Flint vs Percussion testing

[Larry Pletcher]

Steve and I completed the testing that was meant to accompany the percussion tests we did this past May.  At that time we reported the results of timing the small Siler percussion and the mule ear with the same lock plate.  We now can add the results of the two small Siler flintlocks.  Each lock was lubed, and a new flint was installed.  A Chambers 5/16 liner was installed in the barrel. The photos shows the fixture with a flintlock installed and ready.  (You may not be able to see, but the liners used in these tests were equipped with two small holes drilled in the surface so that a spanner wrench can be used to do rapid changes.  This idea came from Dan Pharris who helped in the earlier breech tests.)

The methods used remain the same.  To summarize, a contact is placed below the sear so that when the solenoid strikes the sear, it closes the contact and begins the computer’s timing routine.  A photo cell located at the muzzle stops time.  This means that the time we measure includes:
a.  lock ignition
b.  barrel ignition
c.  barrel travel time to the muzzle
The only variable that is changed here is the lock.  

As before the test includes 15 trials for each lock with averages calculated.  The chart below shows the results.  I included the results from the percussion locks to make comparisons easier.

Mule Ear-----------S Siler Perc. ----------stock Siler Flint ---------Mod. Siler Flint
.010--------------------.011 ---------------------.068 ---------------------.072
.014--------------------.012 ---------------------.068 ---------------------.066
.010--------------------.010 ---------------------.059 ---------------------.068
.016--------------------.010--------------------- .060 ---------------------.072
.010--------------------.010 ---------------------.071 ---------------------.072
.011--------------------.010 ---------------------.066 ---------------------.071
.009--------------------.012 ---------------------.069 ---------------------.078*
.019--------------------.010 ---------------------.062 ---------------------.044**
.023--------------------.008 ---------------------.059 ---------------------.064
.009--------------------.012 ---------------------.086*--------------------.065
.013--------------------.025 ---------------------.071 ---------------------.056
.019--------------------.013 ---------------------.069 ---------------------.075
.016--------------------.009 ---------------------.090*--------------------.062
.010--------------------.012 ---------------------.076 ---------------------.070*
.016--------------------.008 ---------------------.073 ---------------------.071
------------------------------------------------------------------------------------
.0137------------------.0115 ---------------------.0698-------------------.0671 (avg)

* --- Indicate knapping flint
** --- In this trial the photo cell may have triggered on the pan flash instead of the muzzle.  The time is a reasonable for the lock by itself.
 
Temperatures during the percussion test varied from 56-60 degrees, and humidity dropped from 60 to 42% during the testing.  Yesterday during the flint tests, the temperature was 68 degrees and the humidity 60%.

Drawing conclusions remains somewhat sketchy at best.   A few things need to be mentioned first.  The data was drawn from only one of each type of lock.  Would a different mule ear, or a different stock small Siler, or a different modified Siler have changed the results; I don’t know.  So we must qualify any conclusion by admitting our data comes from only 4 locks.  With that in mind, a couple of things stand out in my mind.

First the modified flint Siler’s performance was only marginally better than the stock Siler.  I expected a larger difference.  Perhaps bells and whistles don’t actually improve the actual numbers but only our impressions.  We think it must be faster because it has a roller frizzen, for instance.  

The mule ear lock was not faster in this test than the side-hammer lock.  This was a surprise to both Steve and me.  We expected the short throw to mean faster times; maybe the throw was so short there was no distance to accelerate.  (The spring was VERY stiff.  More thinking is needed here.  Can a mule ear’s throw be so short that it doesn’t have space to accelerate?)

At first glance, the difference between the percussion and the flint trials was larger than expected.  After thinking about earlier tests, these numbers seem typical.   Consider that times recorded for Siler locks in many previous tests run from .0370 to .0450.  My test bed Siler  was consistently under .0400.  We timed barrel ignitions a number of times between .036 to .046. ( I included the .046 for those who bank their prime away from the barrel.  If you prime next to the barrel .036-.038 is a better figure to use.)

