By Craig Brougher

(For other information about sealing leather pouches -click here)

There has been a long and extensive thread in the MMD this month of June, 2009, about treating pouches in player pianos. It's a kinda fun topic and I enjoyed reading about the different methods of sealing leather. Basically, the sealants recommended in the letters ranged around these, primarily:

Carter’s Rubber Cement

Egg White

PVC-E glue, thinned

Neatsfoot Oil and Mink Oil

Pure Silicone Grease, thinned

I'll take these individually because each is worth a comment, pro and con. But let it be known that it's not just the sealant that's important, but also the method. I'll also talk a little about the leather you're using and what NOT to use, as well as which leathers are the best.

Pouch leathers today are far different than original pneumatic leathers of yesteryear. Back in the heyday of player pianos, it was very important to have tight leather, unsealed, to begin with. Much of Playerdom’s famous pouch leathers came from registered herds of Scottish highland sheep, bred especially for their skin as pneumatic leather. These were usually 2-year old longhair sheep, and the skin was particularly dense, but soft, like lambskin. If pneumatic leather isn't very soft as well as tight, all the sealant isn't going to help much. It has to balloon with a bare breath and so it can't have dried glue around its hole rim, or it's just like stiff leather. It's better to have somewhat porous leather and feather light flexibility and softness, than a good seal and a little stiff.

Today's lamb market is different. Lambs are grown for their wool and their meat, so they are watered a lot, grain fed, stalled, and then slaughtered for food. A by-product of this is their hide. So now what we have to work with (at least in many cases) is a grade of leather the old timers would have thrown away as unsuitable!

Some of the very thinnest pouch leather you will ever find will be found in the Ampico. That's because the Ampico has such long tubing runs. For example, in the model B, 5.5 ft is common. At 3.5-4” H2O, these valves still operate reliably while an early Duo-Art valve for instance, won't even budge under the same circumstances. So you can see that this pouch leather must be special.”

Well, it was. It was what was considered 'select',” or grade A to begin with. Then it was sanded down to about .0055 thick. After that, the pouch leather to be cut for primaries was sealed with thinned-down rubber cement although they didn't use Carter’s. That was just a suggestion for field techs who might have to replace one. They much more likely used shoe sole cement like “Tiger Brand,” which is stabilized. We'll get into that, later.

Before those pouches were cut, every skin was marked on a light table so that no areas which showed pin holes or thinned areas were used. Secondary valves (the block valves) used the same leather, but were not sealed leather, to my knowledge. Since the sealer still stiffens the leather, there's a trade-off between sealant and flexibility. Bleeds were determined approximately for the pouch nipples which were 3/4ths of the maximum distance away from a pouch. They were also scaled for gradual porosity in the pouch leather as well. That was then standardized, and all valves got the same bleed. But if bleeds were that critical, then you would have a great variety of sizes, and the bleeds would be sized according to the tracker bar tube length. Such was never the case with any company! If you have problems like that, then you have valve problems which far transcend the bleeds. Changing a bleed is just a work around, then.

Carter Rubber Cement

This has been a popular way of sealing pouch leather and there's nothing wrong with it, initially. The only problem is that rubber cement is, in a way, the liquid form of rubber bands. Natural gum latex is used in each one. And since they no longer put shoe soles on with rubber cement, age stabilizers are no longer necessary in most rubber cements and rubber bands (interestingly, dye colors in rubber bands will preserve the rubber a little). But what happens to rubber bands happens to rubber cement, and the thinner and more sparse the cement is, the quicker it deteriorates. If you extruded a type of rubber cement into tubes, inflated it onto rollers cured it, then sliced it off with rotary cutters, you'd have rubber bands. Now, what happens to rubber bands in, say, ten years’ time? But curing links the polymer chains, making it last far longer that cement.

But if that isn't bad enough, then try this; roll your rubber bands out to about .0005 in. thick, dry that, slice them up, and see how long they last. Did you say, 2 months? Well, if you're lucky and live in Arizona!

If you use Carter’s, the only way to do it is in the flat skin, very thinned out, brushed on, thinly and sucked into the leather pores through a flat screen with vacuum pump. Dry and talcum. But when the pouch has been laid and you 'putty' the cement onto it with a finger or a brush, you pile rubber cement 10 times thicker than it needs to be around the rim of each pouch where it's glued. And, heaven forbid, if you think you can just reseal the old ones this way which had already been rubberized. You stiffen them far worse. that's why Ampico primary pouches stand up like little donuts or little craters. They are 'frozen' by hardened rubber cement. Those would be far more sensitive pouches if you replaced them with new bare unsealed leather, although I don't recommend that either.

