开云体育

On Thu, 21 Aug 2003 22:23:06 -0300, you wrote:

Try a dry run of that for me with "Ignore Toolchange" turned on in the
logic settings. Its the only script your using and it may be causing the
parse error. I'll try to repeat it here.

Parse errors fixed by turning off tool change - problem only appeared
with Rel 10. Strangely - it doesn't happen with that collet file
though?

It doesn't happen in mill either - just made a test file with 5 tool
changes - no problems ?

Just looking at the tool change macro should safe Z be better suited
as safe X?

if SafeZ > z then StraightTraverse x, y, SafeZ , a, b, c

Something I noticed yesterday, you need to be careful in lathe with G0
moves, My old controller did them in order ie a line of

G0 X2 Z5 would do the X move first then the Z and
G0 Z5 X2 would do Z first

G1's etc went in a straight line.

Mach2's compound moves can easily crash the tool into the stock,
particularly returning to tool change position on jobs with smaller
diameters nearer chuck. I've had to add extra lines to try and
circumvent this. That's the main reason it's important that display
matches actual part & tool moves.

--
Steve Blackmore

Steve:

Try a dry run of that for me with "Ignore Toolchange" turned on in the
logic settings. Its the only script your using and it may be causing the
parse error. I'll try to repeat it here.

Thanks,
Art
www.artofcnc.ca

Isak:

Sorry, this one is not possible. The file is Mapped so that any size can be
read. This means a 10Gig file should load fine. The downside is the file is
always open.

Art
www.artofcnc.ca

Hi, Art.

I'm creating a G code on my Cad computer, then upload it to the Mach2
computer.
This way, avery time I update my program I have to close the file (in the
mach), upload it and reopen the same file again.

Is it possible to update the program so that the G file won't stay opened
during the interpertation.
I would suggest the each time you open a file, the whole file will be read
into memory and then closed.
A new program reload button can be added.

Thanks, Isak.

On Thu, 21 Aug 2003 18:58:56 -0300, you wrote:

Good GOD!!!!, This can't be the Steve we all know. Did that letter have a
note of happiness in it?
Seriously, Thanks Steve. Its good to know things are getting there. Keep up
the great work. I will try to find the other annoyances, your input has been
some of the most valuable on the list in terms of finding some of the Turn
bugs. Between you and Jerry the Turn program is getting better fast. It may
even be a beta some day.

I am happy. BUT

Sorry to burst your bubble - just tried a thread program, got several
parser errors, failed to load and also "Error 0x8000fff
(IActiveScriptParse::InitNew (()" followed by an OCX crash - it's
repeatable. Worse thing is Mach2 process will not close, so unless you
do a hard reset, machine blue screens :(

here's the file

stock size 15.5 dia 30mm long
(got a few offcuts of nylon - doesn't damage tools ;)