If you add the lock time to the barrel time (.0370 + .036), our numbers seem to fit.  This addition makes me think that time measured from sear to muzzle in the .060 - .075 range are typical.  I might add that in adding these together we are including the pan flash twice:

<-------------sear to pan ignition--------->
                                                      <-------------pan ignition to muzzle------>

Since I have to use a photocell to start and stop the time, I can't avoid including it in both times.

From the data we gathered and using the limited number of locks, it would be safe to say that percussion ignition is about 3-4 times faster than flint.  The numbers could be interpreted more tightly than that, but the variables involved with flint edges, when you choose to knap, etc, make me hesitate to do so.


Here is the overall equipment.  The Apple IIe and interface are in the background, with the fixture in front.


The fixture here has the flintlock installed.  Omitted here is the wire tie that was used to retain the barrel.    The pistol is rotated up on the steel pin for cleaning and loading.


Looking closely you may see the three holes in the liner.  The outside two are for the spanner wrench.


This shows the contact below the sear.  The solenoid closes the contact when the sear is triggered.  The other end of the contact is attached to the lock-retaining screw.


Here is the photo cell that triggers on the muzzle flash.


I hope this all makes sense.  I'll probably get this together from beginning to end on my web site.  I'm uncertain as whether to send this to MB.

Regards,
Pletch

[David Price]

Larry,

There has got to be a special place in heaven saved just for you.  The time that you spend testing to make shooting more understandable   for all of us is unbelievable .  We all appreciate it.

See you at the CLA.

[Canute Rex]

What David wrote goes double for me.

I am thinking further out, though. So, percussion ignition takes roughly 0.012 seconds and flint roughly 0.055 more than that. The question in my mind is, what is happening to the position of the muzzle during the first 0.012 seconds vs the next 0.055 seconds vs. the next 0.5 seconds after that?

We all move, but how much and how soon related to ignition? If most of the muzzle movement happens after 0.075 seconds from the trigger pull, or before the sear even releases, then the ignition time is less important. If that .012 to .067 second window is the "sweet spot" where we get most of the wobble, then it is most important.

Ok Larry, in all your vast amounts of spare time  , try this: Get a rifle that has exchangeable percussion and flint locks. Equip it with a pressure cell on the trigger, a sear trip sensor, a photo sensor on the pan, a photo sensor on the muzzle, and an accelerometer near the muzzle to measure barrel movement. Have someone shoot a number of shots offhand, both flint and percussion. Work up a chart of muzzle movement over time compared to trigger pull, sear release, pan flash, and the ball exiting the barrel.

That would give an idea of where in the ignition sequence most of the muzzle movement occurs. Or perhaps it is relatively uniform throughout.

I'd like that by next week.

But seriously, shooters gun builders, and lock makers would all benefit from an understanding of the timing of the human interaction with the mechanical system.

It is amazing that people can and do apply such high tech analysis to traditional firearms. People did high speed motion analysis on figure skaters and found out that some of their rock solid beliefs about technique were completely backwards, Now they jump higher and spin more. Larry debunked the "priming powder away from the vent" idea. What's next?

[hanshi]

Got me to thinking about this "wobble window" question.   I've found I shoot as well or better with my flinters than with my capguns.  I'm thinking it's the style of the rifle rather anything to do with ignition in my case.  My best off-hand rifle is the most muzzle heavy and difficult to hold for very long.  But I shoot my best off-hand groups with it, consistently.  For me, I don't think a few hundredths of a second have any effect.  I believe it must be the gun style/length, etc.  But then, I'm not a particularly good shot but I am SLOW! 

