of Washington Phillips's zithers
|first session accompaniment section|
|third session accompaniment section|
|notes on my use of tick marks|
|notes on my use of split-level bridges|
|strings: gauges, materials sources, and fabrication|
The purpose of this page is to document the details of the stringing/tuning configurations of the two Phonoharp zithers I have configured with the intent of reproducing the sound of the instruments played by Washington Phillips in his 1920s recordings.
I hold to the notion that his instrument was just as we see it in the photo, i.e., that it was composed of both the zithers, joined together to make one. Possible motives for his doing this are discussed in the WP section of the Clips Page. However, I don't feel he has nearly all of either of them strung, so I used a single zither for my experiments.
Over the course of his recording career, there is considerable variation in the accompaniment section of his instrument. I explored the two extremes, the configuration of his accompaniment section at his first and third sessions (Dec., 1927 and Dec., 1929, respectively.) I believe his accompaniment configuration at his middle session (Dec., 1928) to be nearly identical to that of the third session.
As far as I can tell, it sounds like his melody section remained consistent throughout his three sessions.
The stringing configurations I arrived at for first session accompaniment section, third session accompaniment section, and melody section are all documented in detail below, along with various other comments about the instruments. Though the intent of this page is documentary rather than instructional, the information at this site will enable anyone who wishes to do so to re-configure a 4/30 fretless zither in the fashion of Washington Phillips's instrument.
You will notice that this setup required an extra bridge that sits out on the instrument's front. That is because some of the notes in WP's Bb chord are too high for any string stretched the full length of the zither to come anywhere near making it without breaking. (This isn't speculation; this statement is substantiated by experiment. e.g., an .010" string snapped at around F#''', essentially a full octave shy of the mark.) Another comment on the F'''' note in the accompaniment section is that it is audible in some session 1 recordings, though not as prominently as at session 3. I only tried to duplicate it on the session 3 instrument.
Incidentally, relative to either of the instruments involved or either of the two sessions in question, there is no significance to my using a Phono 4/30 for the first session setup and a Celestaphone for that of the third. When I began the third session experiment, I had already re-configured a 4/30, and I wanted to compare the physical aspects of the two different instruments in regard to the possibility of their having offered WP motive for creating his double-zither instrument. I refer here to the fact that the melody section of the 4/30 and (especially) the accompaniment section of the Celestaphone are both crowded to the point of making them more difficult to play than the corresponding sections of the other instruments. In my opinion, this comparison added much support to the theory.
As was first theorized by Gregg Miner, octave-tuned melody pairs turned out to be "the key to the kingdom" of the WP instrument's melody section sound. The melody tuning of the Phono 4/30 and Celestaphone (and virtually any chord-zither), as prescribed by the manufacturers, is two diatonic octaves of C. The WP melody section is tuned in two diatonic octaves of F. We were all stumped for a time not only by the higher pitch of WP's melody range but also by its range, which was nearly three octaves. It was a breakthrough realization that octave-tuned pairs increase the diatonic 15-course melody section's range by a full octave, from its usual two octaves to a full three.
My use of tick marks to indicate octaves is purely relative. I don't know which octaves they indicate when properly assigned; I only know that each additional one indicates ascent of pitch. I begin octaves at C, and again I don't know if this is proper. The best way I could think of to positively identify the range of pitch I use was to illustrate it relative to the guitar, and clicking the thumbnail image to the right will open a "map" image which does so.
Note: I use the small "b" for the "flat" symbol, as I don't have that character in my assortment of fonts. It gets the idea across.
As can somewhat be seen in the images of the instruments, I made special bridges for the melody sections of these instruments. They are described below, and details can be seen in the images. First of all, I do not mean to suggest that I think WP equipped his instrument with special bridges; I don't think he did and the photo, though always open to question, doesn't offer any evidence that he did. The point of my making these bridges was simply to make it easier to approximate certain details of WP's playing, particularly those numbers on which he plays a lot of single-note melody lines rather than octave pairs. However, this can be done pretty easily on a Celestaphone with standard single-level bridges, as not only the pairs but the two strings of each pair are spaced comfortably far apart; both are spaced much wider than the melody section of the 4/30 chord-zither.