N10 G21
N20 G90 F150.0 S700 M5
N30 G0 X15.0 Z40.0
N50 M6 T0
N60 G43 H0
N70 G0 X7.5 Z32.5 M4 M8
N90 G1 X7.5 Z15.2063 F150.0
N100 G1 X10.7937 Z15.2063 F150.0
N110 G0 X15.0 Z40.0 M5 M9
N130 M6 T5
N140 G43 H5
N150 G0 X9.7303 Z36.3858 S300 M4 M8
N170 G0 X7.1907 Z34.978
N180 G32 X7.1907 Z19.6977 F1.0
N200 G0 X9.7303 Z19.6977
N210 G0 X9.7303 Z36.3858
N220 G0 X7.1511 Z34.9561
N230 G32 X7.1511 Z19.6757 F1.0
N250 G0 X9.7303 Z19.6757
N260 G0 X9.7303 Z36.3858
N270 G0 X7.1115 Z34.9341
N280 G32 X7.1115 Z19.6537 F1.0
N300 G0 X9.7303 Z19.6537
N310 G0 X9.7303 Z36.3858
N320 G0 X7.0718 Z34.9122
N330 G32 X7.0718 Z19.6318 F1.0
N350 G0 X9.7303 Z19.6318
N360 G0 X9.7303 Z36.3858
N370 G0 X7.0322 Z34.8902
N380 G32 X7.0322 Z19.6098 F1.0
N400 G0 X9.7303 Z19.6098
N410 G0 X9.7303 Z36.3858
N420 G0 X6.9926 Z34.8682
N430 G32 X6.9926 Z19.5878 F1.0
N450 G0 X9.7303 Z19.5878
N460 G0 X9.7303 Z36.3858
N470 G0 X6.953 Z34.8463
N480 G32 X6.953 Z19.5659 F1.0
N500 G0 X9.7303 Z19.5659
N510 G0 X9.7303 Z36.3858
N520 G0 X6.9134 Z34.8243
N530 G32 X6.9134 Z19.5439 F1.0
N550 G0 X9.7303 Z19.5439
N560 G0 X9.7303 Z36.3858
N570 G0 X6.8737 Z34.8023
N580 G32 X6.8737 Z19.522 F1.0
N600 G0 X9.7303 Z19.522
N610 G0 X9.7303 Z36.3858
N620 G0 X6.8341 Z34.7804
N630 G32 X6.8341 Z19.5 F1.0
N650 G0 X9.7303 Z19.5
N660 G0 X9.7303 Z36.3858
N670 G0 X6.8341 Z34.7804
N680 G32 X6.8341 Z19.5 F1.0
N700 G0 X10.25 Z19.5
N710 G0 X15.0 Z40.0 M5 M9
N730 M6 T0
N740 G43 H0
N750 G0 X7.5 Z30
N760 M30

--
Steve Blackmore

I think it would be fine to just place it here in the files section
for archival purposes. BTW I've been reading throught the Mach2
manual and Its very cool. I like the examples and
layman sensibility in the style. Very helpful to beginners. Too many
documents assume prior knowledge. Great job! and I will never accuse
someone of being "Murky" again :-)
My 3 cents worth before taxes.
-Lee



--- In mach1mach2cnc@..., "John Prentice"
<JAPrentice@i...> wrote:

Appendix 3 of the manual is a long list of codes for screen

controls. It was intended to help people designing their own screens
(rather than modifying existing ones). I have never referred to it.
It is a pain to maintain as the screens are updated and will
inevitably be wrong for all OEM systems. The information in it can be
obtained at anytime by running my new utility Mach2ScreenTweak.

Would anyone be upset if I saved 6 pages by deleting the Appendix

(and replacing it by a definition of the format of .SET files and
reference to ScreenTweak perhaps)?

"Speak now or for ever hold your peace"!

John Prentice

Isak:

Make sure the config/State tab is set correctly to ABS or REL in its IJ
settings. This is the most frequent cause of this error. The REAL connection
tolerance of IJ arcs is opened very wide at the moment to assist me in
knowing why a user gets this error.

Thanks,
Art
www.artofcnc.ca

Good GOD!!!!, This can't be the Steve we all know. Did that letter have a
note of happiness in it? Did I read actual success? Quick, someone get some
of what Steve's smoking and send directly to me for testing! I promise I'll
be responsible with it. Now I KNOW I'm getting somewhere....

Seriously, Thanks Steve. Its good to know things are getting there. Keep up
the great work. I will try to find the other annoyances, your input has been
some of the most valuable on the list in terms of finding some of the Turn
bugs. Between you and Jerry the Turn program is getting better fast. It may
even be a beta some day.

Thanks,
Art
www.artofcnc.ca

Hi,
When I use the G3 command with I and J paramethers I get an error message
which says that the radius of the starting point and the ending point is
different.

There are cases that it is impossible that those radiuses would be exactly
equal.
Is there a way to set the tolerance of the circuit radius ?

Thanks, Isak.

I gave a complex answer to a simple question. The simple answer is
that you can't test MACH1/2 in closed loop because it doesn't do
closed loop.

OK, I asked and I got the answer.
It's not a real problem, however. I already decided to go with servos.
I just wondered if anything good could be done with steppers and
encoders together.

