The naked Grafton

In my review of the Grafton plastic alto I said "I reckon a repairer averages three overhauls on these creatures; the first out of ignorance and sheer curiosity; the second out of disbelief that any instrument can be such a pig to repair; and the third just to be really sure that the thing really exists and isn't just a terrible nightmare. The truly masochistic (or perhaps forgetful) might tackle four."
So, having done three, how is it that I came to find yet another one of these beasts on my workbench?

The client, Pete Thomas, tells me that it was my idea to overhaul his Grafton. I, on the other hand, having previous experience of these horns, can't imagine saying "Hey Pete, your Grafton could do with a repad - let me have a go!" - but as Pete is as honest a chap as the day is long I very much fear that he may be right. In which case my reason for taking on a fourth Grafton overhaul must surely come back to being forgetful or truly masochistic (I can't remember which, which is a pain).
In any event I ended up putting myself through the pain and torture that goes with the territory, and decided that it would be worth writing about my experiences for two reasons. One, to share with interested parties the inner workings of this unique instrument, and two - to have a handy written reminder never to undertake such a project again.

I'll admit that it's taken me rather a long time (ahem) to complete the job, and for the most part I've had some very good excuses...such as couple of books to write and a pressing backlog of work.
I started well enough, with good intentions and grim determination - but as the weeks rolled by I found it became harder and harder to steel myself against the foibles of the horn's design, and I have to admit that there have been times when I've resorted to some pretty feeble excuses to avoid putting myself through the frustration of working on the Grafton.
Not having my 'lucky socks' on today was one, as was being unable to make a cup of tea that was up to my usual standards (poor quality water, I reckoned). Having 'a bit of a headache' was a common one - but in all fairness five minutes working on a Grafton while you had a mild headache would leave you with a raging migraine.
I wasn't the only one making excuses though - by the time I finally cleared the backlog enough to allow the Grafton to take up all of my time (slightly hastened by Pete ringing me up to tell me he'd booked a recording session in which the horn was to be used), the skies darkened, the temperature plummeted and the entire south coast of England was buried under a couple of feet of snow.
The gods were on my side too.

Eventually I ran out of excuses and time, and got down to the very serious business of making the Grafton a full-time priority.
I had a fresh pair of lucky socks on, the water had been tested and approved - and the met office were forecasting a little light wind and a spot of rain.

So what's all the fuss about?
Well, before a screwdriver can be taken to the horn there is a need for some careful preparation.
What must be understood is that the mechanism of the Grafton is quite unlike that of a standard saxophone. The main stacks are pivoted on rods (one for the lower stack, two for the upper) and these in turn are held in place by means of a pair of lugs that attach to bosses on the body. You can see one of these lugs in the main photo, dead centre of the body...the upper lug has been removed and would normally sit on the boss just below the crook socket, to the left of the body octave key tube.
The springs are coiled, rather like those found on trumpet water keys, as opposed to a normal sax where the needle spring are wedged into the pillars. Should you be unaware of these design 'features' and merrily remove one of the lugs holding a stack in place and withdraw the pivot rod, the keys will fall off the instrument and the springs will go flying off across the room.
This is itself is something of a disaster - but assuming you were able to find them all you'd then have to work out where they go. This is a particularly nasty 'sting in the tail' - the springs aren't completely interchangeable. Some have longer legs, other have longer tails - and they're all individually tensioned. If you fit the wrong spring to the wrong key and only find out about it once the instrument has been reassembled, you'll very quickly fill a couple of swear-boxes.
If you manage to lose or damage a spring you'll have to make a new one. It's not impossible, just time-consuming and fiddly - and you'll need the correct gauge of spring wire. It's also handy to have a spring winding tool, as found in many a brass repairer's workshop, for making water-key type springs.
Even better is to have some spare springs. I was fortunate enough to have been given the remains of a Grafton some years ago after it had been dropped - it came in very handy!

