I need some help with a probing problem

[steve4460]

Well I used a shopbot probe to scan in a rudder for a sail boat and I need to make a 2 part mold from that.Can anybody help ?.

[gerald_d]

Why don't you take the mold straight from the sample rudder?

[steve4460]

We can't take the chance to damage the rudder in any way . How would you go about that anyway without spending a lot of money for the material to do that ?. Plus the shopbot shoulde be able to do that and we would like to figure that out .
I have millwizzard to make the mold but after scanning the file in I can't seam to be able to open it up in mill wizzard or convert it on ether one of my two computers .The probe to surface translator is making a file that is 4.12 GB big after running all night ,and that dosen't look right .Turbo Cad says that the 3d face record is missing . If you have any ideas that would be great . I can send you the file if you like to help , that one is only 170kb .
Thanks
Stephan

[gerald_d]

I'm positive that you will find a "mold release" coating that is affordable and that will just wash off the original rudder with water. But the regular boatbuilders here will be able to give better advice on this.

[billp]

Go to www.smoothon.com (http://www.smoothon.com). They have ALL the info you need to make molds. They also sell a sicicone mold release in a spray can which you could coat the original rudder with, and then pour a mold around it...Some people even use PVA which is water soluble for the release agent.

[Brady Watson]

Stephan,
It seems to me that you are trying to create a mold to make FIBERGLASS parts...Is this correct? If this is the case then you do not need the Bot for this project.

Buy the videos from FiberGlast (http://www.fiberglast.com) and learn how to make the molds with good visual examples and material references. You are not going to get the expertise necessary from just info on message boards. It is important that you actually see it being done to fully 'get it'.

-Brady

[steve4460]

Hi Brady

That is correct we would like to run that in to some MDF since we have that laying around here. That mold rubber stuff and whatever else we need to do that is quit expensive . So thats way we would like to mke it with the bot since the MDF is free .Plus I would like to learn how to do things like that with the bot so that I can make molds of other stuff too.
Got any ideas ?.

[Brady Watson]

You have mail.

[bleeth]

Steven-Gerald is correct when he says that you can make a mold without damaging the original part. Wax the heck out of the part, spray it with PVA, and then spray it with molding Gelcoat. After that you lay it up with several layers of chopped strand and reinforce as necessary with ribs usually made of cardboard tubes or foam or plywood struts which are attached to the strand with hot melt or whatever (even bondo) and then glassed over with some fiberglass cloth. You can then pop your mold and clean off the wax with Acetone. There are enough marine manufacturers in your neck of the woods that getting hold of an FRP resin and cloth supplier should be no problem. Forget the "rubber mold" system. It is not applicable to your use. I would feel safe using the above system whether the original part was glass or laminated wood and building FRP boats was my thing for over 20 years. For very large or shapely parts (as in whole decks or hulls) you glass in a couple of ports into which you can either blow air or water to help the release step. For a smaller rudder (which it must be or you wouldn't be able to fit it on a bot) You will be able to pop the part from the mold by banging on the mold with a good dead blow hammer after it has properly cured. Take it easy on the MEK-P so you don't get your resin too hot and it will be a done deal.

Dave

[steve4460]

Hi Dave
That sounds like the way to go, however we wanted to keep the cost down on buying all these extra resins etc since we have the MDF and the bot , and I would like to learn how to make molds with the bot .After talking to Brady I learned that there will be quit a bit of learning and some major bucks for the software involved. So we will do some number crunching and see what happens.

Thanks .
Stephan

[rgustav]

Hi Stephan,

If you want I should be able to get your probe data into a usable DXF format. When I was goofing around with our probe a few months back I wrote some programs to read the probe data files and do some conversions to alternate formats. I could dust them off and look at your data this weekend if you want, just PM me if you're interested.

Later,
Ryan

[gerald_d]

Stephan, I don't want you to forget about the idea of probing/botting altogether. It is just that making a one-to-one copy of a rudder is far more efficiently done the traditional way. However, if you wanted that rudder made at half-scale, then the...

probing / number-crunching / numerical smoothing / indexing the mold halves for registration / roughing out the moulds / stress-relieving the MDF / finish cutting / rough sanding to shape / treating the MDF / finish sanding / re-sealing / etc.

....is the tedious route to follow.

[steve4460]

Well the way it turns out , I have to make a more detailed scan of the rudder . The first time I used 0.500 probe with 0.500 stepover and that isn't good enough . So I will have to run the bot for about 40 hours to get a new scan .Well it is beginning to sound like that we will pull a copy of it using the traditional way .I have to see what my buddy thinks of all that .
Thanks

Stephan

[mikejohn]

Surely a rudder is a fairly smooth surface?
Can't you scan at intervals and interpolate the intermediate lines/tool paths?

...............Mike

[steve4460]

Hi Mike

How would I go about doing that ?.
interpolate the intermediate lines ?.

