Design approach help

Pretty good at LSI, IC & PCB design is it?

LMAO !!!

 

Just use open solids, right?

Is it dark inside? Did you let the smoke out?

 

This is the sort of SW thinking that I need to develop. Thanks for a good
tip. Of course, it'll just take time to build familiarity and model
creation skills/techniques. I'm just trying to get some real work done in
the context of learning. This thread has been quite useful. Thank you,
all.

I say the following without being an expert user, of course. I could be
wrong.
I wish SW had better text and graphics functions. It seems that putting
non-extruded text on surfaces is simply impossible or convoluted, at best.

I now understand that, if you mentally detatch from the idea of zero
thickness text you can simply create very thin extruded or extrude-cut text
on just about any surface. Still, it feels like I should be able to very
simply and quickly add "painted" text to parts. The same applies to
graphics.

I'll look into this as an excersice. Perhaps another VB program to take
button names/coordinates and apply the right text and/or graphic to the top
of the keycap at the assembly level. It would surely make for significantly
lighter-weight button part files and the approach could be much more
flexible and easier to use. Gotta think about this one.

Sounds like another job for VB!

More reading for me to do. I also thought that this is not unlike a single
injection molded part (it IS an injection molded part, isn't it!). So, what
approach do people take to complex IM parts? Do you design bits and pieces
of the part and then glue it together in an assembly or do you do it all in
one shot?

I've alredy had assemblies self-destruct. I learned my lesson years and
years ago. I never edit the same file more than a couple of times. Before
I sit down for a new session, the current working directory gets a full
backup. Disk storage is cheap these days.

Well, it's more about a practical consideration, I have to get a design out
within the next couple of months. That means mechanical, electronic and
software design. Then, hopefully, a couple of months later, I'll have the
time to go away for a week and learn what I've been doing wrong!

-Martin

 

?

The problem with long posts... its easy to miss a critical line: My post
started- Beyond the issue of SWx using data from other CAD packages (which
you correctly point out is real-world, and all that is left to debate is how
magnificently/piss-poorly/somewhere-in-between SWx handles it per each CAD
vendor )...

Martin, as I read it, is designing something new, and matt made a great
point that he might not want to start in one package and transition to
another when it could all be done faster in a single environment.
To be fair to the origin of the thread, we should wait for someone to post
about a legacy problem before opening that can of worms (and I won't be
commenting on that thread if it should happen, because I am hired to make
new stuff, not transtion legacy data... as a matter of courtesy I try to
keep quiet when I have nothing constructive to add)

No, you were not wrong (sorry Cliff), and I am grateful for that
contribution that you made to the dialog here - frankly, I would not be
capable of what I am currently able to do had you not challenged me to get
outside of my comfort zone. That nudge, after an eye opening API
presentation by Cholly from SWx that showed me how it all worked behind the
scenes, really forced me to push my understanding of how mainstream CAD
works and what I could do with it if I just changed my methodolgy and
mindset. That is what I like about this NG over the eng-tips one - I like
the theory discussions (learn how to fish -vs- "here's the fish")

Good question, but my answer will not satisfy an idealist. I work in
SolidWorks because that is the predominant modeler in my marketplace, and
what I can get paid to work in.

I bet you would agree that, ultimatley, I am a designer, not a CAD driver.
I'm actually pretty darn good at the design part (that distant sound is my
own horn tooting) - the getting-good-at-CAD part is incidental and, I hope,
temporary based on the era that I live in. 20 years from now we will all
have a good laugh about the primative CAD we were forced to use.
I bet you would also agree that it is a shame that driving CAD is part of
any of our jobs... our talents should be 100% focused on making products,
without having any of our time spent learning new software or beating the
limitations of legacy systems. I welcome your thoughts on either of those
statements if I am missing something.