So the first order of the day was to ensure that during the dismantling process, each spring was carefully retained with its key.
The keywork also makes use of spacers. These are nothing more than plain bits of tube that fill the gaps between adjacent keys. They're unlikely to ping off into the wide blue yonder when the pivot rods are removed, but keeping them attached to their adjacent key means fewer headaches on reassembly as you try to guess which spacer goes where. Some can be almost, but not quite, identical in length...and if you fit them in the wrong place you'll end up having to completely dismantle a stack to correct your error. As we'll see later on, this is not a good thing.
I came up with very simple method of covering both bases by threading lengths of soft solder wire through the keys. This kept the springs and any spacers neatly attached to the keys.

Lastly, I found it helpful to take snapshots of the action, both before it was removed from the body and shortly after...laid out 'post-dissection' as it were.
Such reference points proved to be useful while dealing with this unique instrument.

As with any overhaul the first port of call is the inspection of the body. On a typical sax I'd check the body for straightness, look for any dents or damage, check the pillars for straightness, alignment and wear and just make sure that everything was where it was supposed to be.
Much of this is academic with the Grafton - it can't be bent (unless it was made that way), it can't be dented...it would simply break...and there's not a great deal you can do about the alignment of the fittings as they don't really fit that well anyway. So in fact the tests are; is it all there and is it all in once piece?
A very common problem is that of the bell brace detaching from the body. This is usually the beginning of the end for these unique horns. Once this joint breaks, and goes unnoticed, it puts a lot of strain on the bottom bow joint...and when the bottom bow joint cracks off it usually means the end of the horn.
It's possible to repair the bell brace, but it requires a specialist (and rather poisonous) glue that welds the parts together. Using a standard epoxy glue or a superglue is a waste of time - it'll hold for a while before the acrylic simply shrugs it off.

Once that's checked it's on to the action.
It's here where the fun and games start. When the keys on a sax begin to wobble it's for one of two reasons - the key barrels are worn through use or they were not that accurately made in the first place. Fortunately there is a technique called swedging (pronounced sway-jing) that compresses and stretches the key barrels to take up the free play. It's a technique that's very labour-intensive, and therefore expensive, but it's an effective fix for a worn or badly made action.
The problem is that the technique relies on there being sufficient 'meat' on the key barrels in order to get the tools on. On most horns this isn't a problem, but there are some older designs where holes are drilled through key arms and there's no barrel at all. There are techniques for dealing with such designs, but the Grafton goes one better by making extensive use of this design feature as many as three times on a single key.
As if that wasn't bad enough, the Grafton has another 'gotcha' up its sleeve. Should you find a key barrel that looks like it needs a bit of swedging, you might well find that the key barrel has a casting seam on it. You could certainly get the swedging pliers around it, but the results would be horrendous - the best you could hope to achieve is an oval key barrel.

Perhaps the only way to deal with the problem of wobbly keys - and in the case of the Grafton it's down to poor manufacture rather than wear - would be to ream out all of the keys and barrels and fit oversized rod screws. This would certainly ensure that the keys sat snugly on their pivots, but it would be a complete waste of time.
This is because of the way the rods themselves are mounted onto the horn.
A typical sax has pillars - and a typical lower stack will have five or more pillars on which the keys are mounted. Each pillar supports the rod screw, which keeps it nice and straight and prevents it from flexing. The Grafton has far fewer supports (it doesn't have pillars), the bottom stack has just three and these themselves are by no means tightly fitted to the body and aren't all that accurately drilled either. In short, the whole action is almost 'suspended' on the instrument.
What this means in terms of getting the thing to work is that you can't rely on a key cup coming down in exactly the same place each time. For sure, it won't move far, but a fraction of a millimetre is enough to allow the pad to leak.

Having worked out that the action is pretty much a lost cause, it's time to consider the pads.
The original factory-fitted pads were notoriously spongy. There's a good reason for this - a soft pad will compress slightly as you press the key down, which causes you to press down rather harder than you normally would. In so doing you are using the squishiness of the pad to compensate for the play in the action. It means the horn will play, but it feels bloody awful.

