Designing a knee, trying to get it right

[Aussie Yeoman]

Following on from this thread:

viewtopic.php?f=1&t=170292

I decided I would have a go at designing a knee to play around with articulating on paper, and get something working right in the planning stages before (one day) transferring to steel.

The example is the knee of this statue:



My first supposition was that the cop is too shallow, so in the sketching I made it deeper. I had originally made the proportions of the height (height of the cop to height of the point of articulation) of the cop fairly similar to that in this picture. However as I played around with where the articulation points needed to be in order to move the lames smoothly and with minimal gap through the flexion/extension, the rivets got a lot closer together.

As a result the cop tapers in height a lot more in my sketch than in the sculpture. At the moment I'm not fussed by the wing.

It took about half an hour. I used baking paper to play around with the lames. It's very see-through, takes pencil well and more importantly, pencil erases off it exceptionally cleanly.

I ended up with this:



I hope you can make sense of all the gray lines.

When I first finished I was quite pleased with it because the joint, when flexed, was not 'Michelin Man-like', and there was little gapping through the whole range of motion.

Then it struck me that the distance between the rivets and the second lames when extended, might not be enough to accommodate the patella when the rivets are sitting around the organic joint of the knee.

I guess I should have quick sketched the cop over a photo of my knee to see what the proportions should actually have been. But then again I might have just snuck in with a win on this one.

I'll now throw to the audience to see what others have to say.

[Mac]

Dave,

It's important to consider how the cannons are going to articulate with that. Try including the demigreave and some portion of the cuisse.

You have done a fine job of optimizing the flex, but it may be too much of a good thing. You may want to settle for less motion and more secure "lock".

Mac

[Mac]

Here's a 4 lame knee that's behaving perfectly.

Goll 2845

This one is strongly related to your statue. The lower part is as full flex. The upper part looks like it could do the same.

Goll 3020

This one looks a bit stingy, but I thing there is more motion left in it. Lames 2 and 3 do not look like they have flexed as far as they can.

Goll 3020


[Edit] I went back and put in the Goll numbers
Mac

[Aussie Yeoman]

Thanks Mac, as always.

What're the Goll#'s for those examples? I'm a bit slow at first, so it took me a little while to realise that in the second example, the wing is caught behind the uppermost lame, impinging movement. It is nice to see a working example of something much like the statue. And I just noticed, how strangely atypical are the hinges that join cuisse and wrap-plate (does that have a name?). I think the third example has the same!

It seems to me my cop might benefit being a bit taller from rivet to apex of cop 'bowl'. Thoughts?

[Aussie Yeoman]

I found them by right-clicking on the image and clicking on 'inspect element':

ref_arm_2845
ref_arm_3020

So of course the bottom two have the same hinge...they're the same artefact!

[Mac]

You know.... I was not looking at the hinge, or the numbers, so I completely missed the part where they were the same object.

I'm pretty sure that the "hinge" is not original to this cuisse. It's one of those thingies that sometimes holds tassets on 16th C armors.

Mac

[ManOnFire68]

Awesome idea for a thread. I'll be following this.

[Tom B.]

Here are links to the full PDF's for those legs that Mac posted:

PDF for Goll 2845

PDF for Goll 3020

[Tom B.]

Mac,

While looking through Goll's photos a few months ago, I was surprised to see extent of use of sliding rivets in late fifteenth century German style knee articulations.
There are several examples where every articulation rivet seems to be a slider while some only use slots in the top and bottom connections.

How does this impact the design?
I would assume we initially design for correct fit / shape without slots, is this correct?
How do we determine the correct slot length and location relative to the initial rivet hole?

Here is just one example they also can be seen on the A60 & A62 legs at the KHM.
You also can see them on the examples Mac posted above.
ref_arm_2884

[Mac]

Sliding rivets in knees are on my list of things I would like to have a better understanding of. I will probably put them on the Florian knees after they are hardened.

My current understanding is that they really come into play when the knee is bent and you rotate your foot in or out. This is a different motion than the "inversion" and "eversion" of the foot at the ankle. It happens at the knee, but only when the knee is bent.

