idw sketches keep changing themselves

[Anonymous]

We're currently running IV w/ R6-SP2. Our company makes hydraulic cylinders, and we like to show the welds at the main-to-endlug and the rod-to-rodlug locations, as well as any port fittings that attach to the main. While IV was kind enough to create weld features, they can't approach our groove contours. So, we fake a lot of these weld seams in the drawing. Most every top level drawing shows these welds in a view that is both sectioned (1/2 or 3/4) and usually broken (to shorten the cylinder length to the sheet size).


I was recently taught the neat trick of tying a sketch to a view (select view, then hit 'S' or the sketch button) and until now had just been creating free-floating sketches and 'eye-balling' them to approximately the right location. My problem is that with these view-dependant sketches, we'll create a weld seam and (paint) fill it in to look like a cut weld. But if we save the file and get out, then come back and we'll find the sketches have gone askew.


Typically, if we painted a closed sketch made up from projected lines and sketched lines, the projected lines are no longer present and we've lost the painted area (our sketched lines are still present, just not the projected ones that close the area). Additionally, we'll frequently have a curve that just blows up - it may not be tied to any projected points, but it behaves as if one of the end points stayed still and the other moved out to the un-broken length.


It is very aggravating to have to re-edit these sketches everytime we re-open a drawing. Is this a bug related to sketches on sectioned/broken views, or am I missing a trick somewhere? I'll see if I can get a few pics and post them in the CF. But for now, anyone have any insights? This problem is not all the time on all drawings, and I'm starting to track the files where it keeps recurring in order to find a trend or common trait. The files that have problems, consistently have the same problem, and so far, most of the files have this problem.


ps, sorry for the long post, I'm notorious for being long winded when writing (and yes, I edited this down before posting it) 🙂

[Anonymous]

You can project geometry from the view to tie the
sketch to when you make the sketch with the project geometry tool; is that the
way you are doing it? Should stay put FAIK.

~Larry

style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">

"reeds_93me" <
href="mailto:rstrickland@parker.com">rstrickland@parker.com> wrote in
message
href="news:f140604.-1@WebX.maYIadrTaRb">news:f140604.-1@WebX.maYIadrTaRb...

We're
currently running IV w/ R6-SP2. Our company makes hydraulic cylinders, and we
like to show the welds at the main-to-endlug and the rod-to-rodlug locations,
as well as any port fittings that attach to the main. While IV was kind enough
to create weld features, they can't approach our groove contours. So, we fake
a lot of these weld seams in the drawing. Most every top level drawing shows
these welds in a view that is both sectioned (1/2 or 3/4) and usually broken
(to shorten the cylinder length to the sheet size).
I was recently taught
the neat trick of tying a sketch to a view (select view, then hit 'S' or the
sketch button) and until now had just been creating free-floating sketches and
'eye-balling' them to approximately the right location. My problem is that
with these view-dependant sketches, we'll create a weld seam and (paint) fill
it in to look like a cut weld. But if we save the file and get out, then come
back and we'll find the sketches have gone askew.
Typically, if we painted
a closed sketch made up from projected lines and sketched lines, the projected
lines are no longer present and we've lost the painted area (our sketched
lines are still present, just not the projected ones that close the area).
Additionally, we'll frequently have a curve that just blows up - it may not be
tied to any projected points, but it behaves as if one of the end points
stayed still and the other moved out to the un-broken length.
It is very
aggravating to have to re-edit these sketches everytime we re-open a drawing.
Is this a bug related to sketches on sectioned/broken views, or am I missing a
trick somewhere? I'll see if I can get a few pics and post them in the CF. But
for now, anyone have any insights? This problem is not all the time on all
drawings, and I'm starting to track the files where it keeps recurring in
order to find a trend or common trait. The files that have problems,
consistently have the same problem, and so far, most of the files have this
problem.
ps, sorry for the long post, I'm notorious for being long winded
when writing (and yes, I edited this down before posting it)
:)

[Anonymous]

Yes - sketches tied to a view are utilizing the project geometry tool, but as I've indicated these 'projected lines' are dissappearing sometimes when we re-open a file. This is what occurs when we lose our 'painted' areas.



As for the curves that blow up, they are generally just 'eye-balled' into place and do not utilize projected lines. I'll try using them to tie down these curves, as I've noticed that the 'painted' area curves I've created (not projected) usually remain in place, just not tied down after the projected lines disappear.



As an added note, I believe I am utilizing the project geometry tool correctly in the idw (use in ipt files is not a problem at all). I understand the requirement to select the projected lines and then RMB 'Done' to have them actually project. This is how I was able to create a closed area for my 'painting'.

[Anonymous]

So as I understand this properly you are projecting geometry from a model thru a view to your sketch in an IDW?
Is this correct?

[Anonymous]

I've ran into this numerous times with breakout views and haven't found a solution for it to this date. I know how you feel about updating the idw and yes it is very frustrating. I have an idw file that everytime I edit the parts list it changes all my view lables, go figure? What might possibly be an issue is if you have any bad stuff (the little red cross at the top) going on in the assembly file, if you resolve these it might help. I've also found break out views on a section view will cause hatching to disapear. I feel your pain and your not alone.

