 
> Try Arc Buddy 1.5, I've incorporated the ability to use in Gedit and
> instructions on how to do so.
> 
> JT

URL?
 
> On 9/4/2014 7:58 AM, Gene Heskett wrote:
> > Greetings;
> > 
> > Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in
> > that menu.
> > 
> > gedit is version 2.30.3, but the gedit tools is 2.30.0 according to
> > synaptic.  Is there a repo I need to enable in order to get the
> > uptodate tools package that does contain the 'new' option?
> > 
> > Running it from the axis interface, the data it generates is not
> > saved anyplace, outputting this error in the cli I started lcnc
> > from:
> > 
> > File ended with no percent sign
> > can't open /tmp/tmp55fjrP/arcbuddy-sje.py
> > 
> > Since it does not unfocus even when you click someplace else, the
> > only way I can use it is to generate the code, write it down, and
> > hand copy it to the mdi interface for testing after arcbuddy-sje.py
> > has been quit so I can change focus and type.  That is doing it the
> > hard way. When I found the instructions about how to use it with
> > gedit, which would also be handy, only to find my 10.04.4 LTS
> > version of gedit cannot be configured according to those
> > instructions.
> > 
> > Thanks.
> > 
> > Cheers, Gene Heskett
> 
> ---
> --- Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> 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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[Gene Heskett]

On Thursday 04 September 2014 10:10:37 John Thornton did opine
And Gene did reply:
> Gene,
> 
> I haven't looked at Arc Buddy in a long time but that feature looks
> like it can make it actually useful. I just made it work on 10.04 but
> the instructions are terse at best. It needs a little refinement to be
> actually useful IMOH. I think the ability to select what you want to
> send would be an important addition. For example if your just starting
> your G code you may want to have it send the G0 line that moves you to
> the start point of the arc however if your already at the start point
> you might only need the arc itself.
> 
> I'll see if I can upgrade Arc Buddy to include this feature.

I just did use a copy/paste (highlite its output string & MMB paste) from 
two different workspaces to build a piece of code I then embellished to 
make a subroutine out of it, adding an adjustable X variable to the 
arcbuddy-sje.py output, and an air carve looks ok.  I need to make 4 such 
pieces of wood to hold a pair of huge 15k uf, 60 volt Sangamo capacitors 
in a power supply I'm building for the lathe motor as I don't have those 3 
legged clamping cap mounts, so I'll lay them in 2 pieces of carved wood, 
and cap them with 2 more pieces.  All this because I don't trust the 140 
volt surge ratings of the smaller 100 volt caps I excavated from a failed 
kilowatt per channel Phase Linear amplifier thats about 30 yo.  With this 
transformer, those 100 volt caps have a no signal voltage of 113 volts.

These Sangamo's are also about that old, but had power on them for about a 
decade more in their past history. I'll put 7 watt night light lamps 
across them for charge equality and power down bleeders.  But the first 
thing they'll get when powered up again, is a watch with an IR thermometer 
for a temp rise of more than a degree above ambient, in which case I'll 
have to go borrow a power stat and creep back up on the forming voltage.

All this before they get connected to one of the Pico Systems pwm servo 
amps.

Any help on actually making it a sub-tool of gedit, when my version of 
gedit lacks the external_tools/new capability would be nice.

And finally, thank you John!

> JT
> 
> On 9/4/2014 7:58 AM, Gene Heskett wrote:
> > Greetings;
> > 
> > Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in
> > that menu.
> > 
> > gedit is version 2.30.3, but the gedit tools is 2.30.0 according to
> > synaptic.  Is there a repo I need to enable in order to get the
> > uptodate tools package that does contain the 'new' option?
> > 
> > Running it from the axis interface, the data it generates is not
> > saved anyplace, outputting this error in the cli I started lcnc
> > from:
> > 
> > File ended with no percent sign
> > can't open /tmp/tmp55fjrP/arcbuddy-sje.py
> > 
> > Since it does not unfocus even when you click someplace else, the
> > only way I can use it is to generate the code, write it down, and
> > hand copy it to the mdi interface for testing after arcbuddy-sje.py
> > has been quit so I can change focus and type.  That is doing it the
> > hard way. When I found the instructions about how to use it with
> > gedit, which would also be handy, only to find my 10.04.4 LTS
> > version of gedit cannot be configured according to those
> > instructions.
> > 
> > Thanks.
> > 
> > Cheers, Gene Heskett
> 
> ---
> --- Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> ___
> 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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Setting of HAL pin from gcode

[andy pugh]

On 4 September 2014 15:52, Marius Liebenberg  wrote:

> motion.analog-out-00 => mycomp-float-in.

