Pictures of my hoop-structure round pavilion

[Steve S.]

For Charlotte, others:

http://www.forth-armoury.com/temp/tent/tent.htm

I undertook to design and construct this tent about 8 years ago now, I guess. I designed it when I was single. It is too small to fit our queen-sized oak bed inside. A queen-sized mattress will fit on the ground, but not up on a bed without hitting the side walls.

I chose the round style of pavilion as it is my favorite style of medieval tent, and it seems to include a wide range of historical eras. I wanted a round pavilion with some sort of internal support structure because I wanted a tent that would be usable at Pennsic without taking up a large footprint of space with guy ropes.

I was inspired by an article I found on the web (no longer available) by a living history group called Circa 1265, and a fellow named Andy Goddard. http://www.bumply.com/Medieval/author.html

Andy and I had long discussions about two round pavilions that he and a friend made, using speculative construction techniques to create a truly round pavilion utilizing an internal "hoop" frame to support the roof.

Andy's design, as I recall, utilized hoop pieces with holes evenly spaced all around, and his roof had canvas loops which would be fished through the holes, and then a cord pulled through all the loops to keep them from pulling back through the holes in the hoop. The arrangement was much like vervelles on a helm.

His friend had a different arrangement. His hoop was wide but thin, basically bent planks held together with a somewhat complicated staple, pin, and wedge arrangement. The top edge of his planks were cut with crenelations, and his roof line had canvas loops with toggles. The toggles would catch in the crenelations, securing the hoop to the roof line.

I decided on a simpler approach. My hoop was simple steam bent pieces of wood, about 2" x 2" in cross-section, kept in alignment at their ends with a metal pin. The tension of the canvas trying to collapse the circle would keep all the hoop pieces together - the pin just kept the ends aligned together. We drilled the hoop vertically and installed oak dowel pins at regular intervals, protruding downward, like an upside-down hat rack. My roof has matching canvas loops at the same intervals of the pegs in the hoop, on the seams. Likewise the wall, which is made in two pieces, has loops which wrap around the hoop and catch on the same pegs. Thus the wall loops pull in one direction, and the roof loops pull in the opposite direction, equally balancing the hoop.

The hoop was made of segments of steam-bent ash, which I had bent to shape by, as I recall, Bethel Furniture in Bethel, Maine. My wife and I made a fixture to cut them to the proper arc length and then drilled and doweled them.

The tent itself I modeled in 3D CAD software, Unigraphics, and I used it to unwrap the conic sections into flat patterns. These I sent to Tentsmiths to use to sew the pavilion. I considered buying the canvas and attempting to sew the pavilion myself, but I figured I would screw up the first one and by the time I did it twice it would be less expensive to have an expert do it for me.

Neither I nor Andy have any historical evidence for the hoop structure. There are pictures of round pavilions that show decidedly round roof lines, no guy ropes, or guy ropes which come out from under the valence at the roof line. These pictures could be artistic license or error, or they could be indicative of real occurrences that can most naturally be explained by an internal structure.

My objective of reduced footprint was realized, as when erected with the walls (which I can leave off, if I use guy ropes) the tent does not need guy ropes at all, the walls forming essentially an infinite series of guy ropes all the way around the pavilion.

I think just about everyone agrees that a ropes-only tent is the simplest design, and by this virtue alone the most plausible. It certainly makes for the most portable tent, with one needing only to carry the canvas, the ropes, and the center pole. Even the center pole could be omitted if one was willing to cut down a tree from place to place.

Andy and I agreed that the idea of perimeter poles, that is, vertical poles near the walls, are probably not historically correct. There are plenty of pictures of round pavilions shown with the doorway open and not one shows a hint of perimeter poles, even though they show the center pole.

I have to say I am extremely pleased with the way the pavilion turned out. It is one of the few projects I have done and had it turn out right the very first time. The tent is extremely sturdy, and the round shape makes it shed winds very very well. The steep roof makes it shed rain very very well.

Steve

[Steve S.]

Another thing that drew me, and others, to the internal structure design is that it produces a perfectly round shape, with a perfectly taught conical roof, as frequently seen in illustrations. With a ropes-only pavilion, it is very hard to run the guy ropes out far enough to hold the roof that tight.

Notice this picture of a ropes-only pavilion and how, in spite of the large reach of the guy ropes, how the roof somewhat sags.

http://www.forth-armoury.com/temp/tent/MTT11.JPG

Notice how the crows feet in this example tent to gather the fabric, pulling it towards the guy rope. Ropes-only pavilions just seldom seem to "pop" with the circus-tent tightness often presented in period illustrations.