As rubber cement gets old, it also gets hard. Find a rubber band that's 10 years old. It's brittle. You bend it ever so slightly, it snaps into lots of little pieces. Suppose that rubber band was only .0005" or less thick to start with. You'd have a hard time finding one 10 years old. It would be dust. So resealing old rubberized leather pouches stiffens them further. Even though the rubber cement gets hard, if it's in the pores, it's still sealing for a long time. But remember, the factory probably did not use rubber cement sold for office supplies. Tiger Brand Shoe Sole Cement, or any vulcanizing rubber cement for tire patches, containing about 6% polyisoprene rubber is stabilized rubber cement by a variety of methods and far better than any office supply you can buy, but open to air, it will still get hard. that's the stuff they use to patch inner-tubes with.

[When I was still in grade school, some of our art projects were put together with Carter’s Rubber Cement. Those were kept in a cigar box at home. When that box was gone through just a few years later, everything that was glued with it was now separated. It only took a few years.]

Egg White

You can buy a pound of egg white powder for about $25. It is an excellent sealer, for a long while. It was for a long time the traditional photographic emulsion on paper. But it is proteinous, and it slowly hardens and cracks when bent. That happens faster than leather because it is not a structured, layered long molecule collagen. Albumen likewise stiffens pouch leather even more than rubber cement but it doesn't deteriorate anywhere near as fast. It is a moderately long-lived sealant, but as the pouch operates, it slowly looses more and more vacuum. there's nothing wrong with that. it's normal.

Since egg white tends to stiffen leather, it should be painted on the leather first, before the leather is ever punched out. Any time you apply a drying surface sealant you never apply that sealant after the pouch is applied to the pouch board and dipped. That is an absolute. However you apply a drying surface sealant, you will always make the film thickest around the rim of the leather where it must be the most flexible.

Egg white is an excellent sealant, provided you don't need especially flexible pouches. Because egg white is a surface sealant, and must be put on thicker to be useful. So it's good for large pouches. I would never use egg white for reproducer pouches. And thin egg white stays wet too long and hardens the leather by that reason, alone. Wet some pouch leather sometime, let it dry, and then test it in a test vales setup. You might as well see what I'm talking about, first-hand. Now there are ways to make seemingly unworkable things work, otherwise, and no one is fully aware of all the techniques. Still, I'm getting to what really works well for even beginners without fancy tricks. Bear with me.

PVC-E Glue, thinned

PVC-E is what we call 'plastic' glue,” or white craft glue. It will seal leather up cartridge tight. It also makes leather very stiff. In my book, it is unusable for sensitive reproducers, although it will work for cut-off pouches on stack cut-offs, and stuff like that, as long as you talcum powder it heavily because that leather will remain sticky for a long time, otherwise. It will seal leather so well that it will preserve it. I would think that for things like Reproduco pouch pipe chests and the like, this method would work well. It might also work for rolled pouches and small pneumatic bellows pouches, such as the Schulz player. So it has some possibilities, as long as the leather is well-powdered. I personally don't use it.

Neatsfoot Oil, Mink Oil

These deserve mention only because I warned Durrell Armstrong years ago about them, and that warning is still extant, today. All animal oils, including lanolin, draws moth larva. It has an irresistible odor for them. But what's worse is, all animal oils evaporate slowly, as well as spread by surface tension. When vacuum is drawn into a player, the neatsfoot oil gets pumped right out of the pouches again, but where does it go, you ask? Well, unfortunately, it settles on all surfaces including the metal valve seats, valve support plates, and stems inside your player. that's just the first layer, however.

After that, warm, moist air which also coats parts is drawn below the lighter oily film and then stays put longer, because the oily film is stable and won't allow the moisture to evaporate quickly. So you now get a verdigris growing on tin-plated brass valve parts which otherwise would not have happened. The metal's surface tension draws the water molecules stronger than the oil molecules, so the oil stays on top, trapping the water. Besides all of this, an oiled pouch doesn't stay sealed longer than perhaps a year or two at the most. You can test them 2-3 years down the line, and It's like you hadn't done anything. But it will soak into the wood and prevent a good, tight glue bond for new pouches. BAD IDEA!

Pure Silicone Grease

Pure silicone grease, by itself, is far too viscous to use to seal pouches with. And like every other sealer, you can use too much. Luckily it is a space-age product, used on satellites in outer space, and doesn't evaporate. it's also a very large molecule and doesn't soak deeply into wood, rendering it unglue-able, as do animal oils. It's a very friendly material when thinned with just about anything. Experiment and see which you like the best. You can buy Dow Corning 111 Pure Silicone Grease at most outlets that sell transmission products like gears, belts, and pulleys. (Craig thins Dow Corning 111 with trichloroethylene. See: Thin it just enough that it's easy to apply with a small brush. NOTE: If you're using a lifter disk, see: and remember that the sealer doesn't get applied until after the disk has been installed.)

NOTE from John A Tuttle concerning the thinning of Dow Corning 111:
The mixture I'm using to seal pouches for player pianos is:

2 grams Dow Corning 111
1 fluid ounce M.E.K.
1 fluid ounce Xylol (or Xylene)

It takes about an hour for the 111 to become completely suspended in the thinner. Initially, I shake it up about once every twenty minutes until the solution is completely clear looking. When I use it, I give the bottle a few shakes because there is some settling. I use a soft fine brush to apply the mixture. One coating seals the leather nicely. Two coats seals to about 85-90 percent. (Obviously, it depends on how porous the leather is to begin with.)