Right now, SWx provides me a HUGE customer base... if people start asking
for VX, I will definatley look at it, but I have never heard a single
request for it. I have the responsibility of doing design work in my
customers CAD package so they can continue to work on the products I design
for them and make changes without having to hire me again or invest in a
whole new infrastructure just to change the data they paid for. So, I am
not the one to convince.. it is everyone else
For now, I have to put VX in the same category as Rhino... it might be
great if I worked in a vacuum, but I don't. I had the pleasure of meeting
Bob McNeel (one of the nicest guys you'll ever meet) and even he was
gracious enough to understand that I couldn't use his product based in my
business. However, I have recommended Rhino to companies who are not
restricted by my business's model, based on the force of the glowing reponse
from other Rhino users (and the folks I suggested test Rhino have come back
and said they LOVE it) - and if given a blank-page company looking for a
recomendation, I will probably recommend at least a peek at VX based on your
endorsement. But for me to try VX would be a waste of time... the best that
could happen is I would love it and STILL couldn't use it, so the learning
curve investment just doesn't make any sense for me.

However, I am glad that you have found something fantastic for you that
gives you a competitive advantage over the other guy, and look forward to
you making so much money that the entire industry has to shift to your way
of operating or crumble into dust (actually not meant to be a facetious
statement... I am a Darwinist through-and-through, and hope that the best
package squashes the inferior ones so we can all be designers again and
forget about this other crap once and for all)

regards,
Ed

 

Ed,

Elequently and succinctly put. I think it's safe to say this applies to most
of us.

There's alot more to a business than what CAD system you use, but using the
wrong one can be costly . With the pace of product development, and the
inherent real time collaboration between client and contractor, you both
have to be on the exact same page. The CAD system is what drives this
process in the design and engineering world. It would be foolish to use
something that didn't expose you to the largest market. No matter how much
better it was at some things.

Regards

Mark

 

Very well put. One night, a couple of years ago, I was very frustrated with
decidedly inferior electronics hardware design tools. Tools costing $10K
that are not much more than refined manure. Anyhow, I typed and printed a
short line that hung on the wall and try to refer to as often as possible
for guidance: "We need better tools". That's the secret. Now, in that
case, I spent two months writing code to make this bad tool usable. We
couldn't just dump it.

As an amateur musician I also refer to that world for analogies. I've been
known to say that CAD tools must be designed from the stanpoint that design
can be approached as a performing art. You are trying to express your
vision and ideas --or your clients'-- and get this from mind to paper (or
model) as fluidly as possible. If you have to spend a significant amount of
time preoccupied with the mechanics of the program (memory issue, types of
entities to use to avoid problems, shortcuts, tricks, etc.) rather than have
your mind in the creative "zone" the tool has failed to deliver what we
truly need. Most CAD tools I know do fail miserably at this.

Tool authors need to understand that we don't go to Home Depot to buy a 1/2
inch drill bit, we really want a 1/2 inch hole.

But, things are getting better, aren't they?

-Martin

 

Why do you think I'm spending so much time with MasterCAM, Mark ?

In the next few years I believe CNC Software will dominate like they never
have before and this will open up a lot of oportunities for me. Lots of
companies had a huge head start (DP Technology, etc) and blew it cause they
never learned how to market. With the realease of MasterCAM X they
are going to find themselves loseing market share... slowly and then
quickly.

Thankfully I waited until now but I have got a lot of learning to do on
something I have tried to shy away from for a long time.... MasterCAM's
post langauge.

Timing is everything and I do see the handwriting on the wall.

jon

 

Ed,


I believe trying VX would be the same kind of eye opening experience for you
because it would show you how things should work and how transparent solid,
surface and wireframe should be and is NOT in SolidWorks.

I understand what your saying about having a market with a well established
product like SolidWorks. I don't expect you to change from SolidWorks to
VX. Honestly, I would rather have you stay with SolidWorks but be more
vocal on the changes SolidWork needs. IMO, SolidWorks needs a shit
load of them... and yes working with legacy data (something your not
involved with, as per what you said above) is certainly one issue. Needless
to say this is a huge issue for shops that I have contact with. I don't
expect you to support something your not involved in. Perhaps I can find
someone else who really wants this change in SolidWorks and wants to be
vocal about it ? ;>)


What I do think would help you tremedously is seeing and *feeling*
how the tools work in what I consider to be a much better conceived and
implimented product... don't get me wrong it's not perfect and some things
in VX don't work / need work.

