Official Luthiers Forum!

I use them all the time, but not on the CNC machine. I prefer something indexed to the machine and solid if something causes the vacuum to drop. I used a big piece of Starboard, surfaced it with a big spoilboard surfacing bit and drilled holes through it in a grid. I put T nuts on the back sides of the holes and have something of a tooling plate. It works out very nicely--I hold jigs in place with nylon bolts that won't break a bit if I happen to accidentally hit one.

Dave

Which brings up another question regarding surfacing - when you surface, there are regions of the table that are out of reach of the gantry, does that cause any problems?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

It can if whatever you're fastening down to the table extends further than the surfaced area--for instance, if you have a fixture bolted down that extends a little too far, it can keep it from sitting flat/square to the machine. I use a pretty good sized surfacing bit--considerably bigger than any cutter I use for machining parts--so there is room to spare even though the surfacing still doesn't quite go to the very edge of the table. I haven't had any problems with it.

Dave

I discontinued use of plastics (HDPE, UHMW etc.) because they tend to warp when machined. You have to use the really thick stuff to prevent that.

_________________
"I want to know what kind of pickups Vince Gill uses in his Tele, because if I had those, as good of a player as I am, I'm sure I could make it sound like that.
Only badly."

Oops...double post. Been having connectivity issues today. Sorry!

_________________
"I want to know what kind of pickups Vince Gill uses in his Tele, because if I had those, as good of a player as I am, I'm sure I could make it sound like that.
Only badly."

Those are both polyethylene (UHMW is short for UHMWPE), which is definitely warpy in all incarnations. It's very, very difficult to guarantee any 'in place' stability of anything machined in PE. You can do it, but I doubt anyone here's looking to normalize their fixtures for a day in an oven before and after machining. Acetal (Delrin) and acrylic (pleixglass, lucite, et al) are much more stable when machined, but they're also more expensive. Acrylic, in my experience, is especially stable post-machining compared to other plastics.

Aluminum moves more than I'd like, but it can be accounted for to a good degree with careful machining. I think the fiber filled phenolic/epoxy composites tend to be pretty dimensionally stable but they can swell over time from moisture absorption. Unless you're going to go for hot rolled steel, I think those are the best bets.

Double sided fixtures have a lot of risks for CNC work. They can sit unevenly, can move in the XY plane between uses, and have a really good take off probability if the seal is lost. Normally you'd only need to worry about the part sliding around on the fixture and ducking the part when the vacuum is lost

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

I've been using thick stuff for that reason. The Starboard is less warpy than the UHMW stuff I had used before--it's not as dense--but it's still rarely completely flat when I get it. I've had good luck in getting it very flat and staying very flat (since it's bolting down to something more solid). Still, since I'm working in relatively small dimensions and since I'm not really worried about being off by .001 or so as far as a jig is concerned (since I'm only using these as clamps anyway), for my purposes I'm quite pleased with the plastic I've been using. It's been stable enough once machined and bolted down to the machine's table.

If I were doing aerospace components that had to be perfect down to many digits below the decimal point, I'd pay the extra bucks and go with a more stable material for jigs. There's no doubt that other materials are superior in terms of stability. The question is a matter of how much stability you have to have for your particular application versus what you're willing to spend.

As it is, I'm cutting anything from pearl inlays/pockets to carving necks. The inlays are more critical to precision than anything else, and it's the CNC itself that's really governing what happens there. And the level of precision that I need is nowhere near what you'd need to build an airplane. The vacuum jig that's holding the material down just needs to hold that material down, and in a plane that is hopefully in sync with the machine's plane. The level of accuracy in what I'm cutting is undoubtedly affected more by movement in the wood that I'm cutting than in the material I use in the jigs. When you get down to it, the standard that I'm measuring against is how accurate I am with a jeweler's saw and a hand held router. Mind you, I'm pretty good with both, but the CNC puts me to shame regardless of what my jigs are made with.

Bottom line to me is this: I think that it's OK to worry about the precision of your jigs and machines relative to the requirement of the job. There's enough slop built into the hygroscopic nature of the material most of us are cutting to negate a lot of the precision that we are capable of achieving nowadays. There is an economic factor to weigh into the equation too. If you have an operation that really does need super-dooper precision, by all means use the most stable materials available that are within your budget. On other operations, balance everything and see what makes sense.

I'm a rookie to CNC (and having a blast!), but I've been working in engineering and business for a long time. It's all a balance. If it weren't for the economics of it all, I'd always use the very best material available for any process. But since I have a limited budget, I tend to spend more on the materials going into the instrument itself and choose to skimp (within limits) on things that are only indirectly involved with the instrument. Bob and Don and others out here have insight and knowledge that I'll never have when it comes to CNC work, and I respect and admire their work and opinions. But at the same time, when working within a tight budget (as I am and many others are too), I have to put things in perspective. When the jig I'm building is designed only to hold something in place, and be flat within a reasonable tolerance, I don't really have to worry about it all that much. With a non-CNC tool, I'd have probably been using MDF for the jigs if I was using a jig at all, so darned near anything is going to be better....

