DIY setup out of frustration
- Thread starter Dohle
- Start date
LorenFL
Well-known member
needless to say I'm feeling much more confident as I did before.
Glad to hear you're making some progress with your setup!
This thread inspired me to revisit my own setup. I'd never actually measured anything, so I broke out the feeler gauges.
My action at the 12th fret is pretty low already, so I'm not going to target shaving the saddle any more. I was checking with .078" (it was just an easy feeler stack to get) The A string was actually tight! Little bit of wiggle room on the low strings, but definitely not more than .085-.090 except maybe on the G.
First fret action I checked with .020". A string was right about there, some of the other strings were WAY tall!
I did all of my filing with a folded piece of 220 grit sandpaper! It's likely widening the slots just a little bit, but I was concentrating my efforts on the bottom of the slot, and it worked just fine.
Given that I'm not planning to do any more saddle adjustment, I was going for the best intonation that I could get, and I got it within 3-5 cents on the first fret, and within 2-4 on the second fret. Intonation at the 5th and 7th and 12th fret is even better. I'm pretty happy with that.
The one string that was giving me problems, and is still a little further out than the rest is my E string. I actually brought that slot down far enough to create a buzz! If I really pluck that open string, it buzzes against the first fret. Time will tell if that's a problem when I'm actually playing. If it is, I'll have to decide if I want to "fix" the nut, or maybe take a file to the first two frets and shave them a tiny bit. (I've never done that before!)
It sounds SO much better! Chords sound more harmonious. I had some wild differences in intonation between different strings. I might still be a few cents off, but at least all the strings have the same error and are in tune with each other!
I've got a new set of Worth Brown strings coming. Will be interesting to see if they sound as good as a lot of people think they do.
Graham Greenbag
Well-known member
Reading the above made me recall that I now aim to set the strings just a little higher, my target is 0.6 mm (0.024”), and if I overshoot to 0.5 mm it doesn’t matter. I too use feeler gauges under the strings (string to first fret gap) but as the strings’ readily nudge out of the way (upwards) it’s so very easy to get false readings (that indicate that the gap is bigger than it actually is).
Last edited:
LorenFL
Well-known member
I'm more concerned with getting in tune than the specific numbers, but it was interesting to see where I was/am.
I forgot to mention last night... one of the things I picked up from this thread, as dumb as it might seem... I had it in my head that I needed to loosen the strings and lay them to the side to do work on the nut slots. It never occurred to me that there was enough stretch in the strings to just lift them out of the nut and move them over! I did that last night, and it allowed me to work MUCH more quickly making fine adjustments to each individual string. I didn't even have to retune the string half the time, it was still in tune.
Now I'm going to have to start to research saddle compensation. I've got the action as low as I can get it at the nut, and pretty darned low at the 12th fret, and all strings are just slightly sharp, even at the 12th. I can probably improve that at the saddle by changing the effective length of the string. Something else to experiment with, but I'll have to get a saddle blank to play with.
I forgot to mention last night... one of the things I picked up from this thread, as dumb as it might seem... I had it in my head that I needed to loosen the strings and lay them to the side to do work on the nut slots. It never occurred to me that there was enough stretch in the strings to just lift them out of the nut and move them over! I did that last night, and it allowed me to work MUCH more quickly making fine adjustments to each individual string. I didn't even have to retune the string half the time, it was still in tune.
Now I'm going to have to start to research saddle compensation. I've got the action as low as I can get it at the nut, and pretty darned low at the 12th fret, and all strings are just slightly sharp, even at the 12th. I can probably improve that at the saddle by changing the effective length of the string. Something else to experiment with, but I'll have to get a saddle blank to play with.
LorenFL
Well-known member
Oh yeah... for reading strings with feeler gauges. You don't need much of a gap between the string and the feeler to allow the string to "ring" or even just "buzz" a little. If you use a gauge that's just slight smaller than your target size... you can whittle away at it until the string goes from "ringing" on a very light pluck to "buzzing" because it doesn't have enough clearance to ring.
Measuring that way, you're never contacting the string, so you're not pushing it upward and getting a false reading.
