Pat Sparks BOMARC Build

[LSP_Kevin]

Some of you will be aware of Pat Sparks' BOMARC thread, wherein he is designing said beast for 3D printing. Being without a 3D printer for a while, Pat has asked me to test print his files and provide him with feedback, and with his permission, I've decided to start a separate thread for this, and building the subsequent model. I'm hardly a 3D-printing wizard, so expect to see lots of failures and frustrations!

 

To see the original CAD designs, please refer to Pat's thread, as I won't duplicate them here:

 

 

Pat has continued to revise his files based on my efforts at printing them, but the first item on the build plate was the forward nose section:

 

 

 

This printed without any issues the very first time, which is a rarity for me! Emboldened, I decided to tackle what is probably the most complex part of the model, and probably the single most complex part I've ever tried to print - the centre wing section:

 

 

Close, but there are some serious issues towards the trailing edge, and I'm calling this one as 80% success. So with Pat's advice and a revised version of the file, I gave it another shot with a lot more supports:

 

 

Getting there, but still some minor distortion along the trailing edge:

 

 

The camera doesn't pick it up so well, but the distortion is much more obvious to the naked eye, and exists in both axes. Still, that's a 95% success in my estimation.

 

In the meantime, I printed out the nose cone. This worked flawlessly the first time, but came out too small! I somehow got the scaling wrong. Second time was a charm, however. I didn't get any photos of it separately, but did get one of a test assembly using what I'd printed so far:

 

 

The whole thing is 34cm high at this point, with the entire tail section still to come. The distortion in the trailing edge is also much more obvious in this photo. Over on Facebook, Roy Sutherland suggested printing the centre wing section with the leading edge down, so I thought I'd give it a try. I just used auto supports, and was not surprised that it failed:

 

 

 

But at least the trailing edge is straight! I can only give this one a 60%, however, and it's basically a write-off. I'm currently reprinting it in this orientation, but with more supports - closer to what I used for the second version. I've got my fingers crossed for this next one, but if it's not 100% (or 99% at least), I'll try to figure out what went wrong and iterate through the process again. Everything's printable with the right supports, and that's what I need to get better at. This centre wing section is a 9.5 hour print, so it's not exactly a quick learning curve.

 

Stay tuned!

 

Kev

 

[Finn]

Kev, hope the next print comes out within acceptable limits. If not perhaps try to print the outer wing sections as separate parts, that way it may make it easier for the machine to print the main centre wing section.  

 

Jari

[LSP_Kevin]

6 minutes ago, Finn said:

Kev, hope the next print comes out within acceptable limits. If not perhaps try to print the outer wing sections as separate parts, that way it may make it easier for the machine to print the main centre wing section.  

 

Jari

 

I'd have to get Pat to modify the files first, but if the next print fails, #2 is usable with some work. However, my goal here is to work out how to get the best results from Pat's designs, and get better at 3D printing in the process. A 1/32 scale BOMARC is just a bonus result!

 

Kev

[LSP_Kevin]

Well, the 4th print is done, and while it's miles better than the 3rd one, it's sadly not 100%, and not really usable.

 

 

The trailing edge prints perfectly in this orientation, and even though the outer wing panels have fully formed this time, they still exhibit significant distortion - probably from insufficient supports.

 

 

I can certainly add more supports and try again, but you'll notice that the wing probes don't print at all well in this orientation, and that's something I have no idea how to solve. The print fine the other way, but that messes up the trailing edge! The best approach would be if the outer wing panels and wing probes could be printed separately, and I think Pat is considering those modifications.

 

At a practical level, though, I could easily carve out two good wing probes from the previous prints and graft them onto one that had everything else at 100%. But I'm more interested in learning from these failures than stressing about getting a perfect part.

 

Let's see what happens next!

 

Kev

[Out2gtcha]

Nice Kev, learning along the way here.

[Dandiego]

Kev, I  share your quest for 3d printing knowledge.  For me, resin is semi- expensive. If there is a problem with a print I will try to salvage what I can and fix whatever didn't print well.

 

Those outer wing segments look problematic. I would think that if they were separate pieces the overall print would be easier to print. Also the more complex the print the more problems will arise. Pats design is amazing. Far nicer than anything I can do. For me, I have found that the simpler the shape the easier the print. The engine nacelles could also be separate, making a simpler piece to print and therefore less waste if something goes wrong.

