CNC Conversion

A forum for contacting other builders of Ken Hankinson designs. These designs are now a part of the Glen-L family.

Moderators: Bill Edmundson, billy c

meb46
Posts: 7
Joined: Thu Sep 06, 2018 9:18 pm

CNC Conversion

Post by meb46 » Wed Oct 17, 2018 5:52 am

Having spent the last couple of weeks making my eyes go square from looking at Solidworks hour upon hour, I have found the migration of the Barrelback 19 (stretched to a 20' version) plans into a digital format and enabling me to CNC cut the frames to be a pretty frustrating process. I'm fairly capable with Solidworks, but have found most methods of measuring off the plans and transferring into digital format to really lack the required accuracy, and then when in digital format really requiring extensive modifications to ensure stringers and chine logs actual correlate into the correct places.

Has anyone completed the task previously to a suitable level? At the moment, I am contemplating looking at alternative designs that are more focused on CNC approach, but wanted to check if anyone has had success on creating CNC files off the Barrelback 19' plans?

Excuse the heathen approach of CNC, its very non-wooden boat building process, but its a preference and capability I have that really enables accurate and precision manufacturing.

Keen to hear others experiences and how they made CNC Frames make sense.

Thanks

User avatar
sproggy
Posts: 233
Joined: Sun Jan 01, 2017 12:04 pm
Location: Welwyn Garden City, UK

Re: CNC Conversion

Post by sproggy » Wed Oct 17, 2018 6:48 am

I know this isn't what you want to hear but these boats really don't need precision manufacturing. I started modelling one of the Glen-L designs (modified Overniter) in a ship design application but the measurements didn't come over well or fair correctly. So I gave up and did my modifications the old fashioned way.

Because pretty much everything (frames, chines, sheers, battens, keel) gets faired during the build they end up not the same size, at least on one face, as they are measured to on the plans. So transferring plan dimensions to CAD simply doesn't work unless you do a lot of work to account for the effects of fairing.

Think how far you'd have got actually making your boat if you'd spend your CAD time in the workshop :wink:

User avatar
NAMEngJS
Posts: 76
Joined: Wed May 27, 2015 10:21 am
Location: Metairie, LA

Re: CNC Conversion

Post by NAMEngJS » Wed Oct 17, 2018 8:03 am

I will agree with sproggy on this one as well. CNC is a great tool to have but in the end the amount of work that will be done with the CNC where absolute precision is required is not there. In shipbuilding there is a term called mold lines. The mold line is the exact line of the ship at an infinitely thin sliver. since material has a thickness that has to be accounted for and this thickness is thrown to one side of the mold line or another. take the screen shot below:
fairing.JPG
If we look at the frame on the right of the screen and the purple line defining the batten line we can notice that the purple line intersects the frame outer edge on the aft side. Forward the curve of the batten takes it further inboard than the forward edge of this frame. the material on the frame will have to be faired away. The CNC my be able to get you started but the continuous bevel on the forward side of this frame would be difficult to reproduce and the chances of every single point of contact and/or every bit of fairing required cant be take out of play by the CNC.

Again, it can be useful if it helps you identify how the build will go together and/or eliminate errors that dont involve expensive lumber. However at some point you are going to have to attack the build with 80 grit paper to fair something.
-Juan Suarez

All men dream, but not equally. Those who dream by night in the recesses of their minds, wake to find that it was vanity: but the dreamers by day are dangerous men, for they may act on their dreams with open eyes, to make them possible.

User avatar
Roberta
Posts: 6404
Joined: Tue Jul 31, 2007 5:23 pm
Location: East Troy, Wisconsin

Re: CNC Conversion

Post by Roberta » Wed Oct 17, 2018 8:18 am

Since wood species and density plays such a big part on how the boats ultimately get finished, it is really difficult to predict how any given parts need to be cut in order to accurately cnc them. You could build the boat the old fashioned way and scan the parts as built into a cad program and still the parts may not accurately assemble in the same way. How the shears and chines bend is largely dependent on the wood used and it's similar properties and one piece may arc very differently than corresponding parts making frame width, height, and setpoints fit for a given arc, but not to another. The arcs drawn on plan sheets may or may not match actual arcs of the bend in the woods used, making it difficult to accurately predict frame dimensions and setpoints. That is why wood boats are as individual as the builders and so much of the wood ends up on the floor. A sharp eye for fairing and the expectation that in process adjustments is necessary. A cnc made part may or may not be the best approach. In many cases some cnc parts could be beneficial, but these wooden boats are unlikely to go together like a Revell Authentikit. Best to expect personal involvement and enjoy the experience of a wooden boat you hand crafted.

Roberta :D
Roberta "Queen of the Boat Builders"
Built Zip "Oliver IV", Super Spartan "Jimmy 70", and Torpedo "The Glen L".

User avatar
mrintense
Posts: 3312
Joined: Thu Jun 10, 2010 11:22 am
Location: Austin, Texas
Contact:

Re: CNC Conversion

Post by mrintense » Wed Oct 17, 2018 10:57 am

I agree with Roberta on this as well. I think that far too many of us are engineering types and it feels wrong to go into something with having everything to the umpteemth fraction of an inch. But the reality of boat building is that it is a hand crafted endeavor and like Roberta said, the wood may not cooperate regardless of how accurate you are .

