Neck Relief

Time to bring together many of the thoughts in posts I have made about neck relief. In this thread I am going to look at the following:

1. Is relief necessary or even advisable?
2. How much relief is needed?
3. How do you measure neck relief?
4. What do the "experts" have to say?

And I'll likely end up adding other topics related to neck relief.

What I won't be doing is identifying who says what. That's because I want to be able to critique some of the things others have said and many bass owners get defensive when I disagree with their manufacturer, or even their favorite setup guide. I'm not interested in getting into a squabble here. I will be presenting my opinions backed up with my reasons for those opinions. Feel free to disagree, but do so with some backing facts. And by the way the argument "Mr. Bass Idol says it's so" is not a valid argument unless you also provide his reasoning, in which case let's not bring his name into it, just his reasoning.

A word of caution. I will be referencing some ideas that you may associate to a particular person, and I may be in disagreement with some of those ideas. It can be difficult to dispel myths without that association to a specific becoming apparent. But I am in no way disrespecting the source, only pointing out something that needs either correction or clarification in my opinion. Once again, I am trying to keep personalities out of this.

Now on to the real task at hand.

 

I am not going to start by defining 'neck relief'. I'm going to assume you already know.

So let's start by looking at the question about the need for neck relief. This need is often illustrated by a picture something like this:


where the grey area is the vibrating string. This illustration intends to show that the string vibrates in an arc and to maintain proper string clearance along the length of the fretboard and avoid buzz, the neck needs to be bowed in a similar arc. It seems logical, but it's not as depicted. We'll get to that in a minute.

I have surveyed 30 makers and technicians to see what they have to say about relief. Amongst those are a range of sources from artists who make high-end hand made instruments, to some of the most popular factories, to popular on-line guides and technicians. Of those surveyed none advocated a truly flat fretboard (nor do I). One notable manufacturer that is often misquoted as advocating a dead flat board actually says that you should adjust his instruments "almost dead straight". So despite what is commonly perceived, this manufacturer does not advocate no relief at all. Another very notable maker has been quoted as saying that he adjusts the neck on his basses to be "straight with a bit of relief". Again none of those surveyed advocates a truly flat neck.

So let's assume some degree of relief is advisable. Let's go back to the question of "why". For this I will use a simplified and exaggerated structure (you may have seen this in another thread or two):



Here is a 5 fret neck with a nut (in blue) and a string (green) which rests on the saddle of the bridge (black). We start by fretting at the 4th fret:



For the purpose of this study we will assume that the clearance over the 5th fret is the minimum necessary to avoid fret buzz. The amount of clearance is represented by the red diamond. Now if we fret at the 3rd fret, we discover that there is not requisite amount of clearance over the 4th fret:



We need to raise the 3rd fret a little to get enough clearance:



And now of course, when we fret at the second, there is not enough clearance at the third fret, so we need to raise the second fret:



And when we raise the second, we'll also need to raise the first:



And of course we need to raise the nut as well. What we end up with is a layout like this (the red line being a line along the tops of the frets):



There's the relief curve. It is needed to provide a consistent clearance between fret and string along the length of the fretboard. No matter where we fret there will be a consistent clearance over the frets. The alternative would be to raise the bridge to give the necessary clearance at the first fret, but using the minimum we have defined here it would look like this:



The bridge has been raised substantially causing very high action in the upper frets. And if you work this backwards you will find that you now have unnecessarily high clearance over the 5th fret when fretting the 4th.

So the need for relief is because a certain amount of clearance is needed to prevent the vibrating string from touching frets further up the board. The minimum clearance needed is over ht next adjacent fret and when that is applied along the board we end up with a curved surface, i.e. relief.

 

In the previous post we can see the reasoning behind the use of relief in the neck. The question remains if you really need relief. The answer is definitely maybe.

