开云体育

Hi:

At popular request..Close file, reload file buttons have been added to
main screen for next release. Close will (Duhh) close the current file.
Reload will load the last loaded fil ein a session. This is to make it
easier to resave files from CAM programs without having to navigate menu's.

Thanks,
Art
www.artofcnc.ca

Steve:

Pulse width. As long as it is longer than one interrupt period, it will be
grabbed, if its smalled than one int period, it CAN be missed, but it would
simulate your problem almost exactly.
Thanks,
Art
www.artofcnc.ca

Isak:

I have found a problem in there. This would happen only in full IJ circles.
I have fixed it for the next release.

Thanks, Sorry for the bug..

Art
www.artofcnc.ca

On Tue, 02 Sep 2003 16:48:01 -0300, you wrote:

I'm wondering, as my pulses are quite narrow, that it's sometimes
triggering on the "wrong" edge?

As long as a pulse is longer than 1 interrupt time ( 40us at 25Khz), it
shoudl grab it...

By pulse do you mean leading edge to leading edge, or merely the
actual pulse width, if the latter that may well be the problem, I'll
need to check but from memory they may be much shorter than that.

--
Steve Blackmore

Isak:

I will look deeper and get back to you.

Thanks,
Art
www.artofcnc.ca

It is in a G2 full circuit move (G2 x0 y0 i10 j0).

It seems that the CV light has no effect on the behaviour.

I changed it by G61/G64 MDI command, and confirmed that the light,
infact, goes on and off.

I've tried with CV off and with CV on.
In both cases there is a sudden stop and then sudden move.

Maybe there are some files which are not replaced during the upgrade
process.

Thanks, Isak.


--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

Isak:

Is this in a G2,G3 move? On mine , both axis slow to a stop

before

correcting, then both axis accelerate. (Unless CV is used in which

case it

works as you describe.) Is the CV light off when doing this?

Thanks,
Art
www.artofcnc.ca

Steve:

> No, I think you misunderstand - If commanded is say, 400 and actual is

420 it takes ages. If commanded is 400 and actual is 400 adder starts
instantly and doesn't do this crazy counting to many millions.

Even this should not matter. The only numbers that are important are
"Normal SPinCount" and "Current SpinCount"...

I've just done a little more testing and it may well be that there is
some false triggering, the spindle speed is absolutely constant,
measured with both a mechanical and an external hand held tacho.

The external rev counters (which I borrowed) don't always agree with
Mach2!!

The speed Mach2 senses sometimes jumps slightly, the accel/decel
lights are always cycling, even though the spindle isn't varying in
speed.

This is a good sign, the acc and dec lights mean that one cycle is
slightly longer and the next slightly smaller. When your measureing to 20us
in resoution, no speed variation would benecessary for this to happe,the
lights are gone in the newest version...
>
? Perhaps the slot is too wide, or more

likely narrow?

I'm wondering, as my pulses are quite narrow, that it's sometimes
triggering on the "wrong" edge?

As long as a pulse is longer than 1 interrupt time ( 40us at 25Khz), it
shoudl grab it...

The system is always counting the number of interrupts from leading edge

of

the index pulse to leading edge of the next pulse. It averages these over

a

5 rotation period at all times..

Would it matter if the pulse is inverted? (then triggering on trailing
edge).

Nope, no matter, as long as the negative portion went low for at least
20us. (I think, I will check this...Just make sure low active is set if it
goes neg and hi if not..

Can't do any more tonight, but will test spindle pulses some more
tomorrow, with much longer time period on freq counter, only did a few
one minute tests at 400, 500 and 600rpm.
--

Thanks for the tests,

Art

Steve Blackmore

On Tue, 02 Sep 2003 13:26:57 -0300, you wrote:

If actual spindle speed is not same as requested speed spin adder
keeps going for ages before first cut starts, the bigger the
discrepancy, the longer the spin adder counts for.

Strange. When the spindle speed is commanded, it simply sets a frequency
for the spindle axis and doesn't use it again.