But I've another question now, when you say MACH isn't a closed loop
system, what do you mean ?
Will this affect Mach performances if using G320/340 or Rutex
drivers ?

And which driver does perform better ? G320, G340 or even Rutex ?

Thanks for your reply

Paolo

One quick line on Spindle control. There are also VFD controls out there
that work by serial command. This now works in Mach2, it is possible to
create a macro to send a serial command such as "Speed:3600" to a serial
connected VFD or PLC.

More on this as the macro facility gets worked on. It is also possible to
send out binary commands to PLC controllers and such through a second
printer port.

Thanks,
Art
www.artofcnc.ca

John:

I am looking into a solution for this. It is time for smaller screens to be
available. This will get top priority very soon. (I'm just finishing up a
few things first..)

Thanks,
Art
www.artofcnc.ca

I agree, kill it. Mach2ScreenTweak will be released (with Johns permission)
into the release setup file. So it will appear on all systems. It is a great
piece of work and I'm very happy to include it with Mach2.

Thanks John,
Art
www.artofcnc.ca

Seems like a valid trade.

db

--- John Prentice <JAPrentice@...> wrote:

Appendix 3 of the manual is a long list of codes for
screen controls. It was intended to help people
designing their own screens (rather than modifying
existing ones). I have never referred to it. It is a
pain to maintain as the screens are updated and will
inevitably be wrong for all OEM systems. The
information in it can be obtained at anytime by
running my new utility Mach2ScreenTweak.

Would anyone be upset if I saved 6 pages by deleting
the Appendix (and replacing it by a definition of
the format of .SET files and reference to
ScreenTweak perhaps)?

"Speak now or for ever hold your peace"!

John Prentice

[Non-text portions of this message have been
removed]

__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software

"deruida" <deruida@...> wrote:

A good friend and fellow engineer and I want

to produce a medium sized cast iron 3 axis milling machine that will
be rigid enough for steel and fill the gap between the knee mills
and the cnc machining centers.

[That's an ambitious project! Are you doing the casting here in the US, or
offshore? What will it do that a knee mill won't? What will it lack that a
VMC has? I didn't know there was much of a gap there.]

We want to produce it with a machine

controlled variable speed spindle, but need a little advise on the
pros and cons of different methods. THANK YOU IN ADVANCE FOR
HELPING OUT A RELATIVE NEWBE. Dan DeRuiter.

[There's some fairly comprehensible info on VFD speed control here:
]

Andrew Werby
www.computersculpture.com

Just to let everyone know, I sucessfully made some parts today using
Mach2 lathe :)

Thanks Art <G>

Tool offsets, including those with negative Z values and positive X
seem to work OK.

The display is screwed, it's showing part and tool offsets giving a
misleading apperance and toolpaths, but parts are actually turned OK.

The code contained three tool changes, a reverse taper and three
external IK arcs and a parting off operation.

Next task is to try some threads, I have an outstanding job with both
external and internal fine metric threads (camera lens adaptors).
Tools for this job have negative value X offsets so will test
threading and negative X axis offsets.

I'll try a test job first though with a much smaller OD and see how it
goes.

Nearly forgot - Another old outstanding problem still exists in that
every time job is run Mach2 adds a space to the end of the file, not a
problem until you start to run hundreds of parts and wonder why the
file got bigger <G>.

Thanks again Art - it's getting there!

--
Steve Blackmore

--- In mach1mach2cnc@..., "pavel59" <pavel@k...> wrote:

I'm sorry to repost, but maybe my previous message was not clearly
understandable.

My question is pretty simple:
I need to test MACH2 on closed loop.

I would like to try using encoders+steppers+MACH2

MACH1/2 is not a closed loop software program. Few of the available
hobby grade programs are. (EMC running with LINUX and some external
boards is the exception). In order to close the loop there are some
complex timing and loop gain issues that require either dedicated
hardware or processors.