You could then take it a step further and go up one step at a time with the feeler until you find the point at which the string won't even buzz... that should be pretty close to "actual" string height. Not that it matters. What matters is that it is in tune, doesn't buzz, and feels right.
Measuring that way, you're never contacting the string, so you're not pushing it upward and getting a false reading.
You could then take it a step further and go up one step at a time with the feeler until you find the point at which the string won't even buzz... that should be pretty close to "actual" string height. Not that it matters. What matters is that it is in tune, doesn't buzz, and feels right.
Saddle compensation is pretty straightforward. I like to start with a square saddle blank, with the top edges swiped once or twice across 320 sandpaper so they're no longer sharp. Cut the saddle to the correct height for the action you want before you start, since action changes will impact intonation.
The key relationship to get correct is the spot where the string breaks over the saddle - called the witness point. To compensate a saddle, you file an angle along the front edge of the saddle under a given string, which moves the witness point back (away from the neck) making the effective length of the string longer. As you move it back further, fretted notes will slowly become flatter. You check the accuracy by looking for a correct note at the 12th fret. So you just file a bit, une the string, check the note, and repeat until the 12th fret plays true.
Once the note is true, it's good practice to file the back side of the saddle under that string until your string breaks cleanly down to it's anchor point. This is called relieving the saddle and will make for a cleaner and more stable tone than leaving the saddle squared off behind the witness point. You don't need to file it to a sharp edge, just enough such that the string bends cleanly down instead of essentially having two different bends (one over the witness point and one on the back edge of the saddle).
After all four strings are compensated and relieved, you can take the saddle back out one last time and run it over some very fine sandpaper (I use 800 grit abranet) to give it a polish and ensure there are no sharp edges.
...and then you're done!
Of all the setup practices, compensating a saddle is a good one for highly discerning individual players to get good at, since getting dead-accurate intonation is somewhat personal in nature. If I set the intonation correct for my playing, but you typically fret heavier/lighter than me, you may find the intonation is not accurate for you. This is more true for ukes than many other stringed instruments since uke strings are soft and scale lengths are short relative to the displacement of the string. I can set a uke at 2.5mm action and it will be at 2.5mm action no matter who plays it, but my dead-on-accurate intonation may be 5 or 10 cents off for someone with a very different fretting technique. Of course, it's probably a safe bet that many people wouldn't even notice that kind of variation (and it's probably a safer bet that the vast majority of ukes out in the wild are off by more than that anyways).
The key relationship to get correct is the spot where the string breaks over the saddle - called the witness point. To compensate a saddle, you file an angle along the front edge of the saddle under a given string, which moves the witness point back (away from the neck) making the effective length of the string longer. As you move it back further, fretted notes will slowly become flatter. You check the accuracy by looking for a correct note at the 12th fret. So you just file a bit, une the string, check the note, and repeat until the 12th fret plays true.
Once the note is true, it's good practice to file the back side of the saddle under that string until your string breaks cleanly down to it's anchor point. This is called relieving the saddle and will make for a cleaner and more stable tone than leaving the saddle squared off behind the witness point. You don't need to file it to a sharp edge, just enough such that the string bends cleanly down instead of essentially having two different bends (one over the witness point and one on the back edge of the saddle).
After all four strings are compensated and relieved, you can take the saddle back out one last time and run it over some very fine sandpaper (I use 800 grit abranet) to give it a polish and ensure there are no sharp edges.
...and then you're done!
Of all the setup practices, compensating a saddle is a good one for highly discerning individual players to get good at, since getting dead-accurate intonation is somewhat personal in nature. If I set the intonation correct for my playing, but you typically fret heavier/lighter than me, you may find the intonation is not accurate for you. This is more true for ukes than many other stringed instruments since uke strings are soft and scale lengths are short relative to the displacement of the string. I can set a uke at 2.5mm action and it will be at 2.5mm action no matter who plays it, but my dead-on-accurate intonation may be 5 or 10 cents off for someone with a very different fretting technique. Of course, it's probably a safe bet that many people wouldn't even notice that kind of variation (and it's probably a safer bet that the vast majority of ukes out in the wild are off by more than that anyways).
