THE WEAKLY-PLAYING DUO-ART
By Craig Brougher
There was a thread in the MMD concerning an individual who decided to fix a weakly playing Duo-Art and was having some trouble, so he decided to inquire as to what he might do to take care of this, on that forum. He received a number of replies from well-wishing and gracious folks who wanted to help, but from what I read, there wasn’t anyone really able to do that, because they could not tell him how to measure or check anything, first. Knowing before-hand that he probably didn’t even have a vacuum gauge, how could they do any more than bump along with him in the dark, anyway?
Any authority on the Duo-Art, who has made any checks and tests on well-playing instruments knows one thing, for sure. The Duo-Art cannot feed back any vacuum pressure information to adjust itself for valve travel losses, as do the Ampico and the Welte. Since the Duo-Art system is a presumptuous one which controls everything by the roll irrespective of the condition of the player, the builders designed into this player both a proportional amount of leakage, and valve leakage. The proportional leakage is the spill valve which is directly connected to the expressions and the valve leakage is normal seepage through a very thick suede leather, of about .055-.060 thick. The more leakage you can design a player with, the more forgiving it becomes and the less critical it is. But Duo-Arts at best are very critical.
First of all, the valve seepage in a stack increases or decreases according to the vacuum pressure keeping them closed, so in one sense of the word, that leakage is also “proportional.” But unless this graduated leakage is present, the Duo-Art will never play well, sounding like a real person.
I tried to regulate a model A Steinway grand Duo-Art once which had been restored with patent leather (ie. leakproof) valves. The rebuilder was very proud of his choice, claiming to the owner that such a valve would never leak, and so the player would be perfect! The only problem with the idea was that the rolls never knew just how “perfect” this rebuild actually was. In truth, it didn’t leak a whit! And the owner, from that time forward was never satisfied with it, nor was I. Now I knew exactly what the problem had to be and asked the owner if in fact that was the case, and she told me “Yes, it is.” So I “simulated” a leak in the supply going to the stack with a small metal bleed inserted behind a felt punching. This of course was not much help, because it was a constant and could not get tighter with increased vacuum. It did however sound fractionally better (different) to her, and she thought at the time that it improved it. I knew better, but I also know human nature, and what she wanted was something that sounded different for awhile. I didn’t want to tell her the most sublime nuances of the instrument she would never hear until someone removed these valves and installed the correct suede leather (there were also other issues with these valves, but let’s stick to the main one). I tell this story only to make the point in record time, rather than getting all technical.
Specifically, a “pinhole leak” is actually a very precise amount of leakage, being the amount of overall leakage a #70 drill will make in foil. If you believe that a pinhole will prevent a Duo-Art from playing correctly, you are wrong, but the uninitiated might take that wrong information and decide to rectify their pinhole by getting their player cartridge tight, totally ruining its response. This was one of the letters to the fellow who was hoping to make his Duo-Art play strongly, again. While I am hoping it was just an unfortunate metaphor, it still could be taken at face value and acted upon.
Were you to test every single Duo-Art stack that you restore, you would discover that the most perfect stacks seem to have normal leakage equivalent to, say, several 3/32” dia. hole leaks per side (bass and treble each) measured at 10 inches of vacuum. This number of holes each side of the stack is between 1-1/2 and 2. This obviously was the manufacturer’s goal as well, because every piano with that amount of valve seepage regulates and plays ideally.
That is no pinhole. That is between 3 and 4 - 3/32” dia. holes in the stack (not to mention the rest of the player), which alone equates to 45 times the leakage of a pinhole. That’s almost 1-1/2 orders of magnitude—and considered perfect. It’s invisible to a Duo-Art. But that said, think how tight that actually is per valve. That means that each valve’s total seepage when sealed is maybe a #75 drill (that’s a guess. Anyway, it’s super small). You can’t decrease suede leather seepage much more than that. That was apparently the factory’s ideal natural leakage, working backward to the ideal. They knew that leather would also slowly dry rot, too, but also that at the same time it would “pack.” So until the leather became porous enough to delaminate and fluff up, it would retain its factory tightness for years.
