In the story that I am writing (medieval fantasy) I want one of the factions to have access to a remote weapon capable of penetrating heavy armor (we are talking about steel plates and chainmail, gentlemen) but not a siege, that is to say , that a soldier could take it with him, was able to shoot several times with a single "charger" and was also quick to recharge.

If we go back to history, there was a weapon that soldiers wore and that was feared for its power: The crossbow. With metal bolts and a sufficiently large tensioning force, a crossbow could penetrate the plate armor and cause, in addition to great pain and probably bleeding, the incapacitation of the wounded soldier. There is a BUT, of course, and that is that while an archer can take between 3 to 5 seconds to "recharge" (as long as he has the arrows at hand) a crossbowman can throw himself away for 15 seconds if he is trained and is experienced at one minute or more if it is not (that is, if you give me the crossbow to me, for example) depending on the crossbow, and its tensioning power, it is even possible that it can not be tensioned manually, if not required of a kickstand that provides the user with the strength to place the rope in the tensioned position.

enter image description here

enter image description here

enter image description here

This makes it powerful, but also extraordinarily slow to load, which is why the crossbow troops were always covered or used for defense, between the walls of the castle.

Here is an example video, a boy who has built himself the crossbow. It begins to load it on the second 17 and is ready to fire at 33, but it is not stressed, nor in the middle of a battle ... also as you can see the rope is not as "hard" as it would be a real crossbow, which indicates It's not going to be that powerful, but it shows my point

Handmade crossbow

And here you can see a trained crossbowman using a kickstand on a crossbow that you can not tighten manually, notice how not only it takes several seconds between one bolt and another, but the third bolt chokes him a little because he can not Place the goat-foot correctly at the first attempt

trained crossbow man

We agree that the crossbows would gain much more mortality if maintaining precision, they could load faster; Looking on the internet you can see crossbows with loader, or crossbows of repetition (which are perhaps even more elaborate and laborious to handle than usual). This is where my doubt enters.

There are several types of rifles, including the lever:

Shotgun lever mechanism

In which as you can see in this model that I leave here above, the cartridge to shoot is introduced into the chamber with a lever movement, while discarding the one that has just been fired. So:

Would it be possible to use a similar mechanism for a crossbow with that movement to hook the slack rope and reassemble it? Could it be possible that with this movement the next arrow will also be loaded to shoot in position? We know that because that is exactly what it does in the rifle, but could I pick up a bolt from a dispensable magazine so that once the bolts were finished, you could discard the magazine, put another in position, and continue firing quickly? How would it look?

Without further ado, I hope I have not been too much lead and have aroused some curiosity in your minds. Thanks in advance for the attention.

Edit 1: I am well aware of the existence of the Chinese repeating crossbows, but they look pretty bulky and impractical to me, having the magazine in the face and having just one arm holding the whole thing with shaking pulse. I am looking for something that allows you to quickly cock the crossbow but also allow more precision. Also as said above this is a fantasy setting so even though in medieval times manufacturing was nearly impossible, this is different

  • 1
    $\begingroup$ Crossbows couldn't penetrate plate armor. youtube.com/watch?v=gwKNBB9Ww2E $\endgroup$ – sphennings Nov 11 '17 at 21:21
  • $\begingroup$ @sphennings Some corssbows couldn't penetrate plate armour. The heaviest man-portable arbalests could potentially, though they were bulky and took forever to reload with a crank mechanism. They were generally used by teams of two, with one shooter covered by the other with a giant shield. So more akin to a mortar or machine-gun team, in some ways. Those could never be portable enough, or with a fast enough reload mechanism to be used like shotguns, though. $\endgroup$ – Eth Nov 13 '17 at 16:17
  • $\begingroup$ @Eth Can you cite a source for your claim about penetrating plate armor? I'm unsure what you mean about the similarity between a wildlass spanned crossbow and a machine gun, or mortar. Did you just mean that they were fielded as a team instead of being an individual weapon? $\endgroup$ – sphennings Nov 13 '17 at 16:25
  • 1
    $\begingroup$ A friend of mine once competed in an SCA speed archery competition with a 70 lb pull crossbow named "baby". He came in second place to a lady using a long bow. The secret was that he was reclining in a chair so he could use his legs for the draw (and couls reach his beer). He also used the crossbow to empirically test effectiveness of chain versus mail. A 70 lb crossbow penetrated the plate, front and back, and then continued right through the hay bale behind the armor. Chain actually protected better, dissipating the force like Kevlar does. $\endgroup$ – pojo-guy May 26 '18 at 5:43
  • $\begingroup$ * SCA = Society for Creative Anacronism. They do medieval reenactments and reproductions. All of their gear is reproducible with medieval technology. $\endgroup$ – pojo-guy May 26 '18 at 5:47

10 Answers 10


Modern mechanical engineering + youtube = win.