[northmn]

A couple of things.   First if you throw out extremes ther is no significant difference between the two Siler flintlocks.  They are the same.  I read once that a good English flintlock was a fast as a poorer percussion.  I have always questioned that because the flintlock has to kick open the frizzen, spark, have the sparks set off a priming powder and then have the primer set off the main charge through the touch hole for the duration of the flash.  The percussion has the same or smaller throw to hit the cap which explodes into the powder. The lock time should be shorter.  I also feel that the alternate locks like underhammers and mule ears may not have been popular due to no significant increase in performance.  One even wonders about the new inlines?  Bench shooters theorized that they could improve with alternate ignitions, but if the better shooters think that way and use them the scores will be better becasue they are better shooters.   One thing I have a theory about in shooting flintlocks (and also longbows) is that the human reaction time may be such that between trigger pull and ignition there may be a recompensation such that the rifle is put on target.  Howard Hill claimed that the recurve was "too fast" for him.  I think he could actually move the bow on target after release.  Whats to say flint shooters after a lot of shooting do not do the same thing? 

DP

[Larry Pletcher]

I have been away from a computer for a time and should have answered sooner.  Thank you all for your comments and kind words.
   
David - you have been more than kind with your comments and support.  Thank you.

Canute -  I can't grind out your request in an afternoon.  Actually I can't do it at all with my current setup.  I can only deal with two inputs.  With the elaborate setups in real labs the multiple input experiment wouldn't be difficult.  My spectuation would be that again the human would be the  huge uncontrollable variable.  And, each volunteer would have his own unique, but measurable reaction in the testing.

Flintr - a couple of my friends believe that shooting a flint gun makes them better perc. shooters.  They're perception and yours may be true especially for those who shoot flint more than percussion.

Northmn - I certainly agree with you about the difference between the 2 flint locks.  When we ran our numbers we did in fact throw out the highs and lows to see how they compared with the reported averages.   The problem is that the owner of the modified lock expected better performance; I thought so too.  In fact the difference was insignificant.  He paid a whole lot more for questional improvements in performance.

This brings me to a point that I have wondered about for a long time.  Until this type of testing was done, any perception we had about have been based ONLY on human senses.  Sometimes we wanted one type to be better than another
- my mule ear is faster than your side hammer
- my flint lock is faster than than your lock
- roller frizzens are faster
- my barrel ignites before the flint gets to the bottom of the frizzen.

None of these are detectable by human senses.  The variable alone are bigger than the generalization made.   When we want it to be faster, our mind tells us it is.

An example that came from this spring slow motion shows another oddity.
By counting frames (at 5000 fps) we could tell that a lock with a roller frizzen was third out of 10 in mechanical time but tenth out of 10 in total ignition time.  So fast mechanical time does not always produce the best ignition time.  The more we study this stuff, the more strange things we find.

Any way - just when you thought you had it figured out ---

Regards,
Pletch
 

[Dan]

Thanks Larry.  The community owes you for such endeavors.  All I can do is offer free grouper or shrimp if you wander down this way. Goes well with beer I've heard.

[Daryl]

Thanks Larry - always good stuff - above and beyond the call of duty!

I-too was amazed at the mule-ear times. Too - I'd always thought the underhammers were faster than drum or bolster-type cap guns and that this was the mail reason target shooters gravitated to them in the past and including modern-made bench BP guns. Perhaps it's more the simplicity aspect with heavy barrels and not a speed deal.

One thing I've noticed, is my 14 bore seems to have about the same ignition time as my Ruger #1's have had. As you know, the #1's are about the slowest of modern action types - perhaps somewhat faster than a Sharps or rolling block, though.  Of course, this boils down to human perception, but there is a difference in speed of ignition between the #1's and a Mauser just as a Remington is faster than a Mauser & a Martini is faster than a Remington.  Shoot enough and it's noticeable. This speed aspect of the 14 bore is one of the prime reasons I think it's so easy to shoot accurately at 100 meters & beyond.  Even the Musketoon seems easier to bench accurately than either of my flinters.

[northmn]

I think one advantage of the roller in the frizzen is that it does make for a more reliable kick for opening when hit.    One of the things done to improve speed was a stirrup and a shorter throw spring, commonly seen on percussions, whether that does increase speeds is now questonable.  If one takes the fastest times for the flintlock we still see more than three times the speed of a percussion.  Just from this data alone I would be comfortable with any conslusions that percussions are faster even some of the slower ones.  The bench shooters always claimed faster speed for the underhammers, inlines and mule ears in their writings (custom inlines have been around before the modern ML craze).  If Joe Blow wins with something based on a theory like that it becomes fact.  As I was told once.  Granddad really wasn't stupid.  He used what he used for a reason.