Yes, exactly that.

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

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Steppers losing position

[Viesturs Lācis]

Hello!

I have a strange situation:
Stepper motor of Z axis on a cnc router is losing it's position -
doing a series of numerous holes in material will end up with various
depths of holes.

The strange part is:
1) for each time, when machine is running, it will lose position only
in one direction; if holes are getting deeper, then they only get
deeper, not the other way;
2) the direction of "drift" will change randomly as machine is
restarted and only then; simply running another g-code file will not
change the direction of "drift";

There is Nema23 stepper motor and Gecko stepper driver.
Motor cable is checked for loose contact, step/dir signal lines are
checked for loose contact, ballscrew and linear rails are checked for
any damage to cause extra resistance in particular direction, the
clutch between motor and ballscrew is checked not to slip.

I am totally out of ideas, where to look. Has anyone ever experienced
anything like that? What puzzles me most is the consistency to
maintain direction of "drift" and yet randomly change, when whole
system is restarted, so any suggestions will be appreciated.

Viesturs

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[John Thornton]

Gene,

Try Arc Buddy 1.5, I've incorporated the ability to use in Gedit and 
instructions on how to do so.

JT

On 9/4/2014 7:58 AM, Gene Heskett wrote:
> Greetings;
>
> Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in that
> menu.
>
> gedit is version 2.30.3, but the gedit tools is 2.30.0 according to
> synaptic.  Is there a repo I need to enable in order to get the uptodate
> tools package that does contain the 'new' option?
>
> Running it from the axis interface, the data it generates is not saved
> anyplace, outputting this error in the cli I started lcnc from:
>
> File ended with no percent sign
> can't open /tmp/tmp55fjrP/arcbuddy-sje.py
>
> Since it does not unfocus even when you click someplace else, the only way
> I can use it is to generate the code, write it down, and hand copy it to
> the mdi interface for testing after arcbuddy-sje.py has been quit so I can
> change focus and type.  That is doing it the hard way. When I found the
> instructions about how to use it with gedit, which would also be handy,
> only to find my 10.04.4 LTS version of gedit cannot be configured
> according to those instructions.
>
> Thanks.
>
> Cheers, Gene Heskett


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Setting of HAL pin from gcode

[Marius Liebenberg]

Thanks Andy
Do I understand it correctly. I have to send the value from gcode to an 
analog pin that is connected to my float input in HAL to get the actual 
value. Scaled obviously.

I.E

motion.analog-out-00 => mycomp-float-in.



On 2014-09-04 16:32, andy pugh wrote:
> On 4 September 2014 15:17, Marius Liebenberg  wrote:
>
>> I understand that one can get a value from HAL into a gcode sub routine.
>> I am trying to set a HAL pin from my gcode but it seems not to want to
>> happen.
> The #<_hal[hal.pin.name]> format is read-only, so the easy option is
> unfortunately not available.
>
> To set a HAL pin you need to use the G-code "analog output" or
> "digital output". See M62 etc:
> http://www.linuxcnc.org/docs/html/gcode/m-code.html#sec:M62-M65
>

-- 

Regards /Groete

Marius D. Liebenberg
+27 82 698 3251
+27 12 743 6064
QQ 1767394877


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Setting of HAL pin from gcode

[andy pugh]

On 4 September 2014 15:17, Marius Liebenberg  wrote:

> I understand that one can get a value from HAL into a gcode sub routine.
> I am trying to set a HAL pin from my gcode but it seems not to want to
> happen.

The #<_hal[hal.pin.name]> format is read-only, so the easy option is
unfortunately not available.

To set a HAL pin you need to use the G-code "analog output" or
"digital output". See M62 etc:
http://www.linuxcnc.org/docs/html/gcode/m-code.html#sec:M62-M65

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

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Setting of HAL pin from gcode

[Marius Liebenberg]

Hi All
I understand that one can get a value from HAL into a gcode sub routine. 
I am trying to set a HAL pin from my gcode but it seems not to want to 
happen.
Firstly is it possible to do and secondly, how do I debug to see what is 
happening?