The way that most of the users are doing closed loop servo control is
to use one of the Step to Servo motor control modules (GECKO 340;
Rutex 990H). The loop is closed and controlled in the motor control
module by taking a pulse input and measuring the distance traveled
with a rotary encoder on the motor or screw shaft. The hardware will
keep moving the servo motors until the error between the input steps
and the actual position is zero or it faults if the error becomes too
great.

Much like the feedback in audio systems, too much gain or too much
phase shift in the system will cause rapid oscillations. The servo
encoders need to be on the motor or screw shaft because there is no
backlash between the driving device (motor) and the encoder.

Here is what happens if there is any more than a little "looseness"
in the system: The control unit measures the position of the encoder
and compares that to the expected position. Because the encoders are
distinct counts the error circuit will never be completely stable and
will "hunt" for a zero point by jumping up or down an encoder count.
In a tight system this presents itself as a gentle "hum" that causes
no measurable movement. In a system with the encoder on the table
having backlash, the hunt becomes more violent and causes the table
to start moving rapidly back and forth the distance of the backlash.
This will cause overshoot because of inertia in the table and
increase the error and oscillations. There have been reports of
machines shaking so violently they walk across the floor!

The only cure is to build a system with zero backlash using zero
backlash ballscrews or to measure the position of the motor shaft and
ignore the backlash (or compensate in the software for it.)

I gave a complex answer to a simple question. The simple answer is
that you can't test MACH1/2 in closed loop because it doesn't do
closed loop.

Thanks John for the information. I don't claim to be an expert in
the CNC world, but it is just another automated machine like the
ones that I have been designing for the past 13 years. I am
partnering up with two fellow engineers who are just as eager (or
foolish) as I to create the milling machine I talked about in my
previous post. We have chosen to use Mach2 as our software of
choice due to Art's incredible dedication to his software. I have
been using Art's program since sometime back in the Mach1 beta era
to run my 3 axis cnc router that I built for the sole purpose to
learn the world of CNC. My partners as well as I have extensive
experience in the CATIA CAD world, and one of them has many years
using CAM software to program the machining centers he used to work
with. The other partner has the buisness sense side of things and I
have the background in the custom automation field. I do have a
good friend that is a master electrician to help out in the controls
side of things. Through my career I have made several friends in
the custom machining industry that we will be using to produce the
first few machines. We will be using servo motors most likely
paired up with the Gecko 320 drives. We tossed around the
possibility of using the Rutex brand of drive because of their nice
motherboard back plane approach with the options of the relay out
cards. Both Gecko and Rutex drives are good products from the posts
that I have read, and Mariss, like Art has incredible integrity when
it comes to his product. Part of the reason why I quit my previous
job was because the owner was only interested in making big money
and could care less about his customers! I have a problem with that
logic, because it is the customers that make it possible to do what
we love in our jobs. That's why we will be using Mach2 combined
with the Gecko drives. I thank you all for your support, and I am
sure that before we are finished I will be having many more
questions that need answers.

Good mrning John,

I re-sized the screens to 640 x 480 and the font's remained the same size in
all the Button's as it was in the original screen sizes. This makes for
some very interesting button labels since all the fonts that are not in a
button did in fact resize to an appropriate font size for the screen size.
This may not make sense, but I am sure Art will see the problem by just
taking the standard 1024.set and using ScreenTweak to scale it to 640x480.
Every thing is perfectly positioned and all text boxes contain the proper
fonts but not the buttons.

John Guenther
'Ye Olde Pen Maker'
Sterling, Virginia

Appendix 3 of the manual is a long list of codes for screen controls. It was intended to help people designing their own screens (rather than modifying existing ones). I have never referred to it. It is a pain to maintain as the screens are updated and will inevitably be wrong for all OEM systems. The information in it can be obtained at anytime by running my new utility Mach2ScreenTweak.

Would anyone be upset if I saved 6 pages by deleting the Appendix (and replacing it by a definition of the format of .SET files and reference to ScreenTweak perhaps)?

"Speak now or for ever hold your peace"!

John Prentice