One can play a bass without relief, but it won't be set up optimally. If you are a heavy handed player that like high action you will reap little benefit from relief. Conversely, if you like low action, play with a lighter touch and want to play cleanly (no buzz), the right amount of relief is highly beneficial. But the lower you want the action the more critical it is to have very even fretwork and just the right amount of relief. You can overdo the amount of relief too and it will result in less than optimum playing conditions. But regardless of how you play, what strings you use, what tunings you use, to get the best possible performance from your bass that right bit of relief is a very significant part of the optimum setup.

I want to take a moment to talk about the term "optimum". I am using it to mean "just right" in the sense that more wouldn't be as good, nor would less. A bit more may be very workable for you , or a bit less. But it could be better. Maybe not enough better to be worthwhile for you, but in a nit-picking way and for those with OCD optimum means the best with all factors taken into consideration. So you can not have an optimum setup when the frets are un-even, though you may have a "best-that-it can-be" setup with the factor of uneven frets considered. So, "no-compromise" is the way I will use the term "optimum".

So that brings us to the question about what is the optimum amount of relief? It depends. It's like asking "how long is a piece of string?". It will depend on a number of factors (the ones that determine what is "optimum"). That includes the behaviour of the strings, how you pluck them, the type and mass of the string and on and on and on. I cannot answer the question, but I can point at the factors involved and offer some advice on how to deal with them. And there are lots of others that provide the same advice. The trouble is that there seems to be a whole lot of disagreement out there. So I will proceed by looking at other sources and providing my evaluation.

Let's start by looking at how it is recommended that you measure relief. It should be really straight forward, but of the 30 sources I surveyed there is a surprising disparity. 27 advocated using as string as a straightedge and measuring the gap under the string as a way of determining the amount of relief. Of the 3 others, 1 advocated using a real straightedge and 1 did not specify how to measure the relief. I think that using the string as a straightedge is effective with a specific limitation. The actual gap is usually measure by slipping a feeler gauge between the string and a fret, and it is difficult to determine if the gauge is deflecting the string as you insert it. You want the gauge to slip in there with no gap and without deflecting the string. If you are careful and have good eyesight you might be able to do this, but a small deflection of the string may amount to a couple of thousandths of an inch, and that is significant if you are chasing the optimum amount of relief. If you use a real straightedge it will not deflect and you will get a reading you can be more confident about.

The next question is how you use the string or straightedge to determine the amount of relief. Once again there is not a consensus amongst the sources I polled. Since most of the sources advocated using the string as a straightedge, I will look at those sources to see how they go about it. In all these cases it is advised to fret the string at two different points (such as the fist and last frets) and measure the relief somewhere between the two. But what two frets? They don't agree.

First we should ask why we are fretting at all. We could just measure the gap between the first fret and the underside of the string, then the last fret, and attempt to calculate what the gap would be in the middle assuming a truly flat neck, then compare that to the actual measurement at that fret. The difference would be the amount of relief. But we can simplify the measurement by fretting the string near either end and finding the gap somewhere between them.

Now you might expect that everyone would advocate fretting at the first fret. 25 of those surveyed used the first fret. Two advocated using the second fret, and 3 did not specify at all. And at the other end, 13 recommended fretting at the last fret while 11 advised to use the fret where the neck meets the body. We'll examine the reasoning behind both in the next post.

Then there is the question of where the measurement of the amount of relief should be made. This varied from source to source as well, some of which is related to the source's choice of which frets the strings are fretted at. In some cases it was recommended to measure at the 6th fret, and others as high up as the 9th fret. As we will see later where you measure will provide different results, so chasing the optimum is a challenge.

In the next post I will provide an analysis of where to measure and where our straightedge should be resting.

 

Here's the question - If I am using the string as a straightedge, do I fret it at the fist and last fret, or the first and 15th? Let's start by assuming you will fret at the first - but wait, there's not universal agreement about this. Of the 30 sources I surveyed, 25 said to fret or capo at the first. 2 said to fret at the 2nd fret and the remaining 3 did not specify. We'll ignore the last 3 since they seem to be ignoring the question.