There's another reason why the pads feel so squishy, and that's because of the way they're fitted.
Saxophone pads are normally held in the key cups by means of a glue of some kind - typically shellac. Over the years there have been a few variants on this method, the most notable being the 'Snap In' pads as fitted to some Buescher saxes. This design consists of a small boss permanently fixed to the key cup over which the reflector (or resonator) 'snaps' on.
In many respects the Grafton uses a similar system, but this time the reflector is screwed into the boss. This method of mounting pads is exactly the same as that used on flutes, just on a far larger scale. On the face of it this shouldn't present too many problems - after all, the Buescher Snap In pads work just fine. They're traditionally fitted without any shellac, relying instead on the reflector to hold the pad in position. This tends to make setting the pads a little trickier as many repairers like to use shellac as a 'variable shim', allowing for small adjustments to be made to the pad angle by melting the glue and forcing it into specific areas of the key cup. Without this option other means for adjusting the pad have to be used, such as shims (small pieces of paper carefully placed in the key cup underneath the pad to raise certain areas of the pad).

This method should, in theory, work on the Grafton - but it's not that simple.
Note the card disk fitted inside the key cup - it's really quite thick. The pad is very thick too, easily the thickest pad I've ever seen on any saxophone, bar those fitted to Adolphe Sax's original instruments. The thickness of the card disk added to the thickness of the pad presents quite a few problems with regard to setting the pads - not the least of which is that it's practically impossible to get the pad warm prior to levelling and seating it.
It also helps to explain why the action on the Grafton feels so terribly spongy.
Here's a cutaway diagram that shows what's going on inside the key cup when the pad is fitted.
The crucial part is that large air gap - it really shouldn't be there.
In order for a pad to work well, and to retain its seat for many years, it should be firmly fixed and supported in the key cup. On a standard saxophone the area occupied by the card disc and the air space would be filled with shellac - though the pad would be sat deeper in the key cup and the space between the bottom of the pad and the key cup would be very much smaller than that shown here.
This would mean that when the key cup was pressed down against the tone hole, the pad and cup would act as a single object, resulting in a nice firm closure with a positive feel.
On the Grafton you have everything working against you. The thick, squishy pad is bad enough - but as the key is pressed down and the pad makes contact with the tone hole, the pad flexes as it's deformed and pushed back into that air gap. This is what gives the horn's action its characteristically vague feel.

You might be wondering why such a thick pad and the accompanying card disk was necessary. It's all to do with the angle of the key cups. The way in which the action is designed means that when the key cup is parallel with the tone hole rim, there's quite a large gap. It's this gap which determines the thickness of the pad.
You can change this angle (and thus the required thickness of pad) to some degree on a normal horn by carefully bending the keys - but this isn't an option on the Grafton for a couple of reasons. Firstly, the amount you'd have to bend the key cups means that the cup would end up being pushed too far forward and the pad would barely cover the tone hole - and secondly, it's largely impractical to bend keys when they're not mounted on the instrument...and you'd have to be spectacularly brave to attempt to bend a key on a Grafton while it was fitted to the body. The acrylic body is very fragile indeed, and the amount of leverage required to bend a key cup arm will be more than enough to snap off the action mounting lugs.
So this leaves the repairer having to work with the given pad thickness.

The whole point of taking on this job in the first place was to improve the feel of the horn, so the air gap would have to go and the pads would need to be rather firmer than those originally fitted. There are no pads of the required thickness made these days, so they'd all have to be backed up with thick card...but this time the card would have to cover most of the pad cup base to ensure that the pad lay flat and completely supported.
I made the decision not to use shellac. It would have had made the job easier, but I wanted to retain as much of the horn's originality as possible.
I chose a medium-firm pad for the job. This decision was based on the fact that it would be almost impossible to use heat to set the pads - by the time the cup had taken on enough heat to get through a thick layer of card and a thick pad, the lacquer on the key would have been a charred mess. This means the pads would have to be cold seated...so they'd need to be firm enough to stay flat once levelled, and soft enough to take a seat without heat.
Too firm and the pads simply wouldn't cope with the approximate nature of the build quality, too soft and they wouldn't retain a seat without some heat behind them first.

