PDA

View Full Version : Aspire and 2 rail sweep for fingerboard



banjomanwv
04-21-2015, 08:09 PM
Im trying to figure out how to use a two rail sweep to make a compound radius fingerboard, or a straight radius even, anyone know how to do it?

Brady Watson
04-21-2015, 09:26 PM
Post a pic of the fingerboard shape(s) in question.


-B

Ajcoholic
04-21-2015, 10:16 PM
Im trying to figure out how to use a two rail sweep to make a compound radius fingerboard, or a straight radius even, anyone know how to do it?

To model the finger board, in either a constant or compound radius, I would do this...

- draw the edges of the fingerboard. dont connect the ends together, just draw the tapered edges from the nut to the tail end.
- draw an arc of the desired radius at the nut, going from the ends of the sides you drew and the desired height/radius.
- choose the two sides as your rails, with the arrows pointing from the nut to the opposite end.
- select the arc you drew as the object to sweep.

for a compound radius board, do NOT select the "scale cross sections with width" - which will flatten the radius as the board tapers wider down the length from the nut to tail end.

for a constant radius board, I would sweep along two parallel lines drawn from the widest part of the board, with an arc drawn at that end, and then trim to the actual tapered edges of the board. That will give you a true constant radius all the way down the board even with the taper.

Ajcoholic
04-21-2015, 10:31 PM
I just did a few qucik samples. notice that the constant radius board has the same curve at both ends. the compound radius has the narrower end where the nut is, as a smaller radius and the body end a flatter radius.

nut end of example
http://i83.photobucket.com/albums/j307/ajcoholic/nut%20end%20of%20fretboard_zpswbfcj1oq.png (http://s83.photobucket.com/user/ajcoholic/media/nut%20end%20of%20fretboard_zpswbfcj1oq.png.html)

body end of fretboard
http://i83.photobucket.com/albums/j307/ajcoholic/body%20end%20of%20fretboard_zpscsykapct.png (http://s83.photobucket.com/user/ajcoholic/media/body%20end%20of%20fretboard_zpscsykapct.png.html)

http://i83.photobucket.com/albums/j307/ajcoholic/fretboards_zpstawwanas.png (http://s83.photobucket.com/user/ajcoholic/media/fretboards_zpstawwanas.png.html)

Ajcoholic
04-21-2015, 10:35 PM
The only thing you will also notice, is that with the compound radius board, the thickness at the edge of the board will remain uniform as you go from nut to body end.

With the constant radius board, the thickness at the edge will decrease from the nut as you go along the board to the body end, but the thickness at the center down the length will of course stay the same thickness.

You prob already knew that... it only comes into play if youre binding the edges.

banjomanwv
04-22-2015, 12:11 AM
Ok, I got it......before I read this I figured it out BUT........I thought it was the opposite as you explain as far as the thickness? I may be wrong but if your doing a compound radius of lets say....7-12.................the farther you go the flatter you get which means the nut part is actually thicker on the binding sides than the last fret is.

banjomanwv
04-22-2015, 12:12 AM
I think the straight radius stays the same on the side.

banjomanwv
04-22-2015, 02:39 AM
Now to the next question. How do I fret it along the same radius line?

Brady Watson
04-22-2015, 07:34 AM
Now to the next question. How do I fret it along the same radius line?

Profile toolpath with ON option. Straight lines as vectors. Use Project on 3D option.

All these questions are covered in the tutorial videos and help file...

-B

MogulTx
04-22-2015, 09:14 AM
I think you need to look at the fretwire you are going to use. if it has a curved face but a flat insert- then you just want to mill a flat bottomed slot through the fretboard. If it is a curved material with a curved bottom that is an offset radius from the actual fret, then you COULD use a curved slot that mimic's the fretboard's face. But that is a lot of work to figure that out and to tool path it- and preserve safe Z that will correctly mill an arched slot and not nick the board.... ( and will take a FINE little bit, too, so it is going to be slow speeds!)

