The Ballista

20100309093824(1) This video shows an attempt to hit a concrete paving block.   Stats are:   draw length 45″, draw weight 5000 lbs, bolt weight 6985 grains,  velocity 314.8 fps, energy 1537 foot lbs.

The chronograph was checked with 3 shots from my .25 cal.  Kodiak air rifle and all the readings were in the normal range for that particular piece of artillery.  Today’s velocities seem well founded.  All of which leads me to conclude that the 401 fps reported the other day is a highly improbable number.   Grovel, grovel….retract, retract,….. sorry about that.

I have not sighted Firefly in yet.  There is not much point in doing that until she has several more shots under her belt and the bundles settle down.  In the meantime we can have some fun blowing up bolts on the 1/4″ steel plate.  Who knows, I may get lucky and actually hit one of the paving blocks.

Here are three more videos of some admittedly wild shooting.  Below them are photos of the damage done to the target.  All the bolts are very close to 7000 grains.  Range is 53 yards.

20100309103750(1) 312.8 fps,  1520 fpe.

—————————————————————————————————————————————————

20100309104353(1) 280.5 fps,  1223 fpe.  Probably slower because the fins were oversize on this one.   It looks a bit  wimpy because of the way the bolt bounced out,  but check out what it left behind in the 1/4″ plate.  Penetration is 3″.

—————————————————————————————————————————————————

20100309124009(1) 307.7 fps,  1471 fpe.    Ouch!  my trusty old utility trailer didn’t like that.

—————————————————————————————————————————————————

Clearly I need to start working on sighting in Firefly.  No luck on hitting one of those pesky concrete blocks yet.  Tomorrow Moriary!  tomorrow! ……

I awoke this morning with a healthy dose of skepticism brewing about that 401.1 fps reading from the chronograph.  To make matters worse, the Canadians had decided to grace us with one of their arctic blasts and the snow was coming in sideways for a couple of hours.  By mid-morning the sun made a begrudging appearance and the snow stopped.   It looked like a good time to try another shot and see if yesterday’s “record” velocity was even plausible.  Unfortunately the serving on the string was in disrepair  and after the shot we observed the situation seen below.

A few days ago ago I had one of my imaginary  Roman busy-bodies sitting on my shoulder warning me to take care of the string serving.  But did I listen…..

Probably this blow out of the serving happened during the power stroke because the velocity was down to 270.9  fps.  I am fairly certain this number from the chronograph was accurate because I had made a point of shooting my Kodiak air rifle through the screens several times before making the ballista shot.  The readings for the air rifle were a consistent 648 to 654 fps.  This is perfectly normal for this particular .25 cal air gun and gave me confidence the chronograph was working accurately.  The 270.0 fps on Firefly was disappointing of course, but at the moment I am considering it the result of the blown out serving.  One correction from yesterday  is for certain,  the draw weight for that supposed 401 fps shot was actually 5000 lbs,  not 4500 lbs as reported.  Apparently the zeroing dial on the dynanometer had been inadvertently rotated because it was reading 500 lbs when it should have been set at zero.

The target seen in the following video is in shadow,  and it is hard to make out all the action.  What we are seeing is today’s 270.9 fps shot, with its slow poke 7000 grain bolt and 1140 foot lbs, take on the 1/4″ mild steel plate that is installed on the rear of my sand trap.  Range is 53 yards.    20100308134735(1)

If you play this video on slow it seems as if the bolt is moving through the air at an angle with the tip down.  I am not sure what might be causing this.  Is it because the 2870 grain tip is too heavy for this length bolt and the whole projectile trundles along nose down?  Perhaps it is because the bowstring gives the tail of the bolt an upward flick at the end of the power stroke.   Maybe it is some perturbation coming from the damaged serving.  At the moment I have no explanation.