Would you like to mount an arrow magazine to a bow for rapid fire? Sure.

A crossbow with an arrow magazine? No problem.

Medieval woodworkers could replicate this. Of course the bow wouldn't be compound, and the crossbow wouldn't be recurve, but the magazine mechanism is simple. It simply needed to be invented. If you are interested in insane weapons, I strongly encourage you to browse this guy's channel. The number of gaping holes in the car door he uses as a target should be a hint.

Could it be possible that with this movement the next arrow will also be loaded to shoot in position?

Yes, as shown in the videos, also the chinese repating crossbow.

once the bolts were finished, you could discard the magazine, put another in position, and continue firing quickly? How would it look?

Most likely not, disposable magazines wouldn't work in a medieval setting, because manufacturing things is expensive and time-consuming when you don't have roboticized factories, injection molding and metal stamping hydraulic presses. In a shotgun, you can insert new shells into the slot to replenish the magazine. This is a much better solution. You would refill your magazine as you go by inserting arrows into a hole or a slot.

Would it be possible to use a similar mechanism for a crossbow with that movement to hook the slack rope and reassemble it?


You want to apply as much force as you can to the string. If you cock it with one arm, you will only have one half the draw weight you would have if you used two arms to cock it, so it will be a wimpy crossbow. But if you watch the crossbow video above, it doesn't take much time to press the butt against your chest and use both hands to pull the string.

Now if you want to penetrate armor, then it is all about kinetic energy. A 7.62×39mm bullet has 2kJ kinetic energy, which is ENORMOUS. It is equivalent to the potential energy (mgh) of a 2kg brick falling down 100 meters onto your face. This does some damage.

I wondered about the power output of arms on a crank, which lead to this. This study has astonishingly high numbers, because it includes wheelchair basketball players, who have serious arms. These guys would need to crank 400W for 5-6 seconds at full sprint power to generate 2kJ. People with less athetic upper body would need 10-20s, and it would be strenuous. This is a bit moot however, as I doubt a spring that can hold 1-2kJ of energy and transfer it to an arrow could be manufactured with medieval tech.

Also the crossbow needs to accelerate the string and the springs in addition to the arrow, so a significant part of the energy is not transferred to the arrow.

TL/DR: googling crossbow kinetic energy gives figures in the 200 joules range, like a .22 bullet. This will hurt, but it won't shoot through plate armor...

But of course someone had to make a youtube video of medieval armor against medieval crossbow, then against a 1kJ air rifle...

  • $\begingroup$ I think the draw of the string and tension you are capable to give to it is what gives power, as seen above, the goat-foot allows you to draw with only one hand strings that you could not draw with both arms, due to the leverage, if you could get the same leverage while also loading the next bolt from the magazine like in the videos, you would get a very powerful weapon $\endgroup$ – Isa_YY Nov 12 '17 at 0:03
  • $\begingroup$ For a spring, stored energy is integral of force times displacement. So you can put 2x more energy with 2x the force OR same force but 2x displacement. However when you pull a bowstring, at first it is very easy, then it gets harder, and when it is completely pulled is where you have to pull the hardest (as with all springs, force is proportional to elongation). So if your arm can pull, say, 20kg, you will only use all your strength in the last bit of the pull. For the rets of the pull, you use less strength, so there is untapped potential.. $\endgroup$ – bobflux Nov 12 '17 at 9:36
  • 1
    $\begingroup$ The goat-foot uses variable leverage (it's quite a smart tool): at the beginning, when pulling is easy, it has low leverage, but the harder it gets, the more leverage it adds. An optimum goat's foot would exploit the full strength of your arm during all the displacement (not just at the end) so it is easy to calculate it has a potential x2 advantage in stored energy. It also gives you a much longer displacement, which also increases stored energy. But if you use a tougher spring and use a goat's foot with both hands instead of one, then you'd again have 2x more strength thus 2x more energy. $\endgroup$ – bobflux Nov 12 '17 at 9:41
  • $\begingroup$ The chinese repeating crossbows were both recurved and composite. Why do you think your crossbow couldn't be? $\endgroup$ – Rekesoft Nov 13 '17 at 10:26
  • $\begingroup$ My point is that it would be quite difficult to have variable leverage with a shotgun-style pull handle. But if you allow the mechanism to stick out like a goat's foot, then yes: i.pinimg.com/originals/57/21/87/… looks pretty good, but not similar to a shotgun... $\endgroup$ – bobflux Nov 13 '17 at 10:31

What you are looking for is a repeating crossbow (or a Zhuge crossbow) which was invented in ancient China and used in war.


