Myths dispelled

Paul Vallandingham, the man, the myth, the lawyer who could successfully argue that Thoroughbreds are faster than other breeds of horse because their coat is made up of hairs are actually teeny, tiny wings. :wink:

Sean

 

[Stophel]

flint...percussion...doesn't matter.

Wheel locks are fastest.
:wink:

 

[Mike Brooks]

flint...percussion...doesn't matter.

Wheel locks are fastest.
:wink:

DITTO, Fastest gun I ever shot. :thumbsup:

 

[Black Hand]

...... Here are the actual numbers for the 15 trials:

Mule Ear-----------S Siler Perc
.010--------------------.011
.014--------------------.012
.010--------------------.010
.016--------------------.010
.010--------------------.010
.011--------------------.010
.009--------------------.012
.019--------------------.010
.023--------------------.008
.009--------------------.012
.013--------------------.025
.019--------------------.013
.016--------------------.009
.010--------------------.012
.016--------------------.008
------------------------------------
.0137------------------.0115 average in seconds

But are the differences statistically significant? Not according to my statistics program. The T-test has a P value of 0.1622 which indicates the difference is NOT significant (significance starts at 0.05).
Mule ear - mean 0.01367 SD 0.001147
Siler - mean 0.01147 SD 0.004033

 

...... Here are the actual numbers for the 15 trials:

Mule Ear-----------S Siler Perc
.010--------------------.011
.014--------------------.012
.010--------------------.010
.016--------------------.010
.010--------------------.010
.011--------------------.010
.009--------------------.012
.019--------------------.010
.023--------------------.008
.009--------------------.012
.013--------------------.025
.019--------------------.013
.016--------------------.009
.010--------------------.012
.016--------------------.008
------------------------------------
.0137------------------.0115 average in seconds

But are the differences statistically significant?

This stuff is "significant" to almost nobody.

It's like two people arguing over whether it's 61 or 62 degrees outside. Nobody can discern the one degree difference and it is insignificant. And, this stuff is measuring thousands of a second? Really :shake:

 

[gizamo]

Ding, Ding, Ding,,,,,

We have a winner!

I've been thinking the same thing as Carl... but his comparison of two guys arguing about it being 61 or 62 degrees outside ~ hit the nail on the head.... :grin:

giz

 

[Larry Pletcher]

But are the differences statistically significant? Not according to my statistics program. The T-test has a P value of 0.1622 which indicates the difference is NOT significant (significance starts at 0.05).
Mule ear - mean 0.01367 SD 0.001147
Siler - mean 0.01147 SD 0.004033

I agree completely. Quoting from my earlier post: "It is our opinion that there is no advantage in the use of this mule ear lock over this side hammer lock on this day- as we would have expected. We also understand that individual locks differ in the strength of springs, so we consider this simply a test of one lock vs another."

If this write-up would have been for an article I would have called it statistically insignficant. The surprise here is that there is no significance difference between between the 2 percussion locks. I think most of us would have expected the mule ear to be significantly faster. I still wonder about its extremely short throw???

Regards,
Pletch

 

[Leatherbark]

What if one took two identical rifles. One flint and the other percussion and matched the powder charges to get the exact same muzzle velocity and find a way to time when the hammer is tripped and when the ball hits the target at a certain yardage...........Which ever one hits the target first is the fastest.......................Bob

 

[Larry Pletcher]

Larry, snipped. . .
A couple of years ago, When you were at Friendship, and I was joined by my brother, Peter, Pete had arrived first, and had carefully inspected your printed photos of various locks you had tested. They were tacked to a railing on the East side of the Gun Maker's Hall of Fame, where you were set up.

One of the pictures clearly showed a flintlock with the flint wrapped in leather, rebounding, or Bouncing back off the frizzen after its first hit, and then striking the frizzen a second time, as the frizzen was already beginning to open. He pointed it out to me, as he was aware then of the "war " being held here about lead vs. leather wraps. The picture clearly showed the flint bouncing, rather than scraping steel off the face of the frizzen, proving my concern about using leather as a wrap.