So why would 2 sources recommend fretting at the 2nd fret rather than the 1st? Both were makers of basses and said that they intentionally leave the 1st fret high to provide a bit more clearance over the second fret. Why they felt that was necessary was not explained, though I suspect that they believe that neck relief did not provide enough clearance on their instruments. With the first fret a little higher, they recommend stopping the string at the 2nd fret to take that little extra height out of the equation when setting relief.

So for the most part we can accept that we will fret the string at the first fret. And we need to fret it at the other end as well allowing the string to be a straightedge at fret-top level. As pointed out earlier, of the 30 surveyed, 13 recommended fretting at the last fret, while 11 suggested the fret where the neck meets the body (usually 15th, 16, or 17th fret), 2 suggested other locations (we'll look at those later), and the rest did not specify.

Disregarding the odd 2, there is almost an even split between those suggesting the last fret and those recommending the fret at the body joint. What do the 11 have against using the last fret? There are 2 principal reasons. Since relief is controlled by the truss rod, it makes sense to do your measuring where the truss rod has an effect. Since the heel of the neck is bolted to the body, the truss rod will have no effect in this area - it is being held flat to the neck pocket. The truss rod will only affect the free part of the neck from the first fret to the body joint.

The second reason for not using the last fret for the seat of your straightedge is to avoid any potential "ski jump" from influencing the relief setting. A ski jump is actually a kink in the neck occurring just outside the neck-to-body joint. The truss rod cannot correct such a kink and since we are using the truss rod to control relief, we should be measuring where it has its effect. It is the same reasoning that has us fretting at the first fret - so that we take any anomalies at the nut out of the equation. Here's a diagram that may help:



The heavy black line as a neck with a kink around the 14th fret (red arrow). We've fretted the string at first and last frets and at our measuring point (7th fret - green arrow) it looks like we have good relief. But we don't. The board is flat from 1 to 13, has a kink at 14 then is flat again from 15 to the end. When we set the string height at the bridge to give us the string height we want at the 12th fret, anything from the 14th fret to the end of the board will buzz or choke completely (yellow line).



You may not have a ski jump, and if you don't you can use the last fret as the end point of your straightedge, but it seems advisable to take the possibility of a ski jump out of the equation.

So why do some ever suggest using the last fret as the end point of the straightedge? Some basses never develop ski jumps, so that is not an issue on them. And I think that some makers would not advise you to use the fret at the body joint "in case there is a ski jump" because that is an admission that their model can develop this problem. Not a good admission if you want to sell your basses. But, ski jump or not, there is a good point to be made about measuring in the area that can be adjusted and not including the areas that cannot.

So now we have established the end points of our straightedge. Measure the amount of relief half way between them, right? Not necessarily. Not all necks bend in the same curve. It can differ from maker to maker, model to model and even within the same make/model. It has to do with the profile of the neck, the wood it is made from, and the action of the truss rod. Studies I have done on relief profiles on a number of brands and models has shown that there is a significant amount of variance. In 34" scale basses alone I have found that the maximum amount of relief was found at the anywhere from the 5th to the 8th fret. If you are using the 1st and last fret for your straightedge, the mid point is the 8th fret, but when using the fret at the body joint as the straightedge termination, the mid point is the 6th fret. Given all of this a good choice would be to find the fret where the relief is the largest - that will give you the best view of the relief on that particular neck.

Now we've established the end points of our straightedge and have found the maximum gap between fret top and underside of the string. As mentioned in a previous post, using a feeler gauge under the string is slightly problematic in that it may deflect the string slightly and give a false reading. If our eyesight is sharp enough we may be able to detect this slight deflection, but mine is certainly not sharp enough any more. I could use a magnifier, but instead I use my ears, which still work pretty well despite years of playing in bands. I slip a feeler gauge between the fret and the string then lightly tap on the top of the string. If I hear a "click" I know there was a gap between feeler and string. I try again with the next thicker gauge, and continue this way until there is no click. At that point I know that the real reading is somewhere between the last two gauges. If I hear no "click" starting out, I step down one gauge, and continue as above.

