Painful lesson with slide crooks

[gocsick]

TLDR: Color me surprised .. Today I learned when you take the dents out of a smashed tuning slide crook it might not end up the same width as when you started.


A while ago I picked up a $100 flugabone. It was in exactly the condition you expect for that price. It needed some brace repaired and replaced, and the leadpipe was replaced and soldered on wonky. That was pretty easy to sort out. The biggest issue was the main tuning slide crook was partially crushed and one of inners had a huge dent and everything else was ovaled.

I got it playing by turning a straight mandrel on the lathe to fix the inners but I didn't do anything to the crook.. because I didn't have many dent tools because they are expensive and not straightforward to make. Recently though I came across a set of worn out Gage balls. 2 balls from 1/4" to 3/4" in 1/32" increments at a machine tool surplus sale for cheap.

So today pulled the inners off the crook, and started running balls through and got it mostly round. The bore is 0.515" so I stopped at 1/2". Still had small dents but that's OK.. by me. put the inners back on, get them approximately parallel and in plane and find out they are 1/8" too wide to fit in the outers. I had to do some careful bending and slight ovaling to adjust the crook. It isn't symmetric anymore, but it fits well enough. That took way too long. It was supposed to be a relaxing day off.

[the elephant]

When you work the inside of a crook, it opens up. When you work the outside of one, it closes up.

Anytime you hammer on it, the metal gets thinner. No metal is removed; rather, the metal is MOVED. When you impact the surface, it becomes very slightly thinner, and the metal is moved away from the impact site. You can sort of control this by *how* you strike the metal.

This becomes a genuine problem when working on the Z-60 "Dent Machine" (actually more of a dent iron) on a bottom bow that has been very badly mishapen. If you do not constantly check your work, the ends can easily close up on you so that it is more than a 180º bow, and the tuba cannot be put back together as it came apart. This is a genuine problem with horns like Miraphone, where non-adjustable braces are used in a lot of the assembly.

When I work on severely flattened crooks or bows, I will make sure that they fit back onto the horn correctly, then I solder brass rod stock from the right opening to the left, on both sides. I then rough it out after annealing it, then try to get it as close as I can before removing the rods and finishing the dent work. Nowadays, if it is that bad, I just replace it. My labor costs about ten times what the new part costs. The customer is usually pretty happy to pay for a $15 part rather than $150 worth of my time just fixing what will forever after be a formerly bashed-up part.

[the elephant]

I think I described that pretty poorly. If you don't get me, let me know. I will try to do a better job.

[bloke]

Don't feel too bad.

Experience has taught me how to minimize effects such as this and sometimes even eliminate them, but I also have boxes and boxes of old stuff for when I can't.

Realize that when people show off their work, they're not showing off their screw-ups... and notice how clean (staged) their work areas are

That said, some some of these videos on Facebook (in my view). are really bad screwups in progress, but apparently those videoing themselves do not share my opinions of their procedures.

To address your specific concerns, some ways to counteract this are to continuously coax things' geometry back to where it was originally while simultaneously repairing them, another is to temporarily install a cross brace, yet another is to remove dents from the part while it's still mounted on the instrument. Annealing can be helpful, but it can also be a hindrance. Only annealing the portion of a part where repair work is needed is probably pretty wise.

More and more, there are no replacement parts to be ordered, whereby we have to be more and more like Cuban auto mechanics.

[arpthark]

I was working on a very badly bent and crinkled Conn 24J mouthpipe that I annealed and finally teased all of the wrinkles out of.

Upon mocking it up for final installation, it was slightly warped from where it was originally. Trying to gently coax it back into the right shape, the soft annealed brass went CRUNCH. Annealed brass behaves a lot differently than non-annealed brass, I suppose.

The good news was that the geometry was right, it was just now badly ovaled, so I soldered a crossbrace on it and have been running balls through it again after annealing.

[bloke]

Whenever possible, I try to repair tuba mouth pipes on the instrument, because of just what you describe, and EVEN IF (for whatever reason) I'll need to be removing the mouth pipe from the instrument later.

[the elephant]

Like Joe, if I think I will have to take a horn apart to clean it up, I will do all the dent work before that happens, so that when I use the dent machine, all I have to do is iron out ripples and waves. I never work on dented parts in the dent machine. The very name irks me, as it's not really intended for use in removing dents. It is for finishing work, or so I was taught.

I also only spot anneal, because if you anneal an entire crook, it can easily become really weirdly misshapen. I anneal the edges of the dent and into the depression.

I also agree with Joe in that we are having to fix stuff more than replace parts these days. Back in the old days, this was standard because labor was cheaper and parts were relatively more expensive. We even spot-plated mouthpieces and repaired damage to them on house horns. Later, when that became too costly, we just put new mouthpieces into house horn cases when the old one was damaged in any way that might cost some time.