[Anonymous]

**NOTE** I have added to the CF a pdf to show what I've got. I've given detail views to help show what the idw looks like before (left) and after (right) we add our welds to the assembly. In the top right details (E & J) I've moved my 'free-floating' sketch away from the view, so you have an idea of what's being done. For the top left (C & F), I utilized a view dependant sketch to project the weld groove then added a top curve and filled in the area to show the weld. For the two bottom welds, I again utilized view dependent sketching on the lower left (B & H), and free-hand sketching on the lower right (D & G).



1) What typically occurs is the projected lines in C&F will not be projected when I re-open the file, hence I lose my painted in weld. All I have is my little bubble curve on top.



2) For E&J, not much happens, I just have to manually slide my free-hand sketch to about the right place if I happen to have moved my view. No big deal, just a slight aggravation and there is the fact that it's not totally accurate.



3) for B&H, I project the vertical lines to anchor the endpoints of my curve. That seems to be alright, but once I've finished the curve and turn off visibility for the groove, I find that the groove lines become visible each time I re-open the file. ARGH! I see it did it again before I printed the pdf. You should only see two vertical lines and the curve that bends below them - just as if you were looking at a weld seam in the real world. Though it's odd, the vertical lines up the center, where the two components meet, never gets turned back to visible when re-opening the file - it's only the curves.



4) D&G are the real fun. I've used a view dependent sketch, and didn't project anything. I just free-handed the weld seam in place and made my groove lines invisible again. This is the situation where the curves running vertically get blown up. Usually, the centerpoint and one of the endpoints get moved way off to the right of the view (as if reacting to the full length of the cylinder rather than the broken view position). And do the groove lines become visible again? Sometimes yes, but sometimes no.




And in answer to accu-cad, yes you have it correct. The prefered/recommended method we are trying to use is to start an idw, create our views, create a sketch dependent/tied-to a view, project what we can and put the weld in place.

[Anonymous]

I looked at your pdf, but I'm not sure I see the
problem. You know that the detail view is another view, and a sketch from one
view won't show in a new view, right? Might be easier if you used text/arrows on
the pdf to describe the problem. Jumping back and forth between the post and the
pdf, once I find the views/details you refer to I've
already forgotten what you said about it. My shortcoming, I know, but it's
hard for me to follow. Or maybe you could post the file and mark up the
idw.

~Larry

style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">

"reeds_93me" <
href="mailto:rstrickland@parker.com">rstrickland@parker.com> wrote in
message
href="news:f140604.4@WebX.maYIadrTaRb">news:f140604.4@WebX.maYIadrTaRb...

*NOTE** I have added to the CF a pdf to show what I've got. I've given
detail views to help show what the idw looks like before (left) and after
(right) we add our welds to the assembly. In the top right details (E & J)
I've moved my 'free-floating' sketch away from the view, so you have an idea
of what's being done. For the top left (C & F), I utilized a view
dependant sketch to project the weld groove then added a top curve and filled
in the area to show the weld. For the two bottom welds, I again utilized view
dependent sketching on the lower left (B & H), and free-hand sketching on
the lower right (D & G).

1) What typically occurs is the projected
lines in C&F will not be projected when I re-open the file, hence I lose
my painted in weld. All I have is my little bubble curve on top.

2)
For E&J, not much happens, I just have to manually slide my free-hand
sketch to about the right place if I happen to have moved my view. No big
deal, just a slight aggravation and there is the fact that it's not totally
accurate.

3) for B&H, I project the vertical lines to anchor the
endpoints of my curve. That seems to be alright, but once I've finished the
curve and turn off visibility for the groove, I find that the groove lines
become visible each time I re-open the file. ARGH! I see it did it again
before I printed the pdf. You should only see two vertical lines and the curve
that bends below them - just as if you were looking at a weld seam in the real
world. Though it's odd, the vertical lines up the center, where the two
components meet, never gets turned back to visible when re-opening the file -
it's only the curves.

4) D&G are the real fun. I've used a view
dependent sketch, and didn't project anything. I just free-handed the weld
seam in place and made my groove lines invisible again. This is the situation
where the curves running vertically get blown up. Usually, the centerpoint and
one of the endpoints get moved way off to the right of the view (as if
reacting to the full length of the cylinder rather than the broken view
position). And do the groove lines become visible again? Sometimes yes, but
sometimes no.


And in answer to accu-cad, yes you have it correct.
The prefered/recommended method we are trying to use is to start an idw,
create our views, create a sketch dependent/tied-to a view, project what we
can and put the weld in place.

[Anonymous]

See what happens with the sample file I put in
CF.

style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">

"reeds_93me" <
href="mailto:rstrickland@parker.com">rstrickland@parker.com> wrote in
message
href="news:f140604.4@WebX.maYIadrTaRb">news:f140604.4@WebX.maYIadrTaRb...