If you consider a standard saw's path- ( to create fret slots).... and if you had sanded the fretboard down and needed to deepen the slot- it would HAVE to be a flat bottom slot. I don't now or a way to align closely enough to allow the slot to be reworked with an arch to it... so I am guessing that the fretwire will be a standard roll item that you would install into a flat bottomed slot, and you would either pre-arch it or it will take the shape of the fretboard. ... (I have not built a banjo- nor installed frets- so I am responding from my knowledge of instruments and without he benefit of the latest manufacturing trends.... so YMMV.)

banjomanwv
04-22-2015, 09:31 AM
We do inlay for a living, mostly in the banjo community. It usually takes a .024 bit for fret slots and while it can be done flat, we would like to prebend the wire and follow the curvature of the board

steve_g
04-22-2015, 10:44 AM
Following the curvature is no problem when done as Brady suggested
SG

barrowj
04-22-2015, 10:57 AM
Andrew,

I never seem to be able to view your files on Photobucket, any ideas. I seem to spin forever then just blank.

Joe

Brady Watson
04-22-2015, 11:52 AM
FYI - most fretboards have the metal put in straight without curvature - on guitars anyway. So, projecting onto 3D surface is only required if you need the .023" bit to follow the curvature of the neck. Not sure how banjos are done, but most guitars go in straight.

-B

banjomanwv
04-22-2015, 12:20 PM
Brady the problem with that is the bits only cut so deep. The fret slots need to be .090 on a banjo as standard. With a 8 radius it will hit in the middle of the board and by the time it hits the sides its already deep in the middle. by the time you go .090 deep on the edge it makes it like an accordion. If you start with a board that's .200 it leaves .160 at the last fret in thickness, you then have a slot .090 deep...that's leaves .050 and that's not much so the more meat I can leave in the middle the better

banjomanwv
04-22-2015, 12:22 PM
Brady it worked just as you said, thanks so much...

Ajcoholic
04-22-2015, 10:54 PM
FYI - most fretboards have the metal put in straight without curvature - on guitars anyway. So, projecting onto 3D surface is only required if you need the .023" bit to follow the curvature of the neck. Not sure how banjos are done, but most guitars go in straight.

-B

The only guitars with a flat board are classical. All others (acoustic or electric, bass guitars etc) have varying radii generally from 9" to 12"

Frets of course have to be curved to follow the curve of the board. Most companies run their wire through a bender that matches the boards radius.

In most commercial slotting operations, a .023" saw blade just cuts a straight slot through the board. The fret slot is deeper than the fret tang at the middle, but that usually gets filled with glue anyhow.

If you really need the slots cut the same depth as the tang than you can certainly do so as stated.

Ive made a whole lot of fretboards over the years for various guitars - but I have a small table saw set up with a slotting blade and do it the old way (not CNC). I radius by sanding, although I do machine contoured radius blocks on the CNC router. Im not production, but I used to make 4 to 6 custom guitars a year.

banjomanwv
04-25-2015, 09:14 PM
I still don't have it. Its cutting a compound radius but its not the correct one. I make my two rails and then click on both radius lines I have drawn. then it runs but its not right. how do I make sure the one radius which is 7, is at one end and 12 on the other?

banjomanwv
04-25-2015, 10:19 PM
I noticed when I pick my radius like the tutorial says then I can change it without changing it so to speak.
Scale to exact height changes the radius and then the "combine with the current shape" has two options...merge or add
that changes the radius to.....not for sure what to do?

Brady Watson
04-27-2015, 08:02 AM
Your settings should look like the attached pic. Don't scale height or width. Make sure your cross section vectors are already the same width as the rails and height you want them to be in Z.

After selecting the rails, click on the first cross section and assign it to the top rail starting point. Then the 2nd section and the top rail end point. Hit apply to create the component. This will give you a sweep that starts with the first cross section & ends with the 2nd.