Steve

[Charlotte J]

Steve - thanks for posting that. It looks great. I'm drawn to the slightly slope-y in roof that the spoke gives sometimes, but I think with the right cutting, this could do it too. But we really have to just pick an image and go with it, I think.

[Steve S.]

I'm not sure what you mean by "slope-y". If you mean how the roof is not a true cone, but is more like a Hershey's Kiss in shape, I agree, this is easily done.

Instead of making the roof panels triangles, the long sides of the triangles need only be cut as gentle concave arcs. I wish I had done it with mine. My roof produces a pure cone.

Steve

[Galleron]

Very interesting. How long is each hoop segment? Also, can you tell us more about how the hoop segments connect to each other?

[Steve S.]

I'm not sure - each hoop segment is about 3-4 feet long? It's about a 10 foot diameter at the roof line, divided into 8 equal pieces. Whatever that comes out to.

The end of each segment has a hole drilled into it. I have a set of steel pins that I stick in the hole of one, and then impale its neighbor onto it and secure it to the roof with the loops of canvas, and then move on to the next segment.

The pins just keep the ends of each segment in alignment with its neighbor, preventing the ends from skipping off of one another. The tension of the canvas keeps them all compressed inward as a circle, so they do not, in fact cannot "pull off" of the pins.

Steve

[Charlotte J]

Sorry - posting late at night. Yes, that's what I meant. I do want to test it - the shape should logically work, but I sometimes wonder if the weight of the hoop will pull down on the bias of the fabric to make it simply conical.

If anybody else has hoop construction tents to show, I'd love to see more links!

Here is one that I saw a few years back that gave us hope that a hoop can work with oval:

http://community.webshots.com/album/415222863yrIcUq

Will, it's Laura Morgan's tent. I'll bet we can find out more info on it. I'll drop her a line and see how it's held up.

[Mac]

Charlotte et al,

The canopy can certainly be made to have that "Hershey's kiss" shape by making the edges of the panels curved.

I have done this in some of my models.

I will post a picture if I can figure out how to get around the ridiculous size limits of this forum.

Mac

[Mac]

Steve,

That's a nice tent frame!

Do you find that holding the segments together with a single pin allows too much play? It seems to me that they might tend to rotate a bit with respect to each other. Or, is this just not an issue?

Mac

[Charlotte J]

Ah, ok. I’ve never actually seen it in action, but I haven’t seen that many hoop pavilions.

If you upload the pic to another place, you can either embed it as an image here by highlighting it and using the IMG button above, or just by copying and pasting the link for people to follow. If it’s wide/huge, the latter is probably better so the text of everybody else’s posts aren’t miles to the right.

[Mac]

I found a smaller picture of one of my models. this is the one-man tent, or arming pavilion. The swoop in the canopy and the walls was achieved by curving the long seams.

Mac

[somedudeinutah]

Nice tent Steve!

[Gryffinclaw]

WOW I am really interested in the details of how it was made. Nice job. What tentsmith sewed it for you. Would you be willing to post the details so I can make one?

[Steve S.]

Steve,

That's a nice tent frame!

Do you find that holding the segments together with a single pin allows too much play? It seems to me that they might tend to rotate a bit with respect to each other. Or, is this just not an issue?

Mac

Thanks!

I thought about how to join the ends of the segments for a long time. The guys at Circa 1265 were worried about twist and pull-apart, too, and they came up with some sophisticated ways of joining the segments.

I discovered that it is really over-kill.

All you really have to do is keep the ends from "skipping off" of one another, and a simple dowel keeps them in alignment.

With the roof and walls pulling in opposite directions, the hoop is in equilibrium - it does not want to twist in or out on itself. But really, I get the feeling that the arced shaped of the segments makes them not want to twist anyway. Pull-apart of the segment ends is not an issue because the tension and weight of the canvas tries exceedingly well to compress the hoop, thus keeping the segments tightly butted together. In fact, during assembly, I don't put the last pin in, because you can't spread the last two segments apart enough to do it.

Nice tent Steve!

Thanks!

Steve

[Steve S.]

I found a smaller picture of one of my models. this is the one-man tent, or arming pavilion. The swoop in the canopy and the walls was achieved by curving the long seams.

Mac

Cool model!

Steve

[Mac]

Thank you Steve!