This quadrobate tip is made from hardened 4140 (about 40 RC), and penetrated the 1/4″ steel plate to a depth of 2 1/2″.  The tip also went through the piece of 3/4″ plywood behind the plate and stuck out the other side  1  1/4″  into the sand.  The photo below shows how the bolt tipped upwards at the moment of impact.  Not surprising because any slight downward  angle of the bolt would probably be magnified into a full blown spinning upset as the rear of bolt was forced upwards by the impact.

The one way trip experienced by this bolt must have been excruciating.  Note how the explosive deceleration caused the fins to slip their grooves and migrate forward.  I hope to be able to try another shot tomorrow morning when the light is coming in from the East.  (Assuming of course we can get the Canadians to cooperate.)  With a new serving on the bowstring,  we should be able to answer whether that 401 fps shot was for real, or just a figment of my chronograph’s imagination.  I suspect the later.

Here is a quick vid in real time of the terminal results of today’s shot.  The chalk circle is where my esoteric calculations had predicted the bolt would exit.   Range is 50 yards.    20100307132756(1) ………….. So much for my powers of divination.

As for the data from this shot, I find myself hesitant to post today’s results until I can get some independent confirmation of my what my eyeballs are seeing on the chronograph.  However, it has always been customary for me to post any data the same day it is gathered, and so I can see no reason not to post the following numbers:  bolt weight 6996 grains,  bolt length 33 1/2″,  draw length 45″,  draw weight 4500 lbs.,  ……….. now here comes the part that is rocking my world…….  velocity 401.8 fps,  energy 2507 foot lbs. (3400 Joules).   There isn’t time to attempt a duplicate of this shot today.  Also,  I am going to need to get another chronograph to verify the instrumentation.   This level of performance (if it can be positively repeated) puts Firefly’s power potential far beyond what I ever dreamed possible.  All my efforts must now be directed to confirm or debunk this data.  Naturally if some kind of chronograph error is generating these results,  I will make an immediate and grovelling retraction.

The draw length was increased from 40 1/4″ on Friday, to 45″ today.  Perhaps that explains the possible jump in performance.  Past experience with the Gallwey ballista has shown that it takes a dozen or so shots after the washers have been rotated to a new station,  for the bundles to settle down and start producing consistent velocities.   That the draw weight was 200 lbs. less than Friday,  even though today’s draw length was longer, is a natural consequence of the bundles becoming accustomed to their work and losing some of their initial vitality.  That we should see such a large increase in velocity does leave me a tad skeptical.  On the other hand, the shot certainly appeared fast.

The new larger fins on the bolt resolved all the wobble issues we observed on the previous shots.  Here is a photo of what the bolt looked like before heading out on its suicide mission down to the dreaded archery backstop.

I sure hope today’s results are for real.  If they are not,  I fully expect a severe spanking from the Catapult Gods.

A quick addendum for the day:  somewhere I found the energy to assemble another bolt and try for another world land speed record.  Stats are:  bolt weight 7515 grains, draw length 45″, draw weight 4700 lbs, velocity 351 fps, energy 2055 foot lbs. (2768 Joules).  The velocity was down on this shot partly because the bolt used was 519 grains heavier.  The energy was also down from the last shot by 452 foot lbs.  I do not have an explanation for this phenomena.  At this point all I can do is be rigorously methodical and find a way to verify what the chronograph is telling us.

Here is a real time video of today’s test recorded by using the new radio control unit for the camera.  It was able to trigger the camera to capture the bolt plunging through the same target as yesterday.  The only purpose of this remote set up is to save on the wear, tear, and possible perforation of the cameraman.   Like yesterday, range is 50 yards.  20100306160111(1)

Before its saunter downrange,  here is what today’s 33 1/2 ” long bolt looked like.

After the bolt’s momentary disagreement with the archery backstop,  here’s how the fins faired.

Velocity of this shot was down from yesterday as the bundles are still normalizing themselves to the new torque setting.  Stats are:  draw weight 4700 lbs., draw length 42″, bolt weight 6929 grains, velocity 310.6 fps,   energy 1484  foot lbs.