Add or Merge - doesn't matter if there is nothing else on the screen. Default is add.

-B

banjomanwv
04-27-2015, 10:25 AM
Let me make sure I understand this Brady. I make two rails, and in my case they are 2 inches apart to indicate a fret board blank for a banjo.
I click on one, and then the other and then the first radius cross section, then click on the 2nd cross section and then hit the two rail sweep button.
I noticed you have sweep between spans clicked so should I also? And let me say I appreciate your help so so much.

Its hard for me to wrap my head around for some reason.
I don't know if I am assigning the cross sections the proper way or not AND, I notice on your cross sections you have lines coming straight down to be the sides of the fingerboard and I don't have that.

banjomanwv
04-27-2015, 10:26 AM
Oh yeah, its making my board to thin.....how do I determine the Z?
My board can be any size when I start but the ones ive been using were about .220 thick

Brady Watson
04-27-2015, 11:28 AM
Z is determined by the 2d cross section. Make cross sections exactly what you want them to be in 3d.

-B

banjomanwv
04-27-2015, 12:31 PM
When I attach lines to the side of my cross sections going horizontal at .220 (the thickness of my board) I get the message model thickness exceeds material thickness

Ger21
04-27-2015, 12:55 PM
Increase your material thickness. It needs to be slightly thicker than your finished fretboard.

banjomanwv
04-27-2015, 01:13 PM
Guys, I really appreciate all the help and I hope im not being to thick here. Anyway, my boards are around .200 - .220 to start with. So lets say I start with a .200 board, the highest point of the radius will be on the first fret, middle of the board, the cutting height at its highest should be .000 there and slope on each side downward. On a compound radius the board will get flatter as it heads toward the banjo head so the sides will be lower of course.

But if I start out with a .200 board if I mic it when its done right in the middle of the first fret it should still be .200, how do I set it up for that....Ill give a shot of my set up

Does it have something to do with model position in material?

banjomanwv
04-27-2015, 01:31 PM
2490624907249082490924910

banjomanwv
04-27-2015, 01:31 PM
I start with snapshot one. Thats my two rails, 2 inches apart.
I click on those and in snapshot 2 I click on my 7 radius.
then in snapshot 3 I click on my 12 radius.
Snapshot 4 I got to the modeling tab click on 2 rail sweep and reverse rail direction
Then in snapshot 5 I sweep between spans and click apply

banjomanwv
04-27-2015, 01:33 PM
After that im lost with the z and the cutting file IF up to this point what I have is correct?

MogulTx
04-27-2015, 01:48 PM
You should have a very narrow, flat, line down the center of your fretboard from nut to the last fret wire. That will be your max Z in your material. You want that Z dimension to remain constant. And if you want a full radius on the neck ( yes) then it needs to be a line that is immeasurably narrow.. It is only the slopes on the sides that will change in Z. And that is the function of the two rail sweep. The rails are theoretical rails along which you wish to apply a shape ( in this case, a slowly flattening curve). But the center of the fret board is the zero depth of Z. And if I understand the mechanics of the neck, would remain at the same height all the way down the neck (otherwise the strings would rise away from the neck- and you would not be able to fret it without forcing the pitch to go sharp because of the longer distance from the resting string position down to a lowered fretboard face. Of course Nut and especially the bridge may need radius to keep them close in relation to one another, and the radius of the neck.

MogulTx
04-27-2015, 01:55 PM
I do not mess with the two rail sweep often. But unless I am mistaken, you have tagged the two radii at the same end. I think that is going to give you some form of weird curve that you don't want.

Brady Watson
04-27-2015, 02:20 PM
Does it have something to do with model position in material?

Yes it does. Make sure the model is slid all the way down in the block. You need a little wiggle room/ambient material in addition to the model thickness. Even if it is only an extra .005".

If this is not possible, make the legs of each cross section a little shorter. This will still give you the radii you need...and remember - these aren't guitars for NASA, so if it's off .003 to .005" - it won't really matter. Wood will move more than that between AM & PM.