LorenFL
Well-known member
You just saved me a couple hours of research. Thanks!
LorenFL
Well-known member
I'd already decided to order a bone saddle to play with. Decided to fiddle around and see if I needed to order anything else (like a bone nut) before placing my order.
Sure enough, if I thrash a chord with an open G string, it buzzes now. Dang it.
(Side note... everybody talks about the super-glue trick for un-slotting a nut... nobody talks about what If you don't have any super-glue handy? Well, I did have a paper plate and a pair of scissors. I cut a little strip of that thick paper and stuck it under the string in the slot on the two offending strings. Problem temporarily solved!)
Being one who likes to think of such things as a learning exercise, I figured I'd get the feeler gauges out again and see just how low I'd gone. Surprisingly, NOT that far! A and E were right at about .018-.019. (E is the one that buzzes pretty readily) G was at .019. I didn't get a reading on the C, it was more than the others, probably more than .020.
My take-away there is that YES, you can go to far, even if you're targeting .020. Now, worthy of note, I'm using Soprano strings on a Concert uke specifically because I like light tension. That means the strings are going to move more when they vibrate, so it may very well be that something like .022 needs to be my minimum.
But, once I compensate the saddle, I won't need to go crazy low with the nut slots in the name of intonation. So, there's that.
Time to go order that new nut and saddle. We'll see if there's anything to the notion that bone is better than plastic while we're at it.
Sure enough, if I thrash a chord with an open G string, it buzzes now. Dang it.
(Side note... everybody talks about the super-glue trick for un-slotting a nut... nobody talks about what If you don't have any super-glue handy? Well, I did have a paper plate and a pair of scissors. I cut a little strip of that thick paper and stuck it under the string in the slot on the two offending strings. Problem temporarily solved!)
Being one who likes to think of such things as a learning exercise, I figured I'd get the feeler gauges out again and see just how low I'd gone. Surprisingly, NOT that far! A and E were right at about .018-.019. (E is the one that buzzes pretty readily) G was at .019. I didn't get a reading on the C, it was more than the others, probably more than .020.
My take-away there is that YES, you can go to far, even if you're targeting .020. Now, worthy of note, I'm using Soprano strings on a Concert uke specifically because I like light tension. That means the strings are going to move more when they vibrate, so it may very well be that something like .022 needs to be my minimum.
But, once I compensate the saddle, I won't need to go crazy low with the nut slots in the name of intonation. So, there's that.
Time to go order that new nut and saddle. We'll see if there's anything to the notion that bone is better than plastic while we're at it.
OP
OP
Dohle
Well-known member
Time again for a quick update. Today I had a go at the Kamaka whose action at the nut was closer to 1.0 mm than the 0.5 mm I prefer causing all kinds of intonation and playability issues. I started at the C string but quickly noticed that the action at the 12th fret was really high as well. I knew it was a bit higher than on most of my other ukes but I guess I had never measured it precisely before. It was actually over 3 mm so I decided to take that down before messing with the nut slots. I did mention previously that I'm not that fussy about action at the 12th but over 3 mm is a bit too much even for me. I got the action to almost exactly 2.5 mm which is my personal sweet spot. I actually overshot it a bit since I wasn't aiming that low but thankfully I didn't go any lower. Aside from that, it was an easy and quick job.
Then back to the nut slots. I started to get a bit confused since the height at the nut was barely going down when I started filing but I'm now guessing that's because I was widening the nut slot at first rather than actually lowering it. Because of this confusion, I was happily filing away at the nut slot for the C string so I almost overshot that one as well. Thankfully it again landed almost exactly at my sweet spot of 0.5 mm. I had a quick panic moment because I started to hear buzzing when I brought the C string back to tension but that was probably because of the other strings that were slack and vibrating at the saddle or somewhere else when I played the C string alone. The slot for the E string was the tricky one since that was a bit crooked to the left which caused the string to get caught in the slot when tuning the strings so I had to try to change the angle of the slot more towards the tuning post. Luckily that wasn't hard at all. In fact, filing the rest of the nut slots went pretty easily over all. I brought all of them down to 0.5 mm or as close to that as I was comfortable with using the tools I had.