Heck, I'm building banjos so take my opinions for what they're worth. I guess when you get down to it, anything beyond penknife precision in banjo building is aerospace precision in itself!

Dave

Thanks for all of the advice guys.

I'm now thinking I'm going to stick with my Al T-slot table and bolt or clamp the fixtures to it. I'll be ordering a stash of Star Board so start playing around with.

Still haven't purchased the vac setup as I'm flip flopping as to whether I want to buy the complete setup from joe woodworker or start piecing things together . I hate doing DIY stuff that's not directly instrument related but it looks like it would save about $100 to $200 to buy a pump off of e-bay. That of course assumes that the pump will work when I receive it!

Anyone ever try sealing MDF with shellac or something to make it more air tight? Might give that a shot too just to see how it works.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

I've done it and it works, although it didn't hold a vacuum like plastic does. It was still fine. But I really hate using MDF just because of the dust. At least machining the plastic, it's little chips that go all over the shop. They're easy to sweep up. But that MDF dust is killer!

Dave

Dave,

Couple more questions to your method specifically:

For the gasket - is that just a plain old but joint I saw in your previous post or is there some sort of scarf there that I don't see?

Second - do you surface the "bottom" of the jig, i.e. the side that sits against the table or do you just keep it natural Starboard?

I'm working on my fixtures right now and will post a drawing once I have it complete.

Vacuum pump and other goodies are on the way.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Hey Andy--the gasket is just a plain butt joint. I cut it with scissors--there is enough stretch in the gasket material that I pull it to where it wants to overlap, but then compress it into the routed slot. It's really very forgiving.

On the bottom of the jig--I don't surface it. Starboard is consistent in thickness--when you bolt it down to your Tslot table and surface the top, it will be good to go for anything like we're doing. It's easy to go overboard on this stuff--for some applications that's a good thing. For others, its (in my opinion anyway) overkill. It's all still relative--used to be my jigs were made of MDF and I was using a handheld router or a shaper to cut stuff. With CNC, the jigs I'm using are far more precise, but still nowhere near as precise as they could be. Question is, how precise do they NEED to be for the particular application. I'm having no problems even with pretty fancy inlays and fretwork with the jigs I'm using. I wouldn't be getting any better results with more perfect jigs.

As a sidenote--I've gotten parts of my new Raptor already, and the rest of it has cleared customs and should be here by early next week. Woo Hoo!

Dave

Good deal! I know a lot of people have been waiting quite a while for those things.

Final question - whats your spindle speed and feed rate for machining the Starboard?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

For surfacing the top, I ran at about 12K rpm and 50-75 ipm. Probably way more conservative than I needed to be. For the gasket grooves, I ran at 18K and 150 ipm or so. Starboard come up in chips or ribbons and doesn't have a tendency to burn like many plastics do from my experience.

Dave

I have used Corian counter top material. If you have a plastic dealer close by they usually sell scraps by the pound and you can often find material there that works as well. Sign Shops are also another resource for smaller pieces of plastic and XelLite board.

_________________
tim...
http://www.mcknightguitars.com

I finally got far enough along to design my 4-sided neck fixture and have one more question: as you can see from the pic, I'm using a sort of grid pattern to pull the vacuum. The outside channel is for the gasket. Should the vacuum channels extend out to the gasket channel or should they stop short like I have them? Does it matter?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Couple of points.

I realize you may have quickly drawn this to illustrate your issues but I thought I'd comment on what I see as a possible misconception.

1. It looks like you've drawn your seal channel as a half round groove but it should be a square channel to allow the sealer chord some space to conform into and also facilitate the placement and holding of the sealer chord. A square channel will hold the chord into place but still allow the easy removal of the chord for cleaning. If you do it as drawn you'll have to glue it into place.

2. Also...If you do it as drawn, depending on the chord material, when vacuum is applied there might not be any room for the seal to squish out of it's own way and it will prevent the piece from sucking all the way down to the plate. I pretty much use just 3/16" diameter latex surgical tubing now for sealing. With that I have a groove formula. The groove width is 100% the diameter and the groove height is 80% the diameter of the chord.

3. Softer seal material, like sponge chord, will allow a shallower groove, which will allow for less than optimal flatness on the surface being held. Harder material will require a deeper groove. The idea is that you want the seal material fully compressed so that the wood pulls down all the way to the plate surface.

4. I don't think it will matter if your vacuum grooves reach the seal groove or not...although I've never made mine any other way than to make the vacuum grooves go all the way to the seal groove. Your vacuum grooves need almost no depth at all to still be effective. You might keep that in mind because deep grooves are harder to clean out...a minor consideraion. I make mine .03" deeper than the surface.