Got your file - off to the dentist...Will try to look at it later today or tonight.

-B

banjomanwv
04-27-2015, 03:01 PM
Actually Mogul that is what happens, the Z height will only remain the same at the nut end of the board, the rest of the board gets flatter and flatter and if you had an imaginary line down the middle at a constant Z it would get cut away. the nut and bridge are also radius and the strings follow that line. So if your working with a 7-12 radius the nut would be 7 and the bridge about a 14 or 15 or so

banjomanwv
04-27-2015, 03:05 PM
Brady I think I can get it but horse ****, now I cant create a cut file...my mind is shot on this.....it says I have 4 open vectors and does nothing?

MogulTx
04-27-2015, 04:08 PM
IFF the fretboard thickness is only .200 and there is a 7" radius applied to the NUT end of the board and the board is 27" long, THEN the amount of material removal from center of board to edge assmung the peak of the curve if dead center- and anyone who knows banjos, knows a five string is significantly narrower from 5th string peg to the Nut, AND slightly "offset" -which may play into your final design of your neck radius OR you might decide to pre-prejudice your fret boat centerline to make the "spine" be the physical middle of the neck at the nut and the physical middle of the board at the tail- which would make a slightly angled track down the face of the fret board... I don't have my old Scruggs Book here or I could draft the exact length and fret spacing and 5th string peg placement, to show the minor delta to the CL of the board.... HOWEVER... calculating the radius of 7" at the nut and assuming we are working with a 2" wide board, and removing NO MATERIAL at center, we would expect to remove .072" from both sides of the board, leaving a fretboard that is .128 thick at the nut.

THEN: using a 12" radius 27" away, at the tail, the material removal is only .042 on either side of the fretboard, because the shallower radius has less sagitta across the same distance.

IFF you want the two to match as far as edge thickness, you would (probably) have to elevate the tail by .030" so that it takes off more volume OR maybe do a 3D cut file vs a two rail sweep. Is the .128" thickness at the edge pretty standard with radiused banjo finger boards? (seems pretty thin to me. I'd hate to have to handle and glue it up...if it is not already glued to the neck. And if you are going to machine it after gluing it to the neck, then you have a lot of money invested in that part before you begin this milling process- so you would probably make up some "crummy-wood" first to do your tests...?

YMMV, but that is what the math shows to be the delta sagitta between those two ends of the neck and the question becomes how do you work with it / accommodate it?

Hope this is helpful to your processes!

Monty

banjomanwv
04-27-2015, 05:35 PM
A banjo fret board is 19.5 inches long before being profiled based on a modern 26.250 scale and about 2 inches wide for a blank. If doing a radius of 7-12 (and yes, I do center my board on a center line of the unprofiled board) the center of the board at the nut should stay .200 if the board starts at .200 because that's the center point. A compound radius is like a snow cone and the farther you go the flatter you get so the end of the board is not as thick as the top (nut end) of the board. I have gauges to check it by but still need to now figure out how to do the cut file again....lol

banjomanwv
04-27-2015, 05:37 PM
When you said "072" from both sides of the board, leaving a fretboard that is .128 thick at the nut" that's sounds correct but remember at the nut the banjo neck is only about 1.250 wide.....so once its profiled it would be more in the 150-170 range I think

MogulTx
04-27-2015, 06:18 PM
OK. Yes. They are narrower at the nut and if 1.25, and no material removal at center, then the edge would be about .172" so it would be much closer to the same thickness as at the tail , without doing any tilting of the neck to compensate. Sounds like- from your comment above- that you shape it after. (which plays into the cutting plan)

I am sure Brady will solve it quickly and will watch to see the method of thinking it through and the correct method of blending the two radii. ( I watched a tutorial on this very kind of blending of two shapes together- a couple months ago, but don't recall the right way to make the two of them play together. I will go look for it and see about posting a link.)

