Hand pumps of antiquity - Part 2
This is the last day of B.B.'s hunt, so he'll start answering questions again on Thursday. Thanks for your patience.
by B.B. Pelletier
The test fixture Quackenbush made to ascertain pump efficiency. The pump attached in this photo is the smaller one with the 5/8" piston head.
I asked Dennis Quackenbush to tell me how much pressure a common hand pump can generate. He's been making replica pumps for vintage big bore guns for several years. Dennis' pumps are true to the old designs, except that they use synthetic pump seals. The old pumps had either a simple iron or steel piston that was lapped into the bore of the pump tube, or they used stacked leather washers on the end of the piston rod instead of a metal piston. The stacked washers were compressed by means of a nut so the fit could be controlled.
I've encountered several dozen antique pumps, and all of them were the type having the simple lapped iron piston. I only know about the stacked washer type from reading Air Guns by Eldon Wolff. He admits that the leather washer type is much rarer than the plain piston type.
Before Dennis conducted the tests, it seemed to both of us that neither vintage pump design could be the equal of one with a modern synthetic seal; so whatever pressure he could generate with a modern replica would represent a maximum for any vintage pump of the same physical specifications. That turned out to be an incorrect assumption, as we shall see.
We both agreed that the practical maximum force that could be applied would be the weight of the person doing the pumping. These early pumps had no mechanical advantage beyond that which is inherent in a single-stage mechanism. Although it would be possible to generate more force than one's weight by pulling the base of the pump toward oneself or by jumping on the pump handle, it isn't practical to do so--and it would be very hard to do it on a continuing basis.
Mechanical advantage is possible, and there are some vintage pumps that use it, but they're rare compared to the bulk of the pumps we know about. The single-stage manual pump is the most common design encountered in vintage airgun equipment.
Dennis saw where I wanted to go with this experiment, and he took up the challenge enthusiastically. He used two different vintage-type pumps of his manufacture, plus the modern Axsor pump to check efficiency. He provided the following data.
I pumped Quackenbush's test fixture to 500 psi and left it there for four months without air loss. Then, it was pumped up to 840 psi. Because I weigh more than Quackenbush, the higher pressure was easy to obtain.
Test of air pump efficiency
Quackenbush connected the pumps to a nine-cubic-inch test reservoir, to which he had attached a pressure gauge. He then pumped up the reservoir with each pump and counted the strokes required to get to certain pressure levels. He repeated this experiment three times to verify his figures, which is in the box below. Naturally, when you consider the number of pump strokes required to build pressure in the 5/8" diameter piston pump, he's not going to repeat each test 30 times! But, in the repetitions he did, the results were close enough to make the figures believable. If anyone doubts the data, they can buy a pump from Dennis and repeat the experiment themselves.
Next, he made another 5/8" diameter pump without a seal to test the effectiveness of something truly vintage. I felt sure he would not be able to reach the same pressure as a pump with a synthetic seal. This one had only a tightly fitted piston with a thin film of oil to seal it. The results were quite surprising! The pump with no seal went all the way up to the same pressure as the pump with the synthetic seal. Proving, once again, that we aren't as smart as we think.
Actually, it's incorrect for me to say this pump has no seal, for the oil film seals quite well. Dennis achieved 720 psi with the plain piston pump, using approximately the same number of strokes as the other 5/8" diameter pump that had the seal.
Then, on a suggestion from one of his airgunning friends, he tried "rapping" the air into the reservoir--similar to the method used with the Korean-built Yewah Triple B Dynamite shotgun. Rapping means imparting extra momentum to the pump through inertia. With this method, Dennis raised the pressure ceiling of the 5/8" plain piston pump to about 810 psi, but at a high cost to his personal well-being. He said his wrists hurt so much from doing the rapping just once that he had to recuperate for several days thereafter. So, rapping, while possible, is not a practical way to fill vintage reservoirs to higher pressures, and it's doubtful that anyone ever did it more than once.
One additional thing he did benefits this study even more than these test results. He noticed that the piston diameter relates directly to the total pressure achieved. By itself, this isn't such a great discovery because Cardew already published similar data in The Airgun from Trigger to Target. What Quackenbush did for us was remove the complex mathematical formulae from Cardew's work and substitute simple equations in their place.
Cardew says that 1,000 psi might be achieved by a heavy man. Quackenbush demonstrates that the pump stops at a point very close to the weight of the person doing the work...when the formula is applied.
What the data show is that it's possible to calculate the maximum pressure for any single-stage air pump by simply dividing the weight of the pump operator by the area of the piston head. There should be some loss of efficiency due to friction in the pump, but this is most probably offset by some imprecision in determining the person’s body weight:
This allows us to build a "theoretical pump" that can generate whatever pressure we desire within the limits of physical laws. For example:
250 lbs. divided by .19635 = 1,273.24 psi
150-lb. operator using a pump with a 1/2" piston
150 lbs. divided by .19635 = 763.94 psi
Notice that the stroke of the piston does not enter into this calculation. That's because the length of the stroke determines only how much air is being compressed. It doesn't affect the highest pressure that can be achieved. A longer stroke will compress a greater volume of air; a small-diameter piston will allow it to build to higher pressure.
Up to this point, the term single-stage has been used when referring to the pumps of old. Single-stage simply means that the pump is compressing air in one direction rather than on both strokes. In all cases, the direction for compression of a single-stage pump is the down stroke. The upward stroke sucks in more air for the next downward compression stroke.