Once I had done all of the nut slots, I checked for intonation and it immediately seemed much better than before. Likewise with playability, as barre chords specifically felt much more comfortable, and I didn't seem to have caused any buzzing. However, I did notice a very bizarre thing with certain notes, specifically with A# played at the first fret of the A string. My Snark tuner showed that the A# played there was massively sharp even though the A string was perfectly in tune. I started to check intonation by ear by playing the same note on other strings simultaneously to the A string and I didn't notice any discrepancies. To be fair, the Snark tuner occasionally showed the A# note a bit sharp played on other strings as well but it wasn't very consistent. I then checked the same thing with my Peterson strobe tuner and it showed that the A# was mostly in tune, but if you know how Peterson tuners work they can be a bit fiddly at least with certain instruments as the little moving circle things sometimes just go flying all over the place, so that wasn't really conclusive either. I do have to wonder whether there's something going on with harmonization regarding the A# note that throws the Snark at least a bit out of whack. I continued to test the intonation with chords that use the A# note, and I even took out another uke on which I know the A# on the A string is spot on, played them simultaneously, and I still didn't notice that the note was off on the Kamaka. All the chords and scales and everything else I tried seemed to be fine, so I have no idea what's going on there. I'll continue testing tomorrow but to me it seems that the uke is in much better shape than it was before despite that weird A# note.
Then back to the nut slots. I started to get a bit confused since the height at the nut was barely going down when I started filing but I'm now guessing that's because I was widening the nut slot at first rather than actually lowering it. Because of this confusion, I was happily filing away at the nut slot for the C string so I almost overshot that one as well. Thankfully it again landed almost exactly at my sweet spot of 0.5 mm. I had a quick panic moment because I started to hear buzzing when I brought the C string back to tension but that was probably because of the other strings that were slack and vibrating at the saddle or somewhere else when I played the C string alone. The slot for the E string was the tricky one since that was a bit crooked to the left which caused the string to get caught in the slot when tuning the strings so I had to try to change the angle of the slot more towards the tuning post. Luckily that wasn't hard at all. In fact, filing the rest of the nut slots went pretty easily over all. I brought all of them down to 0.5 mm or as close to that as I was comfortable with using the tools I had.
Once I had done all of the nut slots, I checked for intonation and it immediately seemed much better than before. Likewise with playability, as barre chords specifically felt much more comfortable, and I didn't seem to have caused any buzzing. However, I did notice a very bizarre thing with certain notes, specifically with A# played at the first fret of the A string. My Snark tuner showed that the A# played there was massively sharp even though the A string was perfectly in tune. I started to check intonation by ear by playing the same note on other strings simultaneously to the A string and I didn't notice any discrepancies. To be fair, the Snark tuner occasionally showed the A# note a bit sharp played on other strings as well but it wasn't very consistent. I then checked the same thing with my Peterson strobe tuner and it showed that the A# was mostly in tune, but if you know how Peterson tuners work they can be a bit fiddly at least with certain instruments as the little moving circle things sometimes just go flying all over the place, so that wasn't really conclusive either. I do have to wonder whether there's something going on with harmonization regarding the A# note that throws the Snark at least a bit out of whack. I continued to test the intonation with chords that use the A# note, and I even took out another uke on which I know the A# on the A string is spot on, played them simultaneously, and I still didn't notice that the note was off on the Kamaka. All the chords and scales and everything else I tried seemed to be fine, so I have no idea what's going on there. I'll continue testing tomorrow but to me it seems that the uke is in much better shape than it was before despite that weird A# note.
LorenFL
Well-known member
I love a happy ending!
Those basic chords on the first 3 frets sound SO much better when the intonation is correct, don't they?
Those basic chords on the first 3 frets sound SO much better when the intonation is correct, don't they?
SteveZ
Well-known member
An alternative to using slotted files for nut slot adjustment is to use welding tip cleaners. The cleaners are simply mini round files and available in any hardware store for less than US$10. They come in a small container holding many in various diameters. I've used them for years on my instruments with success.
