h a l f b a k e r y
"Put it on a plate, son. You'll enjoy it more."
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
A truly classic problem is how to protect your city's main gate
(set in a city wall, of course) from a battering ram. Classical
answers include such things as putting the gate at the end of a
tunnel, and there are "murder holes" in the walls of the tunnel
through which defenders can shoot arrows
at the attackers
to use the battering ram. Filling the tunnel with corpses tends
reduce the ability to use the ram.
But here is something that might be a new Idea. Consider the
capital letter "D" --and now consider tilting it over so that its
side is upward, and the curve is downward. Let us make that D
from a quite-large stone block, and put it in the ground such
the upward/flat part of the stone is at ground level. The stone
block is as wide as the city's main gate-way, and people walk
across the top of the block to enter/leave the city.
There is a winch and gearing system to affect the large stone
shaped block. When danger threatens the block is rotated 90
degrees, so that its flat side is vertical, and basically flush with
the city's wall. Half the stone block is still below ground level,
and half is above ground level. There is now a great big PIT in
front of the block, occupying the exact place where a battering
ram should be located, in order to attack the gate. The big
stone "D"-shaped block IS the gate, and the thickest part of the
stone block is near ground-surface level --right at the place
where the attackers would be pounding, IF they could get a
battering ram in the right position in front of the gate!
As mentioned in an annotation. [Vernon, Apr 07 2015]
Romans and ball bearings
As mentioned in an annotation. [Vernon, Apr 07 2015]
On moving large/massive blocks
Not necessarily so tough as [MechE] indicates, but exact details are missing. Ah, well. [Vernon, Apr 07 2015]
|This is excellent, and I shall incorporate it into the
very next fortified city I build.
|By the time the mechanism existed to move around a
slab of stone of that size, repeatably, and without
counterbalancing, gunpowder was in widespread use.
|Even ignoring that, you don't have to break the
stone, you just have to break the mechanism holding
it in place.
|wouldn't it be easy to fill the hole, or am I missing
something? Keeping in mind that the Romans built a
mountain sized ramp to get into Masada, I'm not sure this is
a sufficient deterrent.
|If you fill in the hole, you're still facing the flat side
of a quarter-circle of stone.
|If the hole were filled in, you could then turn the
stone a further 90 degrees. This would flatten
anyone standing on the newly-filled-in hole, and
leave the attackers having to attack the curvey-side
of a half-circle of stone.
|I think you'd want more than a semicircle, so the axle wouldn't be exposed to the attackers. Maybe a ( ) which is balanced more than a D
|It seems to me that the attackers need only employ
a sufficiently long
ram to reach over the pit. I think a more workable
solution would be to
stretch a series of steel chains across the walkway
at variable heights, and some of them vertically
going floor to ceiling, which can easily
be bypassed individually by crouching and high-
sidestepping, but carrying a shield
overhead to protect from murder holes would be
without snagging the leading edge of the shield on
the chains, and
charging through with a battering ram would be all
because all the people holding onto the ram would
be forced to go
high or go low at different times, breaking their
stride while trying to
achieve ramming speed. They also wouldn't be able
to attack the
mooring points of the chains themselves very
easily since their hands
are full of ram.
|[MechE], if you google for ["ton block" megalith] (the
brackets represent the search box), you will find that
ancient cultures moved around quite a few large
stones, of similar size to the one proposed here.
|Regarding ram-forces breaking the mechanism, I did
happen to think that an axle-based pivoting block would
be more vulnerable than a cog-track system. If multiple
curved "racks" were carved into the underside of the
block, the half-circular block becomes the equivalent of
|You can now put a significant number of smaller gears
(drive gears, all of them) in the quarter-circular
subsurface area that the block, in both the "up" and
"down" positions, always contacts. The equivalent of a
"pawl" (multiple pawls, actually) could be used to
ensure the block stays rotated to the "up" position. The
pawls could be located near the bottom of the pit, such
that impact-forces against the block at ground level
wouldn't be affecting them much. It may also be
possible to support the block in other ways, to ensure
impact-forces cannot affect the drive-gear axles.
|The other subsurface quarter-circle could simply have
slide-tracks. They would of course be exposed to the
enemy, and while easily destroyed, they also are easily
replaced, after the enemy goes away.
|[theircompetitor], the Romans used that ramp to get
the top part of a siege tower at about the same level as
top of the wall of Masada. They didn't bother with the
fortress gate. For this city, the Idea is to make a
battering ram essentially useless.
|Note that if the hole was filled
in, the block can no longer be rotated, so it doesn't
matter if the drive-mechanism gets broken (unless the
attackers' plan includes digging out the hole again --but
that is why I described some things about protecting the
|Until/unless the drive mechanism is broken, the
attackers still need to break the THICKEST part
of this gate! Siege towers will probably have to be used
instead --which means the gate qualified as "ram-
proof", after all.
|MechE this is one idea Vernon is allowed to pursue
without being harassed with meticulous pedantic
|That's why I was careful in my phrasing. The issue is not
moving the stone, the issue is making a mechanism to
move the stone. Simply hauling the stone up and down a
ramp, or even with a block and tackle is simple.