What I'm really after is knowing that you know how things could be in
SolidWorks... not getting you to switch to VX. I could say the same about
Global Shape Modeling and thinkID.

I don't really have much respect for Bob McNeel but I do accept that
you like him. I think Rhino development is very slow and that Bob
McNeel has made numerous errors in how Rhino has been developed.
If Bob McNeel had his act together, Rhino would be much more
dominant right now and having much more of an effect on
companies like SolidWorks.

Instead, we had the AG Lib debacle, and developers like Gary Crocker
(IntegrityWare) building add-ins where IMO, Gary and his partner
should have been brought on board to develop Rhino solid modeling
tools a long time ago. Rhino 4 is still a year away ! Sad. What I do like
is that Rhino has put one hell of a scare into Alias and Alias has reacted
by lowering prices and getting more quality tutorials and books out. They
even have a learning addition now. All this is because of Bob McNeel !
I'm thinking about trying Alias Studio Tools Learning Addition. I'm one
hundred percent convinced that I would never have this opportunity
and neither would anyone else if it were not for Bob McNeel /
Rhino so I'm more than willing to give someone I don't think very
highly of full credit in this regard. In any case, I have about as much
interest in discussing Bob McNeel / Rhino with you as you do in
discussing how to handle legacy wireframe data with me. :>)

I will say this, Ed.... there is a long way to go even in the products
I like before they are truly complete tools. This applies to VX, thinkID
and to Concepts.

It would be fun to have a discussion with you, me and Joe
Dunne in the same room. Joe Dunne has the background
(ComputerVision) to know how CAD should be. Joe has
seen and marketed stuff that CV had that was just fucking
awesome. Ask him to tell you about it sometime. Might
be a nice break for him rather than the ho-hum stuff
(SolidWorks) he now deals in. ;>)

Take care and let me know if VX does what I know it will.... open
your mind to how things should be / can be in SolidWorks. :>)


jon

 

Because it's a cracked copy & does not work properly?
Because you cannot read the docs?
Because you are full of BS in the first place?
Because you don't know what it does, having no actual
use for it?
Because you got caught yet again posing as a self-proclaimed
"expert" on a BBS with your pants down & someone else has
a maglite?
Because you need someone else's posts to copy again?

HTH

 

And NOW you are going to try to learn something
(after years & years of posing & spouting BS)?

Don't give up your new day job .... lots of trucks ....
HTH

 

Would these be the ones you were touting ?

No wonder you cannot actually use it or put up <GGGG>.

Tried reading the docs yet? Words too big?

 

Idiot.

 

I like that line A LOT, and believe I will be using it in the future

 

'got that from an aerospace engineer buddy of mine. Straight and to the
point.

Feel free to spread the word.

-Martin

 

They don't use metric?

It's no wonder that US exports are so poor ..... nobody else
in the world wants inch parts .....

 

Hi Martin,

I skipped alot of the messages in this thread, but assuming that I am
not missing something I would suggest the following with graphics:

1.
DO NOT put graphics on your part models UNLESS it is going to be a
molded feature! I have done similar things, and it will only leed to
sweat and tears, and not help you meet your deadline.

2
If the graphics are going to be printed on the parts, then the best and
most useful way is to use the part file to create a 2D line layout for
use in a graphics program, where you can layout the graphics
accurately. Ideally this is in Illustrator, and your output is an EPS
or PDF file for the printer. In manufacturing it is useful to have a
drawing with the buttons and graphics together for control, but the
printer has no use for a 3D cad model with graphics on it. Also, as you
mentioned somewhere, you will only make your parts more complex adding
3D graphics features.

3
If you MUST have graphics as 3D features, and your buttons have fairly
simple top surfaces (flat, slightly domed etc.), I would suggest
creating the sketch with the graphic on a plane that projects normal to
the face you want the graphic. I think it is different in SW2005, but
you may only be able to do one character at a time - but you can cut
the surface with this sketch.
Or...
if you need a 3D version where the character is embossed, you can
offset the surface, and then extrude the character using the surface
and offset surface of the button as the starting and ending surface for
the extrude feature. I would then do a parting line draft around each
character. If you try and extrud the text with draft, and the surface
is not a plane, your characters will become more deformed than is
desirable.

