All-Melody Fretless Zither Restoration

The instrument at hand is a simple all-melody fretless zither of a type believed to be from the 1860s or 1870s.

These early fretless zithers are characterized by their having the tuning information and string numbers imprinted in an arc across the soundboard. These zithers also quite often have the tuning pins set into the edge of the frame, a structural and functional error.

Fortunately the subject of our restoration project has the tuning pins set into the block through the instrument's front as came to be the usual. There is no head bridge. Instead, screws serve as individual bridges for each string. It is interesting to note that an idea similar (though not identical) to this was adopted by Oscar Schmidt Company and incorporated into the design of certain autoharp models over 100 years later.

Here is a look at the instrument as we began its restoration. As this type goes, this one's condition was quite good.

However, it had one significant problem, namely that the tail block was severely cracked. This was in part the result of the unwise practice of using one piece of wire to make two melody strings, which puts undue stress on the hitch pins. This can crack even hard maple hitch blocks.

Here's a look at the damage.

But also, the block didn't look to be made of hard maple as it should be. It could be that it was rotten, but it didn't look like it had been wet. Besides, if it had been wet enough to have rotted, the joints would be wide open. As can be seen, they were as tight as the day this instrument was made.

So, I was faced with a treatment decision. After considerable thought, I decided that the tail block must be replaced with a new maple one. After more thought, I proceeded with my plan for removing the old block.

The first move was to make two passes through with the table saw, staying well clear of the joints. Sure enough, this revealed the tail block to be made of old growth poplar, which is quite soft. The decision to replace it was looking to have been a good one.

The outer surface of the block was angled, which I suspect to be a less than great idea. But squaring it off would be too big a design alteration. Remember, I'm not proposing to change it, but only to make it well again.

The next step was to chop out the middle with a hammer and chisel. Nothing fancy, just chop.

Then I drilled through to the inside with a 1/2" spade bit. Again, nothing fancy, just drill.

More chopping followed.

The chopping grew steadily less aggressive and more focused on precision. At length, the hammer was set aside and the chisel was used as a carving tool. I carefully worked down through the old block until essentially a veneer of it was all that remained. Bits of it disengaged cleanly, as can (sort of) be seen in the photo, but the rest was hanging on for dear life.

Here is a close-up look, showing the thickness of the remaining wood of the block. The arrows indicate the joint.

Next I boiled some water and painted it onto the wood of the old block, then soaked a paper towel and sat it on the wood to soften the glue. I left it there for about 5 minutes.

After two such applications, the glue had softened to the point that the remains of the old block were ready to come out. I used this tool to slide in under it. It's a battle-worn piece of .010" spring. I mainly use it for administering glue in places where no other tool can go, but it worked quite well for the purpose at hand.

Here is a shot taken right after the block was out. I stopped at this point and let it dry overnight.

And here is a piece of the block after removal. The side shown is the glue joint side. As can be seen, the parting was a clean one; there are virtually no splinters from the surface it was joined to. Incidentally, the diagonal cut is from a point in its removal when it was to my advantage to cut it, which I did with scissors.

The following evening, the new block was made up, carefully fitted, and installed. I have provided a numbered image with text at the linked page, as it's a little confusing to tell what's going on here.

1. I always clamp to a "clamp board", a board that spans the whole surface (or near to it) of the back . Its purpose is to provide an overall flatness reference, to prevent the block from shifting upward or downward, as can happen if it is glued "in mid-air." It is a piece of 3/4" cabinet-type plywood (smooth face, no "footballs", and flatter than construction grade .) It is covered with wax paper so I don't glue the zither to it. As I was gluing only at the tail end, I didn't need a clamp board that fit the zither's outline exactly, and since I'll probably never restore another one of these, I preferred not to make a new one for it if I could help it. The one I used was for a No. 73 autoharp, and it worked fine. I had to situate the zither a couple of inches away from the tail end of the clamp board for it to be wide enough, and the exposed couple of inches of it can be seen in the photo.

2. As mentioned, the entire instrument is clamped to the clamp board, and that's what these two clamps are for.

3. In addition to merely clamping the top and back to the new tail block as the 4 clamps at the tail end in the image are doing, the block must also be clamped inward. For this I use a thing I call a "clamp dog." Its purpose is to provide an edge parallel to the tail end of the zither to clamp against. This prevents the block from shifting left-to-right, i.e., in the direction of the instrument's width, which it will do if clamped against an edge not parallel to it. Clamp dogs are also usually made from 3/4" plywood, though 1/2" will normally do in a pinch.

4. Here the excess wood of the new block itself can be seen protruding from the zither's tail end.

Next evening, the zither was removed from the clamps and the new block was planed flush overall. I had to round off the corners to flush it up with the worn corners of the top and (to a greater degree) the back.