I made a few 1/12th models about two years ago. I find that they behave very like full sized tents, and are a very valuable tool.

I you, or anyone else are interested I can try to figure out how to post some pics of them. Most of my pics are too big to post here, and I am not geek enough to figure out how to get them on my web site to link to. Perhaps I can have Marianne help me with that this evening.

Mac

[Mac]

Steve,

How long are the pins which connect your frame pieces together? Are they long enough that they are loaded in bending, or are they really just long enough to keep the segments aligned?

Had you considered (and rejected) making the pins as an integral extension of the frame segments? I wonder if short, square section pins would work. Modernly, we tend to think of using round pins to align things. But, in world where saws and chisels are commoner than lathes and drills, the square mortise and tenon might be the thing. It would certainly help resist any tenancy for the segments to get misaligned through torsion.

Mac

[Sean Powell]

Mac,

Photobucket.com is a free service that Liz and I use to host photos online. It has a simple upload tool and you can resize them directly online or provide links to the full-size picture. If you and Marianne need help drop me a line... Or I can throw them on my PB account for you.

Sean

(Who is currently researching tent cost options and just MIGHT be convinced to go hoop round or hoop oval instead)

[Steve S.]

WOW I am really interested in the details of how it was made. Nice job. What tentsmith sewed it for you. Would you be willing to post the details so I can make one?

I would have to go digging. I think I kept all of my plans, but I'm not sure where they are. Originally, Tentsmiths wanted to make the tents and me to provide hoops, but I did not want to get into that business.

The key parameters that I had to juggle were:

1) Maximum available width of canvas
2) Maximum length of steam-bent wood pieces
3) Making sure no segment ended on a seam, and that a peg was spaced on every seam.

Steve,

How long are the pins which connect your frame pieces together? Are they long enough that they are loaded in bending, or are they really just long enough to keep the segments aligned?

Had you considered (and rejected) making the pins as an integral extension of the frame segments? I wonder if short, square section pins would work. Modernly, we tend to think of using round pins to align things. But, in world where saws and chisels are commoner than lathes and drills, the square mortise and tenon might be the thing. It would certainly help resist any tenancy for the segments to get misaligned through torsion.

Mac

My current pins are about .25" in diameter, mild steel. Just round stock from Home Depot. They are about 4" long, with 2" penetrating into each segment.

Absolutely I would do this differently knowing what I know now.

What I would do now is cut a vertical slot in one end of each segment, and cut a vertical post on the opposite ends. Thus I would have a "tongue and groove" arrangement. This would prevent lateral sliding and prevent twisting. Vertical sliding is not an issue because the canvas holds each segment in the same horizontal plane, with gravity bringing all to the same height. I think tongue and groove would be much, much simpler. simpler. All this could be manufactured with a simple saw.

Edited to add: I'd be leery of making the "pins" square protrusions as you seem to suggest. At least, I would not want to make them too fragile. If they break off, or are twisted off, your segment is ruined.

I think rather than a post, a tongue that follows the entire width of the stock would suffice, mating up with a corresponding groove in the opposite end of its neighbor. And, thankfully, post or tongue I don't think it would need to be very tall (and deep), thus limiting the torque that could be applied on to it, making it stronger.



Steve

[Mac]

Steve,

The tongue and groove thing you suggest sounds good. What do you think about making the T and G go the other way? That is to say, in the horizontal plane instead of the vertical.

That would preclude any possibility of the segments shifting vertically, and the last segment would be held in place by the tension on the canvas.

If, after installing all the segments, the canvas was not sufficiently tight, shims or wedges could be inserted between the segments to increase the circumference of the hoop. So long as the T and G engagement was deep enough, this would allow very nice adjustment. I'm thinking maybe 1/2" to 3/4" engagement, with the possibility if adding 1/8" to 1/4" shims to as many joints as necessary.

I think that being able to fine tune the hoop to the canvas is important. Not only might the canvas be a different circumference in the wet vs. dry state, but it might well stretch permanently as the campaign wore on. With a system like I suggest, you could add shims until it became necessary to swap in an oversize hoop segment.

Mac

[Charlotte J]

It seems with the hoop that the prevailing opinion is that the hoop itself was bent and shaped, or at least something created specifically for the purpose, and carried along with the canvas. I have heard the argument that bent saplings found at the overnight site could have been used. If that's the case, I would imagine a less formal join, even wrapping and/or lashing the pieces together, maybe butted, maybe overlapped. Does anybody have any thoughts on the "find hoop material on-site" idea, or does it seem too riskyy, in case there was no wood available? It seems like it wouldn't be hard to make a tent that could work with a hoop or just ropes.