 

One of the characteristics of relief is that it can look very different from one instrument to another. Here I have measured two J-style necks, same dimensions, same materials (maple neck, rosewood fretboard). Both were adjusted to provide the same amount of relief found at the maximum deflection point. A straightedge was laid on the neck spanning the first to last frets and feeler gauges were used to measure the gap between the straightedge and the fret at each position. The results are graphed with an exaggerated vertical axis to make the results obvious, graduated in thousandths of an inch. The horizontal axis is divided into fret positions.


The green line is the depiction of the relief of a Korean made Squier J-bass. The red line is a Mexican Made Fender Jazz. Looking first at the Squier, we can see that the neck did not have an even relief curve at all, rather it had two kinks; one occurring around the 5th fret, and another at the 15th. The Fender had a relatively straight neck from the 2nd to the 8th fret, with most of the curve happening between the 8th and 13th frets. In both cases the area from around the 16th to 20th frets is straight . This is understandable since this is the area of the neck that is fastened to the body - it's not going to bend.

So why the difference? In this case it is largely a matter of the quality of wood used in the neck. The Fender had a good quality of maple in the neck, relatively straight grained with no runout and closely spaced growth rings which is usually an indicator of good rigidity. By comparison, the Squier had wider growth rings, some significant amount of runout, and most importantly a change of direction of the grain around the 5th fret as seen from the side of the neck.

Neither of these necks provided an ideal relief curve, which would be a smooth curve free of kinks. By comparison, here is the relief curve measured the same way on a J-style bass from a high-end maker:



As before we see a flat section from about the 14th fret through to the 20th. From there to the 1st fret the curve is even. Visually, the neck is straight grained with no runout, and very close growth rings. This is a high-quality piece of maple with very good rigidity. And the rosewood fretboard is also straight-grained with densely spaced growth rings.

This all brings into question where we should be measuring relief. In the last case which is closer to ideal than the other two, the maximum relief occurs at the 7th fret. On the Squier it occurs at the 5th fret, while the Fender is at the 10th. If we take an average of the Squier and the Fender, it would suggest that the 7th fret is a good compromise, and when we compare that to our third example which has the best relief curve, it suggest that the 7th fret is indeed a good compromise. And between the Squier and Fender the measurement at the 7th fret is only 1/1000th of an inch away from the maximum deviation, though in different location on the fretboard.

So what we have gleaned so far is that we should measure the amount of relief at the 7th fret with respect to a straight line from the 1st fret to the fret where the neck meets the body. This offers the best compromise while taking the potential for a "ski jump" out of the equation and removing inconsistencies at the nut from the measurements. Though it may be a "compromised" measurement, the allowance being made is very small - in the cases shown it's about 1/1000th of an inch.

 

Works for Jimmy, works for you. But it ignores many of the points I have brought up. Good enough for you both. Not best.

 

Best practice would be to release string tension so you are not overtaxing the truss rod - i.e. let it adjust the neck and not have to fight the strings to do it. OTOH, I often adjust the rod without loosening the strings. If it's hard to turn the nut I stop and slack the strings.

 

There are a few bits in Spector's setup procedure that I disagree with, and an extra high first fret is one of them. Another is stretching the string - it's a poor way to set the witness points and doesn't ensure that "that the full tension has been achieved". Using the string as a straightedge by fretting the 2nd and last fret can lead to bad adjustments. The truss rod cannot have any effect on the neck from the point where the neck meets the body to the butt end of the neck. So you should stop the string at the first fret and the fret where the neck meets the body- that's the area that the truss rod can affect. And in an obvious error, the guide compares the open string to the string fretted at the 13th fret!

 

A capoed string is not an open string.

 

Who plays up there anyway?

Oh yah, I do. And many of my clients. Some are even Spector owners.