Looking for accuracy info from Fine Line Automation machine users

[yackback]

Hello Everyone,
I'll start this out by saying I've been lurking around the forums for a while and finally decided to setup an account and start posting. I really like the community that I've seen here.

I've been looking into the different CNC routers that are being built by the users of this site and I really like what I'm seeing from the Fine Line plans/kits. I'm most interested in the FLA200, but would love to hear from any and all users of the Fine Line machines, regardless of model.

Here are my questions/concerns:

I'm hoping to use one of these machines for the purpose of instrument building, specifically guitar building. The most demanding operation in guitar building (at least where accuracy is concerned) is fret slotting. The distances between fret slots are generally measured in thousandths of an inch. For instance, on a 25.5 inch scale length guitar, fret 1 should be located 1.431" from the nut, fret 2.782" etc etc.

While I know the theoretical resolution can accommodate this sort of accuracy I'm concerned whether the real world accuracy of one of these machines is up to the task.

So basically I'm wondering what sort of accuracy the users of Fine Line machines are having? If anybody has specifically used one of these machines for the purpose of fret slotting I would love to hear how well things came out. I have considered contacting Nate and/or Ahren directly about this topic, but I figure getting the perspective of the community will be most helpful.

I will admit that wood is an imperfect medium when it comes to holding those sorts of tolerances, since the expansion and contraction of the wood with moisture could change even the most accurate cuts. It's just extremely hard to find any information on what sort of accuracy is "good enough" when it comes to fret slotting.

Any info anybody can provide as far as how accurate their machines are cutting would be greatly appreciated.
Thank you
Justen

[ger21]

I will admit that wood is an imperfect medium when it comes to holding those sorts of tolerances, since the expansion and contraction of the wood with moisture could change even the most accurate cuts. It's just extremely hard to find any information on what sort of accuracy is "good enough" when it comes to fret slotting.

For hundreds of years, people have been hand cutting fret slots with perfectly acceptable spacing. Just about any CNC will be more accurate than hand cutting.

Will you get .001 accuracy over ±20". Probably not. The acme screws that most are using have a tolerance of ±.009/ft. Although in practice, most are much, much better.

I've never cut fret slots on my CNC, but I have hand cut them with no problems, and I know my CNC could do it more precisely.

And moisture content shouldn't really affect fret spacing, as wood doesn't move much at all along it's length.

[yackback]

Thank you for your thoughts on the subject ger21. I imagine it's not quite as big of an issue as I'm making it out to be. I had read about wood not moving much along its length, but for whatever reason I was still thinking that expansion in the other two directions could still slightly effect slotting accuracy. Perhaps not. I'm just somewhat worried about a multi-thousand dollar cnc router being not as accurate is one of the template based tablesaw or miter box fret slotting systems available.

I'd still love to hear from any and all Fine Line Machine owners as far as how you're getting along with the accuracy of the machines. Anybody have a really intricate/high resolution project you can show off? And if such a person exists it would be great to hear from anyone who has used one of these machines for instrument making.

Thanks again for the help.

[brtech]

I have an FLA300. There is accuracy, and then there is accuracy. I'm not an ME nor a machinist, my skills at measuring accuracy aren't great, and my skills at resolving problems that I find with the measurements is even worse.

I believe that I can get around .003 accuracy per axis. Right at the moment, my "squaring" is no where near that good, but it's good enough for now. The FLA300 is R&P on X and Y, and the repeatability and resolution TEND to be pretty uniform. Because the basic microstep is .0005, you can play a bit with the setting to dial the resolution you get in practice to match reality. I do this by moving exactly the 49" or so of range I have on both axis, measure with a good quality rule, and tweak the steps per inch setting. It was pretty close right off the bat, which it should be, so I'm probably calibrating out things like tooth spacing, belt stretch and pulley size variation. I then do moves of a foot or so "randomly" within the range to see if I can spot any non linearity. I can't, at least as well as I can measure. I've also used a caliper on a 5" move, and it's pretty darn close.

If I measure diagonals of squares, I'm not so good. I'm somewhere within 1/32 - 1/16 now over a 48" square. That's not .003. My "tweak" for that is homing on the A axis (meaning I adjust the home on the A leaving X stationary). I need to rebuild the mounting arrangements for my home switches to be able to make small adjustments, and right now, that's not the top of my stack.

I thought Z was okay, but my test cuts say I'm awful, which is my current problem. Since it's way off, I've probably done something stupid, as opposed to some alignment or slip or something. It's a screw, so the math is pretty simple.