-- 

Regards /Groete

Marius D. Liebenberg
+27 82 698 3251
+27 12 743 6064
QQ 1767394877


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] gedit tools->external tools->new missing

[John Thornton]

Gene,

I haven't looked at Arc Buddy in a long time but that feature looks like 
it can make it actually useful. I just made it work on 10.04 but the 
instructions are terse at best. It needs a little refinement to be 
actually useful IMOH. I think the ability to select what you want to 
send would be an important addition. For example if your just starting 
your G code you may want to have it send the G0 line that moves you to 
the start point of the arc however if your already at the start point 
you might only need the arc itself.

I'll see if I can upgrade Arc Buddy to include this feature.

JT

On 9/4/2014 7:58 AM, Gene Heskett wrote:
> Greetings;
>
> Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in that
> menu.
>
> gedit is version 2.30.3, but the gedit tools is 2.30.0 according to
> synaptic.  Is there a repo I need to enable in order to get the uptodate
> tools package that does contain the 'new' option?
>
> Running it from the axis interface, the data it generates is not saved
> anyplace, outputting this error in the cli I started lcnc from:
>
> File ended with no percent sign
> can't open /tmp/tmp55fjrP/arcbuddy-sje.py
>
> Since it does not unfocus even when you click someplace else, the only way
> I can use it is to generate the code, write it down, and hand copy it to
> the mdi interface for testing after arcbuddy-sje.py has been quit so I can
> change focus and type.  That is doing it the hard way. When I found the
> instructions about how to use it with gedit, which would also be handy,
> only to find my 10.04.4 LTS version of gedit cannot be configured
> according to those instructions.
>
> Thanks.
>
> Cheers, Gene Heskett


--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Mixing BLDC and brushed motors

[andy pugh]

On 4 September 2014 13:50, John Prentice (FS)
 wrote:

> The BLDC motors suit a 7i39 very well (interfaced by a 5i20). Are there any
> problems in using just two "phases" of another 7i39 to drive the brushed
> motor? I presume that one would drive, say A, with the desired level
> (positive or negative) holding  inputs B and C at zero - i.e. no bldc
> component is needed.

Yes, that shouldn't be a problem.

> The ideal card for the brushed motor would be a 7i40. Would it be
> practicable to have Hostmot2 firmware for both 7i39 and 7i40 co-exist in the
> 5i20? If so does a BIT file exist with this combination.

I don't think that the BIT file will exist, but the only reason it
couldn't be made to exist would be if the firmware won't fit on the
5i20 FPGA.

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

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] gedit tools->external tools->new missing

[Gene Heskett]

Greetings;

Reading the preamble to arcbuddy-sje.py, I have no 'new' entry in that 
menu.

gedit is version 2.30.3, but the gedit tools is 2.30.0 according to 
synaptic.  Is there a repo I need to enable in order to get the uptodate 
tools package that does contain the 'new' option?

Running it from the axis interface, the data it generates is not saved 
anyplace, outputting this error in the cli I started lcnc from:

File ended with no percent sign
can't open /tmp/tmp55fjrP/arcbuddy-sje.py

Since it does not unfocus even when you click someplace else, the only way 
I can use it is to generate the code, write it down, and hand copy it to 
the mdi interface for testing after arcbuddy-sje.py has been quit so I can 
change focus and type.  That is doing it the hard way. When I found the 
instructions about how to use it with gedit, which would also be handy, 
only to find my 10.04.4 LTS version of gedit cannot be configured 
according to those instructions.

Thanks.

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 
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Mixing BLDC and brushed motors

[John Prentice (FS)]

Greetings

 

This is probably a question for PCW but others might know or have dome it.

 

I am looking at a small machine design where two axes will be BLDC motors
and one (or perhaps two) brushed DC.

 

The BLDC motors suit a 7i39 very well (interfaced by a 5i20). Are there any
problems in using just two "phases" of another 7i39 to drive the brushed
motor? I presume that one would drive, say A, with the desired level
(positive or negative) holding  inputs B and C at zero - i.e. no bldc
component is needed.

 

The ideal card for the brushed motor would be a 7i40. Would it be
practicable to have Hostmot2 firmware for both 7i39 and 7i40 co-exist in the
5i20? If so does a BIT file exist with this combination.

 

TIA

 

John Prentice

 

 

--
Slashdot TV.  
Video for Nerds.  Stuff that matters.
http://tv.slashdot.org/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users