 

Just my 2 cents. Still learning. 

 

Dan

[spartacus2000]

Kev agrees with Dan on separating the various pieces. My experience tells me this. The angles of a piece can be good on one side but problematic in some other places. My advice is "Dividi et impera" , less waste of resin and in the thin parts add many supports in order to have a wider printing surface so that the build plate presses with less pressure on the freshly cured but not so resistant layers. Better to consume a few more grams of resin for the supports than to throw away several more for a badly turned piece, especially since those pieces do not seem to be small.

Un saluto

Marco

[scvrobeson]

I appreciate you taking us on the learning journey with you. It's important that people see the good and bad parts of the new tech

 

 

 

Matt 

[Lee White]

6 hours ago, LSP_Kevin said:

Well, the 4th print is done, and while it's miles better than the 3rd one, it's sadly not 100%, and not really usable.

 

 

The trailing edge prints perfectly in this orientation, and even though the outer wing panels have fully formed this time, they still exhibit significant distortion - probably from insufficient supports.

 

 

I can certainly add more supports and try again, but you'll notice that the wing probes don't print at all well in this orientation, and that's something I have no idea how to solve. The print fine the other way, but that messes up the trailing edge! The best approach would be if the outer wing panels and wing probes could be printed separately, and I think Pat is considering those modifications.

 

At a practical level, though, I could easily carve out two good wing probes from the previous prints and graft them onto one that had everything else at 100%. But I'm more interested in learning from these failures than stressing about getting a perfect part.

 

Let's see what happens next!

 

Kev

 

Kev

The wing probes aren't parallel- is that a printing issue or a parallax issue?

[LSP_Kevin]

2 hours ago, Lee White said:

The wing probes aren't parallel- is that a printing issue or a parallax issue?

 

They're bent and twisted in all directions, and one of them has a major layer shift half way along. Totally unusable, sadly.

 

Kev

[LSP_Kevin]

Pat was kind enough to send me some modified files that separated the outer wing tips from the main centre section, so that they could be printed separately. The trailing edge also had an additional reinforcing block added to help retain its integrity, and not have to have supports attached directly to it.

 

 

I was really hopeful with this solution, but alas, it didn't quite get there - mostly due to inadequate supports on my count. There was also a weird resin blow out around the hole on the left side, designed to allow for the tips to be pinned in place.

 

While the main section is eminently repairable and usable, and I could simply reprint the tips with better supports and call it done, I decided to have one last crack at the full print, using some tips and advice generously passed to me by John Colasante. That print has in fact now been completed, so we'll take a look at that in the next update!

 

Kev

[LSP_Kevin]

So, for my one last print, I chose to revisit Pat's second iteration of the centre wing section, but this time add supports everywhere I could think of, and it actually turned out pretty well:

 

 

The trailing edge is not as nice as I'd like it to be, and I've come to the conclusion that it's not possible (for me, at least), to print it in this orientation and achieve a flawless trailing edge. On the left corner of the main wing section, just before the outer panel, you can see where the trailing edge curves upwards, due to my neglecting to put a support right on that corner. So that's a nice little lesson for next time!

 

Removing all those supports was a royal paint, however, and left the trailing edge in a bit of a state:

 

 

This will clean up fine, and I should even be able to sand everything straight. And since is no other distortion or problematic areas on the print, I'm calling this one good to go. I'm even considering grafting the perfect trailing edge sections from one of the other prints onto this one. I'll clean it up first and see how it looks.

 

I'm now printing the tail section, using what I've learning from 6 (!) goes at printing this piece, so fingers crossed I get it right first time!

 

Kev

[geedubelyer]

Is it worth designing in a sacrificial trailing edge that can be snapped off?

 

[LSP_Kevin]

41 minutes ago, geedubelyer said:

Is it worth designing in a sacrificial trailing edge that can be snapped off?

 

 

That's exactly what this version has:

 

Kev

[airscale]

what slicer software are you using Kev? and are you using the auto nsupport feature or adding them yourself?

 

I generally find with chitubox, giving the part about a 35 degree angle and using auto supports is about 95% successful

 

you can always go in and edit auto supports where they hit the work or are oddly placed