Valve Stem Replacement in Rotary Valves

For modern, mass produced rotary valve instruments, replacement rotors are generally available and they will usually fit as they are received or with a bit of hand lapping.  In older models and antiques it's usually a bit more complicated.  

If you haven't spent a lot of time working a lathe, this job may seem daunting at first, but it is a good opportunity to increase you skills in this area.   I was instructed in this technique by Dan Rauch (now Rauch Horns, Oslo, Norway) when I started doing some work for Mirafone Corporation (USA) in the early 1980s.  Dan had been working there full time, but as his own business was taking off, they asked me to take on some of their work.  

The valve making at Miraphone in Germany was very old school and each valve section was made and valves fit individually.  Replacement valves were available, but needed machining to size before lapping them in, so replacing a broken stem was a quicker alternative.  

The case illustrated here is in a circa 1870 American style circular Bb cornet.  The second valve stem had obviously broken off in the past and two different attempts were made to repair it.  First the rotor was drilled into the bearing all the way into the air passages.  When that failed, a smaller hole was drilled into the rod that had been soldered in and another stem soft soldered to that.  It must have worked marginally well at first, but was not functioning as I received it.  The best repair involves replacing the thrust bearing as well as the stem, so the state of the original bearing was not a concern.  

I made a sketch of the valve with all measurements needed for the reconstructed stem.  This can be done by carefully measuring the rotor, casing, bearing plate and stop arm assembly.  I double check the design of the parts on the other valves to make sure that I'm not misunderstanding how they fit.  

The next step is to machine a brass collet to hold the body of the rotor, as seen in the third photo down on the left.  In a lathe with 5C collets, which are the most common size this brass collet can be made hold the valve body.  For the larger Mirafone tuba valves on which I replaced so many stems, I made a collet to hold the bearing stem as illustrated on my page on sleeving valves, then the final machining is done while a hole drilled in the new stem engages with a live center.  Drill the hole only after making sure that the body of the rotor is centered using a dial indicator.  

With the body collet, the machining is easier, being held firmly centered as the hole is drilled and surfaces machined.  Most rotary valves are tapered on both the bearing stems and the rotor bodies, so these collets must also be internally tapered.  I make the larger of these by machining rod stock the length of the body, carefully copying the rotor's external taper.  This can be done by carefully advancing the boring bar with the lathe  set to a very slow feed.  I mark the boring bar at 1/4 or 1/10 inch increments, knowing how many thousandths of an inch (or less) to advance for each.  Then I cut a slit in this collet so that it will hold the rotor firmly in the 5C collet.  

Next, I cut away the old bearing and enough to clean up the surface beneath.  Then center drill a hole to locate the new stem.  In this case it was 1/8" and for the tuba valves I would make that 1/4".  A brass rod is cut to the length needed for the reconstruction and add a little extra for security, which can be trimmed later, and machine a pin on one end as in the fourth photo.  

Next the rod is soldered in place as seen in the sixth photo to the left (I neglected to photograph this before machining).  I typically silver solder this, knowing that it won't deteriorate or pull loose, but in this case it is soft soldered.  The previous repairs involved soft solder and I knew that I wouldn't be able to remove all traces of that, which would mix with the silver solder and make a mess of weaker metal.  Instead, I used silver/tin soft solder, which has a higher tensile strength than lead solder and will not deteriorate with the presence of saliva.  

Where the air passages had been drilled through, I scraped the excess solder flush with the bore.  Then the rotor is put back in the lathe and the stem is machined to about .001" oversize and lapped into the bearing plate.  The taper is achieved by advancing the cutting tool with the feed set very slow, as described above for making the brass collet.  The thrust surface of the bearing can be trimmed .001" at a time, trying the fit after each cut, until is just fits.  If the measurements were perfect, which is very difficult to do, it should take two cuts.  Usually, there is a bit more trial and error involved.  

The stop arm interior was very difficult to measure accurately and not exactly round, so the stem required hand fitting with a file.  In this valve, the stop arm has a internal pin that locates on a slot at the end of the stem.  This is carefully cut by hand.  More typical in modern rotary valves, the stem has one, three or five flat sides for this location.  These can be cut on a milling machine in an indexing jig that is shimmed on one end to make the taper.  I've done it that way, but find that carefully filing them by hand is much quicker and can be just as accurate, especially fitting to abused and worn old parts.  

That's all there is to it and it will function at least as well as, or better than it did before the damage.