Subject: What Do You Mean, "Do The Valves?"

by Craig Brougher

You would probably be surprised at how many player rebuilders, yet today, do not restore the valves! When I was traveling around the country (in my single years) rebuilding some of the very finest "RESTORED" instruments that exist, I would plan an itinerary, then make a circuit, restoring as many player components as I could in a 3-6 month period. I would stay at the customer's home and, right in front of his eyes, show him exactly why his player wasn't performing up to specifications. In one home, I worked day and night for 5-1/2 months, and fully restored, from scratch, 5 reproducers, with my customer, who was a professional woodworker, helping me out.

On one of my forays, I stopped by a well-known rebuilder's shop in which 3 or 4 other rebuilders were also working and was shown a player stack in progress. I was asked by the rebuilder, "We were just looking at these valves and were wondering whether or not to replace them. What do you think?"

I have been under one of the world's most expensive and exclusive concert grands. It was a special order instrument, of which only two, maybe three, were ever made, and which had been retrofitted at the factory for an Ampico. The instrument gleamed in awesome splendor, but didn't play very well. I removed a couple of valve blocks and showed the customer, "Here's your problem." We popped the tops and dropped out the valves-- just as original as the day they were installed. The blocks had never been separated, and the pouches were stiff and old as well.

If a player is anything at all, "It is valves." It is the valves which make them play, and it is also the valves which cause them not to play. To the degree that the valves are tight and responsive is the degree to which the player is able to respond as well. To ignore them is to cheat the customer. To the degree that they are ignored, is the degree that the owner or future owner is being ripped off. Why not look at it honestly, for a change?

When trying to debug a Duo-Art Steinway someone else had restored, I had eliminated all but the stack, and saw that most of the leakage and losses were there. I knew the rebuilder so I called him. He told me absolutely yes, he had done every single valve. I said, "I'm going to have to drop this stack and take it with me. Do you want a say in it before I take over?" He said, "No, that's ok. But you won't find anything wrong there."

I tested it on the bench and then disassembled it. The valves had not even been touched! Yet they did look nice, I'll give him that much. So I called and said, "Sir, you said you had restored every single valve. These valves haven't even been touched." He said, "It would have been a crime to have removed that leather. That leather was just perfect! I know that isn't the problem, because the player played very well after I was done, and the valve leather isn't going to rot away in the next few months."

Here is where they are all so very wrong. First of all, if the finest old valve is placed on a tester with a bubble jar attached and compared to the same kind of a valve with new leather, you will discover a four-fold difference in tightness, simply due to surface dry rot that you cannot see! But even that much extra leakage is not as critical to the player as what is going to happen very quickly under operating conditions:

The "poppet" of the valve which has been covered on both sides with a leather punching is the movable part. A pouch (thin membrane) has to push the poppet away from its inside seat, up hard against its outside seat, away from the chest vacuum. So the pouch has to overcome a "suction," clamping that leather down to its inside seat which sometimes exceeds 40" of water vacuum pressure. What this does to the old, dry rotted leather is to delaminate it and "fluff it," making it a bit looser each time it lifts, and hence leakier.

It is this delamination effect, the pulling apart of dry-rotted leather, which quickly ruins the power and repetition of reproducers, and less quickly but just as surely ruins the operation of every old pumper player as well. It is the leak you cannot hear, and the leak you cannot see. You cannot look at a valve and say, "Why, that's just as good as the day it was put in there." Instead, you have to take your chances.

Granted, many players have original leather. The reason is simple: That leather is still fresh enough to give a few years of good service. And quite often, those players were seldom if ever played. It's bacteria and moisture, with heat which gradually destroys leather. In the midwestern plains where lots of corn is grown, the air is full of mold and food for bacteria to thrive on, so the tanning acids will not fully poison all the bacterial growth, and if the player is played a lot, it will quickly contain plenty of food and bacteria to sustain the colony for years to come.

If the player then goes dormant for many years without being constantly played, the bacteria are able to work faster. But even so, the midwestern plain states and agricultural areas where pollen and mold counts are high seem to rot the valves more quickly. Moving a northeastern player into one of these areas is going to be "curtains" for that performer in short order, IF it has old valve leather inside.

If each valve in a player leaked only as much as a hole made by a shirt pin (#70 drill bit size), the overall effect would be the same as drilling a 1/4" hole right through the pump. Old leather leaks more than that, initially. New leather scarcely leaks at all (if you've selected it correctly and glued it tightly).