It is lever action. Basically, a very strong soldier has to move the lever back and forth which both moves the next bolt in the magazine into firing position and pulls back the string.

I don't know this for a fact, but based on the fact that medieval European crossbows would have a belt-mounted hook that would allow a crossbowman to use all of his leg muscles to draw back the string on a crossbow, I am guessing that the Zhuge crossbow had a MUCH weaker pull than a European military crossbow would have had. On one hand, you gain the fast repeating firepower, on the other hand, you lose a lot of penetration since that has a direct correlation with the strength of the pull. If the crossbow was as hard to pull as a European crossbow, I doubt it would have worked with a simple lever.

This makes sense since the ancient Chinese didn't have a lot of enemies with significant, high-tech armor. They faced (for example) hordes of lightly armored horsemen. Nobody in Asia had developed the level of armor that was used in Europe, so the penetration was not as important.

In your case, I do believe you will face a trade-off between rapid fire and penetration. The more of one, the less of the other.

  • $\begingroup$ Also, the Greek Polybolos - although, this was larger more of a static turret-weapon rather than a hand-held weapon. $\endgroup$ – Chronocidal Apr 9 '20 at 15:36

To break down the more important part of the question, a mechanism to draw a crossbow with sufficient power to penetrate plate armour would imply an enormous amount of energy (typical crossbows could deliver @ 200 joules of energy, even firearms of the 1400's were capable of delivering 1000 joules of energy). This was, in fact, perfectly possible. Crossbows of increasing power and size were developed, with some requiring a ratchet and crank mechanism to draw the bow. While the knight receiving such a bolt would be rather unhappy (to say the least), the bowman would either need lots of friends, a pike square or the protection of a crenelated castle wall to effectively use such a bow without worrying about being impaled by the knight's friends while he is drawing the bow for the next shot.

enter image description here

Crossbow with cranking mechanism attached

What you are looking for is mechanical advantage, where you are not applying a lot of force or energy yourself, but some mechanism is doing the work for you.

The problem is you can't get something for nothing. A common example of mechanical advantage which demonstrates the principle is to think of a block and tackle pulley mechanism. Drawing a rope around a single fully simply angles the direction of your pull, but adding pulleys in fixed and moving configurations can provide mechanical advantage, allowing you to pull or lift even massive loads. However, in order to do so, you end up pulling a lot of rope (the various equations are at the link above)

enter image description here

various examples of block and tackle

Without knowing the size and draw of your bow, or the actual amount of time you want the action of drawing the bow to take place over, it isn't possible to put in numbers (but the equations at the link will allow you to do so, and you can play with the various factors). However, for a heavy steel bow with a high draw weight, a cursory look at the equations for levers and pulleys suggests that any mechanism capable of rapidly drawing a powerful bow will be improbably large and difficult to manipulate. A lever with a very long arm, a huge block and tackle arrangement or even some sort of weird gear train would be needed to generate sufficient mechanical advantage to rapidly draw the bow one handed.

enter image description here

A lever with high mechanical advantage

The reason a lever or pump action firearm can be cycled so quickly (and even bolt action rifles like Lee-Enfield's can be fired at a high rate with a trained user) is the action isn't cocking a giant tension arm like a bow, but manipulating a small steel block, and cocking a relatively small spring for the firing mechanism. The energy is in the chemical propellants in the cartridge itself.

  • $\begingroup$ I came here to address issue of energy and penetrating power. Nice to see it already was addressed, in a way better than I would. $\endgroup$ – Mołot Nov 12 '17 at 10:19

You can't have a tube magazine of a crossbow. The length of the bolts is long enough that it would need to stick out far in front of the bow for it to have any real capacity. This would make it far too cumbersome to be practical.

The closest real world designs to what you are considering were the repeating crossbows of China and Korea. They had a stacked magazine on top of the bow that dropped a bolt into place during a lever powered cocking process.