Attachments
Last edited:
OP
OP
Dohle
Well-known member
Well, the happy ending didn't last. After what I thought was a successful setup, I noticed that the C string was now getting caught in the nut slot when tuning the string and this time it was because I cut the nut slot just slightly too narrowly. So obviously I decided to widen the slot just a little. Unfortunately, that caused the string to vibrate in the nut slot causing that sort of chimy buzz. I gently cut the nut slot again with the narrower file which got rid of the chime but it was too late, the nut slot was too low already. Now there's a very pronounced buzz on the open string on the C whenever I play it with any amount of force. Curiously, there's also a buzz on the C string when fretted so I don't know if I took the saddle too low as well but it definitely isn't very low at about 2.5 mm.
*sigh*
To the luthier it is then...
*sigh*
To the luthier it is then...
LorenFL
Well-known member
Why give up now? You've learned a bunch. You can either fill your nut slots and go at it again, or replace the nut if you want to get crazy. And replacing a saddle is even easier.
You've not done anything that can't be easily fixed.
I replaced the saddle on mine the other day. Ended up accidentally ordering "synthetic ivory" saddles, so my bone experiment will have to wait for another day. But, with a little grinding, filing and sanding, I was first able to replicate my existing saddle with about 1mm more height. Then I beveled the top so that the string contact point was about 1mm further back. Finished by sanding the bottom down to just a hair taller than the original.
End result is that my action is a tiny bit higher (it was too low before), and the intonation is WAY better!
The only fault now is the E string is still a little funky. I could do yet another custom saddle and try to further compensate that E string, but I'm not sure I'll bother. We're talking 2-4 cents here. It's really, really close. The other strings are +2 to 4 cents at the 2nd fret, the E is just another 2 cents beyond that.
Bought some super glue yesterday to fill my E slot. (it's currently got a piece of paper under the string) Trying to decide if I want to do that, or just start with a new nut blank. I guess I'll go the easy route first just so I can say I did.
You've not done anything that can't be easily fixed.
I replaced the saddle on mine the other day. Ended up accidentally ordering "synthetic ivory" saddles, so my bone experiment will have to wait for another day. But, with a little grinding, filing and sanding, I was first able to replicate my existing saddle with about 1mm more height. Then I beveled the top so that the string contact point was about 1mm further back. Finished by sanding the bottom down to just a hair taller than the original.
End result is that my action is a tiny bit higher (it was too low before), and the intonation is WAY better!
The only fault now is the E string is still a little funky. I could do yet another custom saddle and try to further compensate that E string, but I'm not sure I'll bother. We're talking 2-4 cents here. It's really, really close. The other strings are +2 to 4 cents at the 2nd fret, the E is just another 2 cents beyond that.
Bought some super glue yesterday to fill my E slot. (it's currently got a piece of paper under the string) Trying to decide if I want to do that, or just start with a new nut blank. I guess I'll go the easy route first just so I can say I did.
I can maybe relate to why going to the luthier. I'm a perfectionist (sadly, it's taken me 30+ years to realize this) so if I think something is "wrong" (this may even be just psychological) it'll nag me until it is fixed. So in this case, if I had used superglue or whatever to repair the nut; and, even if it works 100%, it'll probably nag me to no end.
LorenFL
Well-known member
Yeah, I get that. But, the more particular you are... setup is such a personal thing... you could pay someone to do it, and still not be happy with the result.
I view this exactly the same as why I choose to do most of the maintenance and repair work on my own vehicles. I know I'm going to take the time to do it right and to my satisfaction. If I make a little mistake, I know I can fix it and I have no one to blame but myself. If I take it to a shop and they mess something up... now I have to take it BACK to the shop and deal with them again. It's often easier and less time-consuming to just DIY.
But, there are things I'm willing to do, and things I'm not. I pay someone to mow my lawn. I'm about to pay someone to paint my house. I'll definitely pay someone to do roofing! And if I had to tune a piano, I'd probably take a pass.
Tinkering with the basic setup on a ukulele... I find it almost cathartic. It's simple, the results are immediate. Just takes a little patience.
