Need help to close a surface object (loft/patch)

Need help to close a surface object (loft/patch)

tom7UBLC
Contributor Contributor
2,510 Views
10 Replies
Message 1 of 11

Need help to close a surface object (loft/patch)

tom7UBLC
Contributor
Contributor

Hi all,

 

I want to model my van and have designed a bunch of sketches representing the profile from the side, rear and front. I have succeed at joining the side profiles but can't seem to find how to close the sides. I have tried loft and patch, to no avail.

 

When I create a loft with the splines on sketches "Body side behind rear wheel wells", "Body front by doors", "Body top windshield", "Van cut bottom windshield" and "Van cut headlights", the resulting surface is rounded on the edges and doesn't join the existing loft, created from the side profiles.

 

My question: how can I close the side of this van so that it meets the existing loft and has the shape designated by the splines, i.e. follows the curve of the side walls?

 

Thanks,

Tom

0 Likes
Accepted solutions (2)
2,511 Views
10 Replies
Replies (10)
Message 2 of 11

davebYYPCU
Consultant
Consultant

I am not a surfacing guy, but know lofts, 

I think you are trying to get the easy button to work.

I got the patch with front door spline to work, but no other rails are accepted.

 

You may do better to reduce the detail, and do wheel arches and bumber bars as separate add / subtract items.

 

Might help....

0 Likes
Message 3 of 11

tom7UBLC
Contributor
Contributor

Thanks for your advice @davebYYPCU!

I could reduce the complexity of the wheel arches as you suggested. I noticed there were also "deformities" at the top, where the side meets the windshield and by the roof.

0 Likes
Message 4 of 11

beresfordromeo
Advocate
Advocate

Hi @tom7UBLC 

 

Do you still need help with this model?

 

I am happy to try to try help you but your attempt to model this type of thing is arguably ambitous and will require a high level of commitment. If you have moved on there is little point anyone else getting involved because I am afraid this is going to be a labour of love, it may be worthwhile and both of us may learn a lot from it but it is pointless trying to do this is a cursory way. Sadly even reducing details may not help, because the absence of those details may prevent other aspects coming together. 

 

If you intend to pursue this type of modelling in the future I would recommend checking out videos by namenotimportant on YouTube. This channel does not post often but the content is always worth waiting for and should give you an idea of some of the principles involved.

 

https://www.youtube.com/channel/UC18IFSIQL8i67ZXQITMLopg

0 Likes
Message 5 of 11

tom7UBLC
Contributor
Contributor

Thanks for getting back to me!

 

Yes, I very much want to model this van. I have made a bit of progress since I posted. I reduced the amount of detail and have got to the point I can loft the whole van. As you pointed out, I would like to get a certain amount of detail in the design so I can do realistic simulations. Should this be done at a later stage?

 

The actual thickness of the van body (i.e. the reinforcing members) is about 50mm. However, if I try to thicken the surface body inward, I can only do so by 0.1mm on some of the surfaces and some just won't (the sides).

 

Help much appreciated,

Tom

0 Likes
Message 6 of 11

beresfordromeo
Advocate
Advocate
Accepted solution

Hi @tom7UBLC 

 


@tom7UBLC wrote:

As you pointed out, I would like to get a certain amount of detail in the design so I can do realistic simulations. Should this be done at a later stage?

 


So I guess we should be clear about the design intent. What do you want to achieve from the design? You mentioned simulation, can you expand on this a bit?

 

@tom7UBLC wrote:

 

The actual thickness of the van body (i.e. the reinforcing members) is about 50mm. However, if I try to thicken the surface body inward, I can only do so by 0.1mm on some of the surfaces and some just won't (the sides).

Thickening a surface to create a solid will only work if the surface geometry is sufficiently prepared to avoid surfaces in the resulting body overlapping. Looking a your model overlaps are very likely because the surfaces are not very well developed at this stage.

 

The way around this would be to create a copy of the surfaces that you have, scaling the copied surfaces about the centre of the model by the appropriate amount and then lofting the edges to meet your original surfaces.

 

I don't mean to deter you from your aim but your difficulties suggest that you will need to think carefully about how you are going to approach this model depending on what you want to achieve. If the intent is to perform some other function and the surface modelling is just a way of getting some geometry for reference you could consider using a model from Grabcad or similar resource.

 

https://grabcad.com/library/mercedes-benz-sprinter-1

 

It your intention is to model this van and in doing so learn about the tools available in the surface environment I would suggest we start by attempting a small section of the van like the front qarter panel so we do not have to take so much into account in the modelling process.

 

 

 

 

 

 

 

 

 

 

Message 7 of 11

tom7UBLC
Contributor
Contributor

@beresfordromeo, apologies for my late answer: didn't get any notification.


So I guess we should be clear about the design intent. What do you want to achieve from the design? You mentioned simulation, can you expand on this a bit?