*NOTE** I have added to the CF a pdf to show what I've got. I've given
detail views to help show what the idw looks like before (left) and after
(right) we add our welds to the assembly. In the top right details (E & J)
I've moved my 'free-floating' sketch away from the view, so you have an idea
of what's being done. For the top left (C & F), I utilized a view
dependant sketch to project the weld groove then added a top curve and filled
in the area to show the weld. For the two bottom welds, I again utilized view
dependent sketching on the lower left (B & H), and free-hand sketching on
the lower right (D & G).

1) What typically occurs is the projected
lines in C&F will not be projected when I re-open the file, hence I lose
my painted in weld. All I have is my little bubble curve on top.

2)
For E&J, not much happens, I just have to manually slide my free-hand
sketch to about the right place if I happen to have moved my view. No big
deal, just a slight aggravation and there is the fact that it's not totally
accurate.

3) for B&H, I project the vertical lines to anchor the
endpoints of my curve. That seems to be alright, but once I've finished the
curve and turn off visibility for the groove, I find that the groove lines
become visible each time I re-open the file. ARGH! I see it did it again
before I printed the pdf. You should only see two vertical lines and the curve
that bends below them - just as if you were looking at a weld seam in the real
world. Though it's odd, the vertical lines up the center, where the two
components meet, never gets turned back to visible when re-opening the file -
it's only the curves.

4) D&G are the real fun. I've used a view
dependent sketch, and didn't project anything. I just free-handed the weld
seam in place and made my groove lines invisible again. This is the situation
where the curves running vertically get blown up. Usually, the centerpoint and
one of the endpoints get moved way off to the right of the view (as if
reacting to the full length of the cylinder rather than the broken view
position). And do the groove lines become visible again? Sometimes yes, but
sometimes no.


And in answer to accu-cad, yes you have it correct.
The prefered/recommended method we are trying to use is to start an idw,
create our views, create a sketch dependent/tied-to a view, project what we
can and put the weld in place.

[Anonymous]

I've added an annotated version to the CF.

[Anonymous]

Your file is fine. Though you are utilizing the weld feature, and we aren't - we are just faking it in at the idw level. We haven't learned how to get the weld feature to work properly with our J-Grooves. I'll check Dotson's and Bliss' sites to see what more they can add to our use of the weld feature. For now though, we are just sketching the welds on the idw and leaving the iam mated but unwelded.

[Anonymous]

Forgot to weld the retract port on! 🙂

I'm curious - how would you handle representing the same cylinder if the retract port was rotated at say, 2 o'clock instead of 12 o'clock as shown? In plain acad, we would draw it as shown and then call it out as 'Not True Rotation'. I've struggled with this problem mightily in MDT where you can't have a sectioned, broken view with one (or more) sections represented as aligned. I fear this is true in IV as well. True, what we do sort of violates real drafting rules, but our old cheatin' ways allow us to show both function and construction details in a compact manner. If we go to IV, I fear our drawings will have to get much larger and busier just accoplish the same thing we did with dumb, 2D sticks (and hatches that behaved themselves).

[Anonymous]

The sections aren't all that complex yet; don't
support curves, but they will properly rotate the section in some cases. For
example, a plate (flange) with three gussets at 120°. In the side view, only
lone gusset will show from the center to the OD as it should be, but section the
front view through two gussets at 120°, it will turn the section view to
show two full length gussets correctly. Other cases, are trial and error more or
less ... least AFAIK.

~Larry

style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">

"tbrack" <
href="mailto:tbrackbill@hamptonhydraulics.com">tbrackbill@hamptonhydraulics.com>
wrote in message
href="news:f140604.9@WebX.maYIadrTaRb">news:f140604.9@WebX.maYIadrTaRb...

Forgot
to weld the retract port on! 🙂

I'm curious - how would you handle representing the same cylinder if the
retract port was rotated at say, 2 o'clock instead of 12 o'clock as shown? In
plain acad, we would draw it as shown and then call it out as 'Not True
Rotation'. I've struggled with this problem mightily in MDT where you can't
have a sectioned, broken view with one (or more) sections represented as
aligned. I fear this is true in IV as well. True, what we do sort of violates
real drafting rules, but our old cheatin' ways allow us to show both function
and construction details in a compact manner. If we go to IV, I fear our
drawings will have to get much larger and busier just accoplish the same thing
we did with dumb, 2D sticks (and hatches that behaved
themselves).

[Anonymous]

Okay, I faked it; don't see the problems you are
encountering but don't know what I'm doing differently. I put an example in CF
for ya. See if that's what you're looking for?

~Larry

style="PADDING-RIGHT: 0px; PADDING-LEFT: 5px; MARGIN-LEFT: 5px; BORDER-LEFT: #000000 2px solid; MARGIN-RIGHT: 0px">

"reeds_93me" <
href="mailto:rstrickland@parker.com">rstrickland@parker.com> wrote in
message
href="news:f140604.8@WebX.maYIadrTaRb">news:f140604.8@WebX.maYIadrTaRb...

Your
file is fine. Though you are utilizing the weld feature, and we aren't - we
are just faking it in at the idw level. We haven't learned how to get the weld
feature to work properly with our J-Grooves. I'll check Dotson's and Bliss'
sites to see what more they can add to our use of the weld feature. For now
though, we are just sketching the welds on the idw and leaving the iam mated
but unwelded.