Another Shaping Day
 

Hi all

Today I consolidated and finished up the deck of the surfboard blank that was cut a few days ago.
It moved around a bit and was not as accurate as I would have liked.
Considering it was a first go to get the feeds and speeds dialed in and generally test how things went, it was a great success.

Here are a few pics of the board covered in foam dust and later cleaned up.

Attachment 12864
Attachment 12865

Here is the Mechmate ripping into the foam on video.

http://www.youtube.com/watch?v=BUWjqCgmubc
http://www.youtube.com/watch?v=7GFSWH69sL8

Lastly the hold down was still lacking and the resulting shaped board had errors nowhere near acceptable.
So like Kobus started thinking about vacuum hold down.
I jerry rigged up a piece of plywood and some closed cell self adhesive tape.
Even with this rough implementation, it worked very, very well.
In fact I had to dail down the suction as it marked the foam.
Watch what happens in my one handed phone video how it sticks to shaped EPS foam.

http://www.youtube.com/watch?v=wAnuRHKGIvM

Lots of room for innovation and improvement with my processes but still felt good about how far I got today.
Frankly still a bit stunned about the power of the vacuum hold down on the uneven foam surface.:eek:

Regards
Ross

it only need little vacuum to hold a workpiece the size you are working on...
I would suggest you double-side tape a small piece of fomica which is >3X larger than your suction cup that would hold thing down without suction mark on your foam.

Cheers Ken

Good idea, will keep that one in mind.
I have some 1mm tape that gets used as hold down quite regularly.
The remaining VHB tape that holds onto the racks, it is only used for special occasions.

Reduced the vacuum to 10 Inches Hg and the marking was no longer an issue.
The marks were occurring at around 20 inches Hg, far too much vacuum.
My trial blanks are very soft, blown around 2004 when the EPS technology was still getting the formula right in the surfboard world.
If I can get no marks at 10 Inches Hg with the soft blank, the new far more dense ones should be fine.

Still working on a better system to hold the blanks but need access to some plastics that are not available here over the December break.

Regards
Ross

Looking good, Ross.
With foam that light you'll need veneer to limit denting, and be sure to glass on a dropping temperature (late afternoon), to limit out-gassing issues. We're heading to 40kg/m3, which is probably 2-3x what you have.
I'm learning from your hold-down issues. Spackle is another way to seal so vac pads don't pull air. Water it down with clean (distilled) water to mayo consistency so it goes on smooth.

Why have you used the Boardcad cutting path, rather than exporting the dxf to a more reputable CAM program?

Hi Red

The CNC cam programs are not as sophisticated as you might think they are.
In short the issue is the blank is modeled accurately in the Boardcad app while the CNC cam apps the blank can only be modeled as a block eg x, y and z.
Even when wire cut blanks or foam blocks are used, more time is required to cut through the extra air or foam in the respective models.
That time is best spent cutting the model surface itself a little more slowly to produce a better finish.

More tellingly the cam apps will produce only a raster toolpath pattern.
This means that the gcode will follow an axis or a specified angle but not flow along the natural outline of the shape like the specialist code does.
Over time as I looked at how the typical CNC software worked I had to rule it out and instead opted to torture myself with the specialist code :).

The blanks I am using a pure rubbish and only suitable for testing.
My recent investigations lead to these guys as an EPS blank supplier http://www.coreindustries.com.au/.
The density and compression strength is quite amazing and seems for the moment ot be displacing XPS in the blown or wire cut blank market.
They have missed the mark by not supplying a specialised fish blank though.

As ever I am open to suggestion and if you can get an alternative solution together.. I'm all ears as their is no denying blocks are cheaper than blanks !!

Regards
Ross

Made me think Red

Quick calculation on the foam density of my softest blank.
Shaped volume 27.3 l or 0.0273 m3 x 850 grams shaped weight = approx 23.2kg /m3.

Regards
Ross

23 kg/m3 will dent even with veneer (I've had 5 at that density). Use hard foam inserts for leash and fin plugs. Be careful pouring epoxy into the plug holes because it heats up when pooled and melts the foam

Good info on the CAM issue, thanks.

The best (most accurate, best finish) boards I had cut were done across the rocker.

Cheers Red

None of my current crop of blanks will see any glass.
They are destined to be destroyed in my experiments on feeds, speeds, cutting depths and hold down.
While this will not relate to you most likely, the stringer cuts and the end plunges that cut off the excess timber are far trickier than the foam itself.

I like your hint of glassing on a lowering temp.