Labor became the limiting factor at that time. Now it seems to be gravitating back toward parts cost being the limiting factor that directs the way a repair is done.

[Mary Ann]

Probably @gocsick can answer this -- what is the physics of annealing?
(because I still remember when I bought my horn I was told it had been "cryogenetically" (sic) treated.)

[bloke]

I'm not a scientist, but I know a very little bit about a very few things...and not a lot about a lot of things.

Annealing prompts alloys such as brass and other things to reorganize and restructure molecularly. (Of course, the ultimate reorganizing and restructuring would be to heat it even more to the point of being molten, but that's beyond what you asked - as well as what would be far beyond useful in a repair .)

freezing:
I tend to suspect is humbug, particularly because every time I've ever read about this, there's an additional equirement by the person doing it to also thoroughly clean the instrument.

Sadly, (though there are synonyms) there is no fancier, more scientific-sounding term for annealing as there is for freezing.

valve alignments:
Many of these are humbug, in my opinion. I've seen high-tech looking spacers installed on instruments that line up valves just about as well as they could have been lined up with traditional materials. Additionally, almost no valves (particularly not piston valves) are going to line up perfectly at all port locations and in both stroke positions -:no matter what. (Some pistons and their casings are remarkably well made, but they're not made by NASA subcontractors, and - just as a reminder - a lot of NASA stuff has blown up.)
Recently, I worked on a very stuffy playing custom-made helicon whereby the pistons had been "aligned" by someone famous for doing valve alignments. The 3rd and 4th pistons were close to the best they could be (but not quite). The first two were visibly down -;more than 1/8 of an inch below the other two. The first two were terribly wrong, whereas the correct position for one and two was / is with numbers three and four. ... I have no idea what the person who did that was thinking, nor what they thought they were perceiving. The washers did look fancy, though. Once I lined all of them up correctly (not with a camera - which can scratch the bore, but with my simple homemade sliding measuring stick) the instrument played much better, whereby the instrument's owner's eyebrows jumped up towards their hairline.

Science can be very technical, but -:when claims just don't seem to make any sense at all - it's probably a good idea to think them through as a layman (even though not a schooled scientist) and try to imagine if there's any possible way those scientific-sounding claims could possibly make sense. If there is, then they are worth investigating further. If they don't, it's probably time to ask an actual scientist for their opinions and to share knowledge (one who isn't trying to sell a product or service, or who is paid to promote a product or premise).

[gocsick]

This actually covers a large portion of a graduate level class I teach, and work hardening and annealing form a huge part of my research activity. Here is the 5 minute version.

When you deform a metal it gets both harder and stronger but it loses ductility and eventually becomes brittle. Annealing restores the ductility but also makes it softer and weaker. Here strength means the amount of force you have to apply to get the metal to permanently deform.. called the yield strength. The physical mechanism behind this was actually a mystery until quite recently compared to how long skilled craftsman have been working metals. It wasn’t until the invention of Xray diffraction in 1913 that a lot of this was worked out.

At the atomic scale metals are actually crystals. The atoms form a nice orderly arrangement. These crystal are actually microscopic, that is what people refer to the grain structure of a metal. Each grain is a tiny microscopic crystal and millions and millions of tiny crystal are “glued” together to make a piece of metal. This is what Xray diffraction proved. Once people realized metals were crystals they tried to calculate what the strength of a metal should be. From the energy given off in chemical reactions we understood what the bond strength or bond energy should be, so they could calculate how much energy it should take to break and reform a whole bunch of bonds. The problem was the actual strength of metals was about 1000X lower than this calculated value. Some clever people speculated that if the crystals were not perfect but instead had defects you wouldn’t need to break all the bonds at once, you could move the defect through the metal by breaking only a few bonds at a time. The classic analogy is moving a large area rug over 6”. If you tried to drag it you have friction everywhere and you need a lot of force to move it. Instead you could move one end 6” making a wrinkle (our defect in the crystal) and push that wrinkle along to the other end very easily. In crystal the equivalent defects in 3D are called dislocations. When you consider the strength to move one dislocation it matched up almost perfectly with the experimentally measured forces. So by the 1930s this “Dislocation Theory” was firmly in place and was used to explain a whole bunch of different phenomena in metals, despite not having any direct experimental evidence. That wouldn’t happen until the 1950’s when the electron microscope was invented and we could actually see dislocations in metals. I have metallurgy textbooks from the early 1950’s that say things like “When the dislocation is finally proven to exist, physicists will sleep easier because of the ever growing mountain of models and explanation that pre-supposes their existence.”