The process of setting the pads was quite complicated. First, the total pad thickness had to be achieved. Artist grade card was used for the backing - this type of card is composed of many thin layers which can be peeled away as required. It's also very dense, which helps to eliminate any compression over time. The key was then fitted to the instrument and the spring set, at which point the pad could be ironed and checked for level. Any further adjustments required the removal of the key and the spring - and in some cases the supporting lugs too. As you can imagine, this proved to be very tedious and time-consuming.
Once the pad had been levelled it could be seated. This would affect the pad height - not by much, but easily enough to create a leak. This means removal of the key again and reshimming as required, followed by reseating.

It became apparent that the amount of flex in the action varied depending on how many keys were fitted and sprung at any one time. On a normal saxophone you can set each key individually - but with the Grafton having rather less support for the rods, and a bit of play in what supports it had, the only way to ensure as much accuracy as possible was to assemble the entire stack each time you wanted to set an individual pad. This is where using shellac behind the pads would have been a boon, enabling small adjustments to the pad level on the fly. It was at this point where I really began to question the notion of adhering to the original specifications...

Setting up the action is an unusual process. The stacks are regulated by means of screws, as can be seen in the photo above and to the right, which have to be turned with the aid of a small spanner. All very 'industrial'.
It's not actually a bad method of regulating the action - modern horns often use small screws to adjust the internal regulation (the relationship between connected keys, as in the photo above), and the Selmer BA had a similar system for adjusting the external regulation (the height of the keys, as in the photo on the right).
The big problem is that once the bell keys, which house the external adjusters for the right hand stack, are fitted to the horn there's no easy access to the internal adjusters underneath the stack key arms. Even if you could slip your spanner in underneath the pads, it's very difficult to get the spanner over the nuts...and there's always something in the way that prevents you from turning the screw. This means having to be extra sure that the internal regulation is properly set up before the external adjusters are fitted.
All good and well in theory, but every overhauled horn will need further adjustments after the play-testing period...and in the case of the Grafton this means dismantling half the horn to do so.

The left hand key stack works on much the same principle.
There's one internal screw adjuster (for the A key) and a tab for the B key (which can be seen just behind the front top F key pearl).
Again, the external regulation relies on a set of screws that are affixed to another part - in this instance the upper stack key guard.
This isn't a particularly stiff component, so you can imagine how hard it is to achieve any kind of accuracy with the regulation when the guard keeps flexing. It all gets a bit 'approximate' - and it's no wonder the horn was originally fitted with squishy pads...it must have solved an awful lot of problems with the regulation. Get it about right, and rely on excessive finger pressure to take up the inevitable discrepancies.
Fortunately, the internal adjusters are accessible when the guard is fitted...so there's a lot less swearing involved when you find you need to tweak the left hand stack regulation a little.

By far and away the most problematical aspect of working on the Grafton is doing 'spot tweaks'.
Let's say you're working on a standard saxophone - you've assembled the horn but now you're not that happy with the tension on one of the springs. Not a problem, you just reach for the springhook and adjust the spring as required.
You can't do this on the Grafton. There's limited access to the springs once they're fitted - the best that you can do is maybe slacken them off a touch by carefully pushing against the spring tail while in situ, but for anything else you'll have to remove the spring.
Unfortunately this will mean removing most of the action...and sod's law always dictates that the spring you want to adjust is slap-bang in the middle of the key stack - just like the one on the side C key cup shown here.
There's a double-whammy on this key stack, it's composed of two separate stacks. The upper rod holds the main keys and the lower rod holds the side trills and octave key pad. When assembled, some of the upper stack springs rest on the lower stack barrels - which presents a major headache if you need to get at anything on the lower rod.
There's no point in even attempting to remove the lower rod on its own...you'd never get it back in place without removing the upper rod (and all the keys). It's even risky to remove one of the lugs - if the rods move even slightly out of line the springs will come unseated...and you'd have to take the entire stack apart to get them all back in place. This isn't much fun.

