Full Version: Fillets again

Hello

I am trying to model a propeller and spinner assemby for a model aircraft, largly speaking I have got the model where I want it.

However I cannot get anything like a sensible fillet from the blade section into the spinner cone.

I suspect that this is probably due to my poor modelling skills but if anyone could point me in the right direction I would be very grateful.

Best Regards

Barrie

Attachments:
2020_PROP_1.3dm

---

Hi Barrie, well there's several things that will make that difficult to fillet, one of the biggest is that the leading edge has a quite sharp valley in a very small area where it comes together rather than being smooth:





That's pretty unlikely that the edge-based filleter could make any sense of that type of juncture between fillet pieces.

One other approach you could try would be to trim away some area to make both a larger hole and to cut back the blade to open up some space and then put in a Blend surface:





The other thing that could potentially give you some fillet surfaces to work with would be to do some surface/surface filleting. That's where you use Edit > Separate to break out pieces into individual unconnected surfaces, then pick 2 surfaces (whole surfaces instead of edges) and then run Fillet. That can create some fillet surfaces that you might be able to use parts of where the edge-based filleter won't have any chance.

The other thing that's difficult about these kinds of fillets are figuring out what should be done on the trailing sharp edge.

Another thing that you could try would be to use either blend or surface/surface fillet as described above to build some initial connecting surfaces, and then use Construct > Curve > Iso to extract some isocurves from them to give you some cross section curves to use in network or sweep to build your connecting piece although it won't be completely smooth.

- Michael

Image Attachments:
barrie_fillet1.jpg  barrie_fillet2.jpg  barrie_fillet3.jpg  barrie_fillet4.jpg 

---

I've attached here what it looks like putting in some fillet surfaces using surface/surface filleting. They kind of swoop off a long ways then you cut them back using Edit > Trim with the "isocurve" option.

Maybe blend is better than fillet anyway because notice how regular constant radius fillets do not naturally line up like you probably want at the trailing edge.

- Michael

Attachments:
2020_PROP_2.3dm

---

Hi Barrie,

Michael pointed out a big issue with your model. If I was in your shoes I would try recreating the blade so that the le is smooth. Also, you want to have a circular arc trailing edge. What I have been doing is making a solid blade and a solid hub or spinner in your case. Then boolean those. Then I check to make sure it is one solid. If you get that far, you can then fillet the edge. I have found Rhino and MoI to be very hit and miss with these things. One thing that seems to be a problem is if there is a seam in this area. I usually have to rotate the hub/spinner so the seam for that part is not in the area you want to fillet. I don't know why this is an issue, perhaps Michael can shed light on it. Lastly, you can experiment with the fillet radius. Usually your te radius will limit the size of the fillet you can put at the root of the blade. So if you want a really big fillet you will have to increase the te radius. You can get this to work, but it is frustrating and takes a lot of trial and error. What I wrote above, is what I have learned with battling with this type of thing for a long time. In Rhino I check the validity of the solid by calculating volume for it. If Rhino fails to calculate a volume for each part, chances are the fillet operation won't work. Also, the final solid (before filleting) should be able to calculate a volume. If not, Rhino would fail to fillet. MoI doesn't have that ability. But from what I saw in one of Michael's recent posts, it looks like volume calculation is coming with v4. However, I don't know if this is a good indicator of fillet success for MoI (as I haven't played with the v4 beta).

Also, many years ago a forum user here named Burrman figured out the NACA airfoil leading edges are elliptical. So if you use the elliptical curve tool to create the le, that will get you the correct airfoil le. I've never had any issue with the fillet failing at the le using this technique. Which burrman taught me here, ages ago. The suction and pressure curves of the airfoils I create using the NACA definition. Never had issues there either. Usually the failure happens around the te because you want a root fillet that is bigger than geometrically possible to create or if there is a model seam in the root fillet area. Any other issue has been because of subtlety corrupt geometry (like Michael pointed out with your model). I've had many models not get created the way I thought and had to redo them, so you're not alone. Usually, the calculate volume trick tells me if the model is good or not. This is because, for me at least, whatever is wrong with the model you can't really see most of the time. Sometimes, I have to just abandoned whatever method I tried to make the blade, and go with a different approach. After years of doing these models, I found just using a loft of various airfoil shapes was the best way. Otherwise, rhino would screw it up. Doing a loft is about the only sure fire way of getting the exact blade shape right. I usually use 13 airfoils along the blade. I position them where they are supposed to be and then loft it. If you have the ellipse at the le and the arc at the te everything should work out. You end up having to redo it all if you aren't happy with the size of the root fillet that you can generate.