Far better to simply take Glen L's plans for what they are and get started building. I can't begin to tell you the number of people who have come on to this forum and wanted to do the CAD / CNC thing and then you never hear from them again. I think it's precisely for the reasons you mentioned, it can be a frustrating experience. On the other hand if you dive in head first, you'll make some mistakes for sure, but usually they are recoverable and you will have far more fun.

Come on in, the water's fine. No toe dipping necessary!! :D :D
Carl
a.k.a. Clipper

Crafting a classically styled Vera Cruise

Clipper's Vera Cruise Build

Hercdrvr
Posts: 947
Joined: Mon May 11, 2015 12:52 pm
Location: McKinney TX

Re: CNC Conversion

Post by Hercdrvr » Wed Oct 17, 2018 11:24 am

Well said Roberta,
Matt B

TomB
Posts: 622
Joined: Wed Jan 18, 2017 4:07 pm
Location: Holland, MI

Re: CNC Conversion

Post by TomB » Wed Oct 17, 2018 3:27 pm

The guys that built Saetta Classic Boats, barrelback transom and Tahoe stem, used CNC for their frames. They imagined selling a kit for home assembly. If they had received orders CNC would have been great.

I have the barrelback plans and studied them extensively while wishing I had room for a build. There are many curved frame parts and making right and left hand duplicates concerned me. CNC would make the pieces quickly and accurately if you have the data to feed the machine. As an alternative, I made templates. A 1/4" of plywood or fiberboard, tracing from the plans, rough cutting and sanding to make the templates look like the drawing got me templates in a couple of days. Then use the templates to make frames, another couple of days at most (I'm now building a bigger boat and saving the barrelback templates for the next build :D ).

Tom

User avatar
kens
Posts: 4708
Joined: Wed Nov 12, 2003 5:25 pm
Location: Coastal Georgia

Re: CNC Conversion

Post by kens » Wed Oct 17, 2018 6:11 pm

Maybe I'm old-school, maybe I'm expecting immediate results,
But I have to ask why do guys take a set of plans, and then, try to interject them onto CNC, CAD, or any numerous computer program???
My gosh, the time it takes you to input all the info, then twiddle with it, then fret about it, then cut the parts, then start the build, then put a shim in it SOMEWHERE and to think that you started off with a full set of plans that are already known to work.
I must ask why?
My perspective is to take all the time involved for inputting the CAD/CAM/CNC ,,,,,,,and just build the damn boat!!!!

I apologise for my rant.
Oak is over rated, everything about it takes extra time; then it warps, splits or checks !!! :roll:

Hercdrvr
Posts: 947
Joined: Mon May 11, 2015 12:52 pm
Location: McKinney TX

Re: CNC Conversion

Post by Hercdrvr » Wed Oct 17, 2018 8:36 pm

Amen to that!

PeterG
Posts: 699
Joined: Mon Dec 03, 2007 7:08 am
Location: Connecticut

Re: CNC Conversion

Post by PeterG » Wed Oct 17, 2018 9:16 pm

So I think if one wants to work in a CAD or 3D design program that's OK, especially if it's a learning experience and helps one understand the complex shapes and curves with a boat hull. It's a good tool for design and visualization as well as repetitive production. Done well it will get you really close to what you want to produce. It's all in the effort you're willing to put into it. The vast majority of us don't find it very helpful and don't have the skills for it on our projects which are usually one-offs. I have even seen it done badly on some boat designs where the structure wasn't faired properly. I was taught old school lofting on hands and knees in a shipyard mold loft (they are strictly CAD now) that's a lot of fun on a 425 foot ship drawn in full size profile and plan. Those skills helped me tremendously with my boat project but weren't really necessary. That's my position on CAD efforts. Bravo for trying CAD, just understand it's not the end all be all for boat design or construction, just another tool. This is from someonene who is more likely to grab a sharp hand tool than a power tool...
Murphy's Law: Anything that can go wrong, will go wrong.
Griffin's Law: Murphy was an optimist.

meb46
Posts: 7
Joined: Thu Sep 06, 2018 9:18 pm

Re: CNC Conversion

Post by meb46 » Wed Oct 17, 2018 9:30 pm

Team - Many thanks for your valued input, and to be honest, very much the response that I expected given the nature of the builds. As what was touched on briefly, I do use the Modelling process as part of my visualization process and to work through the understanding of the structure, and helps me work out my build process and technique, not to mention modelling out some modifications.

I run a couple of shipyards in Asia, and have a pretty sound understanding of building boats in Aluminium and Steel, but the classic wooden boat build process is new. Typically my vessels are 100% modelled before any metal is cut, so the CNC/3D CAD habits are hard to break :)

Thanks again for the input, much valued!