You want to know if it's possible, not how to do it - Say the missile shall have an energy of 1000 joules (like a rifle bullet). Energy put out has to be put in at some point, and we have no blackpowder to produce it for us. Energy is work, in this case, so we can put it as Force * Distance. Force is the force you bring to bear, and Distance is the distance for which you do so. So 1000Joules = 1000N * 1m = 100N * 10m ... etc. 10 N (Newton) is the force you need to lift a 1kg stone in earth gravity. Apply a force like lifting a 10kg stone for ten meters? We see that our goal was too high. Let's approach from the other side: Shotgun racking distance? Say 20cm = 0.2m. Force you can bring to bear in shotgun -racking position (from start-of-rack to end)? Say 100N - Work done: 0.2m * 100N = 20Joule - that makes for a hardcore BB-gun, but is not a weapon of war. We could pump more, maybe even utilise both directions, then every pump would get us 40 Joules - 25 pumps and we're at 1000Joules ... but this will take time.

This is not dependent on missile form (balls (which would be great for a nice magazine-design), arrows (great long-distance performance), disk (nice spread)) - a 1000 Joule aluminium BB will be hella fast, a 1000 Joule steel spike will be slower but that is all for the weapons designer to decide, the possibility of the weapon existing is given by Physics.

For your own ruminations: In easy linear cases (Force goes linearly from min to max along the distance) , we can use (min+max)/2 as Force). So a bow that needs practically no force at begin of draw, then over a draw of 1m is steadily needing more force to a max of 400N, can be thought of as having stored energy from Work = (0N+400N)/2 * 1m = 200J.

What we are calculating here is only the absolute maximum of energy that could be put out. The real value will usually be lower because of losses to friction et al. Even good springs have losses on the order of 5%. Calculate any mechanism you introduce as voiding about 5% of your stored energy. This accrues fast for complicated designs (ratchet, spring, trigger = 3 mechanisms...)

The only way around this is by using energy stored in advance: Pre-tensioned springs, black powder, pre-pressurized gas, pre-spun-up wheels, pre-fabbed magic, ...

If we are deep into the realm of fantasy-engineering, we could just look at the power needed: Energy = Power * Time so for instance 1000J is 2000Watt * 0.5 seconds; Human cycler's legs can output about 400W, so we only need 2.5 seconds to accrue 1000J - how that energy is transferred from the contraption on your legs to your handheld repeating crossbow is a question for the dwarves :-) - But how fantastic would a battle line made up from people spinning away on ergometers look? Possibly not.


The closest modern day equivelant I can find is this ( https://www.arrowinapple.com/crossbow/swiss-crossbow-makers/twinbowii/ ) a lever action compound crossbow, though I couldn't comment on it's effectiveness at armour penetration etc. It certainly would fit into a fantasy type structure.

There is also the latchet crossbow ( https://youtu.be/M0m5udFoPnA ) though bear in mind this is a very lightweight bow, as any higher draw weight would require a larger mechanism to fire.

I would suggest some sort of similar mechanism to this, but maybe with a pulling-motion, so you latch the claws onto the string, then push down with your foot while pulling up with the handle to place the string into position (kinda like a chest-expander with a crossbow attached to one end ;) )

I hope this is still a project you're working on, otherwise I just wasted my time. :D


Ok so we are in a fantasy setting which is excellent as it removes a large amount of the difficult physics as it can explained away as magic.

In such a setting your clever dwarven like artificers can create good springs and strong gears that would allow for a hand cranked crossbow to be fed from a bottom fed bolt magazine.

Pull the lever and it turns a small gear which, with cunning craftsmanship and a little bit of magic dust, turns the larger gears to recrank the bow and load the next bolt. Real world this would have to await the development of some very strong metals and probably hydrolics or powered cranking mechanisms.

Also depending on the level of magic in your fantasy world production of such a weapon could be very expensive perhaps reservered for the royal guard or the mages personal retainers.

Personally from a maintaining the balance persective, I would make it so the magazines are only be able to hold a small number of bolts and would probably take a significant amount of time to reload or change, say the magazine held four bolts that it could loose quickly, reloading time would take the same amount of time required to bring it inline with a normal crossbow being fired four times.

Also there not going to be disposable, lots of expensive and cunning working in that little box as making them to easy to replace would again significantly effect game balance.

As to penertation of heavy armour, your going to need magic again. Crossbow were not that effective against armour, they would normally be used, as mentioned elsewhere, by the defenders of a castle who would be sending bolts down in to the great unwashed and, more importantly, lightly armoured masses outside the castle walls.