Cutting the slots will be just like cutting rivet slots for anything else. You begin with holes in the usual places, and then cut back at an angle so that the articulation will remain "locked" at it's greatest flexion. You do this by removing one bolt of the temporary assembly, and sliding that pivot point back a bit while maintaining the lames in their flexed position. Mark a point at the new pivot location Repeat this with all the pivots, taking care that the new locations will not let the articulation gap. You can than connect the dots, as it were, and the articulation will still function; but now it will do so with that pivots at any point along their slots.

Mac

[Tom B.]

Thanks Mac
I look forward to watching this on the St Florian legs when you get there.
I see how this motion would be benificial and have noted its absence in my current leg harness.

[Mac]

The other thing I think it will accomplish is letting the cuisse rotate a bit with respect to the bent lower leg. This may be useful with those very high gothic cuisses, which will try to rotate outward when you sit.... that torque being supplied by the upper lames jamming into you groin.

Mac

[James Arlen Gillaspie]

I may draw some flack for saying this, but let me hasten to say first that the late German 'gothic' style is my favorite armour style. Now, having wrapped myself in the flag, let me say that I am also very fond of its opposite in so many ways, the mid 15th century Italian heavy cavalry armour, as well. Their approaches to the ergonomics of the human knee were very different. The first armour in any European style I ever made was a pair of legs modeled as well as I was able to at the time on those of A 62 KMW. The sliders enabled the knees to adapt to the S curve that everyone has, and to handle the direct attachment of the greaves by means of the classic pin through the bottom lame of the poleyn and all the resulting complications. The great thing about it is that it works without one having to understand a damn' thing about ergonomics, even when the knee is of minimal diameter. It gets even better when one looks into the poleyns of A 60 and A 62 and sees that both have straps running down the center to keep the lames from gapping! The Italian style, by contrast, has no sliders, but a nonparallel arrangement of the rotational axes of the poleyn lames in conjunction with loose rivets, that are just loose enough to accommodate the variation in the S curve through the knee in the majority of men.

[Tom B.]

It gets even better when one looks into the poleyns of A 60 and A 62 and sees that both have straps running down the center to keep the lames from gapping!

Wow!
Let me make sure I understand.
Is there a vertical strap on the inside along the centerline, connecting the lames to eachother?
One strap from top to bottom or one for top lames and a seperate for the bottom?
You wouldn't happen to have a picture would you?

[James Arlen Gillaspie]

Hello, Tom, I have long owed everyone my paper on A 62, but there remains the business of putting the finishing touches on it and actually finding a place to publish it. Look at the knees of A 60 and A 62, and you will note a rivet right in the middle of the medial ridge. All the lames of the poleyns are connected to their neighbors, up and down, and to the lame at the bottom of the cuisse.

[Tom B.]

Hello, Tom, I have long owed everyone my paper on A 62, but there remains the business of putting the finishing touches on it and actually finding a place to publish it. Look at the knees of A 60 and A 62, and you will note a rivet right in the middle of the medial ridge. All the lames of the poleyns are connected to their neighbors, up and down, and to the lame at the bottom of the cuisse.

I have often wondered about those center rivets but had forgotten about them when I made my post above.

I for one, would be very very interested in such an article and would buy it how ever it ended up being published.

[Aussie Yeoman]

Much to my delight, I found that the dimensions of my initial sketch are not terribly dissimilar to extant examples from the same time/space as the statue I'm looking to model.

The depth of the cop, where the actual dishing is, is about right. The distance from apex of cop to rivets is about 1/4 inch too shallow, and the distance between the rivets is about 1/8 inch less than is typical.

I tried to redesign the cop to these new dimensions, but found that through the try-it-and-see method of articulations on paper, the rivets drifted back to where they'd been on my first attempt, above. It seems I'm the dunce of the class that can't get anything right. I had a bit of a play overlaying lines on extant pieces:



It's Goll ref_arm_2264 if the above doesn't work.



I'm quite in awe of how simply others manage to conquer the articulated joint, while I seem to struggle. In the above overlay, the lames work really well, but when I try to sketch it of my own invention, it goes terribly.