5. Some woods are easier to hold than others due to porosity. When I have a problem piece like this I simply use wide, thin plastic sealing tape on the vacuum side of the wood, which utterly seals the surface and allows a full vacuum to develop. When I need to do this I program to accomodate tape thickness.

6. Some pieces are simply too small to hold down with vacuum. Depending on the shape, it's best to consider using two-sided tape as a way of holding a piece. This usually requires a very flat, machined surface to apply the tape to and a flat tooling surface. Tiny things like truss rod covers automatically go to tape holding fixtures in my mind...pieces up to the size of a bridge actually.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

Well, it really sucks!

I had a bunch of time today so I got her done - unfortunately before I read your post but it seems to work ok anyway. Found I could cut around 200ipm with no problem whatsoever. I ended up making the gasket channel .2" deep. I am using foam gasket and it seems like .2 is just about right - perhaps .190 or .180 but time will tell. I did use a 1/4" round nose bit but with the channel greater than 1/2 the diameter of the gasket, it seems to have enough room to compress.

Somehow screwed up my depth on the vacuum pocket and it came out .2 instead of the .1 I had planned. Not sure how I screwed that one up. I'll probably do .03 next time though.

Did have a major crash though - first time I made the gasket grove I did it at .135 and that was too shallow - gasket wasn't compressing enough to allow the stock to go flush to the surface so I made some new code to just come in and deepen the channel.

Loaded the new code

I didn't re-zero or anything.

Hit cycle start, machine went to my fixed plate and probed for tool length.

Machine went to where it was when I started the program, off in the upper left corner.

It then dropped to the clearance plane (I think, looked about .250 over my stock) and crashed into a clamp on its way to the work piece. Didn't break a bit but it did narf up the nut. Managed to hit the estop on the keyboard. VFD had an error on it as well.

My thought is that somehow it was using the clearance plane from visual mill instead of safe Z for the traverse.

Spindle still works though so that's good. Nut is a bit narfed up - it collided with the t-bolt for the clamp but it doesn't seem to be vibrating or anything and is working fine. I ran the program again with the machine starting elseware and everything worked fine. I'll probably get a new nut just to be safe.

The other thing is I need to get a real collet wrench!!! The spanner I'm using is garbage!!!

Tried it with the pump and it really sucked. I was able to slide the piece when I really cranked on it though so good thing I've got the dowel pins. The extra volume in my vacuum pocket doesn't seem to be negatively affecting the performance. I'm able to easily pull 22" which is what I've got my bleeder set at right now with my finger over the inlet.

I'll use a really slow feedrate for boring the corresponding holes in the part stock since they'll be held with only vacuum during initial location of the holes.

Thanks everyone!

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

A short video of my machine cutting the fixture:

http://www.youtube.com/watch?v=voXPa9JtsVA

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

That music is to die for....meaning to be shot for.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

Extra volume is good. The gauge only says what the pressure is at the far end of the system, but the pressure underneath your part at various points can be quite different. The more restriction there is between the pump and the leak (your gasket seal zone), the larger the difference in pressure will be. The same thing happens with using long thin hoses attached to compressors- at the far end of the hose you can lose 10-20PSI or more.

Leave some solid areas to prevent the part from deforming, but otherwise hollow out what you can. The difference between a flat-faced grooved plate and a 0.020" pocket with lands is quite substantial, and I've found that 0.05" to 0.08" is the point of diminishing returns for guitar-scale parts. Any extra space you can leave will function as ballast in case of a leak and give the system more time to recover.

_________________
Bob Garrish
Former Canonized Purveyor of Fine CNC Luthier Services

Think not only in terms of holding a flat part down onto a plate with vacuum but also how you can design the part and vacuum plate to mechanically work together to eliminate side movement during a cut. My most successful fixtures are a combination of a mechanical boss and vacuum.

Sometimes vacuum is almost enough to hold a part but because of side loads while cutting the part might be prone shift on the plate during the cut. If possible, I always design a locking boss on the part and a recieving slot in vacuum plate that keys the part into the fixture with the idea that this boss will be milled off on a later operation. A boss will keep the part from sliding laterally on the fixture and the vacuum then only needs to hold the part down while the cut is made.

Other times I can do the first op on a part while holding it with soft jaws in a vice. If the outside profile can be cut this way then the part can be flipped over and set into a depression on the vacuum plate that tightly captures the part around that profile. This too isolates xy movement while still using vacuum to hold it down.

And, of course, part pins with corresponding holes in your vacuum plate will do the same thing.

_________________
I read Emerson on the can. A foolish consistency is the hobgoblin of little minds...true...but a consistent reading of Emerson has its uses nevertheless.

StuMusic

I haven't had any problem cutting a round channel and press fitting (without any adhesive) the round foam cord as you've done Andy. It's pretty much the way I do all my jigs any more. Looks like it's working well for you too.

Of course, anyone who adds Raymond Scott music to his YouTube CNC videos gets a big plus vote in my book too....

Dave