The idea of these bridges is simply that they are stepped in order to put one string of each pair high (accessible) at one end of the instrument and low (access blocked by the other string of the pair) at the other end of the instrument. I made the 15 high-tuned strings accessible at the head end and the 15 low-tuned ones accessible at the tail end. And of course playing in the middle where they cross offers access to both strings of each octave-tuned pair. This is all made clear by the images, which give close-ups of the head and tail bridges and a side view. The latter didn't come out great, so I "enhanced" it, the red and blue lines representing the high-tuned and low-tuned sets of strings.
Strings for all chord-zithers fall into three categories:
bass strings: custom-wound
Gauge information is given below for all strings used to re-configure a 4/30 fretless zither to approximate the sound of Washington Phillips's instrument, with either session 1 or session 3 accompaniment section configuration.
With the exception of certain autoharp models, replacement strings for fretless zithers to accommodate even stock stringing configurations have been unavailable for many decades. When I first began to play the chord-zither, I realized that this posed a major problem. No commercial string manufacturer in the world would go near the idea; in fact, most never even bothered to answer my inquiries. It was clear that I was going to have to try and solve this problem my own.
Solve it I did, after much research, and the information in this section of the page represents the fruit of this research.
By far the most difficult part of the unavailability problem was finding a source of proper replacement bass strings for fretless zithers. The best-sounding replacement bass strings are wound in copper. Steel, alloy, and bronze are available but don't sound as good. I contacted Tom Fladmark, an autoharp builder and maker of strings for same, and found he was up for the job. However, he doesn't have copper winding wire, so it is necessary to provide him with it. I found it to be available in small quantities from McMaster-Carr. Their array of gauges is limited but adequate.
Early on, I had tried stripping back a bass guitar string, and the results were awful; it buzzed horribly and incurably. More recently, I tried it again with a different brand of string, and the result was passable. So I would say that my results have been variable; it may work fine or it may not work at all.
smaller-gauge wound strings
These can be made from guitar strings. Actually, even some of the smaller-gauge bass strings can be too, such as the D of chord-zithers that have that chord or those that have a high rather than low F bass string. I have experimented much with different guitar strings, and I think the nickel-wound ones for electric guitars sound best. Bronze-wound strings in this gauge range consistently go dead very soon after installing them. I don't know why.
The best value I have found are the "Rogue" (house) brand nickel-wound strings sold in lots of a dozen by Musician's Friend. The strings sound good, and just as importantly, their service is first rate too.
I have also had good luck with Louie's Juke Joint in New Orleans, who sells Mapes nickel-wound strings in lots of 48. Again, I found the product and service both to be good.
The best deal of all, price-wise, is to buy them directly from Mapes in lots of 144. So far, I haven't taken that plunge, but I probably will. Meanwhile, MF's dozens and Louie's 48-packs are a good alternative.
plain wire strings
I buy this from Mapes in half-pound spools. It's quite inexpensive in this quantity, which is fortunate as they're the only source I know of. Their name for it is dulcimer wire. Service with this company is generally good but occasionally not so great, though they seem responsive to persistence.
Music stores and supply houses sometimes offer music wire in small quantities, but as would be expected it is considerably more expensive.
Every last scrap of information you need to make your own strings for fretless zithers can be found in the Guides to Stringing section of this site. To date, I have outfitted various fretless zithers with over 3500 proper replacement strings, thus I feel confident in saying that you can feel safe in assuming this information to be reliable.
So, if you read and absorb the information at this page and in the Guides, you will know as much as I do about re-stringing your 4/30 fretless zither in the Washington Phillips-style configuration.
To quote Wash himself, "That's all."