Anyway, it seems common in Gothic knees that the lames often do not maintain a minimal gap throughout the range of motion. Perhaps armourers then cared a little less about gapping. The point James makes about an internal leather is fascinating.

I was worried that at full extension, the lames two out from the cop (the red ones in the above) would dig into the patella. Then it occurred to me that viewed from the side, the front horizon of the thigh and shin are not only at an angle to each other, they are also on different planes. Which is to say that extending two lines from thigh and shin would not end up with them intersecting in the patella. So, when the knee is at full extension, the joint itself will have one half neutral or perhaps flexed to a small degree, and the lower lames will be partly in hyper-extension. Am I on to something, or is this irrelevant?

James, you said here: viewtopic.php?f=1&t=132841 that there is a very good reason that the height of the cop at the outside rivets is smaller than the height of the inside rivets. Is it because the knee gains more distance when flexed on the inside of the knee compared to the outside? We never got closure on that issue in the other thread.

Cutting the slots will be just like cutting rivet slots for anything else. You begin with holes in the usual places, and then cut back at an angle so that the articulation will remain "locked" at it's greatest flexion. You do this by removing one bolt of the temporary assembly, and sliding that pivot point back a bit while maintaining the lames in their flexed position. Mark a point at the new pivot location Repeat this with all the pivots, taking care that the new locations will not let the articulation gap. You can than connect the dots, as it were, and the articulation will still function; but now it will do so with that pivots at any point along their slots.

Mac, in what direction is the 'back' in this instance? Away from the cop?

James: in the A62 do you mean to say the leathers prevent the lames from hyper-flexing?

If you have a paper ready, or will be ready to publish, I'd recommend seeking out the team at Freelance Academy Press for digital/hard copy publication. That'd be right up their alley.

[Aussie Yeoman]

Oh, need to add: I've had a browse but can't find a decent one. Does anyone have a good pictures from the lateral and dorsal aspect (from the side and looking down your leg if you were wearing it) of a Gothic knee cop?

[Mac]

That knee is a nice find! If you use that as an exemplar, you can not go far wrong.

In my experience, the important thing about keeping the armor from bearing on the patella is to make the articulations between the canons and the lames be of an authentic nature. It's only when one tries to get more motion out of that joint by rounding the ends of the cannons in that there is trouble.

Mac

[James Arlen Gillaspie]

James: in the A62 do you mean to say the leathers prevent the lames from hyper-flexing?

The leathers appear to be there to prevent them from gapping, unless they are there to lessen wear on one's hose (?). I had the distinct impression they were there to prevent gapping, however, but would like to look at them again to be sure. If I ever get to do that again. Most of the other examples don't seem to have them, which weighs heavily against the antigapping notion. The photo shows a rare exception.
I don't seem to be able to find such leathers in the Munchner Zeughaus examples, or see a rivet in the middle of the knee cop, even though they are of Augsburg make. It is interesting that there are not the liner rivets on the top inside of the lowest lame of the poleyn that occur on the Italian examples, though there seems to have been a liner attached at the top of the cuisse.

ref_arm_2827_000.jpeg (65.05 KiB) Viewed 2041 times

[Aussie Yeoman]

Mac, or James, or Wade or Whoever,

is it the case that the distal end of the demi-cuisse, and the proximal end of the demi-greave start to curve away from the body ever so slightly? This, in my mind, would help prevent the outer lames from extending so far that the inside edges bear on the anatomy.

Why is it, do you suppose, that in the blackened knee above, the rivets for the cuisse side of the knee are 'deeper' into the cop than those on the greave side?

[Mac]

Mac, or James, or Wade or Whoever,

is it the case that the distal end of the demi-cuisse, and the proximal end of the demi-greave start to curve away from the body ever so slightly? This, in my mind, would help prevent the outer lames from extending so far that the inside edges bear on the anatomy.

We see the distal ends of some cuisses flaring out a bit. I'm not sure how much difference it makes.

Why is it, do you suppose, that in the blackened knee above, the rivets for the cuisse side of the knee are 'deeper' into the cop than those on the greave side?

