O-ring mfgs. will usually have a chart, Sterling for one, showing the dimensions for O-rings. Different dimensions for static and dynamic. A 3/4" O-ring will be over that size to fit properly into a 3/4" bore.
Here's a copy of something I saved from back about 1997 by Scott Conger:
_________________
A few years ago, when I was new to the Hobby, I often heard that axle pumps and o-rings in particular, were no *#$@! good, and the
person that managed to build a reliable pump was someone to be held in awe. I was surprised to hear many of the "facts" that were widely
accepted as Gospel. The following is meant to lay the theoretical foundation of how o-rings really work, and present a few facts about
design considerations in both industry and our hobby of Live Steam. It is by no means a complete or detailed article on the subject, but
may shed some light as to why some things work and why some things don't!
FACT: Virtually all published designs that I have seen in this hobby have incorrectly dimensioned grooves or seats when o-rings are
called out. It's not surprising that so many in the hobby have had mixed success.
FACT: O-rings are usually bigger than their nominally published size. An o-ring with a cross section of 1/8 inch (for example) is bigger
than 1/8 inch and needs a wider and deeper groove. Different o-ring compounds cause o-rings to be different widths. This is not normally
a consideration in our hobby, but is interesting to note.
FACT: An o-ring seals due to both compression (assembled squeeze) and distortion between itself and the sealed component(s). Most
designers in the hobby are relying on compression alone, and seem to be guessing at the correct dimensions even at that. The result is a
pump that requires an unusually high force to actuate, causing premature failure of both the o-ring and the mechanical components that run
the pump eccentrics.
FACT: The extra groove width is a calculated width based on o-ring size and application (not just a guess) and exists to allow the
substance that is being sealed, be it liquid or gas, to get BEHIND the o-ring to energize and partially distort it into the tiny gap between
parts. In this case, the pump bore and the piston. This increase in pressure on the o-ring compensates for the usually modest compression
in the design. This way, the o-ring only works as hard as it has too! The same design that pumps water into our boilers at only 125 psi,
will also power uge cranes and other machinery at truly enormous pressures.
MYTH: NO! NO! NO! Based on the above information, o-ring grooves are NOT overly wide to allow them to roll! Period.
FACT: One of the worst fluids for the average o-ring to be in contact with for long periods of time is water! The standard garden variety
of o-ring, including those found in plumbing and hardware stores is made from a compound called BUNA N. The only compound
recommended by o-ring manufacturers for water, hot water, or especially steam, is ETHYLENE PROPYLENE. Unfortunately, this
compound deteriorates in the presence of petroleum oils. Life just ain't fair!
A good compromise is the compound VITON. VITON has a comparatively long life in a water or steam environment, and also, isn't
vulnerable to oil. The tradeoff, however, is it's higher durometer (hardness) and subsequent reduction in sealing ability, which is of such a
small percentage as to be negligible in our case.
MYTH: Well, if BUNA N is found in hardware stores, and is used in plumbing, then it's just fine for our pumps.
FACT: The reality is, that this compound of o-ring is used in plumbing because it is cheap, it swells, and most plumbing applications are
static, i.e.: they don't move. In the presence of water, over a year's time, BUNA N will swell 15% at room temperature and up to 60% at
elevated temperatures. Great if you want the joint to get tighter! Terrible, if your poor engine is trying to shove a piston in and out, with a
badly swollen o-ring on it! I'll bet some of you have repaired some device, only to have the original o-ring not fit back in where it came out
of. Now you know the reason why. I'll also bet that when your pump gave up the ghost, it had a flat spot on it, or even small tears or fuzz
on the o-ring. These defects are common after even a very short time with the wrong compound.
FACT: In the case of axlepumps, the best design has the o-ring mounted on the piston, sliding back and forth in the pump bore. O-rings
are less effective, and often worse than old style packing when the o-ring is part of the pump body, as is the case on many crosshead
pumps.
" Why is that?", you may ask. "Good question", sez I. Remember that part of the seal is caused by distortion. When the o-ring is on the
piston, it drags along the bore of the pump, distorting ever so slightly into the gap. Now think about this for a minute...the pressure of the
water, trying to get past the o-ring, is also distorting the o-ring into the gap in the same direction as the friction is. Neat, huh?
Now, in the case of the converted crosshead pump, the o-ring is part of the pump body. As the piston is pushed in, the o-ring is being
distorted into the pump body due to friction from the piston's motion. Now, stop and think...are you seeing it yet? That's right, the pressure
of the water is pushing back against the o-ring, opposite the distortion caused by friction. These two forces are effectively canceled out, and
the piston simply skates back and forth through the o-ring, creating an ever present drop of water at the packing nut!