DP

[hanshi]

I had always thought my under hammer was faster than most any side lock.  Eye opening to discover this is not necessarily so.  There seems to be so much more to the story.  Though I could never discern the difference, it just seemed to me a caplock would have to be a little faster than a flint lock.  How surprising to find out it was THAT much faster.  Great experimentation Mr. Pletcher.

[powderman]

Not to change the subject.

Just curious, when testing antique style firearm mechanisms,
is it necessary to also use antique style computers?
I'm not sure but the computer in the pictures looks to me
like a very old, original Apple. Complete with 5-1/4" external disk drives!
I go to computer swap meets and haven't even seen them there in years.
That may be a collectors item.

What's the story there? 

[jmforge]

Not to change the subject.

Just curious, when testing antique style firearm mechanisms,
is it necessary to also use antique style computers?
I'm not sure but the computer in the pictures looks to me
like a very old, original Apple. Complete with 5-1/4" external disk drives!
I go to computer swap meets and haven't even seen them there in years.
That may be a collectors item.

What's the story there? 

I was thinking that perhaps Larry is taking this whole "traditional" thing a bit to seriously.   Good stuff, man.  Now here is a strange thought.  Might the heavy barreled underhammer target guns be more accurate in theory not because the action type is inherently faster, but because the combination of weight and mechanics of the action cause the rifle to move less during the ignition sequence?  Newton theorized long ago that for every action, there is an equal and opposite reaction, however minuscule.  That got me to thinking about how the movement of the lock translates into making the gun move during firing.  Would the momentum of a sidelock cause the gun to twist in tiny amount to the lock side and would this movement be more noticeable relatively speaking than the underhammer trying to "lift" the muzzle of a heavy rifle or an inline trying to "push" the muzzle forward and down?  The second question would be whether if this did occur, if it would even be measurable much less have any effect on accuracy?.

[Larry Pletcher]

Not to change the subject.

Just curious, when testing antique style firearm mechanisms,
is it necessary to also use antique style computers?
I'm not sure but the computer in the pictures looks to me
like a very old, original Apple. Complete with 5-1/4" external disk drives!
I go to computer swap meets and haven't even seen them there in years.
That may be a collectors item.

What's the story there? 

D,
You're right it is an Apple IIe - left over from my elementary school.  I'm still using it because of the interface sitting just to the right of the Apple.  It was made in a class run by a physics prof at a local college.  It was designed specifically for the Apple platform.  I wrote the software in Apple's BASIC and use a machine language program from Vernier.   

The problem with changing to a PC platform is replacing the interface and rewriting the software.  It's easier to keep this setup running than to change over.  I'm currently looking at a PC interface that may work, but it handles decimals only to 2 places -- not enough for timing locks.  I want at least 3 dec places and prefer 4 places rounded off to 3.  A beta copy of software  that will address this problem may be  available soon. 

Another ability that I'm having trouble findind is the ability to operate 2 relays a measured time interval between them - for instance fire a relay to start a lock and fire a relay firing a flash .003 sec later.  That  ability is handled by the Apple and interface perfectly.  Finding that in the PC world is tough. 

Regards,
Pletch

[Daryl]

Concerning gun movement prior to discharge - it is a real thing with lock mechanisms a lot faster than a cap lock or flinter. One merely has to bench the gun and dry fire it a few times, noting the sight movement during the parts movement.

  Gun movement during ignition was the reason there was a limit of about 22 to 24 pounds on 'speed' lock spring conversions for Mauser actions.  Keeping springs below 26 pounds increased lock speed, but kept movement to a minimum.

 Now- equate that to a flintlock mechanism - heavy flint, heavy cock and frizzen, lead instead of leather holding the flint in the jaws, heavy main spring, rebounding/bouncing frizzen prior to ignition taking place - all translates into movement during or just prior to ignition.  A short-throw underhammer, striking to the centre of the breech end of the barrel would have the least movement of any lock - perhaps.

