Thanks guy's!!!!


Frenchy: """"""""""What is the code for put the Youtube video inside the forum?""""""""""

It's the embed button, but you have to go just below that and put a check mark in "use old embed code", then copy and paste.. The forum will strip out everything but the embed part...

So it looks like this part of a flash embed tag:

embed src="http://www.youtube.com/v/Yp1NXKwyf6I?version=3&hl=en_US&rel=0" type="application/x-shockwave-flash" width="560" height="345" allowscriptaccess="always" allowfullscreen="true"></embed

(i removed the begining and ending brackets so it shows in the forum)

@Burr many thx!
A little test :=) Works like a charm!
Not from me just taken by hasard! Seems not a plane's engine even there is a propeller :)

Your welcome Frenchy! That looks like the Complex engine model used in a viewport comparison, where MoI's viewport "blew doors" on the 10-20 K opponents!!! :o

Hi Danny,

"""""""""""""BTW you could try a tiny fillet on that trailing edge before the union and main fillet that should get rid of that fillet abnormality."""""""

Yeah, I tried that.. It works with a smaller fillet, but trying to do that huge "2 inch" fillet still produces that buggard up trailing edge...

I think a Variable fillet command will allow it to get the results desired. Like Michael mention, I think the desired result would be to have the fillet "ignore" that corner patch and just meet itself..

Here is a screenshot of a variable fillet in ViaCAD using a "by position option":



This is one in my CAM package that has a notoriously weak fillet engine (It's more of a CAM package, with Simple "solid modeling" adding in:



If I add any value other than "0" in this one, the corner patch buggers up also...

Punchcad produced a nice result!

Funny, Michael always states that MoI can be a nice "companion" to Rhino.. I have been considering buying Rhino to be a Companion to MoI!!! (Never really get it's interface or methods, but the plethera of advanced commands can be accessed for single shot methods, while we wait for MoI to get some of these...

I've been hemming and hawing on my companion to choose, so havnt done it yet... I have been looking at a couple, but just havnt taken the plunge yet.

I also wanted to follow up, because I couldnt let it go.... I made the fillet work in MoI with Danny's suggestion. I had originally filleted the trailing edge, but that puts seams at very small intervals that still messed with the filleter... Here I show getting a good wing root fillet out of MoI by creating a single, contiguous surface for the trailing edge.



Thanks Da.nny

Thanks for showing this clever setup. :-)

Peace,
Rich_Art. ;)

"I have been considering buying Rhino to be a Companion to MoI!!!"
You hit on a sentiment that had come to me through frustration the previous day: I had to do, what I thought, simple vector tracing of some art work. I usually go to "Inkscape" for line work, as AI has very uncomfortable line tools and process - this is what ensued:

" Lets start with a throught-point curve" - oh yeh, don't have it.
"Control-point curve", nope.
"Dammit I have to muk-around with those damn handles again!".
"Now where is the 3 point arc and circle?", ???, "intersect/fillet/blend/rebuild, ???!!!, inferred snapping and dynamic guides, ?!!!!!!."
"Dammit this interface is making my eyes bleed!"
"Note to self: if you do line work go directly to MOI - any type of complex technical drawing(that requires 3d-line-views, go to MOI) - if you want to relax, MOI -- Use AI as its companion".

Warning: feature requests ahead :)
There is one nice feature I miss(from Inkscape), creation handles persist past the point of creation(eg. 3 point circles have the 3 points left adjustable). And line intersection and trimming could be streamlined by allowing the 'brushing' away of intermediate line segments. But relatively speaking, MOI is heaven.

Hi Ambimind - it's still amazing to me that so many 2D illustration programs have stuck with the Bezier handle method instead of NURBS. The Bezier handle method tends to have that odd behavior of bulging out previous sections of the curve.

> There is one nice feature I miss(from Inkscape),
> creation handles persist past the point of creation(eg. 3
> point circles have the 3 points left adjustable).

Probably in MoI this will come as some kind of history related function eventually.


> And line intersection and trimming could be streamlined
> by allowing the 'brushing' away of intermediate line segments.

The tricky part about this is that MoI also has to deal with surfaces as well and not just curves alone - that can sometimes add some constraints for how a particular command is set up to operate. Also the current Trim workflow allows for just splitting objects up into pieces instead of necessarily eliminating any parts - that's a pretty important function to have as well.

So that's why that kind of trim may end up being some kind of additional QuickTrim command in the future, see here for some other discussion on it:
http://moi3d.com/forum/index.php?webtag=MOI&msg=4129.3

- Michael

Another cool demonstration video Burr, I like stirring up thought in people :)

Kudos to you too :)

-

So Trim and Blend are your best friends :)

Ps When you say "pretty close" this can be used in plane industry for make piece of wings? (what is precision needing? )
Or just sufficient for make very nice Scale models ? (for image or handcraft building or 3D Print)

Hi Frenchy,
When I say "pretty close" I am speaking about "matching the curve that already exists"... from the original model.. The last video was just showing a method to have the fillet work in this particular case. (no need to watch me sit there for 5 minutes trying to "match his curve")

If you were building a "precise airfoil", you would probably incorporate the conic tool with blend.

