In-Place (top down) vs. base work feature (bottom up) design

[Anonymous]

Need some advice:

My noob is has fallen into the practice of basing parts upon other parts in
the assembly. Then, projecting pertinent geometry (unadaptive), he creates
the part. Of course, Inventor arbitrarily chooses the location of the origin
of the new part and that often has nothing to do with the "logical center"
of the part.

So he ends up with an assembly that looks good, but is filled with random
"adaptive" workplanes, purple lines, and parts that are not really
associated with their origin workplanes.

Do I just let him continue until he paints himself in a corner, or should I
jump in. 3 times now I've steered him toward bottom up design, shown him the
Walt Jaquith treatise on the "house of cards". Am I being too wary of the
top-down approach and I should let him continue down this path, or should I
intervene before bad habits are formed?

Does anyone have any experiences like this?

[Anonymous]

I vote you start cracking some skulls
now.

We've currently got more work here than we know
what to do with, and have been relying heavily upon contractors to help us with
the workload.  One guy had absolutely no clue what he was doing with
Inventor.  I'd actually doubt the guy ever saw Inventor before - he was
that bad.  He actually spent two weeks modeling and detailing (er,
rather attempting to model and detail) a splice plate - a simple
rectangular plate with a number of holes in it.  He was promptly excused
from the building.  Another contractor we had, on the surface, seemed to
have a decent grasp on Inventor.  He at least got work done.  He quit
(or at least we think he quit, we're not even sure if he was canned),
and a piece of equipment that he had worked on needed to be modified.  One
of my colleagues ended up scrapping his entire assembly and remodeling and
redetailing it.  Why?  Because his models had a complete lack of
intelligence, utilized no intuitive method of grouping parts or assemblies, and
so it would have taken longer to figure out what he did and why he did it that
way than to start over from scratch.

What everything in 3D modeling boils down to is
editability and reusability.  If your "n00b" isn't generating models that
exhibit those characteristics, you're going to pay for it later on. 
Instill good modeling techniques now before his projects get completely
out-of-hand.  Make sure that he/she knows how to properly use
adaptivity.  Stress the use of derived parts and skeletal modeling. 
Make sure he/she understands how Inventor's Bill of Materials and Parts Lists
work.  Spending one day to teach and/or demonstrate these techniques can
pay off big in the long run.

---

Brian R.
Iwaskewycz

style="FONT-SIZE: 14pt; FONT-FAMILY: Verdana">CAD Systems
Manager

style="FONT-SIZE: 14pt; FONT-FAMILY: Verdana">
href="http://www.corefurnace.com">Core Furnace Systems

[JDMather]

I think there is nothing wrong with top down as long as you do it properly. I was doing some on site training with a company where to model their assembly top down worked very well. What I had them do is immediately exit the first sketch and part editing of in-place modeled parts and constrain the origin centerpoint to a logical position. Then we would go back to the part and start creating sketches and features. It is a little bit of a hassle but to get the benefits of top down without ending up with a mess you have to exercise some discipline.
I certainly wouldn't let the newbie continue with his current method - it always much harder to break bad habits than to create good habits.

[Anonymous]

The best way to learn Inventor is by trial and error. It's the way you discover what to do, and even more important, what NOT to do.

Cross-Part Adaptivity (CPA) can be a real pain in the -you know where- when references are lost. Unadaptive projected geometry can be even worse, as you will not get any errors when the base-part changes.

Although the best way would be to let him paint himself into a corner (which *should* be a lesson learned "the hard way") you would be better off to jump in.

-Slap him on the wrist if he does not project the origin point onto his first sketch (I know there's an add-in which projects the origin point on each new sketch)
-Slap him on the wrist when he creates additional workplanes to constrain a symmetrical part into an assembly. Show him how you would center the first sketch around the projected origin.
-Slap him on the wrist when we creates more 2D dimensions than necessary. Show him how to use 2D constraints in a proper way.
-Slap him on the wrist when you see him opening assemblies from Windows Explorer. Explain the major importance of Project Files again 😉
-Slap him on the wrist if there's no grounded part in an assembly
-Slap him on the wrist for each underconstrained part in the assembly
-Slap him on the wrist when he is complaining about slow-down when he has 20 adaptive parts and sub-assemblies. First say that he must turn off the adaptivity of the parts/sub-assemblies when he is done with those parts, then explain him the advantages of Skeleton Modeling one more time.

That's all I can think of right now 🙂

[Anonymous]

I was once told there are three stages to apprenticeships.
The first is learning the right way to do things, usually at college.
The second is going out there on your own and having the confidence to do it.
The third is taking an appentice under your wing and learing how to train them. Apparently you learn the most at this stage.

There's nowt wrong with TTD, particularly when it's done by the DAcc's Generators.
If your appentice wants cross part associativity then teach him about using spreadsheet driven parameters.

You might want to teach him something about Design Intent as well.