[Leatherbelly]

  Does that beat my Commadore 64? You know,the one that the phone sits in it's special cradle,makes a bunch of spacey noises,and ZZap,your on line in just 20 minutes! Ahh,teknology! LOL!

[Larry Pletcher]

  Does that beat my Commadore 64? You know,the one that the phone sits in it's special cradle,makes a bunch of spacey noises,and ZZap,your on line in just 20 minutes! Ahh,teknology! LOL!

Leatherbelly,
Ya this thing is much faster - blazes away at 1 mhz.  The clock speed does allow me to causes delays between solenoids by thousandths of seconds.  Each idividual count in an empty for-next loop is 1/1000 second.  Thats what allows firing the lock and then firing a flash .003 sec later, (a fairly useful ability.)  If I had to use this for anything else I'd give up on computers.

Daryl,
I'm interested in the movements caused by locks.  We saw an unusual thing in the last slow motion video tapes.  This wouldn't have been noticed except for our attachment method to the fixture.  We used the lock retaining screw; we allowed the front of the lock plate to rest on a raised knob.  When the sear was tripped, and the cock fell, the lock plate did a wheelie - you could see it rise.  One could almost make the case that teh lock was coming up to meet the flint edge.  If you think of a drag car rotating around the rear axle, maybe the lock plate is rotating around the tumbler axle.  Out of a gun this is something you can see.  On the gun the differences in mass come into play.  I doubt if the falling cock causes a measurable muzzle rise.

Lynton McKenzie, the late great engraver, reported lock movement in a Bowling Green Seminar.  He said that he felt his Stoudenmayer lock had springs that were too harsh.  He made a milder set and saw increased scores.  I timed the lock with both sets of screws in the late '80s.  It was faster with the harsh springs and more consistent with the mild set.  I think Lynton was "in tune" enough with his gun to perceive this.

The comments above that discuss various lock solutions for bench guns beg a response.  A lock of any kind will cause some movement and a reaction of that movement.  The easiest way to "still" that movement is to increase the mass of the gun.  I doubt that the type of lock makes little difference when it is mounted on a 40-50 lb bench gun.  JMHO

Regards,
Pletch

[jmforge]

 I'm currently looking at a PC interface that may work,

SACRILEGE!!!!!!! Find one for an iMac, you traitor!!!!!

[Larry Pletcher]

 I'm currently looking at a PC interface that may work,

SACRILEGE!!!!!!! Find one for an iMac, you traitor!!!!!

Cuts to the quick doesn't it?  I was the last Mac person in my school.  When we went to PCs I kept the Mac server in my room and ran a small lab of Apples at the back of my room.  Finally I had to give in. 

I really don't care what my platform is if I can keep the abilities I have now.  The tech folks usually hickup when I talk about controlling 2 relays and determine the time interval between them.  That and 4 decimal places in timing are the most important.
Regards,
Pletch

[jmforge]

 I'm currently looking at a PC interface that may work,

SACRILEGE!!!!!!! Find one for an iMac, you traitor!!!!!

Cuts to the quick doesn't it?  I was the last Mac person in my school.  When we went to PCs I kept the Mac server in my room and ran a small lab of Apples at the back of my room.  Finally I had to give in. 

I really don't care what my platform is if I can keep the abilities I have now.  The tech folks usually hickup when I talk about controlling 2 relays and determine the time interval between them.  That and 4 decimal places in timing are the most important.
Regards,
Pletch

LOL...........Used PC's until the end of 2006, which was long after the rest of my family had switched over to Macs.  Since then, I have had 2 1/2 years of trouble free computing.  My experience with cell phones was even worse.  I went though 3 Palm Treo's in 15 months and an equal number of Motorolas in about the same time frame. All failed due to some hardware issue.  I now have had an iPhone for a year with nary a bobble.  I guess that I am an obnoxious convert kind of like an ex smoker.