So the deformation of metals depends on moving these dislocations through the metal. Work hardening happens because when you deform metals these dislocations zip around and run into each other. They get stuck, react, and multiply. As metal work hardens the number of dislocation goes up very dramatically. We measure dislocation content by line length per unit volume, kind of like measuring the amount of yarn in a box by measuring the length of tangled yarn and dividing by the volume of the box. When a metal is as perfect an annealed sample as we can get there is about 10^12 m-2 dislocations… to put that into perspective that means if I take a cubic meter of metal and stretch the dislocations out end-to-end it would reach from the earth to the sun. In a heavily work hardened piece of metal we get about 10^16 m-2 that distance is the sun to the next nearest star (about 1 light year). There are some really simple models that directly link the strength of the metal to the mean free distance between dislocations.. the shorter the distance the higher the strength. In 1938 Taylor proposed strength is proportional to the square root of dislocation density. There are dozens of different models for dislocation evolution with deformation all under the general heading of hardening models.

Annealing is the removal of dislocations when the metal is heated. When you put in thermal energy the atoms jump around and dislocations can actually untangle themselves and anhiliate. In order to do this you need it get the metal hot enough that atomic motion is relatively easy. All the atoms are vibrating, some are shaking more violently than others. The temperature is the average amount of kinetic energy of the atoms. Some have a lot more. My the time you get to 40% of the melting temperature, the dislocation content can be greatly reduced. This is called recovery. If you go to about 50% of the melting temperature you trigger another mechanism called recrystallization. This is where the defected grain structure is replaced by new grains that have a low dislocation density nucleating and growing. Annealing generally aims to produce a fully recrystallized grain structure. There is a time dependence with temperature too. Brass melts around 925C or 1700F. Brass annealing is usually done between 600F (315C) to 800F (425C). However there is a time component. At 600 it can take a few hours to fully anneal but at 800 it is only a few seconds.

At temperatures lower you can get a stress relief. That is going to reduced and homogenize the stresses locked into the metal through deformation, but that isn’t going to soften the metal or make it easier to work at all. That is why those videos where people take a butane or propane torch over a thick bottom bow “to anneal it” are really not doing anything.

Cryogenic treatment is actually very different and a real phenomenon. Whether it does what is claimed to brass instruments, I don’t know. Cryogenic treatment evens out the residual stresses. When you put an instrument together, you get uneven heating from soldering etc. When things cool down all the braces and dent ribbons etc are all pulling different amounts in different directions. So the metal is under different amounts of tension, and some people claim it doesn’t resonate evenly because of this. So when you cool metal is shrinks, but it doesn’t shrink equally in all directions. Remember that grain structure which is a bunch of crystals all stuck together, well each of those crystal is going to shrink a little differently. When you cool something very cold, to liquid nitrogen temperature say (-196C or -320F) it actually shrinks enough and pulls on itself enough that the stresses in the metal exceed the yield strength and you are microscopically deforming the metal everywhere. When it warms back up it goes the other way. The net effect is the residual stresses are evened out and you don’t have regions that are under drastically different stress states than others. So this is an actual industrial process used when you need to achieve ultra-high tolererences in parts that can’t be heat treated after machining. Residual stresses can cause parts to flex or spring when they are machined as metal is cut away. Cryotreating the metal before machining can be used to remove this effect. It also does things to stabilize certain thermodynamic phases in steels, but those are not relevant to brass instruments. I did a literature search and found one reputable paper looking at cryogenic treatment of instruments “The accoustic effect of cryogenically treating trumpets” Jones and Rogers, presented at the 146th annual meeting of the Acoustical society of America 2003. They treated 10 Bach Strad trumpets and compared them against 10 untreated trumpets. Both by quantitative accoustic measurements, and blind testing by professional trumpet players… and concluded there was absolutely no measureable effect of the cryogenic treatment. The effect was significantly less than the instrument to instrument variability between the trumpets.

[the elephant]

I have always maintained that cryogenic treatments affect the metal, but they do not affect the instrument made from the metal.

Perhaps I was not so far off the mark after all…

[gocsick]

I have always maintained that cryogenic treatments affect the metal, but they do not affect the instrument made from the metal.

Perhaps I was not so far off the mark after all…

I think that is the correct take.

[bloke]

some people claim it doesn’t resonate evenly because of this.

...and some other people claim that the column of air within - overwhelmingly - is what resonates.

Mostly, trumpet players are into these sorts of things minuscule-effect/no-effect things, and
the trumpet players who seem to be mostly into these minuscule-effect/no-effect things
seem to be those who I would pass over, when looking down a list to fill out a personnel list for a gig.

Remember, those who do those treatments ALWAYS require a CLEANING in conjunction with the treatment...
...and guess why they ONLY offer freezing the entire instrument and NEVER offer annealing the entire instrument.