banjomanwv
04-27-2015, 06:22 PM
I watch a few but they don't go into enough detail.....I appreciate all the help for sure and Brady has been life saving here...thanks Brady

banjomanwv
04-27-2015, 06:28 PM
model position in material...........I don't even know how to get back there now.......

banjomanwv
04-27-2015, 07:02 PM
Do I have to Join vectors to get to a tool path?

bleeth
04-27-2015, 07:16 PM
Been watching this and I admit I am not a guitar maker but if I understand it correctly the fretboard is narrower at the top than the bottom.
This means you need two profiles. One for each end. Then when you do your sweep (which should be along two straight vectors representing the length) you will have one curve in end view for the start end and a different one for the finish end. Then the curves across the board extrapolated by the software in between the ends should flow smoothly from one end to the other and give you a straight line along the apex.
For a 2 rail sweep along the whole neck the length vectors should be 2 straight lines (although not parallel) and not joined to anything. Make sure they both have their start points at the same end. One will be the drive rail and you need to make sure your profile arcs have their start points on the same side as well. For an experiment mix it up and take a look at the results!

MogulTx
04-27-2015, 07:19 PM
(responding to banjoman)

Not for the two rail sweep .

I do need to go back and look at how you assign one radius to one end and then the other radius to the other... and then apply a slow transition to the radius change... there is a simple function and when I see the video I am sure it will come back to me. That should be all you need... Then you will need to see how the first piece comes out and see if it meets closely to your radius gauges....

banjomanwv
04-27-2015, 08:45 PM
Im doing a fingerboard with engraved pearl of a locomotive...when its done ill work on this....and post pics

MogulTx
04-27-2015, 11:48 PM
OK the Vectric tutorial is for about 3 versions of aspire ago.... the functions are not the same at all. Tried for about an hour now to combine the two radii as a two rail sweep of multiple components.... does not like it at all. If I had done a bunch of these, I feel relatively confident I could go in and poke through the set up and get it done.... but I am going to have to bow to better experience. On a side note- it appears that what Andrew had posted looks pretty darn good and may well have been what you wanted- or close to it. If Brady has difficulty, I would ask Andrew what it is that he did . ( Brady won't have a problem doing it. The man has skillz.)

And I would implore vectric to make a new tutorial, but I realize I just became a "last version" guy and do not have version 8.... I guess I will be upgrading... again... Oh well. Hang on for a minute. AI am sure someone will pop up with the correct profile in a model...

banjomanwv
04-28-2015, 12:11 AM
My aspire is 3.0 so its an old one, I just never used it....never needed to till now

banjomanwv
04-28-2015, 12:20 AM
Im getting there...material set up so I am good there but I still cant make a cut file and Ive cut 10 boards as of now trying this out...don't know what I did before but now when I click on both rails and both cross sections its telling me I have 4 open vectors and does nothing

banjomanwv
04-28-2015, 12:24 AM
Brady how do I know how long to make the little legs on my cross section?

Brady Watson
04-28-2015, 07:37 AM
Check your email.

-B

banjomanwv
04-29-2015, 12:11 AM
I think with Brady's help I have it.

Marc F. Lupien
04-30-2015, 12:27 AM
Hi Banjomanwv,

I carved my very first compound radius fingerboards 5-6 months ago with a toolpath (sbp) file created with a «Visual Basic For Application» program I wrote in Excel. I carved 2 test fingerboards with it and it seemed acurate. If you want a copy, email me at marclupien@hotmail.com. I make no guarantee but it works (some comments might be in french though). Note that I did *not* cut the fretslots on the CNC.

With the recent VCarve pro 8.014 upgrade, I thought it would be a good idea to write a custom gadget to create such a toolpath directly in VCarve pro. It would be a better integration than having to use Excel but I won't have time to implement that soon...

Marc.

banjomanwv
04-30-2015, 04:01 PM
I really appreciate that but I have been able , with Bradys help to now cut my own with Aspire