4.
If you want to only show the graphics for visualization / 3D rendering,
then the way to do it is to apply the graphics in photoworks and
render. That is another can of worms.... Again, if you are looking for
speed, I often simply take a screenshot of the top view (or whatever
view) and use that image in photoshop / illustrator or other graphic
package where I can add the artwork. fast and relatively painless.

I think that will get you into enough trouble for now I think....

Good luck!
Daniel

 

Martin,

Another thing I would strongly suggest, as you seem to be going through
a steep learning curve with SW, is to learn from existing parts /
assemblies.

I found that one of the fastest ways to understand construction
strategies and techniques is to look at existing part file histories.
The beauty of parametric history is you can actually follow the step by
step process someone used to make a part simply by rolling back the
history.

I am sure you can get someone here who has an exhisting keypad part, to
send you the file so that you can see their process and logic. You
would learn very quickly that way. I have learned many very useful
techniques this way. If they were not mentioned before in the thread,
there are some very proficient users with parts you can download. Just
a few from my list:
http://www.mikejwilson.com
http://www.scottjbaugh.com
http://webpages.charter.net/mkikstra/SolidWorks.html

You will also start to quickly see good / efficient technique verses,
shall we say, less than ideal solutions....

Again, good luck!
Daniel

 

Tool authors need to understand that we don't go to Home Depot to buy a

If you want to get philosophical and a little off-topic (or...maybe not) I
am of the opinion that software is partly at fault for some of the
outsourcing issues some industries have seen. What do I mean?

The gadgets being designed today are orders of magnitude more complex than
just a few years ago. At every step of the way there has to be heavy
reliance on software-based design and management tools as necessary and
indispensable tools with which to do the work. As these tools fail to
deliver performance and reliability --or fail the address that I want a 10mm
hole, not a drill bit-- the cost of getting to market grows, perhaps
exponentially at large organizations. So, what do you do? You take your
crummy software infrastructure and move it somewhere where you can afford
engineers burning hours upon hours dealing with tool issues rather than
getting work done. In other words, crappy software may very well have been
behind some of you loosing your jobs to other markets.

It's a gross oversimplification and a myopic look at the problem, so,
please, don't jump down my throat, I know that the whole affair is much more
complex and subtle than this. But, this is a plausible explanation I can
humbly offer from first hand experience with tools that should allow you to
do the work in hours but require days due to their inadequacies. I couldn't
imagine teams of dozens of engineers at high wages being allowed to function
in this manner.

-Martin

 

How timely this is. I spent last night having a gargantuan battle with SW.
It involved many C4-ladden explosions of assemblies, hanging, strange
behavior, closing part/assembly files by itself and SW generally being a
real bitch to deal with.

As a new user, I can't say that I am impressed, you end up mentally so far
away from what you originally got into the swamp to for...

Anyhow, I have now given up on the idea of having these nice intelligently
defined buttons that have a configuration-per legend, etc. To me this would
have been the most elegant approach to the problem. Very much
self-documenting. Easy to maintain. However, I've come to the realization
that SW gives you the tools but does not guarantee that the head is attached
to the handle solidly enough. So, you choose to use the tools at your own
risk and, at any given point in time, a sharp blade might just fly off the
handle. Unfortunate.

I want the text in the model/assembly because...well, it's a damn CAD system
and I want all my documentation to travel with these files. I can do this
with ACAD in 2D or 3D. No problems there. It's just text.

In the end I might discover that the only sensible approach might very well
be to do as you suggest and either document the key-top (yes, they are flat)
text in ACAD, Photoshop or Illustrator. I'll ask my vendor what they might
prefer. This might save me from a lot of pain and aggravation. BTW, the
legends will be laser-etched, and so, the idea of extrude-cutting the
key-tops with text was very appealing because it would make it look like the
final product.

-Martin