Some surfaces behaviour makes me confused

[design-engine]

I love proe... it is so much fun


Edited by: design-engine

 

[Jacoolo]

once You are payed for things You love to do, You can be named lucky

so it is rather batter to love Pro/E

 

[Jacoolo]

ok let`s make this - 1 000 000 long post - come true

the last bathroom appliance investigations. I think I end up with it

sharp edge between surfaces is intended

 

[Jacoolo]

one simple case - I think just to clarify Soliify > Patch funcionality

I made an example as below, where drafted geometry is made by Extrude > Surface > Capped ends

then it is drafted and surfs are copied without bottom cap(surf volume is not capped on bottom side)





then I tried to make Solidify to push Pro/e to combine solid with surface. I was supposed to obtain one solid geom where two volumes intersect each other

I thought it would work with Solidify > Patch(that is why I left one side opened in surface volume)

but no..., Pro/e can not handle this - Patch option is shaded

when I extend this towards inside, so surface volume is inside solid one - no problem. But this is obvious and it is not my intent.



What is wrong? I can swear I used Patch option with following condition before - I suppose in Pro/e 2001!

 

[dgs]

In my experience, WF doesn't offer the patch option unless it's needed. I'm not sure how it determines that it is needed, though. Then again, it seems to default to patch when protrusion or cut work fine. Back in 2001, the patch option was sort of the last resort if the protrusion or cut options failed to work. I always considered a 'patch' as a little bit of a band aid. "I couldn't get it to work right, so I just patched it."


On your situation, Pro|E doesn't do well when the open edges of a quilt are not buried in the solid. Better practice is to trim it to the solid before trying to solidify. Actually, I'd trim the original rather than making a copy, or better yet make a copy of the top of your flat and merge that with the cone, then solidify.


This ties up the loose end of the surface geometry and gives Pro|E some clear direction on what you want to do. With the open end of the cone surface hanging out the bottom, Pro|E isn't quite sure what you're trying to do. The top is clear, built what are you after with the bottom? Should Pro|E assume that it gets trimmed or assume that it gets filled as if it were capped? Better to clearly communicate to Pro|E what you're after.

 

[Jacoolo]

thx dgs

I was trying to avoid all stuff You mentioned above to be done to achieve such "hybrid" volume.

Maybe I expected to much form Pro/E. From my side it should be easy task to do if
I were Pro/E. I would have trimed it(surface volume with open end) to last solid surf it intersects with. Shame Pro/E can not handle even such simple case.

Ok two cents about - Why for hell I need sorta stuff. Well, case is I am working on two housing for new device, and I am lack of AAX.

So far I am more interested in investigation for best parting line position for this device(the line housings gonna meet each other), best solution for front, rear part of it. And of course still confused about best fixing method(housings gonna be made from plastic). So summarizing this - I still play a lot around concepts(more, the best)

So, for this moment I would like my ribs follow automaticly to the newest housing revision(they are done in models, but I control them at assembly level and sometimes ribs loose their references)

So I figured out it would be best to push ribs and another stuff at assembly level. Make it as surfaces and then solidify by Patch in model. My intent was to create these ribs as Extrude feats with the front face described and contoled by datums from assembly. The rear part of ribs would be made with default depth. So I would not care where exactly it should end(that is why I removed capped surfaces)



So this approach would allow me to not reference such extrudes to existing housings, and easly introduce many ideas without geting in to Resolve mode. This in theory. Unfortunately Pro/E is not willing to cooperate with my idea

 

[dgs]

I see what you're getting at, but to my way of thinking, you are jumping through hoops to try to get Pro|E to assume what you want rather than simply telling it what you want. That doesn't work well with my wife, and it doesn't work real well with Pro|E either. In my experience, the better you communicate to Pro|E what you're after, the better the results will be. (That's true with my wife too!

)


Resolve mode is not the end of the world, in fact it can be a good signal that you've violated your design spec. In this case, if you used the inside surfaces I think that the feature would be pretty robust. It might fail if the housing got too small, but that would bee good as the failure would signal that you've made it smaller that it can be.


I do understand your frustration in exploring many concepts when the design keeps changing and you have to keep selecting new references.