. . . snipped again. . . .

My hat is off to you, Larry. :hatsoff: Thank you for all the work you are doing. :thumbsup:

Paul,
I should speak to the evidence from your brother about chatter marks. I'm no lawyer but I would offer a plausible alternate explanation for the "chatter" marks that your brother saw in the still photo. This lock is my "test mule", having been fired probably thousands of times since 1988. During those tests sessions, the lock has fired every kind of flint, agate, etc - bevel up and down. About every possible situation was tested. I'd submit that the cause of the chatter is bevel up-bevel down firings for thousands of trials. In my early years, everytime I tested bevel up and down. Early tests were done with 20 trials each. I think it quite plausible that this lock punishment is the reason for the chatter.

I agree with you, though, that if this were on a gun it would be a reasonable assumption. A shooter would likely try his lock out, find what it likes, and then rarely vary from what works.

To another topic, I think mentioned by others, is the reasons for decision-making on methodology. Almost always the reasons behind testing a certain way was/is to eliminate variables. Each experimenter first tries to eliminate variables if possible. To control a variable is next best. An example of this is using a hot wire to ignite the pan in the March MB article. If I had ignited the pan with a flintlock normally, the quality and quanity of sparks intruduce a huge problem with variables. Using the wire eliminates this issue completely. There are many decisions like this that been made for similar reasons. If a reader sees a curious method used, it may be because of a variable problem.

A last comment that should be made is that that "Good Science is Repeatable". I think of this often when I work. If I think I can't repeat an experiment with at least similar results, I generally review my methods and see if I can remedy the problem. (I have delayed a test session simply to have a humidity level that is similar to an earlier test.) Usually if I don't have confidence in my data, I hesitate to post until I do.

Ok, this has been too long, and I've made comments to various writers. I am encouraged by thoughtful comments. I appreciate level-headed debate.

Regards,
Pletch

 

[Zonie]

Leatherbark
IMO, all things being equal as you suggest, in theory the results would be exactly the same as those shown by Pletch's experiment.

In reality, there could be some differences if two different guns were used because there are no two "identical" guns. The sizes and roughness of the bore, the depth and twist of the rifling, the location of the vent or drum...., all of these things could have some effect.

That is why the method Pletch used is the best approach. It rules out unseen variables.
This gives a totally fair comparison between the Flintlock and Percussion system.

 

[Hawken12]

Hi Pletch I appreciate the work you are doing and you stick to the facts as you know them,without assumptions (Just the facts Ma'am)and you answer with care,thankyou .
Paul ,you write alot ,I also assume you have shot alot,still thru out this exercise you enter a caveat ,its not me ,its what others see\hear .
Seems simple to me Pletch is gonna be in Friendship ,Paul is gonna be in Friendship,Pletch is gonna have his equip there ,bring yer lock an lets see what ya got there ?

 

[Larry Pletcher]

What if one took two identical rifles. One flint and the other percussion and matched the powder charges to get the exact same muzzle velocity and find a way to time when the hammer is tripped and when the ball hits the target at a certain yardage...........Which ever one hits the target first is the fastest.......................Bob

Thats pretty much what we did - except for the travel time to the target. We used the same stock and barrel. The trigger was eliminated - the plunger pushed on the sear itself. Locks were substituted into the same lock mortice. The photo cell stopped time at the muzzle - again the same for each lock. The only piece that changed between sessions was the lock and the necessary nipple and drum. Obviously no drum is needed with a mule ear. Extraordinary means were used to eliminate fouling as a variable.

Our flint trial comes next. Remember we did the flint because we couldn't wait. The next session will use 2 flintlocks a Chambers liner with a .064 diameter hole. Same barrel stock, load, etc. The only variable is the lock. We'll use the same load and cleaning methods as we did on the percussion trials. We'll see the difference caused by a stock lock and a highly modified lock.