Thanks .
Stephan

[rgustav]

If you have a point cloud from the probing, and you have Rhino3D, it's something like:

1) Import the point cload into Rhino (I entered this one by hand)
2) Select all the Points and make a Point Cloud. (Curve->Point Cloud->Add Points)


6506

3) Create various Cross Section curves that cut through the point cloud grid. (Curve->Point Cloud->Point Cloud Section)


6507

4) Select all those curves and create a Surface from them. (Surface->Curve Network) This one is shaded to make it easier to visualize the surface.


6508

5) Export as a DXF, import into whatever you use to make your toolpaths.

Of course accuracy of this derived surface with respect to your real world object is a difficult judge, it's related to the density of your point cloud w.r.t real size and the quirks of the curve fitting/interpolation algorithm being used. For example, what you get on the edges of the surface may not be particularly accurate and may require some hand editing to bring in line with reality.

Later,
Ryan

[steve4460]

That is over my head ,I guess its time to go to school for that.
That sounds like a six month course to learn how that all works .
Thanks
Stephan

[rgustav]

Stephan,

Dude, you're in Minnesota too!

It's not that hard, Rhino does all the work for ya really. It's just a matter of learning how to use the tools. The tools cost though. ;) You can download a demo version of Rhino if you ever feel like playing around with it.

Anyway, good luck with whatever approach you decide to take with your problem.

Ryan

[steve4460]

Jup . I guess there is only about 4 bots in MN.
Thanks for that rino info , I will check it out .
Thanks
Stephan

[hespj]

Aren't all the methods suggested above a blatent infringment of copyright? (Although I guess if the Rhino point cloud needed any significant editing it might be a new creation). It would certainly look like it if the whole hull were being copied.

Am I right in remembering that any bright ideas posted on this forum became the property of ShopBot? I guess that works both ways.....
I suggest they pull this thread.

[gerald_d]

John, before I made my first post, I considered that it was a prototype plug that needs a production mold? I am more contemplative of the fact that they want to mold a rudder and not build it up from a stong core........

[hespj]

Gerald, a "spade" rudder projecting out of the bottom of a boat certainly needs to be mounted on something very strong projecting up into the boat (SS tube) - it's under considerable bending load as well as torque. Rudders hung top and bottom are not subjected to these large bending loads.

Even if one is building a rudder around a strong, load bearing rudder stock, the stock can be sandwiched between two fibreglass mouldings. Spade ruddders are necessarily quite thick - they have to contain a beefy rudder stock - so the two half mouldings would be hollow. Being thick they have complex hydrodynamic shapes which are jelously guarded by designers...........

[steve4460]

This would be a copy of a rudder that gets attached to the back of the boat . From what I understand the company has stoped making this boat for a long time plus there where only 200 made of it .
John what do you mean by rudder stock ?.
My freind was thinking that after makeing two halfs , to join them together and then fill it with concrete and some wire mesh. Is that not how it is done ?.

[gerald_d]

*deleted post* pintles on transom?

[hespj]

"John what do you mean by rudder stock ?"

Stephen, the rudder stock is the part that connects the blade of the rudder to the steering mechanism on deck. On a rudder mounted under a boat the rudder stock is often a stainless steel tube firmly attached to the rudder blade, and extending up into the boat where is has a tiller or a quadrant for wheel steering attached.

On a transom hung rudder it is that part of the rudder between waterline and deck, which is not as wide as the blade, but thicker. This transmits the torque from the tiller down to the rudder blade.

"My freind was thinking that after makeing two halfs , to join them together and then fill it with concrete and some wire mesh. Is that not how it is done ?"

I have seen transom hung rudders made in two halves in female moulds, but I've never heard of anybody filling the mouldings with concrete and wire mesh. Whilst it could be strong, it would be very heavy, and you really should keep the ends of boats as light as possible.

The rudders I've seen made in two halves had the mouldings filled with pour in foam, the foam and any fibreglass above the top of the mould trimmed off (the flat top of each mould is the centreline of the rudder), and the two foam filled halves bonded together. If a decent layup is used for the fibreglass this is quite strong enough. "The difficulty isn't designing it strong enough, but designing it just strong enough" a famous boat designer once said. I'm sure he didn't use concrete and wire mesh.

Having said all that, unless you're going to be producing a large number of these rudders a far quicker way is to cut the rudder out of plywood, using several layers where necessary, and giving the blade a bit of taper towards its trailing edge (not much), round off the corners and then sheath in glass cloth and epoxy. With this method you needn't worry about copyright issues either. (Although your concrete/wire mesh idea adds a certain amount of uniqueness to the copying method).

"pintles on transom?"

Gerald, that's why sailing is so popular, because it has its own words which allow the formation of an exclusive subculture. The words should be pronounced with a peaked cap with an anchor on it.