 

[Jacoolo]

dgs You hit with absolute accuracy 0.00000001 to the points which are familiar for me too(Pro/E, wife

)

And yes, I agree sometimes it is my fault I want things go different(I consider both cases mentioned before)

"It might fail if the housing got too small, but that would bee good as
the failure would signal that you've made it smaller that it can be. " -

exactly this situation I have now. I do not care if something is good or bad from Pro/E`s point of view, because of this project state I am more interest in robust design, try many ideas, bring them at screen fast. If I am each time pushed to crete the same things from scratch because there is one ref missing, my tempreture go crazy(more then oridinary 36C

)

I assume my ideas should go to the trash now

 

[Jacoolo]

another iron challange and - as usual - new problems to solve

I decided to devide whole model in to few major area

*dark grey - major surface
*yellow - second surface,
*green - third surface
*red - "nose" surface



it could be noticed - red lines in pic above - later in design, major surf must be trimmed and fillet surf gonna be added between major surf and yellow and green. That is why it is important to maimtain tangency(at least) in the area marked with red circle

Maintaining this demand tangency transition which is tricky to obtain.

The core aspect for introduced surfaces is the transition between yellow and green surf and "nose". It seems to be easy(maybe it is) but it demands a little strategy to obtain desired results. Of course for this example it depends directly on prepared curves network

 

[Jacoolo]

basic curves network for this iron challange


Now two approaches can be introduced to obtain green and yellow surf( The key demand is smoothe intersection edge btween those two surfaces):

*green surf first, yellow second related to green one

*yellow first, green related to yellow one

Let`s discuss first approach - green surface is done by VSS. Then curve is projected on it and in the end curve through points is made to build "nose" surf




in the end projected curve and curve through points are projected on to Datum plane by sketch to build yellow surf

 

[Jacoolo]

second approach

the yellow surf is made first. At the begiining the sketch with conic arc is placed on to datum plane and other boundries are prepared. when this is ready special curve is projected on such surf to maintain correct transition between yellow and green surface.

Curve through points are created and green surface is made by Boundary Blend



in the end the nose surf is prepared

 

[Jacoolo]

in all cases the back curve for yellow surf(transition area between yellow and major surf) is created by sketced spline with curvature in one end driven with relations




ok, now the problems:

*no matter which approach I use the nose surface has problems with curvature - no constant plot or too high values at one end

*are there any other tricks to maintain tangency transition between major surface and yellow or green one?

Do not hesitate to write down Your opinion. I am curious abot Your way to solve this thing.

 

[design-engine]

anyone in Minneapolis MN? I am speaking last at the Pro/E USER conference in Minneapolis today. This is a big conference. 300 + attendees. This conference could be the largest conference ever. Back in 1997 I spoke at this conference to over 500 persons. My talk is on A-Class surfacing and modeling approach and technique BTW.

 

[Jacoolo]

Bart ,

store the speach within camera, convert to avi and put it at FTP, so the rest of guys sitting thousand of miles from USA could enjoy same with those who are sitting there now)

it is not fair that the guy so much addicted as I am must sit here, and probably half of Your audiance were push to go there because the boss told such)

I wish You it wasn`t this way(half of audiance pushed to attend) ))

You said it first(You performing now in MN), now I am waiting for Power Point presentation))

good luck, do Your best

 

[design-engine]

I have been making the youtube videos. Ill put one together for the a-class surfacing next. This one will compare Pro/E Surfacing to Alias Surfacing with respect to A-Class surfacing... Which is what my talk touches on some. practice for the long beach conference in June.


Edited by: design-engine

 

[Jacoolo]

so Bart when can I expect those movies from Your speech?

I count all sleepless nights I spent waiting for it. do not let me feel left down)


New tutorial from Ragde at Product Design Forum has just appeared - check this out. Worth the effort

here is the link

I suggest also to check Parel`s blog to check some fine designs and tricks for surfacing(unfortunately ALias and SW related)

here is the link to Parel`s blog

 

[[-Skint-]]

New tutorial from Ragde at Product Design Forum has just appeared - check this out. Worth the effort

here is the link

So are you going to re-create this tutorial for Us using Pro-e techniques

.

 

[Jacoolo]

he, funny, I`ve just finished it in Pro/E.

Some things are intresting while comparing both tools.

For example - Ragde use 3 point spline, where mid point is place on construction circle by concident constraint.

This is not possible in Pro/E if You control spline with a polygon(as shown in Ragde tut).

However You can construct spline as following but without suing polygon option. So this is funnu difference between Pro/E and SW.

Next one is the moment while creating 'neck" surface with Boundary blend. I got a feeling Ragde solvesd this in SW using only one curve for first direction, and two edges in second direction, what is not possible so far in Pro/E( I suppose it could be done within ISDX).

So I obtaied same result by VSS with tangency option

 

[Jacoolo]

Lately, during my trip to Barcelona I was inspired to model a panel from interior of Boeing 737-800 series



Yes, it can be noticed, I did not bother with end details and curvature transitions. This was intended.

 

[Jacoolo]

here one can find curvature network, and edges layout





it is really not hard to model this. However it is really nice to see how quite model can be achieved within few basic surfacing tricks.

Any input in pleaed to come