You might wonder why we chose to use a .064 vent. Because of our desire to eliminate fouling as a variable, we chose .064" so we could use a pipe cleaner as a cleaning step.

Regards,
Pletch

 

But are the differences statistically significant? Not according to my statistics program. The T-test has a P value of 0.1622 which indicates the difference is NOT significant (significance starts at 0.05).
Mule ear - mean 0.01367 SD 0.001147
Siler - mean 0.01147 SD 0.004033

I agree completely. Quoting from my earlier post: "It is our opinion that there is no advantage in the use of this mule ear lock over this side hammer lock on this day- as we would have expected. We also understand that individual locks differ in the strength of springs, so we consider this simply a test of one lock vs another."

If this write-up would have been for an article I would have called it statistically insignficant. The surprise here is that there is no significance difference between between the 2 percussion locks. I think most of us would have expected the mule ear to be significantly faster. I still wonder about its extremely short throw???

Regards,
Pletch

Larry,

You should probably double your sample size before you call it statistically insignificant, however I would say the most significant result is the larger standard of error on the Siler. Increased variation is a killer. That said, I'm still surprised that the Siler is even in the ball park with the mulie. Did you take into consideration the length of the hammer/cock travel?

Sean

 

[flinthead]

Thanks for posting your results from the testing. I found it very interesting!

Great posts.

 

[SAJones]

Doesn't surprise me much. As a flintlock shooter, I've had almost the same lock times as caplocks, but seems there is an almost imperceptible delay (Sometimes more than "almost imperceptible", but that's s different problem). Since the theory goes that the heat of the pan flash sets off the main charge, perhaps the time to heat up the main charge would be longer than the explosive flash of a cap which would directly ignite the main charge?

As far as White Lightening vent liners, I've installed two, and noticed no improvement over my modifed (coned) liners. In fact, one is more susceptible to fouling than any other vent liner I have installed. This vent is installed so it is about one radii from the breechface. With all the extra work & special taps required, I will no longer use the White Lightening liners.

 

[Larry Pletcher]

Larry,
You should probably double your sample size before you call it statistically insignificant, however I would say the most significant result is the larger standard of error on the Siler. Increased variation is a killer. That said, I'm still surprised that the Siler is even in the ball park with the mulie. Did you take into consideration the length of the hammer/cock travel?

Sean

Sean,
On the flintlock timing I normally calculate the SD so I get a sense of the consistency of the lock. I also did is with the vent experiments. I didn't run SD on this set because we weren't finished until we timed both flintlocks. I normally would not release results half way through the complete test. I should have waited here until we were done with all 4 locks. Then we'd have 4 locks, side by side, with fast time,slow time, average, and SD.

As you noticed, the Siler was a surprise. Maybe it's surprisingly fast; maybe the mule ear was slower than expected; or maybe we're been fooled all this time thinking that mule ear systems are automatically faster. As you note I qualified our results because while I stand by the numbers, I don't know cause.

Regards,
Pletch

 

[musketman]

What if one took two identical rifles. One flint and the other percussion and matched the powder charges to get the exact same muzzle velocity and find a way to time when the hammer is tripped and when the ball hits the target at a certain yardage...........Which ever one hits the target first is the fastest.......................Bob

The amount of primer used in the flintlock matters too, too much and it will make a "fuse" effect. In my opinion, to be truly equal in the test one would have to use as much priming powder as the mixture found in a percussion cap.