Some disturbance in the bolt flight was still visible.  The tale end appeared to spiral around the central axis a bit.  Tomorrow we will try bigger fins.  How retro.

After much flopping around these last few days,  Firefly was finally made ready for some high energy shooting this afternoon.   Two shots were made before the light faded.   The stats on the first shot are:  bolt weight 6991 grains, draw length 38″, draw weight 4500 lbs., velocity 333.1 fps, energy 1722 foot lbs.    Specs on the second shot with the same bolt, are:  draw length 40″, draw weight 4700 lbs., velocity 337.9 fps, energy 1772 foot lbs.

Here is a video in real time of the second shot.  Range is 50 yards.    20100305173642

Yesterday the bundles had been loosened and then torqued back up to a level higher than I actually intended.  This became immediately apparent as the draw weight went up much faster than usual as the bowstring was being pulled back.   As can be seen in the video, the limbs had about another 20 degrees to go before  they would have been perpendicular to the stock.  In other words, these results were obtained with only about two thirds draw!  Given the high draw weights I decided not to chance pulling it back any further.   By sheer good luck,  the bundles turned out to be perfectly balanced even though they were taken to a higher torque level than they ever had before.

The bolt used here has a blunt tip 3/4″ across.  It punched through the top of my compound archery target (about 3″ of dense foam) on through 24″ of straw bale, and then blew through the 1 1/2″  fir on the backside, apparently stopped only by the tough plastic fins acting like a brake as they tried to squeeze through the hole in the wood.  Here are a couple of photos of the target.

The important advance seen with today’s shots is the huge jump in projectile energy.  The best Firefly has  managed before is around 1000 foot lbs.  Now she is churning out 1772 foot lbs.  The best my old Gallwey ballista could do was around 1150 foot pounds.   The inherent power of the inswinger design is finally beginning to surface.  In truth, I am still reeling from this dramatic increase in performance.  As I have long suspected, serious power with these machines is largely governed by introducing very high levels of torque into the bundles.  How much is a very high level of torque?  That is not such an easy thing to measure accurately because of the enormous amount of friction between the  washer and vernier plate as the twist is being applied.  My attempts to measure the force being applied to the tightening wrench for the washer,  have largely been a failure because the  reading on the dynanometer appears as an indeterminate spike only when the washer is actually moving on to its next station.  Suffice it to say, the torque now being applied to the bundles in the tuning process is a lot more than anything I have attempted before.  It seems wise to stay at this level and not push things too far right now.

The fins on the bolts I’m using probably need to be a bit bigger to counter a slight wobble in the bolt flight.   It may help to shorten the bolt by a couple of inches.  This will effectively increase the spine of the shaft and past experience has shown that this can help stabilize the bolt as well.

I could not be happier with today’s results.    Sure hope the Catapult Gods are pleased with all my labors.   Very tired,  must sleep now.

Goofing around at the forge this afternoon, I knocked out the quadrobate features on this hardened bolt head.

With this 2870 grain business end hooked up to to a 28″ Bamwood shaft, the finished projectile should come in at around 7400 grains.   Now to find a target that will be suitably impressed.

Here is a super quick vid of a cobbled together,  long range remote set up for my Sony camcorder.     20100228230157

In the immortal words of Mr. Spock,  “Crude, but effective.”   It can trip the slow smooth record feature from at least 100 meters away.  That gives a full four seconds of slow motion recording time to capture the proceedings downrange.

Hurry up silly little winch part.  My trigger finger is getting a mite itchy.

Firefly is starting to paw the ground and snort for the race to be on.  Unfortunately her Warn winch needs a replacement shaft coupling, and that won’t be arriving until Wednesday.   No matter, there are always more bolts to build.  Our intrepid race horse of a catapult could also do with a new bowstring.  The current one is not as good as it should be.   I wouldn’t want some Roman busy-body to get all mil-spec on me about it.