No, I think you misunderstand - If commanded is say, 400 and actual is
420 it takes ages. If commanded is 400 and actual is 400 adder starts
instantly and doesn't do this crazy counting to many millions.

I've just done a little more testing and it may well be that there is
some false triggering, the spindle speed is absolutely constant,
measured with both a mechanical and an external hand held tacho.

The external rev counters (which I borrowed) don't always agree with
Mach2!!

The speed Mach2 senses sometimes jumps slightly, the accel/decel
lights are always cycling, even though the spindle isn't varying in
speed.

I've just put the frequency counter on the pulse output and it agrees
with the external rev counters, it has an in built clock and the
freq/time matches - there are no missing pulses I can detect.

As I said previously, the pulses are absolutely clean with nice square
edges - they are a full 5V, how does Mach2 count these? How
wide/narrow can it tolerate? Perhaps the slot is too wide, or more
likely narrow?

I'm wondering, as my pulses are quite narrow, that it's sometimes
triggering on the "wrong" edge?

The system is always counting the number of interrupts from leading edge of
the index pulse to leading edge of the next pulse. It averages these over a
5 rotation period at all times..

Would it matter if the pulse is inverted? (then triggering on trailing
edge).

Can't do any more tonight, but will test spindle pulses some more
tomorrow, with much longer time period on freq counter, only did a few
one minute tests at 400, 500 and 600rpm.
--
Steve Blackmore

--- In mach1mach2cnc@..., "P. J. Hicks"
<hickspj467@c...> wrote:

Art,

In view of the ever growing complexity I am wondering if a 'popup'

help reminder system might be added to Mach1-2 similar to the little
boxes that appear under the cursor in windows apps.
<snip>

Dear PJH and everyone:

Art will probably input to this one as well but it is an area which
obviously links with printed documentation and I have been toying
with some ideas which I am happy to share. But first it would be
interesting to hear what everyone feels is the main need. The
problem, as I see it, is that while it is easy to have a "tool-tip"
pop-up scheme that might say, for example, when you hover over
Program Run button "Click to display controls for running your part
program" this does not seem to me to be much use.

So what are the things you all find you need to know when your hands
are oily, the spindle is running and the manual is on your bedside
table?

John Prentice

Isak:

Is this in a G2,G3 move? On mine , both axis slow to a stop before
correcting, then both axis accelerate. (Unless CV is used in which case it
works as you describe.) Is the CV light off when doing this?

Thanks,
Art
www.artofcnc.ca

Hi Art,
Maybe I didn't explain myself correctly.
The problem is with the other axis (the one that dosen't need
backlash correction).

In a full circuit the axis which need the correction is naturally
deceleratase and accelerates.

The 2nd axis runs at full speed, then stops, then the 1st axis
corrects, then the 2ns axis starts at full speed.

Thanks, Isak.





--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

Isak:

Still checking, but I cannot repeat it. When I set backlash on

and command

a circle, for example, it rotates to the 90 degree points, ramps

down,

backlash corrects, ramps back up and continues....
I continue to test...
Art
www.artofcnc.ca

Steve:

> If actual spindle speed is not same as requested speed spin adder

keeps going for ages before first cut starts, the bigger the
discrepancy, the longer the spin adder counts for.

Strange. When the spindle speed is commanded, it simply sets a frequency
for the spindle axis and doesn't use it again.

Let me explain the variables in use here, it may aid in troubleshooting
this. Since your pioneering, you should read the following a couple of
times....

Normal Spin Counts....

The system is always counting the number of interrupts from leading edge of
the index pulse to leading edge of the next pulse. It averages these over a
5 rotation period at all times.. When you command a G32 move, it stores the
Normal count and displays it as Normal Spin Counts..