Do not seal any pouch this way that requires a lifter disk, until AFTER the disk is applied. I use hot hide glue for that, and also to put the pouch down to begin with. I also talcum the pouch, once the solution has soaked down into the leather and is now dry. As far as tightness is concerned, no leather pouch needs to be cartridge tight. There is a point of diminishing returns and it happens quickly. You can test this yourself if you have built some sensitive testing equipment like I invested in 30 years ago, or more. I spent the time required to try out all this stuff to see what really was the best, overall. I'll happily put my money on silicone grease.

"Now this is far different than a 'silicone' sealant!" That is a rubber caulk. Do not use that, under any circumstances. What silicone grease does is, it seals only in the pores of the leather, but unlike Carter’s Rubber Cement, does not start getting firmer and firmer over time. Nor does it evaporate. And, it protects and preserves leather even longer than PVC-E would. It also masks the smell of natural oils to insects. You will not find even one moth hole in a leather pouch sealed this way. It also has far less stiffening effect on leather than any solidifying sealer you can possibly use.

So this is the leather sealer that has it all in my opinion, and while it's rather expensive, a little goes a long way. It does not soak into wood or leather. It does not evaporate. It leaves the leather as soft as it was to begin with, as long as you don't overdo it. It preserves the leather, and it repels insects. It can also be squirted onto the leather or brushed on with a sable brush, after the lifter is applied. It dissolves in lacquer thinner and other things, and can be made as thin as you like, and it will NEVER deteriorate.

The Most Important Thing in Pouch Work

If I was going to mention the most important thing in pouches, it isn't perfect air-tightness. It is leather quality, defined as a combination of air tightness, light weight, consistent thickness, and softness. What we mostly use today is already less than anything the original player manufacturers would have used 'even' the cheapest outfits could not afford to utilize the stuff we are stuck with for the most part, today. As I have already pointed out, there are few animals raised mainly for pneumatic leather. For example, a Duo-Art accordion bellows recovered in the typical pouch leather of today would survive here in the Midwest, unprotected and outside in room air, at the most 8-12 years. When Aeolian first covered those bellows in pneumatic leather, they lasted 40-50 years. I suspect that was albumen coated leather, but the majority of our pneumatic leather today won't last that long with any coating.

Now there is kangaroo hide. It is extremely tough, almost untearable leather, and it will last several lifetimes, inside of a player piano. The reason I don't use kangaroo on reproducers is because it is stiffer leather because it is denser. It is however, very tight. But for pipe chests and organ work, nickel pianos, and most player pianos, it is unexcelled. So now we have the need to pick the right leather for the job, right? We need to look at all aspects of the subject, first. Columbia Leathers also carries a Burgundy-colored leather with good longevity but it is an organ leather, designed for uses other than small pouches operating quickly at low reproducer pressures.

There is also a type of zephyr skin that is dry-tanned, very much UNLIKE the junk we were used to seeing at PPCo. That was oil-tanned gut. It varied hugely in thickness and then was laminated together to form sheets, which later on, in service, delaminated! It caused a lot of misery along the way. At any rate, the good zephyr skin is still made and comes from Germany— of course, and makes an excellent pouch skin, too.

Even at that, whether it's zephyr gut, or soft-tanned lambskin leather, all hides have pin holes and thin places, and unless you mark these out before you punch, your pouches are going to leak. don't expect any sealer to fix bad leather. It won't.

The more porous leather is, the more sealant you need to use. The more sealant you use, the stiffer the pouch will become. Pouch leather needs, first of all to be soft and flexible. Second, it needs to be reasonably airtight but not cartridge tight. I have found by experimentation that as long as the bleed's total cu. ft/min leakage is 3 times more than the pouch cipher, the valve usually works perfectly. that's because the factory sized the bleeds, taking into consideration gradual increases in pouch porosity.

So remember, low mass and flexibility is the most important thing and tightness is next. If a player has primary pouches, these then become the most important set of pouches in a player —NOT the secondaries. And if you want to know how well you're doing, test them. Put a 6-7 ft. length of tubing on the pouch nipple, to a trackerbar nipple that you can fit, singly, into the test tube's other end, run the pressure down to 4" or so, and see what works, sensitivity-wise. Then raise the vacuum to a usable figure and start testing for best repetition. The smaller the overall bleed losses, the slower the repetition. The greater the overall bleed losses, the faster the repetition (and again, there's a point of diminished returns here, as well. So everything is a trade-off, from the leather you decide on, to the sealer you use or don't use, as the case may be.

Craig Brougher

I Don't..

This page was last revised on January 9, 2017 by John A Tuttle

Page Design by Player-Care

Grand 16 IconGrand 32 Icon

407 19th Ave, Brick, NJ, 08724
Upright 32 IconUpright 16 Icon
Google Adsense Ad