 

[Leonredbeard]

Sorry, Stophel and Mike,
Larry's high Speed video of my wheellock shows that it is not the fastest. The original Manton beat it by quite a bit.
Yes the wheel spins very fast, but the puny sparks from the pyrite took a loooong time to light the powder.
The speed of a wheellock is one of perception. The small wheel spins on the center of its axis. It makes almost no indication of movement by feel. Try your wheellock in your gonne without prime and load and then do the same with ANY flinter. The difference in feel is astonishing.
The flinter has two large pieces of steel spinning from one end and slamming down to a stop.
This gets your attention whether you have tuned it out of conscious perception or not. On a wheellock there is not enough indication that any action is taking place. You pull the trigger and the gonne fires. I seems fast but it is not.
Go to Larry's website and watch my wheellock in action.
volatpluvia

 

[Mike Brooks]

Go to Larry's website and watch my wheellock in action.

I'll do that. :thumbsup: Well, even though you busted my bubble they are still the funnest gun ever to shoot!

 

[Dphar]

ignition is not the only factor in the speed of the gun going off. Ignition is part of the overall process. All of the various things that speed the gun going off must align for the relative types of "speed" we are considering.

For instance, take a percussion with a drum and a patent breech, like some of the CVA rifles. The fire starts in the nipple, ignites the powder in the 1/8 or 3/16 channel in the drum, then burns into the base of the patent breech, which is 3/8 of an inch diameter and then the main charge in the 50 caliber barrel. It has to burn it's way through a twisted angled funnel for the gun to go off. If this takes .07 seconds, how much slower is it than a direct feed into the main charge without all the angles and channels.



Good patent breeches do not have a "twisted funnel"



On the other hand an underhammer with a percussion nipple (or mule ear) with a nipple screwed right into the barrel with an old fashioned flat breech plug, in theory should go off faster. The fire is introduced directly to the main charge.

But then there are other factors. Does the hammer fall squarely on the nipple to give best "seal" to the system?
Does the more tightly packed powder against the nipple face in an underhammer or mule ear slightly slow ignition where as the looser powder in a drum of patent breech flash channel speed ignition? The powder in the flash channel being loose propagating the flame faster? This was thought to be the case historically. How it would test with modern equipment we do not know and I am not sure how to even test this with sufficient repeatability shot to shot.
It is thought that drop tubing powder into a barrel or cartridge case slows the ingnition cycle somewhat as opposed to loose powder which it wouls seem would allow faster initial flame spread through the charge and a less uniform burn from shot to shot. I.E. the drop tubed powder is more uniformly packed and gives a more consistent burn from shot to shot.
It is possible that the packed powder burns faster but still more uniformly needs more research but I have found some information as I was typing this but it needs to be studied before further comment.


That being said, I once heard of a fellow that intentionally placed the nipple on an underhammer approx 1 inch forward of the breech face so his powder charge columne would be ignited in the middle. So instead of burning from breech forward, it would burn both ways from the middle of the powder charge. I also wonder what that would do to speed of powder consumption and pressure. For instance, would you get a dangerous pressure spike? could you get more velocity?

This would require testing. But most rifles historically were not made in this manner and the old time target shooters were, if anything, more serious than todays since they often shot for significant quantities of money they had wagered. $1000 or more in the 1870s and 80s was a lot of money.

Mr Pletch has done a magnificent job in his experiments. I always have that extra, what if question. For instance his experiments on priming close to or away from the pan. What kind of vent did he use, would it make a difference had he used a flint falling on a frizzen creating several sparks instead of just a hot wire at one point. I had always heard that a prime should be slightly more toward the side away from the vent hole because the sparks of the flash powder burn up and away, not around corners so the powder needs to back against the far wall of the pan to throw sparks into the vent hole. On large musket locks, a fence or gate was often used around the flash pan. Did that also help keep the pressure/fire and sparks to the vent hole? Could such a concept help on a smaller lock?



and that brings to mind another question, what ignites the charge, fire from the flash pan or the sparks thrown from the prime powder, or both. Would the presence and proximity of the bottom of the frizzen help direct pressure and fire sideways into the vent hole? (thereby possibly speeding ignition)