These old models are being made by "guys tracing bitmaps", so no such thing as precision or exact.. I think the true design would have "precise plotting" of everything.. Not just sketching.

thx for "precisions" ;)

I just tried Inventor Fusion on this. (It can read 3dm files direct by the way). After doing a boolean join, the filleter got it right first time. Here is a screen shot:

Nice.. Inventor fusion looks like Autodesk 123D...

Peace,
Rich_Art. ;)

I believe that will be the same in the free 123D ;)
http://www.123dapp.com/

Pleas bear in mind that I am new to MOI !

However - based on previous tuition by Michael, Burr and others- her is my latest attempt..............


And the model itself............


This does expose my somewhat cumbersome construction techniques. In particular how to replicate the rear face of the front fillet when constructing the rear fillet ?



As before - all advice most welcome.


John

Hi,

When I went through this not long ago I tried Autodesk 123D and it doesn't import or export 3dm. Maybe someone knows if I'm wrong about that now. MoI can easily do this fillet task though. It's just of matter of figuring it out. It took me awhile too but what Burrman is showing is essentially the same thing I arrived at.

Make your wing a solid, make your plane a solid. Union them. Then fillet will work fine on the intersecting curve it makes. The reason I make the wing and plane a solid first is that sometimes there can be issues with either one. So this way you find which is messed up. But burrman did it all in one op which is fine too if you have good geometry.

What I do and what could work for you is make a leading edge (le) and trailing edge (te) curve. Then make an airfoil cross section out of two curves. A pressure side (bottom) and suction side (top). Sweep the airfoil curves along the two rails. Then see if it will union into a solid. If it does you should be good to go. I found lofting like burrman was doing can create some problem areas. Sweep gives you a very smooth surface. But lofting could work for you. You don't have as much twist as a propeller has. It looked like the lofting was working fine in burrmans video. I've never experimented with sweep when you have a non-constant chord like you do. You might need a height rail too. Michael or burrman would probably know. Bascially you would have to make sure the airfoil is scaling appropriately. I don't have that problem because I have a constant chord distribution.

Autodesk 123d is a bear to download and install. Also the interface is redonkulous. I think they were hitting the c-pipe when they came up with that. I prefer MoI. The other thing I liked about MoI was that it forces you to model more like you would build. Yes other software may work in more situations but thinking about how its made and modelling it that way seems like a good idea to me. My experience was all the failures I would have with MoI were in situations were you couldn't make it anyway. If you can make it MoI can model it.

I did have the issue with tolerancing however. That required an unintuitive switch to the mm unit system then back to the meter system I was actually using. Other than that MoI gets an A+ from me. Your fillet is way bigger than mine though, so this probablly won't affect you.

Just my two cent, I can see you're already well on your way to success.

Hi prop_design !

I took courage from your earlier post !

I can truly say that I learnt a lot from this exercise. I found that I was rushing at things - and missing screen prompts - so I took a deep breath and slowed down.

One thing I discovered - I had not really fully understood the original wing geometry.

In the front view the profile of the wing has straight line top and bottom - but in plan view there is a kink in the leading edge. Therefore the wing root section cannot simple be scaled from the mid-wing (or tip) sections.

Yep - accuracy is paramount - if your geometry is out things just will not work.

I will probably go back (at some stage) and start again.

There's an old joke in the UK -City Type to Local Resident --- " How do I get to ?????"

Local Resident to City Type ---- "Well, I wouldn't start from here"

Thanks again for your feedback


John

The problem that MoI hits and AD fusion doesn't, is, I think in the way the fillets are constructed. AD seems to recognise a sharp edge and when it does, it automatically extends the fillets until they cross then trims them back to a sharp edge. It must interpolate the crossing in some way, as the extended fillet surfaces won't always meet exactly. MoI doesn't have this interpolation and gets in a knot. Michael G will know why it works in one case but not another.

PS: just tried it in Rhino, which works if Distance between Rails is selected as the fillet type.

>When I went through this not long ago I tried Autodesk 123D and it doesn't import or export 3dm. Maybe someone knows if I'm wrong about that now. MoI can easily do this fillet task though.


Well just export in Moi3d to .SAT and import this in Autodesk 123D. Works without any problems. If you export from Autodesk 123D to .SAT as well, Moi3D will be able to import it without problems.

I found Autodesk 123D a bit better with filleting.... I really hope Michael can improve the filleting or build in filleting as seen in Autodesk 123D.

Peace,
Rich_Art. ;)