 


I want to use it to show what my camper van will look like when it's finished. The main objective is to show the inside of the van, mainly the top. I'm designing a bed frame that will mount on the ceiling and want to show what it will look like. If the outside could look realistic with a reasonable effort, that would definitely help me too: I could show how the solar will come onto the roof.

 


Thickening a surface to create a solid will only work if the surface geometry is sufficiently prepared to avoid surfaces in the resulting body overlapping. Looking a your model overlaps are very likely because the surfaces are not very well developed at this stage.


"not very well developed": do you mean insufficiently detailed or just poorly designed? I'm trying to understand how they could be more "developed" in the context of avoiding overlaps.


The way around this would be to create a copy of the surfaces that you have, scaling the copied surfaces about the centre of the model by the appropriate amount and then lofting the edges to meet your original surfaces.

This is new to me so I will look into this, hoping to figure it out by myself.


 

I don't mean to deter you from your aim but your difficulties suggest that you will need to think carefully about how you are going to approach this model depending on what you want to achieve. If the intent is to perform some other function and the surface modelling is just a way of getting some geometry for reference you could consider using a model from Grabcad or similar resource.

 

https://grabcad.com/library/mercedes-benz-sprinter-1

 

It your intention is to model this van and in doing so learn about the tools available in the surface environment I would suggest we start by attempting a small section of the van like the front qarter panel so we do not have to take so much into account in the modelling process.

 


I will check out the Sprinters on Grabcad too. Maybe that'll do.

Thanks so much for your help!

0 Likes
Message 8 of 11

beresfordromeo
Advocate
Advocate
Accepted solution

Hi @tom7UBLC 

 

"not very well developed": do you mean insufficiently detailed or just poorly designed? I'm trying to understand how they could be more "developed" in the context of avoiding overlaps.

 

By this I mean, the surfaces that have resulted from your lofts are not leading to smooth transitions. Mostly this is because the loft tool does not have sufficient geometry to drive it and the resulting surfaces may be a little stretched or pinched in places. This is difficult to describe without examples but perhaps you could think of it like a jelly mould. If the mould is too 'tight' in some places, the jelly will not want to come out in one piece. I am sorry if this is a terrible analagy but it is the best I can think of at this time.

 

In the Topology with T-splines video by namenotimportant he discusses how topology is used to drive the important features of a t-spline model. I know you are using the surface environment and not the form tool but the principle still applies. Where surfaces make difficult transitions it is important to give the loft or patch tool more information so that the transition is properly controlled to give you the result you want; namenotimprotant describes this as ring-around-the-detail. 

 

https://www.youtube.com/watch?v=BAQgvASaOPg

 

If you have a surface which performs as you wish it to except for some small aspect of it, it is worth considering cutting the surface at that point with some relevant geometry and then lofting or patching in that area with some new geometry to drive the result.

 

Hopefully you will have time to improve your understanding of surface environement in fusion because I think it is a useful and undervalued part of the software, mostly people avoid it perhaps because of the level of understanding needed to get good results. When combined with advanced sketching techniques however, the surface environment can be really rewarding.

 

I am sorry that I have not been of much help at this stage but if I can help in future please feel free to ask.

 

 

 

 

 

Message 9 of 11

TrippyLighting
Consultant
Consultant

@beresfordromeo wrote:

 

Hopefully you will have time to improve your understanding of surface environement in fusion because I think it is a useful and undervalued part of the software, mostly people avoid it perhaps because of the level of understanding needed to get good results. When combined with advanced sketching techniques however, the surface environment can be really rewarding.

 


I could not agree more. Parametric Solid Modeling was really developed to speed up engineering workflows. Its focus was never to develop complex and visually appealing geometry.

That is indeed the focus of surfacing techniques.

However er, it takes time to develop the proper techniques and habits and a vehicle isn't necessarily the best object to start with.

 

 


EESignature

Message 10 of 11

tom7UBLC
Contributor
Contributor

Thank you so much! You've been tremendously helpful by exposing concepts I didn't know about the surface environment, how it could be used and what it's purpose was: this all goes toward my training :+1

Thanks for sharing the link to namenotimportant's videos, wow!

 

0 Likes
Message 11 of 11

beresfordromeo
Advocate
Advocate

@tom7UBLC wrote:

Thank you so much! You've been tremendously helpful by exposing concepts I didn't know about the surface environment, how it could be used and what it's purpose was: this all goes toward my training :+1

Thanks for sharing the link to namenotimportant's videos, wow!

 


You are most welcome. It is always a pleasure to help someone who wants to learn so my thanks to you for listening.

 

As @TrippyLighting has wisely stated, the surfacing environment rewards those who think more deeply about their approach so in time you will fail and learn and succeed, hopefully I or someone with far more experience like @TrippyLighting will be able to help you in the future and in the process we may learn something new, because each model often brings new challenges and therefore new solutions. As @davebYYPCU also wisely pointed out, the easy button rarely gives good results. 

 

Best of luck with your project.

 

 

 

0 Likes