Anthony

---

Also,

These screencasts, I created awhile ago, touch on the stuff I mentioned:

https://youtu.be/R5DF822U89w

https://youtu.be/jNkNNawmfJ0

Ironically, the green model shown in the screencast is a case that just won't work with Rhino, even though it looks fine. It won't calculate a volume. Also, as mentioned in the screencast, I have had instances where Rhino will calculate a volume but the exported step file is not valid. So if you try to mesh it for FEA it won't work. So generating the root fillet is a huge pain in the butt. Even with years of practice. It's hit and miss with no clear idea of what is going wrong. Michael has talked about getting a new fillet engine at some point. That would be welcomed.

The automation of the blade creation has been worked on from time to time by forum users here. But, as far as I know, nothing has been completed yet.

---

Hello Michael and Anthony

Thank you very much for your comprehensive replies.

I will read the replies in detail this evening and I think I will then be able to get a model that we can use to machine a prop mould from.

I create the blade surefaces with a network surface using two aerofoils, one at the root and one fairly close to the tip and with a wire for the leading and trailing edge.

I think my aerofoil might have been a little crude and rustic at the leading edge, I will get an elipse onto the leading edge.

Burrman is a clever guy, I think he is still around, I bump into him on the Bobcad forum sometimes.

Many Thanks

Barrie

---

I went back and tried fixing that model shown in the screencast. I've tried before but didn't have success. I think I found the problem. Basically the model with the airfoils projected onto cylinders will only allow a root fillet radius of .08mm while the model with the airfoils not projected onto cylinders will allow a radius of 1mm. You can actually see where the fillet breaks in some cases. I just kept trying different fillet values. Basically, I tried a lot more values than I did in the past. The thing is, .08mm is really small and undesirable. That's why I never went that low before. This was all done with Rhino 5, but in the past it seemed to work almost exactly like MoI v2 trial. I've been waiting for MoI v4 to come out, so I can switch from Rhino to MoI.

I don't know if MoI behaves the same as Rhino wrt fillets. Rhino will create invalid fillets with no warning. Calculating volume is how I check to see if the operation was actually valid. As mentioned before, with Rhino 5, I have had some models calculate a volume but still have an invalid step file exported. So this whole root fillet thing is a giant pain in the butt.

But the main thing is that you have to experiment with root fillet radius values. You also get limited by the te radius. So it's a lot of trial and error, even if you created good geometry up to this operation. One other thing I've found as far as model quality. When you go to fillet that edge, it should be one curve when you select it. If the model has you selecting multiple edges, chances are it's not going to fillet correctly. I also found the blend edge was no different than the fillet edge.

---

Hi Anthony, do you have the model from the screencast posted somewhere?

- Michael

---

Hi Michael,

I just made changes to the file to get it to fillet. I can work on it some more so there is a solid prior to filleting. I don't want to hijack this thread though. I'll create a different thread. I was just trying to share what I have learned with doing the same thing the original poster is trying to do. For me, it's been hit and miss with Rhino 5. This is one of the main reasons I want to switch to MoI. I haven't used MoI since the v2 trial. I went with Rhino because it had some features I needed that MoI didn't have. But it looks like MoI v4 will have those features, so I'm looking forward to switching.

I think what goes on is that, for a given geometry, there is an unknown max radius for the rolling ball fillet. So you end up having to manually iterate. It would be very helpful if there was at least some feedback saying that a good fillet can't be created for the specified radius. Instead, Rhino just puts in a bad fillet and corrupts the solid. Ideally, it would be great if the software told you what the max fillet radius is. That would solve the problem completely. Also, if you know why having a seam in the area you want to fillet is a problem, I'd love to know.

---

Hello Michael and Anthony

Thanks for your help with the spinner prop project, I have got a result that I think I can forward to tool production with.

I used a blended surface, I trimmed a cut out in the cone to 0.75mm bigger than where the blade joined and then trimmed the blade out by 0.75mm and then blended between the two.

I can see that my modelling skills need refining somewhat going forwards.

Best Regards

Barrie

Image Attachments:
fillet.jpg  fillet_trailing_edge.jpg 

---