An order of  broom handles arrived today.  I have ten of them 60″ long,  1 1/8″ diameter.  That’s enough to make 20 bolts for the low, low price of  $1.22 each.  Even better, they appear to be made from laminated bamboo,  Bamwood I think they call it.  It’s mostly very straight and appears easy to straighten when it is not.  Fooling around bending  it in the vise, showed it to be plenty tough.   This is really good news  ’cause I was kinda sweating what I’d use for shafting.

Also, I dug out a new-in-the-box  remote control set up for model airplanes.  By using it to control a  servo poised over the start button for the remote used with my camcorder,  I’m planning to get some good downrange footage of the terminal ballistics on these Bamwood bolts.  (That’s fancy talk for videoing them smash into things, up close and personal)

I made up 6 bolts today,  26″ long and varying in weight  from 6500  to 8379 grains.  Painting them day-glow orange seems to help visibility for the video.  I only managed one shot before the light turned sour. Here is some footage of that shot, slowed down 4 times.   20100223153553

The slow motion flight of the bolt is visible for 70 yards and a small spurt of white can be seen as it burrows  into a  patch of snow,  just outside the veggie garden.  Also,  this video clearly shows the bolt flying wide to the left.  More torque is needed on the port side bundle.   I will try and get more video as the tuning proceeds,  and  then we can see the shots finally straighten out.  Stats on this shot as follows: weight of bolt 8379 grains;   draw weight 3700 lbs; draw length 49″;  additional 7 1/2 degrees added to the top washer starboard side;  velocity 233 fps;  energy 1010 foot pounds.  I plan on getting the bundles balanced and the shots flying true, before applying the higher levels of torque to the bundles.   Nap now. ………

………Wake now.  Plonking in at 8379 grains,  the little bug in the above photo is going to be the fat Albert in the crowd.

However, it is nothing compared to this original beauty I discovered on ebay last evening.  Makes a chap think of hardwood sabots, it does.

…. and so anyway, we got off one shot today with the great big heavy harpoon of a bolt we made yesterday.    Before it exploded,  yawing through the plywood on the sand trap, it showed us the following:  10054 grain weight,  208 feet per second, 965 foot pounds of energy.   Slow because of its weight, but not wimpy exactly. In gun world, that’s about the same energy as a .44 magnum.  The good thing is that I only increased washer rotation 15 degrees beyond the tepid twist we had in the last round of testing.   In other words, there is a lot more velocity left to harvest as we continue to increase the torque.  We will have to stop when the bundles  max out with unacceptably high draw weights.  Today’s  draw weight was 3500 lbs.   What is an unacceptably high draw weight?   Now, that is a good question.  Also a dangerous one if we get overconfident.  It is time for some new safety protocols.  More on this later.

Below we see the broken bolt from today’s test laid out on the machine.

This Nessie of a projectile yawed into the target sideways because the starboard bundle was torqued a tad tighter than the port one.   This  shows up in bolt flight as the bolt swooping up and to the left,  relative to the bolt groove.   Confirmation of this disparity between the bundles was evidenced by measurements taken with the new Wheatie meter.  This Mk. 2  model is no longer paunch powered, but has its very own yellow strap and ratchet.  The spring balance replaces the calibrated bruising meter I have packed around for years.  In this photo,  200 lbs of pull  is being applied to the starboard limb to get it to lift away from the stanchion.  By selecting the correct size pin gauge to go into the gap that is formed between the limb and the stanchion,  a relative comparison can be made of the torque in each bundle.  For example,  from the above set up we found that when 200 lbs was applied to the starboard limb, it lifted away from the stanchion .240″.  On the port side it lifted .264″.  Hence the starboard side is more highly torqued.  The closer we get to balancing the bundles, the closer we get to making this baby sing.

My dalliance with super heavy bolts will have to wait a bit.   I happen to have a bunch of 1 1/8″ broom handle material.  (Thanks Richard.)   It is nice and straight and should make good shafting for a dozen or so bolts in the 6,000 grain class.    That is tomorrow’s task.