|Gearing it to rotate around an axle, and have the gearing
function repeatably and reliably is another story. Ditto
producing the motive power other than through brute
force, and the mechanism to lock it in place. You don't
see that sort of equipment at the sophistication this
would require until the industrial revolution.
|^ That's me with the axle; Vernon doesn't mention one at all. Either way, as written, a fully semicircular chunk of rock is physics-ally a poor implementation: in order to get it into a defensive position you have to actually lift tons of stone (in terms of center-of-gravity); destroy the chock and it slips right back.
|I've modified the main text slightly, to more clearly
describe the result of rotating the block 90 degrees.
|[MechE], regarding mechanical sophistication, might I
remind you of the Antikythera Mechanism (linked)? The
Romans had/used ball bearings (link), although I'm not
implying here that they are appropriate for this Idea. It
is just that some things considered "modern" aren't
necessarily all that modern.
|[FlyingToaster], I mention chock (as "pawl") in an
annotation above --and plural, not singular. With
enough chocks in enough places under the rotated
block, they could protect the axles of the small gears
also mentioned in that annotation, and be too
numerous for any significant impact-force of a battering
ram, against the block, to break them --the impact-
force would be spread too thinly.
|There's a bit of a difference between the 6" cast
bronze gears in the Antikythera mechanism and
ones capable of handling 20 tons.
|I'm not saying the mechanism is impossible,
although it is difficult, and not in keeping with the
period of fortifications you are talking about. Yes,
the Romans were better at this than the middle
ages, but heavy gearing of this sort simply wasn't
happening. Wooden gearing, the most common in
both Roman times and the middle ages isn't going
to be strong enough. Stone gearing isn't
significantly stronger, stone is tough under impact,
but teeth would tend to crack under load.
|Bronze and iron castings on the scale you're talking
about still fail as often as not, even with modern
|And consolidating sponge iron on that scale isn't
going to happen, so iron doesn't happen until you
get a puddling furnace during the industrial
|Not to mention the issues of all of this mechanism
being buried under a 20 ton rock you only move
once every several years, which is going to lead to
some issues with rot (wood), rust (iron) or spalling
(stone). Bronze won't corrode so bad, but as I
said, not strong enough in any size that can be
|And, as an aside, lead ball bearings aren't going to
handle 20 tons of rock.
|The best possible mechanism for moving stone on
this scale, during this time period is a windlass,
but that doesn't work well for causing rotation
unless it can be pulling on the far edge. Which in
this case means the windlass is either pulling rope
under and has all the rot issues mentioned unless
it's human accessible (and therefore a weak point)
or over in which case either the belly of the d-
shape is out, or the windlass is.
|Your best bet for the locking mechanism would be
heavy wooden beams through openings in the
stone, and that's only going to be as strong as the
beams, less the weight of the stone already on
|A drawbridge has the same basic concept (block
the opening with something that covers the ditch
otherwise), and an iron-bound drawbridge out of
heavy wood will be just as resistant to a ram. This
is especially true since the inner gate was rarely in
a straight line with the outer gate, making the
maneuvering go a large ram rather difficult.
|This is also why most actual recorded uses of rams,
contrary to Hollywood, were aimed at the stone
walls rather than the gates.
|Why not simply shape the gate in a delta fashion that
deters Rams with a glancing blow?
|[MechE], in another annotation I thought I made it
clear that a lot of gears could be used, which would
dilute the weight of the block that any one gear
experiences. Of the materials you mentioned, bronze is
the one most likely to be appropriate. If you have a 20-
ton block and 40 gears, then that is only 1/2 ton per
Such weight-dilution could also make the other thing I
described (carving "racks" of gear-teeth into the
underside of the block) more durable.
|With respect to using a ram against a wall, you are
talking about a castle wall, not a city wall, which was
usually a whole lot thicker than a castle wall. And this
Idea is about a CITY gate.
|Like I said, make the gate into a V-shape so that you can reinforce it and simply can't hit it straight on. Use the power of triangles, man.
|Um, city walls were usually thinner than castle walls,
when there was a difference. And yes, city walls had all
of the same tricks to keep a ram away from the gate as
the castles did. If the city walls were thicker, it was due
to a stone/rubble/stone construction, which isn't a lot
|And it doesn't matter how many gears you have. If
they're hand finished, most of the load is going to be on
one, because they won't mesh uniformly.
|Regardless, if you agree that the minimum gear size is
more than a couple of inches thick, then the technology
didn't exist to cast bronze in that thickness. It still really
doesn't today. Modern large bronze gears are cast very
carefully with a thin web and the gear is machined into a
rim that is as thick as possible. The best versions are
cast in a centrifugal mold, such that the rim is under very
high pressure during casting to force metal into the rim,
and voids and gas bubbles out to the center.
|I disagree with your load comment. If you were talking about high-strength steel alloys then the loading wouldn't distribute well. But in the days of malleable pig-iron, not so much a problem.
|I see for this maybe a geartrain made of steel-clad large wooden or even stone gears.
|Starting with a horizontal cylinder of rock, if you slice the bottom and top off, you end up with a ( ). The bottom portion could be built back into a cylinder in a framework of metal, much lighter than the original stone, incorporating runners. The mess can then be slid to rotate the cylinder.
|Or just orient the cylinder vertically, like some kind of normally-open revolving door.