TTD is very useful for knocking up a quick & dirty design. But that has to be converted into a clean and sleek design so others can
follow and understand the Design Intent etc.

You might want to point your appentice at Kevin Schneider's blog http://mfgcommunity.autodesk.com/blog/kevin/ He makes some
intersting points about Smarter Designs etc.


--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs



"John-IV8SP1" wrote in message news:5218989@discussion.autodesk.com...
Need some advice:

My noob is has fallen into the practice of basing parts upon other parts in
the assembly. Then, projecting pertinent geometry (unadaptive), he creates
the part. Of course, Inventor arbitrarily chooses the location of the origin
of the new part and that often has nothing to do with the "logical center"
of the part.

So he ends up with an assembly that looks good, but is filled with random
"adaptive" workplanes, purple lines, and parts that are not really
associated with their origin workplanes.

Do I just let him continue until he paints himself in a corner, or should I
jump in. 3 times now I've steered him toward bottom up design, shown him the
Walt Jaquith treatise on the "house of cards". Am I being too wary of the
top-down approach and I should let him continue down this path, or should I
intervene before bad habits are formed?

Does anyone have any experiences like this?

[Anonymous]

Thank you Brian, JD, Teun, and Duncan. I will print out these and share them
with the engineering mananger. Along with my role as technical leader in the
group comes some need for discretion toward new hires. 😕

Please I would also welcome anyone else's comments as to the way that
they've handled these situations decisively and discreetly. Thanks in
advance.


"John-IV8SP1" wrote in message
news:5218989@discussion.autodesk.com...
Need some advice:

My noob is has fallen into the practice of basing parts upon other parts in
the assembly. Then, projecting pertinent geometry (unadaptive), he creates
the part. Of course, Inventor arbitrarily chooses the location of the origin
of the new part and that often has nothing to do with the "logical center"
of the part.

So he ends up with an assembly that looks good, but is filled with random
"adaptive" workplanes, purple lines, and parts that are not really
associated with their origin workplanes.

Do I just let him continue until he paints himself in a corner, or should I
jump in. 3 times now I've steered him toward bottom up design, shown him the
Walt Jaquith treatise on the "house of cards". Am I being too wary of the
top-down approach and I should let him continue down this path, or should I
intervene before bad habits are formed?

Does anyone have any experiences like this?

[Anonymous]

The "technical leader" lags behind with Inv8sp1 ;~)

Sorry, I couldn't miss that opportunity.



--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs

[Anonymous]

Hehe but now it's out of my hands - corporate is now dictating the migration
schedule and I'm advocating migration to 11. Cheers!


"Duncan Anderson" wrote in
message news:5219750@discussion.autodesk.com...
The "technical leader" lags behind with Inv8sp1 ;~)

Sorry, I couldn't miss that opportunity.



--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and
proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs

[Anonymous]

From IV8 to IV11...you're in for a shock! 🙂

[Anonymous]

That's assuming I haven't been on the beta teams, which I'm not saying that
I have been.

I'd like to keep the discussion centered around modeling practices in this
thread, however. ;~)


wrote in message news:5219775@discussion.autodesk.com...
From IV8 to IV11...you're in for a shock! 🙂

[Anonymous]

My co-worker just did a very small project with IV, and he used Top-down, with all kinds of adaptivity to do his design. He first asked me what the proper method should be for doing the design, and I started to explain the skeletal method.

Unfortunately, there was just too much to tell him to get him up to speed soon enough. The project due date was looming, so he had to just jump in and run with whatever looked best to him.

Luckily, he got it right the first time, and didn't need to go back and revise the model too much. On the last day of the project, things started falling apart just as he finished the final piece of the puzzle (with scissors and a hammer).

He exported the model, made some drawings (in AutoCAD, for some reason), and got the project done. It was a completely custom job, so it will never need to be edited, thankfully. If it was going to be a library part, I would have stepped in and helped him much more than I did.

[Anonymous]

John,

This sounds like a very good excuse for you and your apprentice to go on some official training for Inv11sp1DWF. After all, you need
to be "officially" shown all the new goodies and how best to use them and your apprentice needs showing the best way of using Inv to
get the maximum benefit from the investment.

You might even be able to persuade your bosses that it's worth paying you to go for AICE.


--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs



"John-IV8SP1" wrote in message news:5219789@discussion.autodesk.com...
Hehe but now it's out of my hands - corporate is now dictating the migration
schedule and I'm advocating migration to 11. Cheers!


"Duncan Anderson" wrote in
message news:5219750@discussion.autodesk.com...
The "technical leader" lags behind with Inv8sp1 ;~)

Sorry, I couldn't miss that opportunity.



--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and
proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs

[Anonymous]

Hmmm all good food for thought. But why training for sp1DWF?


"Duncan Anderson" wrote in
message news:5219915@discussion.autodesk.com...
John,

This sounds like a very good excuse for you and your apprentice to go on
some official training for Inv11sp1DWF. After all, you need
to be "officially" shown all the new goodies and how best to use them and
your apprentice needs showing the best way of using Inv to
get the maximum benefit from the investment.