I haven't even mentioned surface finish, compression set, modulus, resilience, permeability,... but you probably already have had a belly
full, right? Well, enough for now. I'll be glad to entertain comments or questions. Call me at (813) 525-1341.
P.S. For those that may scoff, I didn't dream this stuff up. It can all be found in the Parker-Hannifin Corp. O-ring Handbook. It's a lot
longer than this article, and I highly recommend it to any one who has trouble falling asleep. Real technical assistance can be had by
contacting:
Parker Seal
O-Ring Division
2360 Palumbo Dr.
P.O. Box 11751
Lexington, Ky. 40512
(606) 269-2351
Greg Lewis, Prop.
Eyeball Engineering — Home of the dull toolbit.
Our motto: "That looks about right."
Celebrating 35 years of turning perfectly good metal into bits of useless scrap.
There will be lots of different opinions on this, but here's what I do.
The pump ram should be a close slide fit in the bore. A little play can be allowed, but the bore needs to be smooth because of the "O" ring. A reamed fit is OK. It also helps if the ram is made from stainless, but bronze is also OK.
I make the "O" ring groove so the ring just slides into the groove (no axial play). I measure the pump bore and make the groove depth so that it is .004 to .007" less than the "O" ring thickness. You need to have that much radial "pinch" so the "O" ring can seal in the bore.
One "O" ring is enough. More than one just adds more "drag" to the operation of the ram. You can put the "O" ring on the end of the ram, or you can put it on the end of the bore, either location works, just don't put an "O" ring in both locations at the same time. Greg's logic is also correct when he explains why it is better to put the "O" ring on the end of the ram.
What Greg says is right. I also use Viton for all my "O" ring applications and the Parker book is a good source of information. Just don't use any of their groove dimensioning information. If you do, you will find the seal too tight to move. The other thing that is important to remember is that rubber expands a lot with increases in temperature so in some applications you may have to allow for this.
I double down on the Parker O-ring Handbook recommendation. I've had a copy for years and rely on it.
The problem for hobbyists is of course that it's often difficult to locally source a small quantity of o-rings (or ONE) in a specific size and compound and a "that'll git'er done" ring from the hardware store won't do the job (or do it for long.)
on all my engine's with axle pumps I use Viton O-rings, I machine the slot so that i have .003 compresion on the o ring, and a slot width of 150%
of the O-ring width .So if you are using a 1/8" wide o-ring I make the slot 3/16" wide . use only 1 o-ring,as two one will run dry and fail , and fill the slot with grease before you install the o-ring. the grease keeps the o ring soft.
the reason for the 150% slot width is that as the pump cycles back and forth the o-ring rolls back and forth as direction changes.Thus no flat spot wear
on the o-ring. I learned this little trick from the Master Ben Nixon (steam Pump Man)years ago .on average I usually change the o-Ring out every 3 years or so if it needs it or not,during check valve cleaning.
This is just my way ,I am sure there are other ways but I found that it has worked for may years on my 3/4" , 1" , and 1 1/2 steam engines , and have never had any axle pump failures.
Joe
The hand pump has no o ring because it should have packing material in the nut on the pump that the ram slids into. Either graphite or expanded Teflon will work.
As for the axle pump. It takes a standard o ring. I use Viton as others have noted.
i like both...properly done. The packing material of graphite will last almost forever...and if it does happen to start to leak, just tighten it up and keep going. A tender pump that fails may or may not shut you down for the weekend...but if it is packed..you should never be shut down.
I also have o-ring water pumps that have run for years (generally following Joe and Ben Nixon's recommendations).
An o-ring that fails requires disassembly of the pump.
Admittedly, a properly applied o-ring will last a very long time...but a bit of dirt in the inlet can cause a lot of damage....hence Joe's sage advice...
no...not suggesting that you use o-rings AND PACKING...nor should you use two o-rings.
Thanks all for the info and advice. I'm headed to the Parker O-ring guide.
10 Wheeler Rob: how many wraps of graphited yarn would you expect on the hand pump? 2 full wraps or so? It makes sense now that this is how that pump works. The packing was missing when I took it apart.
Kozo Hiraoka wrote a very good article on o-ring usage. His premise was that the commercial o-ring guides make the o-rings fit too tight for live steam models. He offered tables for o-ring use of both metric and imperial sizes. A very good guide. It was published in Live Steam magazine and may also be in the back of his "Building the New Shay" book. My shop copy doesn't say what issue of Live Steam it was in. The steam brake pistons on my Shay were too stiff using the commercial tables, but when I made the piston grooves per Kozo the brakes work fine.