Now it waits until the queues up a G1 move at the requested vel/rot
feedrate, but blocks the move from happening. The buffer begins to fill with
the G1 calculated move (single or dual axis..)
Now, when the buffer has a certain fill factor, the engine variable
Threading is set to true. This informs the engine to trigger the move n
counts of interrupts from the leading edge of the index pulse. (This is your
entrance angle. Maximum entance points is the number of interrupts between
rotations scaled to a set interval. Typing an entrance angle scales a start
point from that variable. So a zero degree entrance angle means the move
will start at leading edge. This means you have to have the tool at least
one Acceleration period from the stock before calling the G32.

Now the move begins at the pre-determined velocity of the axial move
previously queued. But the number of rotations is now measured AND compared
to the stored "normal spin counts" variable. If the number of counts of the
next rotation is higher (the spindle has slowed), the adder is incremented
to a number calculated to add up and overflow the number of times the
previous rotation was too high in counts.On each overflow, the next
interrupt is rejected and no movement from the buffer occurs. This means if
the first rotation of cut is 100 interrupt counts longer than the "normal
average", then an adder of xxxx is calculated and this begins to be added to
a counter on each interrupt. During the next rotation, this counter will
overflow 100 times evenly spaced through the entire rotation. The axis motor
will miss a pulse every overflow, thus bringing it down to the correct speed
for the spindle. It will be a smooth transistion as the "missing" pulses are
simply stretched into the pulse train for an even time count. Obviuosly,
though, if the adder value gets too high, the number of "correction" pulses
will overwhelm the actual pulse stream and some roughness could occur.

That, is the design. As I've typed the explaination, I've thought of a
couple of things to check. But read this carefully, because it may give you
a clue by the numbers as to whats going on. If the adder climbs for example,
this would mean the engine "thinks" the spindle is slowing down on every
rotation from the time of "normal" storage. Look at normal vs current spin
counts. If current is ever twice the normal, even for a split second, your
very intermittantly missing pulses from the spindle. Current spin counts and
normal spin counts should never vary real widely and not much at all if the
bit has not met the stock yet. Try a cut in the air and see if the normal
and current remain close. (Close is a relative number, a differance of 500
would be 20mSec difference or 1.2RPM ) , So if this difference ever gets too
high, then we're not reading the spindle speed correctly. But set speed vs
real speed should have no impact whatsoevcr to the adders and these numbers
as they are not involved. Jerry's experience shows a greater entrance
stability so no false triggering on the leading edge, this should be all
thats required for a good repeatability on multi pass, your results have me
suspicious that the adder is going too high, which would indicate either
pulse instability of speed instability. Can you turn off program control of
the spindle speed and select one manually? This would aid in diagnosis. Dry
air tests are all thets necessary to get the adder numbers and such..

> The single pass thread was pretty good and the pitch was close, but it

looks like Z isn't reacting quickly enough, the pitch at the start is
slighly different than when it settles.

This sounds not too bad for a single pass, but the start being different
from the settled pitcfh should only be 1 rotation long. This would be
typical as the speed cannot start correcting until 1 turn after the
slowdown. I am suspecting a very intermittant missing pulse form the sound
of it...

Overall I'm pleased, it looks like maybe bit of tweaking and we're
there<G>.

Thanks for your tests Steve, I think in the end this exercise may get us
alot closer to good threads and its a good treatise on how Mach2 does it
threading in any event. (In case anyone was curious...)

Thanks,
Art

Art,

In view of the ever growing complexity I am wondering if a 'popup' help reminder system might be added to Mach1-2 similar to the little boxes that appear under the cursor in windows apps. I envision those little banner type popups that could be enabled or disabled as desired and have the ability to be written by the user/or not. I once wrote a 'C' program that popped up a help box when the focus was on a variable then F1 was hit; on the first time it allowed you to write the help banner text and on subsequent hits it pops up the box. You could also edit the text by going to a 'memo' file.
Just a suggestion from an amatuer. PJH

--- In mach1mach2cnc@..., "ozzie34231" <Ozzie@c...>
wrote:

For those wondering what the tool toucher is, it's just a piece

of

ground 10mm stock

<snip>

Steve Blackmore

I've good success using a small adjustable parallel bar between

the

work and the tool. Get within about 3/8", adjust, measure, and set
machine dro.