(had to leave this for an hour while I put the kids to bed, hope I'm not now repeating something!)


The new Mythbusters series has the heart of what I think you want. They built a repeating crossbow with the magazine underneath rather than on top. Since the whole thing was operated remotely without any huge motors it must not have been too hard to cock.

They used the stack feed of a handgun or rifle, though, not the tube feed of a shotgun. As others have said, tube feed is out of the question due to the length of crossbow bolts.


In all the designs of crossbows I'm amazed that no one has come up with a mechanisms that uses the weight of the crossbowman to cock the crossbow. Most soldiers weigh more than the 130 pounds of the draw weight of a typical war bow, especially if you add the weight of armor. The closest was the Greek Gastrophetes, but that was extremely complicated, not suitable for mass production. The Qin bronze crossbow trigger lock seemed to be very efficient with its ability to be automatically cocked as you slide the bowstring back into it. Playing around with a paper clip I came up with a simple mechanism for cocking the bow itself. Take a long rod of iron or bronze and wrap it around your wooden stock that has the trigger lock. Bend the two ends over the bow and then down towards the far end of the stock. Then bend the two ends of the rod towards each other to form a flat U and hammer weld the two ends together. At the far end of the stock you cut a angled slot for the metal U to fit into at full cock. The soldier slides the bow up to lock the string on the trigger lock and then, putting front of the stock firmly on the ground presses with his foot on the center of the bow pressing it down to slide down the stock until the U slips into the slot on the stock. Then up comes the crossbow, an arrow is placed on the trigger lock and the top of the bow and fired. I'm sure there are many design improvements that could be made but I really don't see any real reason this couldn't work. Simple, easy for a blacksmith to make, fast reload, powerful and accurate, and taking minimal training for a fresh recruit to use. I'm really surprised something like this was never developed.


Since this is on google, I felt like chiming in because I love crossbows half as much as Shadiversity loves Castles. First off, the Arbalest penetration hasn't been tested for a long time because it's a super heavy siege bow more than a basic crossbow with a 2 bolt reload at two minutes. It's not designed to be field used, which is where light and heavy crossbows are used. You need to worry about making magical metals or composite and wood that can reach these forces without hurting the operator. Chinese Zhuges were dropped in favor of larger crossbows because they lacked penetration, and anything like Saracen bows are utter garbage to be used as a source.

Secondly, while doing the math to state penetration capability, everyone forgot in the last two years where Longbows and Crossbows at wargrade strength were fired on reproduction (good and mediocre) breastplates. The first is a reference to Lindybeige at a convention where a 130 pound longbow harmlessly bounced off a breastplate, the crossbow videos were tested by Skallagrim where bothmediocre (arms street iirc) lamellar and plate breastplate were shot at by a 300 lb and 1000 lb crossbow repeatedly, and failed to penetrate completely. They can not individually penetrate plate armor, especially after the 15th century. Projectile weapons were never the main stay of an army for centuries because they lacked the penetration needed to pierce basic layers of cloth and boiled rawhide armor. Longbows and Genoan Crossbowmen became mainstays because of the effectiveness of their soldiers more than their weapons, similar to antiquated Crete.

There's a thing floating on the internet that even mentioned at Agencourt the English longbowmen not killing French knights with armor penetration, but by downing them and the men suffocating under their armor. I also want to remind people that a carrier plate is not designed like plate armor was, and that a modern crossbow can pierce it because it is so much more powerful than medieval ones.

  • $\begingroup$ Hello. I don't really see how is this supposed to answer a question about loading mechanism. Please take the tour to get more familiar with Stack Exchange format. This is not a forum. This is Question and Answer site. $\endgroup$ – Mołot Sep 27 '18 at 11:24
  • $\begingroup$ Welcome to WorldBuilding.SE! I agree with Molot, it looks like you've completely ignored what OP is actually asking about (the loading mechanism), and instead focused entirely on his brief description of the crossbow's armour-piercing capabilities. Your knowledge is impressive, and I appreciate that you're trying to help, but please try and focus on helping with what the OP actually wants help with. $\endgroup$ – F1Krazy Sep 27 '18 at 11:36
  • $\begingroup$ I'm going to hold off on voting to delete this for now, because this information could still be helpful to the OP, but I'd just like to make sure you're aware that this answer is currently in the deletion queue and may still be deleted by other high-rep users unless you edit it to address the central question. $\endgroup$ – F1Krazy Sep 27 '18 at 11:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.