The gcode Boardcad produces is a good enough approximation but not great.
Zoomed in you can see wobbles in the cut lines.
Using an open model approximation of the bezier curves when converting into a NERBS surface supposedly improves the modeling.
So far I have not played with that (will though).

Already I have found the bottom cuts better at a lower stepover value than the deck.
The domed deck spreads out the toolpath and causes less recutting of the previous passes.

The EPS also accumulates most heavily in the middle roller on the Z slide, if you are doing any thinking on wiper brushes.
Eventually it spits off a ribbon of compressed foam as it builds up.
I cleaned the Z slide with compressed air quite successfully but get the impression the wipers would need to be pretty stiff to work as the bond is strong.

Regards
Ross

Who would have thought that reading the instructions actually works:)

Jed you might be interested in this.
The open model approximation of the bezier model produces far better Gcode.

Attachment 12868

Attachment 12867

A significant improvement on the deck code in particular.

Regards
Ross

Cheers Ross

I'll send you an email on that one as the tool path looks MUCH cleaner. I haven't looked at BC lately but know they have a beta version out at the moment.

Very Cool Surfboard
 

Just figured this would be appropriate with this build log.

http://cncrouter-info.blogspot.com/2...come-more.html

Very cool surfboard made from bamboo i believe.

Wobbly Supports
 

Hi All

I have been working on my foam hold down system recently, it sure is crawling along but I am slowly getting there.
The objective for my supports was to have them rotate around a point on the top plane of the support.
First up it occurred to me that i needed a ball joint that I found and extracted from a machine support leg.
The hex nut at the ball end was also sanded down to allow the top to rotate through a larger range of motion.
Attachment 13025

Then I machined a half moon in the top support and a mechanically fixed bottom retainer with a push fit lip
Attachment 13026

This is the support being assembled with the bottom retainer in place
Attachment 13027

Finally my wobbly support fully assembled
Attachment 13028

The plastic is UHMW poly and is self lubricating, so the joint rotates nicely.
The top will get a piece of sandpaper glued to it to trap the foam in place.

I am currently experimenting with a vacuum hold down system that uses some commercially available 2.5 bellows vacuum cups.
The cups grip the shaped foam beautifully easily bridging the cutter striations.
To find these was a great effort as the cost from most suppliers was through the roof.
Eventually I found a supplier that sold into a large packaging market and held stock at a fair price with good availability.

Now to patch these bits together into a working system !!

Regards
Ross

Nice work Ross !!

Cheers
Tony.

Surfboard Vacuum Hold Down
 

Hi Tony

Got to love the Mechmate if you can imagine it (and design it) you can have it.

Attachment 13029

This is one of three units needed, missing one vacuum cup mount as I am still short some parts as I work on my solution.
Been a while since I have done some alloy so had to put my alloy head on - so different to timber and plastics.

Regards
Ross

You sure that is not for a breast feeding mommy ?:D

Ha Ha, in all the time I was making it never thought of THAT :D
But now you mention it, I'll ask around and see if there are any takers.

Regards
Ross

MM = MOTHER's MILK not Mech Mate ;)

Ross;

THATS IMPRESSIVE.. Nice job. I think I am going to have to steal your idea for a little project I am going to do. Very nice.. Did you cut that aluminum with your MM?

Best regards,
Miguel

Hi Miguel

Yes the machine made these.
As a rule use a fairly high feed rate, cut with very shallow passes and always use a cutting lubricant of some sort.

I have found dry cutting is an instant pathway to failure.
Others using something more fancy than a two flute carbide end mill might have other solutions though.

Regards
Ross

Ross: What kind of bit are you using to cut the aluminum?

I was planning on trying to use this gizmo to see if it made any difference to the cutting. All it requires is compressed air and it will cool off the bit. I am not sure if enough tho. I don't mind using lubricant at all, but i was trying to keep any fluids from the machine, although the amount of lubricant to cut small pieces like that cant be that much.

If you do try out the vortex tube for cooling, inform us how it work, I was also looking into it as I found out that some shops are using these and dry cutting to cut on costs of coolant and its recycling.

Hi Miguel

I use ordinary carbide 2 flute cutters. Had a cutter specifically for alloy but killed it instantly doing my first dry cut test.

Some people use compressed air, just standing there with a gun and giving it a squirt.
After all air is a fluid and it is just as much about moving the heat and chips away as much as lubricating.
Some cutting fluids are oil and water emulsions , the water evaporates when heated and provides the cooling to the parts.

The cooling gun is a great idea as it also extends its use into the plastics and polycarbonates that have a real habit of melting if they are cut too aggressively.