Jerry

Yep! me too. Mach2Mill manual Fig 7.10 shows technique in Mill
context. Corrections screen is handy for adding in the gage
thickness.

John Prentice

--- In mach1mach2cnc@..., Art <fenerty@a...> wrote:

John:

The 100 pulse rate will occur if a multi-media call is made to

Windows. If

you had explorer open, its possible a small multimedia call was

made (for an

animated icon or somethin..) this could give you 100 Pulse freq.

Just a thought.

Yes thanks. I would not have run Mach2 on a machine with other
software and testing it in the office did not think of implications.
Probably explains my previous 100 report as IE is on the computer
all the time.

Don't think it explains Repeat though. Will keep looking for
consistency and report.

John Prentice

Art - a few observations from some test threads:

If actual spindle speed is not same as requested speed spin adder
keeps going for ages before first cut starts, the bigger the
discrepancy, the longer the spin adder counts for.

I tried three different strategies:

A mutiple pass with 29 deg entrance angle and G0 moves between passes.

Multiple pass 0 deg entrance angle with G0 moves.

(I also tried above with G1 moves, but after 5 mins aborted after tool
was crawling back to Z start at 1mm per minute - it was a 1mm pitch
thread)

Single pass thread.

Both the multipasses were rough, looks like either radial start
position was not the same, or lost steps / positioning in Z - very
hard to tell which it was as the end result is the same with multiple
cuts that don't quite line up.

The single pass thread was pretty good and the pitch was close, but it
looks like Z isn't reacting quickly enough, the pitch at the start is
slighly different than when it settles.

Overall I'm pleased, it looks like maybe bit of tweaking and we're
there<G>.

--
Steve Blackmore

All right Art, way to go!!!!!!!!!

dg

Message: 6
Date: Mon, 01 Sep 2003 19:41:39 -0300
From: Art <fenerty@...>
Subject: Re: Run From Here?

Servo:

Already there. The "set next line" key preps the cycle/start button to
start from the line you were on when you pressed it. This is different

from

RunAt where the whole file is executed.

Art
www.artofcnc.ca

For those wondering what the tool toucher is, it's just a piece of
ground 10mm stock inserted into a piece of nylon and the lot held in
the chuck, fasten one multimeter lead to it, the other to tool

holder

and slowly jog in until meter beeps on continuity setting. Much

easier

than squinting down a loupe at the gap!

--
Steve Blackmore

I've good success using a small adjustable parallel bar between the
work and the tool. Get within about 3/8", adjust, measure, and set
machine dro.

Jerry

Hi:

I would like to convert my dfx file with mach2, push a button and

tell my plasma cutter to cut, is this going to be possible. Do I
have bad downloads, am I downloading the wrong trial version,are
there some settings to be made, What do I do. Also would like to
know how to set dwell time or pause rate for plasma cutter. Thought
I could use dwell time for spindle as it is set up nice but looks as
though the spindle does not turn on and off for each cut.

I don't know why the different responces on each version, this doesn't
normally happen, but to get the spindle to turn on and off for each cut,
check the "plasma mode" button on the DXF screen import. It will generate a
spindle on at the start of each cut and a spindle off at the end before
moving.

I suspect that its a matter of settings. Sometimes, reinstalling a new
version wipes out any settings you might have made. You can stop that from
occuring, by saving the Mach2Mill.xml file when your done setting up. After
installing a new version, copy that file back into the Mach2 folder and your
settings will revert back to your settings.

DXF's can be tricky across different programs, so usually its a matter of
trial and error until you get a sequence down pat. The DXF's should import
unless they are defectiove in some way.

Good luck,

Art
www.artofcnc.ca

John:

The 100 pulse rate will occur if a multi-media call is made to Windows. If
you had explorer open, its possible a small multimedia call was made (for an
animated icon or somethin..) this could give you 100 Pulse freq.

Just a thought.
Art
www.artofcnc.ca