The timing was from priming powder ignition how its started and where might make a difference. But so long as it was done the same every shot the test is valid since it was testing relative vent speeds. The guard used in modern times to protect the shooter to the right. Its not likely to increase pressure in the pan though it could make it hotter I doubt its a factor.
While sparks going in the vent either from the frizzen or the igniting priming surely occurs at times I would hesitate to assume these are the primary source of ignition. The priming is very close to the charge especially with a good liner so radiant heat is as more likely a source of ignition as sparks from either source since it is the same every time and is less random. Thinking a pressure wave is entering the vent requires understanding that 1. the pressure is pretty low and 2. the vent will not accept a low pressure wave from entering since this would require compressing any air in the vent. Supposition but I doubt there is significant pressure intrusion into the vent/barrel of a loaded FL.





My theory is that when the powder burns, the pressure, sparks and fire created move in a 360 degree, three dimensional radius, unless prevented from doing so by hard surfaces. for instance the pan and it's edges cause the pressure and fire to go upward. Would the mere presence of the bottom of the frizzen act like a forward fence and help redirect that pressure and flame?

The other question, which Pletch's experiment (In the March Muzzle Blasts) seemed to answer was whether that pressure and flame runs around corners, such as from being next to the barrel and making the 90 degree shallow turn into the vent hole. I would think that the powder and charge in the barrel and vent channel would act slightly like a hard surface that would not give to pressure until ignited either by spark of flame. We've all had flashes in the pan. The pressure of the powder burning wasn't sufficient to ignite the powder. A flame being thrown out with that pressure, shot up and outward, and yet didn't ignite the powder. Is it that some minute spark must be thrown into that hole to make ignition faster? If we were to draw a three dimensional pan surface and then draw radius' out in every direction up and out from the shape of the pan, where would the most overlap of the radius be? And more particularly, where on the flat of the barrel would the heaviest concentration of spark or flame be located? In the March Muzzle Blast Article, the charred marks from sparks on the index cards, seems to show a darker tee-pee shape on the card, right about where the vent should be located. That tee-pee shape to my thinking comes from the sparks being thrown up and away from the edges of the pan.

Flashes in the pan generally, almost always, occur as a result of a clogged or dirty vent or too little powder in the pan. Murphy's law also intrudes. I have had one unexplained flash that I can recall it was a shot at a deer, I just figured that fate frowned on this deer dying at that time.
Having fouling around the vent seems to "insulate" the main charge. I think this is why many of the best locks have vent wipers on the pan cover. The flame still produces radiant heat and it takes only about 400 degrees to ignite BP. The burn temp of black is about 5 times this so it produces a pretty hefty dose of radiant heat.
However, if the paper I just started to read is correct, and I passed it on for confirmation by an expert, the ignition *may* be from cast off micro-globs of molten saltpeter.

Now that creates another question, if the flame is so directed, up and away from the pan surfaces, is there an optimum pan shape for even faster ignition? Most pans have rounded edges. even on the exterior side away from the barrel flat. Some have a more square edge to that far end of the pan. Does it make a difference. Most pans run perpendicular to the barrel and run nearly the full width of the bottom of the frizzen. Would the gun go off faster if the pan didn't extend so far from the barrel and had a slightly more bowled shape to concentrat the sparks at the vent hole? ie should the pan be shaped more like a half of a bowl, actually a parabolic shape would seem to be required.

I had a cva gun years back that had a shallow oval shaped flash pan. and it went off reliably, but very slowly and required that it be full of powder, or I would just get pan flashes. Obviously not the best shape.

It was probably a slow lock and a poor vent design. Buying a low cost factory made flintlock is not generally a good idea from the performance standpoint

Now cutting lock/ignition time by .005 seconds may not make a difference in a patent breech like TC guns use. The fire still has to negotiate that twisted funnel to ignite the main charge.

Dan

Reading the full text of the material I mentioned above revealed it was not granualted powder so the packed burn rate verses unpacked is not relevant to the discussion