I view this exactly the same as why I choose to do most of the maintenance and repair work on my own vehicles. I know I'm going to take the time to do it right and to my satisfaction. If I make a little mistake, I know I can fix it and I have no one to blame but myself. If I take it to a shop and they mess something up... now I have to take it BACK to the shop and deal with them again. It's often easier and less time-consuming to just DIY.
But, there are things I'm willing to do, and things I'm not. I pay someone to mow my lawn. I'm about to pay someone to paint my house. I'll definitely pay someone to do roofing! And if I had to tune a piano, I'd probably take a pass.
Tinkering with the basic setup on a ukulele... I find it almost cathartic. It's simple, the results are immediate. Just takes a little patience.
Yes, I agree that the perfectionist attitude only covers some things. But, it doesn't mean that I must personally do the work. As long as the work is done competently and I think the result is good, I'm happy. In the above case, a nut with a repair would not be acceptable for me (maybe it is because I like the ukulele); so I'd want a new nut put in its place.
OP
OP
Dohle
Well-known member
I'm afraid I have to go with clear here. I would love to continue to experiment myself but replacing a nut is another thing I'd have to learn. Maybe in the case of a cheaper uke I would continue trying myself but I just can't justify learning on this expensive an instrument. I know I would be just second guessing myself at every turn as I have been doing so far. I know that the luthier will do a satisfactory job once I give him the specs I'm after, so that'll be that and I can have peace of mind. That's not to say I won't be doing work on other ukes in the future as I've definitely learned a lot and know what mistakes to avoid (to a certain extent at least).
I do have to say that I'm a bit distraught about all this affair. I now have three K brand ukes, two of which have needed a setup from a luthier and one that I had to setup myself. Of course that says more about the shops I bought them from rather than the quality of the brands but still I just can't properly express how frustrated I am.
I do have to say that I'm a bit distraught about all this affair. I now have three K brand ukes, two of which have needed a setup from a luthier and one that I had to setup myself. Of course that says more about the shops I bought them from rather than the quality of the brands but still I just can't properly express how frustrated I am.
SteveZ
Well-known member
If you decide to have the nut replaced by someone, suggest you get the old nut back. It would give you a chance to practice refilling and reslotting on a spare item.
OP
OP
Dohle
Well-known member
Ok, time for another update and a much more positive one. After a good night's sleep, I decided that I'm not going to give up after all. I thought to myself that if I can just get the nut removed I can maybe shimmy it up similarly to the saddle or maybe even replace it. Removing the nut was the most nerving part for me beforehand because I was afraid I'd damage the uke but it actually wasn't that bad. I had to work on it for maybe 15 minutes but it came off quite easily in the end with no damage to the uke. So, I shimmied the nut and saddle with a bit of cardboard, testing out a few different heights, and eventually I got the action at both ends to a pretty good height with absolutely zero buzzing. What a huge relief that was. Granted, the intonation at the first fret isn't quite as good as I'd want it but it's a slight improvement nonetheless. Playability is more noticeably better than originally. So I'm basically back to square one with little improvements, and that's not a bad thing at all after yesterday's catastrophe. I could of course get back to taking the nut slots down more carefully but I'm not going to do that. Instead, I have ordered a bunch of nuts and saddles so I can make new proper ones for the Kamaka instead of relying on my cardboard solution (currently the nut isn't even glued on). And additionally, I will repeat the same process for my KoAloha which has a really low action as well currently. Not as low as to cause any serious buzzing but the nut and saddle could definitely use some improvement.
I think I got a bit too upset yesterday because of the failure but now I'm determined to get these ukes in shape. Thanks for the encouraging comments so far, LorenFL particularly.
I think I got a bit too upset yesterday because of the failure but now I'm determined to get these ukes in shape. Thanks for the encouraging comments so far, LorenFL particularly.
SteveZ
Well-known member
Keep the faith! The first couple ukes I customized took more than one attempt, considerable cursing and swearing, and more time and effort than I realized. Eventually, it gets easier, faster and almost routine. One thing's for sure, the sound from any instrument seems much sweeter when it's the result of personal effort to customize.