With respect to personal safety, even with the small amount of alloy stuff I have done you quickly get tired of breathing the volatiles comes off the machine when the sprays are being used.
Gave myself a really good headache the other day from breathing the vapour.

The gun seems to be a great idea to replace a misting system, so if you buy one make sure to report back on how they go.
I have seen a price of roughly $280 but expect they might be more expensive here in Australia.

Regards
Ross

Found the first big down side - very large air requirements.
Only found one model that could run on a three cylinder belt driven compressor.
The link to the model above requires something like 25 CFM :eek:

Regards
Ross

Ross, have a look at this site. I have read a few reviews over on Pratical Machinist forum and for the most part they were all positive. The biggest negative seems to be the price. Most of them being machinist just google the patienent number and make one.

http://www.fogbuster.com/

http://www.google.com/patents?vid=US...AJ&output=text

http://www.metalworking.com/dropbox/...Fog_Mister.txt

Hey guys.. I ordered a Exair 5215 yesterday for $200 (eBay). I will let you guys know how it works once I get it.

Hi Russell and Miguel

Russell
Great links for the droplet mister.
The drawings and description are very clear.
Thanks

Miguel
Not sure what sort of air supply you have, but when you get yours let us know how it works and what sort of air supply you used to achieve it.

Generally
From what I understand the adjustable air coolers are very inefficient, that does not make them bad but does up the air requirements somewhat.
The adjustment of the units themselves also will have a big impact on the air flow and temp of the air.
This is referred to as the cold air fraction, very cold air requires lots of hot air going out the hot side of the tube.
This link explains the whole thing very well for others that may be interested.
By no co-incidence this is the supplier I was referring to that made a low flow unit capable of delivering a product that was within the 300 liters per minute maximum airflow I have available.
http://www.stream-tek.com/products/v...ifications.php

One tube can also generate a range of flows and temperatures by using interchangeable vortex generators.
This listing on ebay demonstrates how hot and cold generators and flows can be interchangeable in a kit form using one tube as a basic platform.
http://www.ebay.com.au/itm/290538748...#ht_632wt_1139


How they work - that's up to Miguel :)

Thanks guys great information.

Regards
Ross

IMHO, The easiest+cheapest+guaranteed+time tested+no-brainer way for applying coolant is by applying kerosene to the bit with a brush.
**ps, use PPE while doing it.

BTW, you get to watch every cutting moves while doing it :)

@Ken. Make sure you have a couple spare fingers on the shelves.. :)

Water color paint brush has long handle... Don't worry. the bit won't jump out & bite.

Ken, a paint brush is certainly a cheaper option :)
And I am a big fan of watching what is happening if nothing more than to be quick to the EStop especially with alloy.

I bought a $5 can of RP7 with a plastic tube nozzle today, it's also a good cheap solution. You don't even have to dip the brush ;).
I make up a dam of plastic around the part and it catches most of the chips and coolant.

Hold Down Update
Different day and another revision , driving time is good thinking time.
Definitely going to drill a few though holes on the spoilboard.
Also going to use a hollow 3/8 BSP booker rod to save some more height on the vacuum cups.
I can say without reservation this is getting expensive !!!

Regards
Ross

What is RP7? sounds like my kind of thing :D

Hold down - Three Pieces Done
 

Hi Ken

Attachment 13045

It is a heavier oil than most spray oils in a can like WD40.
The colour is tending towards a light engine oil tone rather than clear.
Really nearly anything will do.

I have machined the three assemblies that will give me the three ranges of heights I will require to hold the blank.
These are approx 90mm, 100-110mm and 35-50mm.
All assemblies will bolt onto the table for positioning.

Attachment 13046

Waiting for some more ball joints, 3/8 lock nuts and some M10 half nuts before I can install the assemblies.

Regards
Ross

I'm following your hold-down journey with interest.

I picked up a burr and mowed some foam with it. It sliced through like butter (especially when I made a mistake with home refs and the bit ran diagonally across the piece!).

It's a small bit - 8.8mm diameter, 45mm long, from memory, so probably too small for a full sized board? My first board was only 300mm long ;-) http://www.cheapesttools.com.au/shopexd.asp?id=8459
I'm still seeking out a big burr. I have a "wood miner" to test that has a 20mm dia and is about 100mm long. But it can only be spun to 10 000rpm (designed for angle grinders).

I'm interested in your locating system. Clamping to stringer fore and aft is common, but problematic is the stringer is curved.