Here is the amount of mess this project has made...seems like a lot of shavings for a relatively small job, but then it usually does.

The new block was stained to match and was ready to seal. The next few evenings were spent applying a coat of sealer to the block and leaving it to dry until the following evening.

The finishing of the tail block completed, next came situating the hitch pins in the "blank" new block. The next few images show the method I always use to do it. First, the distance between the two outermost strings/pins was measured. As the instrument has 20 strings, this measurement needed to be divided by 19, for the 19 spaces 20 strings make. No number on Earth divides evenly by 19 of course, but that's all right. Big tip: this is a LOT easier to do in metric...which would you rather divide by 19, whole numbers or fractions?

Two marks were made on a piece of .010" archival folder stock. (Don't be impressed; what he means is that he cut up a manila file folder.) The distance between the marks represented my best attempt to duplicate something like 13.8746739450233 mm. (What luck; exactly one whisker shy of 14!) The points of a divider were aligned with the marks.

It's not a good idea to situate the hitch pins all in one straight row in the block because they can collectively assume the role of a wedge and split the block. It would be a shame to split our new block; this zither's been there, done that. They should be situated in two or sometimes three rows if there are a lot of them and/or they're crowded. This zither has only 20 wide-spaced strings, so two rows is plenty.

I measured the thickness of the block, and it was just (again, exactly a whisker) over an inch. So I figured if I started at the center point, 1/2", and spaced each row 1/8" from it, the rows would be 1/4" apart and each would be 3/8" inward from the edges of the block's thickness. Those measurements all seemed good, so I made a number of parallel lines on the manila, starting 1/4" from one edge and continuing across the width of the paper. We will call these the "long lines." You'll see in a minute why I made so many.

The distance between the two lines perpendicular to those (which we'll call the "short lines") was the overall distance between the two outermost strings, the measurement we divided by 19. It's a good idea to keep all these lines both as square and as parallel as possible.

I put one point of the divider (which I'll call "point A") on the intersection of one of the short lines and one of the long lines nearest the edge of the paper. With the other point ("point B", of course) on the same long line, I punched both points through the paper. I then proceeded to successively put point A in the hole made by point B and punch both through again...and again and again. The goal was to land point B directly on top of the short line at the other end of the paper from the one where I started. This can take several attempts, and THAT is why I made so many long lines.

The first "run" came up a little short of the target short line. So I adjusted the divider outward a tiny bit. The second run was still a little short. On the third run, point B speared the short line. (Success on the third run is pretty good.) I then did another run on the long line adjacent to the successful one, and marked the points alternately on each long line with little crosses.

I then measured 3/8" from one of the long lines and cut the paper along that line. This was to be the edge I would align with the joint of the instrument's top and the tail block. The paper was also cut approximately 3/8" from the other long line. Length-wise, I cut the paper to a length where I could tape it to the face of the block. With an awl, I punched holes in the block at each of the marked holes in the paper.

Here is the result. I then drilled the holes for the hitch pins according to these marks. (And yes, they do have to be drilled; you can't drive those little nails into hard maple.) As to the diameter of the holes, the pins should fit tightly enough to require a little more effort than driving them into pine or some other soft wood.

The next project was to polish all the metal parts...tuning pins, hitch pins, screws, and bridge wire. I do this with a wire wheel in my ancient tower-type bench grinder. Shown are some crude-but-effective tools that have saved my fingers several pounds of flesh. They're just holders for pins of various sizes. The long one is for tuning pins; it's helped me polish a couple thousand of them. The shorter ones are for bridge and hitch pins.


...and after. You'll note that I take it easy on the threaded parts of the tuning pins; I'd rather leave a little rust than to dull the threads or reduce the diameter any more than I have to.

And here's the old brass bridge wire in mid-polish (in fact, the grinder was running when I took this shot.)

One interesting detail...the bridges of fretless zithers are normally held on by nails, not glued. This one's bridge nails are tiny square ones.

Next, it was time to clean and polish the zither. I use No. 7 rubbing compound, followed by Oz polish. I started with the back. Before...

...and after.

Next was the sides, then the front. I always plug all the holes so I don't fill them up with mung. I use this foam-like stuff...kind of a cross between foam rubber and styrofoam. It's used for packing material. I don't know the trade name for it. Come to think of it, I don't know where I got it. Anyway, it works pretty well. The front, before...

...and after.

It was then time to re-install the tuning pins, hitch pins, tail bridge, and screw "bridges."

One last little detail...the inside of the back was covered with this decorative paper. It's interesting how it isn't textured at all but looks 3-dimensional (well, the part of it that's in focus, anyway.) All right, it's time for strings.

And here we have our finished zither. The ultra-high melody strings are a real bonus to its sound.

The instrument being finished, it's time to add it to the Instruments Directory. Click here to check out its Instruments Directory listing.

Back to Restorations Page