In the end, I'm thinking .005 is the number I should be able to get. It's possible I could calibrate my way to something better, but that's real low on the priority list.

[yackback]

Thanks so much for the info brtech! Forgive me for being a bit of a Noob, but when you are referring to your A axis what exactly are you referring to? Do you have an actual 4th axis with a rotating spindle or does that have something to do with using a motor for two separate rack and pinon's on the X axis?

I love the pictures I've seen of your machine by the way.

[ben_jackson]

When both sides of the long axis of the machine are driven independently, each stepper is controlled by a separate channel. In Mach 3 that's usually the "A" axis and it is slaved to the other matching axis (X or Y depending on how you decide to name it). He's talking about the difficulty in squaring/homing an axis where both sides of the gantry are driven separately (and over different tracks!).

[yackback]

Originally Posted by ben_jackson When both sides of the long axis of the machine are driven independently, each stepper is controlled by a separate channel. In Mach 3 that's usually the "A" axis and it is slaved to the other matching axis (X or Y depending on how you decide to name it). He's talking about the difficulty in squaring/homing an axis where both sides of the gantry are driven separately (and over different tracks!).

Ahh, thanks for the explanation, I had a feeling that was what it meant. Hmm, seems like potentially a lot of extra headaches.

[691175002]

Wood in general will regularly expand and contract about 0.5%. Fret spacing on guitars requires very little accuracy. One could do the length->frequency convesion to determine the exact point at which innacuracy would produce an audiable difference but let me assure you that almost any machine will be at least an order of magnitude more accurate than necessary. People still do this stuff with a pencil and ruler.

The chances of any process that directly involves a person pulling or pushing material through a saw having higher accuracy than even the saddest of cnc machines is near zero.

[zeeway]

Love CNC and love guitars...and I also have a FLA 100. I have made some sting instruments (banjos) and cut the frets by hand. I have to agree that this CNC, when tuned, will do much better on fret cutting than any manual method.

Now having said that, you just cannot just slap the machine together and expect the best accuracy. You have to carefully adjust the bearings to the ways. You have to carefully tune the motors so that when you ask for a move of 22.345 inches (for example), you get it. You have to carefully adjust your machine for squareness.

If you do all that..and have your machine cut all the frets in one machining process, and if there are positioning errors, it is likely they will be proportional as you go down the fretboard. This means the scale will be correct musically, but the "scale length" may be a little longer or a little shorter than you designed...which to me is not much different than having to "compensate" a bridge. Hope all this makes sense. My bottom line here is go for it.

Angie

[studysession]

I have the 2x3 kit that I finished building the week before Christmas.. And I have built guitars with.

I am extremely satisfied with the kit and it's accuracy.

[yackback]

First, thank you all so much for the replies. It's always nice to hear several perspectives and first hand experience.

Just out of curiosity, A few of you have mentioned having slotted fretboards by hand, but also own CNC's. Obviously it's been expressed that a CNC should be more then accurate enough for the task, but is there any reason why you didn't use your machine for the job? Perhaps you didn't own the machine yet, or didn't have the proper bit? Just curious.

Originally Posted by zeeway If you do all that..and have your machine cut all the frets in one machining process, and if there are positioning errors, it is likely they will be proportional as you go down the fretboard. This means the scale will be correct musically, but the "scale length" may be a little longer or a little shorter than you designed...which to me is not much different than having to "compensate" a bridge. Hope all this makes sense. My bottom line here is go for it. Angie

That does make some sense to me, though it seems like random variation in the lead screws and such would lead to random variation in the the precision of the cuts. At least that's how I envision it in my mind. Hopefully that sort of thing would be so small it wouldn't matter though. But yes, I definitely understand how a uniform variation in the positions in each slot could most likely be dealt with quite easily.

Originally Posted by studysession I have the 2x3 kit that I finished building the week before Christmas.. And I have built guitars with. I am extremely satisfied with the kit and it's accuracy.

Sweet pictures. Good to hear somebody else is using the Fine Line machines for instrument building. Did you do any fret slotting for the guitar you posted pictures to, or was that a partially finished neck you already had? Thanks for sharing.

[studysession]

I only made the body. That neck is from my Saga TC-10 which is a TELE-clone. Eventually I plan to also CNC the neck but for now I am just focusing on the bodies and buying the necks.

Thanks for the compliments....