“For years” however doesn’t mean forever. We might be talking about 50-60 years under ideal conditions, but then all the time, the pump is also wearing out its cloth and its flaps and their seats are getting rotten, too. Then what about all the other valves in the system that we haven’t talked about? They too are gradually getting weaker. So are the pneumatics, which were just rubber covered cotton cloth. The rubber gets brittle and cracks where they fold, and so they leak badly. And then the same pneumatics were glued down on leather patches which by now can be literally bumped and they will fall off, they are so rotten. So before the stack valves fail, many other things are first in line.
What happened then in the 70’s and throughout the next 30 years were those “total restorations” (so-called) in which all the player parts were exquisitely refinished, possibly even the outside valve leathers were replaced while the sealing side (the most important) of the valve was almost always ignored. I have even seen “restorations” in which the valve wells still had their original cloth tape sealing up all the valves, meaning they hadn’t even been touched. And if you think that’s something, some of the world’s most expensive “restorations” were delivered to a hapless owner, in which the pump was never even removed from that piano, much less rebuilt, and I’ve seen that to be the case in multiple instances, and quite often neither the pump OR the valves had been touched. Why do I mention this? Because when owners wonder why their piano is suddenly playing weakly, there may be a host of reasons. You cannot say, “It was restored only 10 years ago, so it can’t be a major problem.”
The upshot of this hit-or-miss mentality became a huge disillusionment in the early 80’s and forward. The chickens came home to roost and many player rebuilders lost their businesses because of it. Some, rightly so. But a few were good, conscientious rebuilders whose customers decided not to trust any and they got unfairly thrown out with the bathwater. Ironically however, the former customers of the mediocre had to keep their legend alive otherwise their own “restored” instruments would be dubious. Luckily, these instruments were usually only played when company came over, so those still ancient leather valves were well packed for that unaware buyer. All the time however, I was making noise just like I am now, decrying the mediocrity, while a few of their old customers still believed them even though their own pianos were playing weakly, and to this day a few still may (they would have to be very naïve), but to my knowledge, very few such players are still playing, much less playing according to factory specs. You could say, “they play weakly.” Or actually more like, yearly. The Christmas rolls are the ones they hope to hear perhaps just once more.
What then causes a Duo-Art to play weakly? Well, any leak in the system is all it takes, but far more than a pinhole. If each valve in the system seeped as much as a pinhole (the equivalent of .007 sq. in, the overall leakage would be about the same as drilling a ¼” hole into the pump. The Duo-Art can easily accommodate that. Matter of fact, it can accommodate 2 or 3 of those with very little dropout if you just re-regulate it for its total losses. The performance, while not perfect, will be acceptable to most ears as a 30 year old player.
The most suspicious component in the Duo-Art will be the pump. While you cannot test this properly without a capacity tester, you can still give it the old barnyard check. You remove the supply hose to the expression box and cover it gradually, with your palm. Experience is required here but if after you have sealed it you don’t completely stop the pump from turning (the belt is supposed to slip according to the design specs), then try patching its bellows. What so many rebuilders of pumps fail to notice is that when they make their single fold in the open end of each of the 4 bellows they create a rolling deep crease on the sides of the bellows which strips out the rubber quickly, especially in that “cheap black pump cloth” (and you know what I’m talking about, and to whom I speak). So the simple and temporary repair here is to glue thin cabretta punchings over these tears. That will take 8, 1-3/8” leather punchings and some glue. OK? From there, you can test the pump again, because as often as not, the biggest leak in an old Aeolian pump are the flap seats inside the bellows (feeders). Those you cannot repair, anyway.
As you can see, from here on you will require stuff like masking tape, heavy duct seal clay, and a precision vacuum gauge (Not a Marshalltown gauge). By isolating one component at a time, you will be able to add up all the leaks and after all that work you will probably discover that everything needs to be replaced anyway because too large a percentage of the critical sealing material is as old as the piano itself. It was never really “restored,” to begin with.
Design by Player-Care
407 19th Ave, Brick, NJ, 08724