So here's the trick. You'll need to remove both lugs - but before doing so the barrels are 'wired' together with a couple of pipe cleaners. This prevents the stacks from moving apart and keeps the springs located.
A spare rod is inserted into the stack you wish to work on - in this case it's the lower stack and the spare rod is being fed in from the left.
As it's inserted it pushes out the proper pivot rod...up until it reaches the key you wish to remove, at which point you withdraw the proper rod a little further and hey presto - a single key pops off!
The spring can now be tweaked and the key reassembled. Once in position the proper rod is pushed back in, which pushes out the spare rod. Neat, eh?
It's not quite plain sailing though, as you have to fit that coil spring under tension. You can't get in afterwards and set the spring, so a fair amount of fiddling, cursing, effing and blinding is required... and you have to be careful not to move the stacks around too much. It's still a lot less work than having to remount the entire upper half of the horn.

Once the horn is all back together and blowing as well as its able, it's time to sit back and take stock.
It's often said that it's hard to find a repairer who's willing to take on one of these unique horns, and this article will perhaps go some way to explaining just why that may be the case.
Saxophones, even the best of them, aren't precision machines. They're full of compromises - both in the acoustic design and in the manufacture. To be honest they simply don't need the level of accuracy demanded by, say, an engine or a clock - but they do have to be well-built enough to withstand the rigours of being played.
If you've browsed any of my reviews you'll have seen that I often pick up on build quality issues - pointing out where defects and flaws may affect reliability and suchlike...and if the Grafton was being built today it's a safe bet that it would get a very scathing review in terms of its construction.
The keys are poorly built (just look at that offset hole drilled in the Eb key above), the action is poorly mounted, the spring system is a Heath-Robinson affair...and absolutely no concession at all has been given with regard to maintenance and repair.
The feel of the action is mediocre at best. The coiled springs have very little snap to them, the keys flex when you press them and it's virtually impossible to balance the springs in situ.
Putting a set of good quality pads on the Grafton really showed up the limitations of the engineering. If you blow a horn and encounter what feels like a leak, what do you do? You press harder on the keys. It's a perfectly natural reaction. If you do that on the Grafton, the keys flex, and it leaks even more. The original squishy pads would have hidden this flaw in the design...which is perhaps another reason they were fitted in the first place.
And then there's the body itself.

To be fair this is probably the best-designed part of the instrument. It's unfortunate that the material chosen turned out to be so brittle, but then that's what was around at the time the horn was built.
As you can see on the right, when they break - they really break!
That aside, the body is quite well-built. The tone holes are neat and tidy, the way the action fits on is really quite ingenious (even if it's less than satisfactory in terms of stability and accuracy) and there's absolutely no denying that it's a real looker. This is, essentially, why the Grafton has made it into the hall of fame rather than being a mere footnote in the annals of saxophone history. It's unmistakable, it oozes art deco style - it's the archetypal loveable rogue.

And what of the sound?
Newcomers to the concept of the plastic saxophone often make the mistake of thinking that it will sound, well, plasticky.
Far from it - it sounds like a saxophone...like one made of metal.
It has its own tone, to be sure - just as any sax does - and I suppose the best description of it would be 'slightly fuzzy'. It's at this point where devotees of the 'material makes a difference' point of view would say "Aha! That's because it's made of plastic!"
It's understandable, but it's an argument that falls a bit flat on its face when you point out that many plastic-bodied instruments such as clarinets, flute and recorders tend to sound rather brighter in general than their wooden or metal counterparts. Of course, this has nothing to do with the plastic...it's down to the design of the bore.
In fact the Grafton is a rather good demonstration of how little body materials matter when it comes to tone.
Consider the following; if you accept that material makes a difference then there ought to be a very noticeable difference between the tone of the Grafton and that of a standard brass saxophone. But there isn't - which means that the difference between one alloy of brass and another will be far, far smaller...and the difference a coat of lacquer or plating will make will be even less.

If you own a Grafton alto and would like it repaired, it's not worth dropping me a line. I can tell you right here and now that the answer will be an emphatic, cast-iron, gold-plated, hand-embroidered, fully authenticated and genuine "No!"