Hi Red

Sitting here working out how the new script based beta version of boardcad 2.1 works as your post came up.
Jed looked into it the other day so I felt I was trailing behind, haha
Happy to report it seems to produce even better gcode.

Great link for the ball nose cutters - winner on that one. :)

You don't say how you are generating tool paths (assumming Cut3d as you have said in the past) but I had no problems with a 12 mm end mill and 8.8 is not much smaller.
In the end it just means more passes, closer together (stepover is the generic CNC term).
The length of the cutter will probably be the limiting factor due to the plunge depths to cut the sides, stringer ends or the inside upward curve of the nose on the deck esp. if the router base is still in place.

Some machines used/use the stringer clamps, then came along stringerless blanks and an new senerio was required.
Gripping the stringer ends always struck me as a very effective way to hold the blank, so shame that.
Some code, like boardcad, can also cut the stringer if the code is given the parameters to do so.
My philosophy was to go for broke and ensure I could hold all the blank types, stringered, cut stringered, stringerless and wire cut blanks from blocks.

The design process has been expensive and time consuming.
It is also by no means finished as I have not tested the vacuum cups on the curved striations that the ball nose cutters will leave.
Obviously the bigger the diameter of the cutter and the larger the number of cuts will decrease the height of each striation and improve the available vacuum hold.

Finally, now you have cut some foam, darn noisy hey !!!

Regards
Ross

I use a shop vac to hoover up the foam dust as the cut progresses (feasible for small models). I emptied the vac (gee there was a lot of white dust). Then I made the mistake of assembling the vac for water suction and switching it on inside the house. Tiny white balls everywhere. blown out the exhaust because the filter wasn't in. I was unpopular with the wife, even after I'd finished washing the kitchen floor and vacuuming the toaster etc. :(

Advice I got from the Cut 3D forum was for 10% step over for finishing foam with a ball end, but this may result in an excessive number of passes. Trade offs.

Be Careful about Non Specialist Advice
 

To give some perspective on this one the CUT3D people have advised you to do something like 650 lengthwise passes to machine the blank per side.
A 10% stepover using an 8.8 mm bullnose will only advance sideways 0.88mm per each board length traverse.

Typically a board will be machined lengthwise with something well under 40 passes.
I have doubled that to about 80 passes and it takes around 15 mins to do that.
At 650 passes the machining time will be something around 2 hours per side.:eek:
Production machines like the APS 3000 do the job a bit more roughly using wide stepovers but will do the deck in about 6 minutes and the bottom in 4 minutes.

This is an example of a 66 pass deck cut (50 deck + 2x8 rails).
This could easily be considered over shaping but does it look great. :D
Attachment 13069

What the Cut3D people have completely failed to understand is that the shaped blank is screen sanded as a post CNC shaping process.;)

Regards
Ross

10% stepover is a standard for ballnose, 7-8% for 3D relief and no sanding. So they gave you good advice but for a wrong surface finish.
Get a larger ballnose (25-50mm)so you can get it fast with small stepover for less final hand work.

No fooling you Danilo but I still disagree ... to a degree !!

Part of the truth is we are too cheap to buy the big specialised surfboard carbide cutters, they charge an arm and a leg for them here in Australia.:D
Red is on to a supplier for a 3/4 inch cutter and I have managed to come up with a 20.637mm cutter.

I can't speak for Red but I have allocated a 20 min maximum window per side to achieve the best possible finish with a reasonable cost cutter.
What this means in practice is the 10% stepover is out the window even with the 20.637mm cutter.

10% of 20.637mm is roughly 2mm per pass. The board above allows 50 or so passes for the top surface and then a further 16 for the outside rails.
That moves me along only about 100mm of the 500mm of top surface of the design.
The required stepover even with my new larger carbide burr will be something like 50%. More if the cycle time needs to go down further.
The largest commercially EPS cutter is 31 mm in diameter.
The large stepovers that are commonly used to achieve realistic production cycle times mentioned are no accident

The advise given to Red was offered in a bubble and did not account for scale.
Read it here at http://www.vectric.com/forum/viewtopic.php?f=12&t=13495.
The models x value is only 376mm, so you can afford the luxury of a tight stepover... but not at full scale.

Regards
Ross

@Ross; I think you can achieve faster speeds if you get a bigger Piston Pullback Spring and change your catalytic converter. haha:). Kidding.

I cant wait until I understand everything you mention on your posting above.. I am going to start some reading and try different things with mine.

Maybe you can find a large diameter drill, the method Sean uses for EPS

http://www.mechmate.com/forums/showt...ght=foam+drill