Imanny
Posts: 1
Joined: Mon Dec 31, 2018 5:05 pm
Location: Burkina Faso
Contact:

CNC Conversion

Post by Imanny » Mon Feb 04, 2019 9:41 am

When do you think youll start working on a SX3 type conversion?
I have a G0619 and the CNCFusion Conversion kit already assuming thats the route you will take I would definitely want to be part of that one when and if you do

Also
Have you thought about CNC Lathe Conversions
Theres even less out there on those than on the MILL conversions

Kevin Morin
Posts: 739
Joined: Thu Jan 24, 2008 11:36 am
Location: Kenai, Alaska

Re: CNC Conversion

Post by Kevin Morin » Sun Mar 24, 2019 4:11 pm

While not truly conversant with the actual plans for the many pretty wooden boats in the Glen-L catalog- on thing that seems pretty important in any frame cutting exercise is that the NC router would have to be four or five axis? a 3Axis machine couldn't hope to put a bevel (per station) on the frames' outer edges that I can understand?

As to entering the information into a CAD package- since the very lines on the hardcopy/paper/physical template are wider than a CAD "line" - it seems to me that converting from paper to CAD would require redoing the hull models' surfaces from scratch- fairing the model- checking surface uniformity: THEN using the 'new' CAD model outputting frames from fore or after faces or perhaps a middle or center median of the frame's rough material?

Then a typcical 3Axis router could cut the rough blanks- but without the other one or two axis on the router (?) how could an NC cut path account for the bevel/outer edge angle to the transverse?

As to notching for stringers- unless a VERY fine finish ball end mill were able to reach into the notch 100% and line in some extremely finely rounded corners- the notches for a typical stringer/longeron/scantling seems like it would be kind of wasted effort. And here again, unless the router had a very fine end ball mill of sufficient strength to cut the sloped, very possibly angled and out of plane notches..... it seems hard to understand what cutting equipment would make a decently shaped notch in a 3/4" sided frame?

I guess I can visualize a notch cut with a bit the correct or corresponding size that was an indication of where a finish saw would be used to 'finish out' the stringer's cuts, but it seems a bit of work for so little time "savings"?

On the other hand, if a building mold set were being created- cutting MDF into unbeveled sections/station frames that were beveled and faired on the actual jig- and kept for repeated builds of one hull (??)does make some sense in terms of retaining the frames. But it still seems like an awfully high end NC machine is going to be needed to put accurate CAN/CAM lofted bevels on compound curved hulls' permanent frames?

I've used surface modeling and CAD/CAM cuts metal boats, but there a typical frame is less than 1/4" thickness and since the boat is joined with welding the tiny fraction of an inch of gap at teh edges formed by '90deg' edges on cut outs; can be ignored.

As noted by the many wood builders above; it seems like a place where the effort or trying to cut transverse frames with NC is not returned in time saved or better small boat building- unless you were going to do a bunch of builds off one set of cut files AND you had a more than 3Axis NC router to do the edge cuts??

Cheers,
Kevin Morin
Kenai, AK
Kevin Morin

User avatar
Jimbob
Posts: 729
Joined: Thu Jul 25, 2013 1:07 pm
Location: Sacramento, CA

Re: CNC Conversion

Post by Jimbob » Mon Mar 25, 2019 9:53 am

I had been building furniture, turning segmented bowls, carving duck decoys, and other precision woodworking for some time. For me, I was accustomed to everything having to be perfect.

Boat building is not that precise. It is more like building a house from blueprints. The most important thing that I have found in boat building is that you want everything to be symmetrical and in alignment with other parts. For example the frames need to be aligned with the motor stringers, and each side the same from the center-line of the boat. That precision will save you grief later when you get to decking, cover-boards, and hatch covers. Fairing will be done to those "perfect frames", so I wouldn't spend a lot of time with setups for those frames. I also wouldn't pre-cut your frames for the battens. Most of the notches on your frames are determined as you place your battens. Angles on those notches will follow the battens. Make one frame the traditional way, and I think you will see what I mean. The last piece of advise is to understand the sequence of your assembly. So many things affect other things later.

I made a jig to aid in cutting the notches for the battens with my router. (pic below)

After saying all of that, I must say that now, I enjoy the boat building over the fine furniture building. It is very satisfying to see it all come together, and much less stressful.

Jim
Attachments
P1240107.JPG
Works with a flush trim router bit. Fully adjustable for angles and widths.
Jim Neeley
Building a Barrelback in Sacramento, CA
viewtopic.php?f=15&t=28089#p172969

Aalexanov
Posts: 1
Joined: Wed May 08, 2019 2:46 pm

Re: CNC Conversion

Post by Aalexanov » Wed May 08, 2019 3:00 pm

As to entering the information into a CAD package- since the very lines on the hardcopy/paper/physical template are wider than a CAD "line" - it seems to me that converting from paper to CAD would require redoing the hull models' surfaces from scratch- fairing the model- checking surface uniformity: THEN using the 'new' CAD model outputting frames from fore or after faces or perhaps a middle or center median of the frame's rough material?
You have to do something like that: https://www.shmexpert.com/feed/state-of ... -drawings

Post Reply

Return to “Hankinson designs”