The modern high-pressure air pump from Sweden, in contrast, compresses air in both directions. It's really two pumps nestled one inside the other, both housed inside an outer tube. [No, it's actually THREE pumps, nestled inside one another. And there is a new Swedish pump that claims to be four pumps!] Both pistons have a small piston diameter. The center of the larger pump is occupied by the smaller pump, so the AREA of the two piston heads is comparably small. The larger of the two is the second stage, which operates on the downward stroke--the direction in which the greatest force can be applied. The upward stroke sucks the air into the pump and pressurizes it slightly to begin with, thus making it ready for the second stage to compress it to a very high pressure level. This pump pressurizes a greater volume of air to a higher pressure than its small piston would normally allow because it acts as though it's twice as long as it is.
If a single-stage pump were made with a very small piston--perhaps 3/8", what sort of pressure might be possible from a 150-lb. operator?
Of course, this small piston pump would take a lot longer to fill a reservoir, so we might increase its stroke to help matters. That gives us a long, thin pump that would eventually become too long to carry conveniently. Also, the thinner the pump piston, the thinner the piston rod; although the piston HEAD might be able to compress very high pressures, the rod that connects it to a source of force will eventually become too thin to bear up under the strain.
From this discussion, we see that Cardew was correct. It's possible to generate 1,000 psi and even more with a simple single-stage air pump.
Not only do vintage pumps begin to max out at pressures much lower than modern pumps can achieve, the guns being pressurized cannot deal with air pressures anywhere near these high levels.
The large valve contact surfaces of vintage big bores constrain them to use lower-pressure air. With the state-of-the-art technology available to 17th century airgun makers, it was all they could do to seal valves against pressures ranging from 400 psi up to 650 psi. They used horn and leather to accomplish as much as they did. Today's use of hard synthetic valve heads bearing on thin areas of contact on precisely machined steel seats was simply beyond anything they could achieve.
Also, they lacked the homogenous materials from which to make reservoirs to contain the higher air pressures. Even if they'd been able to make the valves and pumps to work at higher pressures, the reservoirs would have held them back.
Again, I turn to the documented studies that have been done to support this statement. This time, there's even less information than before. The number of ancient airguns that have been intentionally tested to destruction is almost zero. A fair number of them have blown up from too much pressure, but that was accidental; and apart from a big boom and possible injury to the operator, there isn’t much to go on. Cardew does mention one of the folded and brazed butt reservoirs that was tested to destruction.
Amazingly, the antique folded-iron reservoir held until 6,000 psi was developed. When it finally did blow, only one rivet popped, resulting in a controlled exhaust rather than a catastrophic explosion.
When such tests are conducted, the vessel being tested is never filled with air for fear of an explosion. Instead, oil or water is forced in under pressure, so the failure can only result in a safe leak. Unfortunately, this test is exactly the kind of report that a careless person will cite when trying to operate vintage equipment in an unsafe manner.
A case in point
Modern aluminum paintball tanks are designed to contain CO2 safely. CO2 has a pressure determined by temperature. At 70-degrees F, it's just under 900 psi. When the temperature rises to 95 degrees, the pressure increases to around 1,100 psi. Because of this variability, and because of the possibility of even higher temperatures, such as when a tank is stored in a car on a warm day (130 degrees is easily possible), the manufacturers rate the tank up to 1,800 psi. That's the number they put on the label on the tank for everyone to see.
Now, along comes Mr. Airgunner, and he reads this number. Okay, he thinks to himself, if this tank is rated to 1,800 psi, it's safe to pressurize it to that level. He fills it with air instead of CO2. The air tanks rated this high cost much more than the CO2 tanks (wonder why?), so using a paintball tank represents a real savings. If he does this, the tank is filled right up to its design maximu--a level that was engineered by the manufacturer to handle an emergency only. Because CO2 pressure varies with temperature, they make the tank to withstand a higher pressure in case the temperature ever rises unexpectedly. It isn't expected to withstand that pressure all the time, even though it's engineered to do so. It's expected to hold a pressure of around 900 psi when confined. Some folks go by the numbers without appreciating that they're actually subjecting these bottles to a 100 percent overfill. It doesn't end there.
If one person is willing to do that, what's to stop another from pressurizing the same tank right on up to 3,000 psi with the same fill equipment? Maybe he's been using paintball tanks this way for a long time, and there's never been any trouble before. "After all," he says, "these things are over-engineered, anyway."
Yes, they are. And, at 3,000 psi, you're more than 2/3 of the way into that safety margin. Standard pressure vessels are engineered to not fail with less than four times their standard working pressure--which you’ll remember is 900 psi, nominally. Blowup can occur any time after 3,600 psi for a tank having a 900 psi standard working pressure. That's for a new tank. Who's to say at what point an old abused tank will let go?
But tanks have safety valves (burst disks), don’t they? Again, yes, they do. Do you want to trust your life to a small piece of metal that's been sitting in the tank for years? What if the kid at the plant decided to use heavier sheet metal when your tank was made? What if there was a goof when the company placed the order and a stronger material was used without anyone's knowledge? What if someone in the field "fixed" it before you got it (perhaps to keep it from rupturing so easily)?
If modern paintball tanks represent a danger from over pressurization, what about vintage air reservoirs? They don't have burst disks or any other safety devices, plus they've been lying around for decades and centuries with their dubious metal walls containing who knows what kinds of pressure.
Also, the old reservoirs intentionally had grease smeared on the walls near the valve to catch and retain dirt and dust. At 500 psi, nothing is going to happen; but when you get one of these oldies up to 2,000 psi and more, that grease is going to vaporize into a highly explosive gas. Just ask a dive shop what happens when petroleum-based lubricants are used in scuba tanks. Boom!
When we work with precharged airguns of any type, we want to obey all the safety rules that pertain to them. When the airguns are also large bore, we want to be especially careful because they're so powerful.
Part 3 will demonstrate how pressure relates to velocity.