If you would like to check this out for yourself, take a plastic film can and punch a hole in the lid representing the inside valve plate hole. Using a hot glue gun and a drill, fasten securely a 5/32" nipple to the plastic can somewhere and tube it to a bubble jar, to the long tube part going into the water. Apply a suction to the other bubble jar tube with the short tube above the water. The bubbles you see are representative of air leakage. By changing to different leather, you should be able to tell a difference.

The Orchestrion Builder's Manual and Pneumatics Handbook has a large appendix in the back covering all these tests and much more. Also tells you how to build several kinds of bubble jars, and how to hook them up. It also gives you a very lucid explanation of valves and how they work, how to optimize them, and what not to do, and why. If you want to learn about valves, there is probably no better book on the subject. It is available from Player-Care.com.

Just a little more needs to be said about valve travel and the effects of narrow gapping.

As has already been pointed out, the surface of the valve leather facings has a lot to do with the resistance of the valve, too. Some authors still speak in terms of valve gaps, as has been done now for the last 30 years or more. I mentioned the fact that I "blew" my valves to get a resistance measurement, instead of "gapping" them with a micrometer of some sort, or a shimming tool.

The Ampico valve travel machine, which also appeared in some technical advertisements of the day, was designed to measure air travel through the valve gap instead of measuring the gap itself, nothing was mentioned in regard to how much gap the machine deemed ideal.

I realized that I never explained how I arrived at my "ideal" gap. Here's how it was done. Some years ago when I was hiring girls to do things like this, I would notice, even after rechecking their settings, that some Duo-Arts didn't play as well as others. I was using the old gapping techniques then. I tried not to let any of those pianos leave the shop until I had fixed them. So I would drop the stacks and retest.

I stayed up nights, sometimes, trying to figure this out. I finally hit on the solution to why these pianos could play their zero or first intensity well, but suffered during full power, and particularly noticeable in the single pianissimo passages as well. The mezzo-forte and average loudness playing seemed to be just fine. It was the nuances in the very low end-- just above the set point-- and the levels at the highest intensities that suffered the most, to my ear.

So I changed the valve gaps and after regapping a set of valves half-way, started "blowing" down through them. I immediately noticed small differences that I couldn't gap! I decided that it was the suede differences, since the gaps were identical. Then I got the idea of blowing through an empty valve plate and comparing it to a valve. That night was the turning point of valve regulation for me.

I also discovered that valve losses (travel losses) do not somehow follow linearly with a corresponding increase in gap. I explain this effect by saying that a narrowly gapped valve has a bit more proclivity to touch both seats at the same time as it is being balanced and hoisted by the pouch-- always at an angle. So because it is very slightly "jammed" that way, as some of them travel upward at an angle, it requires longer to seal. This time adds to the travel-time and travel losses. If that turns out to be incorrect it doesn't really matter anyway, because at any rate, you are regulating its zero or first intensity with this set amount of travel, so the extra percentage of travel loss is cancelled out. Bingo-- it becomes a moot point anyway!

The Duo-Art is the perfect instrument to prove this fact with, because its expression regulators do not try to keep stack pressure constant at all! Unlike what some authors think about the Duo-Art regulation system, these regulators were designed for one thing: To keep the zero intensity at a specific level _inside the expression box_ only. This, of course is a lead pipe cinch simply by looking at the roll coding of a Duo-Art. There is enough pressure reserve inside the expression box to supply about two or three notes enough zero intensity to play. If you want to play 4 notes, you have to add an intensity, so it's better to add two intensities and put the soft pedal on for safety margins-- which is the same practical idea as what you'd do for the zero intensity, in most cases.

The reason that narrow valve gaps affect the soft nuances noticeably is because of the slowing down of very low pressure air. One might think that a narrow gap would work for low intensities, but that is only so for your "ZERO" intensity-- the intensity you have set everything to. So naturally, that intensity has to be ok. From that intensity, however, the piano does not seem to make as noticeable a change from level to level as it does when there is a bit less resistance in the valve gap.

Since I cannot tell anyone a "valve setting" this way to use, all I can do is say that it always gaps greater than .032". But air resistance-wise, it is a bit beyond the throttle point of the valve.

If you really want to learn valves, set them with a shim gauge at .032" and listen to the piano perform. Then pull the stack and reregulate the valves my way, replace it, and try it all over again. You'll see what I mean! A certain amount of air restriction is still necessary by the valve. You need to "feel" that valve face when you blow into it. But there is a setting, somewhere between .032" and .035" that is just too restrictive for that reproducer to work well at. You'll see what I mean. So trust your own senses and try it. You'll like it! You'll see why Ampico went to all the trouble of measuring air flow instead of gaps!

E-Mail to: Craig Brougher
Phone No: 816-254-1693