You might even be able to persuade your bosses that it's worth paying you to
go for AICE.


--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and
proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs



"John-IV8SP1" wrote in message
news:5219789@discussion.autodesk.com...
Hehe but now it's out of my hands - corporate is now dictating the migration
schedule and I'm advocating migration to 11. Cheers!


"Duncan Anderson" wrote in
message news:5219750@discussion.autodesk.com...
The "technical leader" lags behind with Inv8sp1 ;~)

Sorry, I couldn't miss that opportunity.



--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and
proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs

[Anonymous]

Because DWF extension is only available for subscription customers and your company might find it cheaper to go for subscription -
the incremental steps are easily learnt from the What's New - than paying for re-training every three to four years when they decide
to upgrade by three to four versions.



--
Duncan
"Humour ... is one man shouting gibberish in the face of authority, and proving by fabricated insanity that nothing could be as mad
as what passes for ordinary living."
(Terence 'Spike' Milligan K.B.E., 1918-2002)
www.autodesk.co.uk/inventorjobs
ps. I'm not a salesman, just trying to think logically.



"John-IV8SP1" wrote in message news:5219980@discussion.autodesk.com...
Hmmm all good food for thought. But why training for sp1DWF?


[snip]

[albertemi]

I need some advice. I use Solid-body modeling (TOP DOWN) with a base sketch controlling my assembly. I just got a message from my colleague complaining that my models cant be modified easily and they don't understand what I have done. I came to realize later, my colleagues are used to bottom-up (traditional way) modeling techniques with a motherload of constraint. I have moved away from the traditional approach a while back as I have gotten very accustomed to TOP DOWN mixed with some traditional constraint relationships between parts. Now, the time I personally save and I can take a large scope of work is completely wasted if, some adjustments need to be made by others not familiar with TOP DOWN. My first reaction is frustration but then again my models must be editable by others. What shall I do? Does anybody experience this? It seems I have to model the traditional way no matter how I feel about it.

[cadman777]

I totally converted to TopDown a little over a decade ago. But I used a number of other methods, including BottomUp for 10 years before that. There's NO WAY I would return to any other method of modeling, unless Surfacing is involved or some odd-ball other kinds of modeling.

So my advice to you is, tell your associate to LEARN TOP-DOWN MODELING or 'tough beans' you ain't changing for him. If you want to be 'nice' about it, then show him how to edit your TopDown models and explain to him why his method is inferior in almost every case.

If you work for a corporation, and they don't have any modeling standards, then maybe now's the time to create a set of modeling standards that everybody who works there can learn and use. However, I doubt there's a one-size-fits-all TopDown modeling method, b/c depending on the kind of modeling you do (AEC, Mechanical, MEP, Surfacing, etc.), your TopDown standard may change. In fact, it may change depending on what kind of model you're creating in any given discipline. So if that's the case, and you can't arrive at any 'company standard', then give the guy a bottle of GreyGoose or Glenlivit12, and ignore him. And if management gets involved...tell them if they want to do it BottomUp, DO IT THEMSELVES or FIND SOMEONE TO REPLACE YOU, and then 'pack up and walk'.

[BenoNZ]

---

@albertemi wrote:

I need some advice. I use Solid-body modeling (TOP DOWN) with a base sketch controlling my assembly. I just got a message from my colleague complaining that my models cant be modified easily and they don't understand what I have done. I came to realize later, my colleagues are used to bottom-up (traditional way) modeling techniques with a motherload of constraint. I have moved away from the traditional approach a while back as I have gotten very accustomed to TOP DOWN mixed with some traditional constraint relationships between parts. Now, the time I personally save and I can take a large scope of work is completely wasted if, some adjustments need to be made by others not familiar with TOP DOWN. My first reaction is frustration but then again my models must be editable by others. What shall I do? Does anybody experience this? It seems I have to model the traditional way no matter how I feel about it.

---

It seems you have replied to a topic from 2006 somehow?

Anyway, this does sound very frustrating. I would suggest to your collogues that they learn both methods so they can work with you and help them to understand your methods.  If they know the software I see no reason why they couldn't understand quickly how it top-down works.

[albertemi]

I just joined this team. My manager advised me to align myself with everybody else. So that's that ... 
Thank you, for taking the time to respond.  

[BenoNZ]

---

@albertemi wrote:

I just joined this team. My manager advised me to align myself with everybody else. So that's that ... 
Thank you, for taking the time to respond.  

---

It might be worth mentioning to your manager about how things can be improved then? It's hard to speak for your situation. 

I do remember when I worked for one large company with numerous Inventor users that had methods that could be improved, suggesting it didn't go down very well.

[albertemi]

Yes that happened to me too, suggested improvements to no avail. Talking to the manager might work but changing the habits people have is close to impossible. Ill keep a low profile for now, but it's eating me inside.