[Steve S.]

Steve,

The tongue and groove thing you suggest sounds good. What do you think about making the T and G go the other way? That is to say, in the horizontal plane instead of the vertical.

That would preclude any possibility of the segments shifting vertically, and the last segment would be held in place by the tension on the canvas.

If, after installing all the segments, the canvas was not sufficiently tight, shims or wedges could be inserted between the segments to increase the circumference of the hoop. So long as the T and G engagement was deep enough, this would allow very nice adjustment. I'm thinking maybe 1/2" to 3/4" engagement, with the possibility if adding 1/8" to 1/4" shims to as many joints as necessary.

I think that being able to fine tune the hoop to the canvas is important. Not only might the canvas be a different circumference in the wet vs. dry state, but it might well stretch permanently as the campaign wore on. With a system like I suggest, you could add shims until it became necessary to swap in an oversize hoop segment.

Mac

Interesting thoughts. Here is what I think:

I don't think fine tuning of the hoop is necessary. My tent is going on 8 years old now, we've probably used it some 20 times over that amount of time? We've not had a problem with fit yet.

When I hang my hoop segments from the roof today, one segment at a time, doweling the ends as I go, the last dowel I don't insert because you can't spread the hoop apart at the last joint to insert the peg. You might be able to spread it 1/2" - enough to jump a tongue into a groove, before your strength gave out. You can't do that and get a peg in though, especially since I'm not sure with both hoop segments strapped into the canvas I don't think you could bend get them both aligned with the dowel and dowel holes and so snap together. So I leave it out.

If the diameter of your canvas roof at the roof line is too small, then you have a problem, of course, But if your canvas is slightly over-sized, I don't think it would be that big a deal - gravity would pull the hoop down until the canvas could not go down anymore, causing an inward compressive, and upward supportive, force vector on the hoop. I imagine you might see some wrinkling along the roof line as the canvas gathers.

Invariably, the only slip action I ever see on the unpegged junction is in the horizontal plane. The end of one hoop segment can misalign with the other in the horizontal plane only. Almost never (in fact, I can't remember it ever) enough for the two ends to skip off of one another, just enough to make a visible bump on the outside of the canvas. The reason there is no slipage in the vertical axis is because the entire weight of the hoop, which is substantial, is pulling downwards, and the canvas, pulled uniformly tight in a circle, suspends the hoop in a perfect plane. Additionally, the weight of the hoop pulls down, but the counter vector of the canvas is both up and towards the center pole, which has the effect of compressing the hoop segments quite tightly.

Anyway I'm pretty satisfied that the primary axis of motion of an unpegged joint is always horizontal, perpendicular to the center pole. Thus a tongue and groove in the vertical orientation would halt that, as well as any twisting which, if your walls and roof pull in opposite directions, is minimal.

It seems with the hoop that the prevailing opinion is that the hoop itself was bent and shaped, or at least something created specifically for the purpose, and carried along with the canvas. I have heard the argument that bent saplings found at the overnight site could have been used. If that's the case, I would imagine a less formal join, even wrapping and/or lashing the pieces together, maybe butted, maybe overlapped. Does anybody have any thoughts on the "find hoop material on-site" idea, or does it seem too riskyy, in case there was no wood available? It seems like it wouldn't be hard to make a tent that could work with a hoop or just ropes.

I suppose this might be possible, especially if you knew precisely the length of pole you needed. Perhaps you could mark this length on your center pole as a permanent reminder, and then go gather up saplings of appropriate thickness, cut them to length, and then lash them together end to end, and then bend them into a hoop, and lash the final ends together. This would probably result in visible bumps along the perimeter of the canvas at the binding points.

One nice thing about permanent hoop segment pieces is that they do nest together, making packing them not an atrocious thing. If you have a cart to carry the canvas and center pole, the segments would not be much of an extra burden.

Steve

[Larmer]

Interesting. Thanks for posting.

[Mac]

Steve,

Your argument for vertically oriented tongues and grooves is compelling.

Do you ever consider retrofitting your hoop with them? You could make the tongues as splines, and glue them into slots. It would be a relatively quick job with a router or a table saw.

Mac

[Steve S.]

I'm afraid to screw it up.

I might consider making new hoop segments.

Steve