That looks rather more by way of accident than by design.

Mac

[James Arlen Gillaspie]

I think I just realized that we might be having a case of being divided by a common language. When I say 'gapping', I mean that the lames could move right past each other and create an opening. Is this what you mean by 'hyper-flexing'?

[Kristoffer]

Are you sure Mac? On the pictures posted earlier in the thread it appears the cuisse side rivets on the lame closest to the knee are further in then the greave side ones..

[Aussie Yeoman]

Yep, that's exactly what I mean James.

[Mac]

Are you sure Mac? On the pictures posted earlier in the thread it appears the cuisse side rivets on the lame closest to the knee are further in then the greave side ones..

I need to see a picture of what you mean. I think we're talking about the same thing... but I can't be sure.

Mac

[Mac]

I think I just realized that we might be having a case of being divided by a common language. When I say 'gapping', I mean that the lames could move right past each other and create an opening. Is this what you mean by 'hyper-flexing'?

There's too many things someone might mean by "gap". When I contemplate of the sort of "gap" where the lames fail to lock, and open up in a way that might admit a weapon, I think of that as a "crash".... but I certainly don't expect others to recognize that expression.
I think we need to settle on terminology.

Mac

[Aussie Yeoman]

Gah! James, I just realised I put the wrong link in my post above. Now edited.

What I meant to say was: James, in this thread: viewtopic.php?f=1&t=132841 there is a very good reason that the height of the cop at the outside rivets is smaller than the height of the inside rivets. Is it because the knee gains more distance when flexed on the inside of the knee compared to the outside? We never got closure on that issue in the other thread.

[Kristoffer]

I need to see a picture of what you mean. I think we're talking about the same thing... but I can't be sure.

Mac

https://lh3.googleusercontent.com/-8RcM ... 0_003.jpeg

Take this knee as an example, the rivets on the first lame on each side of the cop. It looks to me like the greave side has the rivet further out towards the edge then the cuisse side. This is the opposite from what I expected because of the top of the poleyn usually is wider then the bottom.

Am i;

1. Making sence?
2. Missing something?

[Mac]

I'm still not seeing what you're seeing. I think I need the proverbial "circles and arrows".

Mac

[Kristoffer]

Ehm, nevermind. Its my late night brain playing tricks on me. The knees being bent the way they are with the top lame still inside the cop got me confused. I withdraw my previous ramblings.

[Mac]

Cutting the slots will be just like cutting rivet slots for anything else. You begin with holes in the usual places, and then cut back at an angle so that the articulation will remain "locked" at it's greatest flexion. You do this by removing one bolt of the temporary assembly, and sliding that pivot point back a bit while maintaining the lames in their flexed position. Mark a point at the new pivot location Repeat this with all the pivots, taking care that the new locations will not let the articulation gap. You can than connect the dots, as it were, and the articulation will still function; but now it will do so with that pivots at any point along their slots.

Mac, in what direction is the 'back' in this instance? Away from the cop?

Right. That's not very clear.

I mean that the slots originate at their "normal" pivot locations, and them proceed deeper in to the lame, as opposed to closer to the edge of the lame.

Start with a normally articulated lame. Remove one temporary pivot bolt, and replace it with a rivet or a punch or anything that will function as a pivot and is easy to withdraw. Flex the lame fully, and draw a line on the lame along the edge of the cop. Use something that makes a thin line, like a pencil or a fine point marker. That line will be your reference. Withdraw the temporary pivot and slide that end of the lame back into the cop, while maintaining the same angle of flexion. Make a mark through the empty pivot hole in the cop onto the lame. Slide the lame back a bit farther, and make another mark. Keep sliding the lame end back and making marks until you a row of dots that seems about the right length for the slot. So long as you kept the lame at the same angle, and pressed tightly into the inside of the cop, any of those dots could serve as pivot locations.

Return the lame to its original position and replace the pivot screw. Remove the pivot screw from the opposite end of the lame and repeat the procedure with that end. You now have a line of dots at each end of the lame. Any dot from one end will work with any dot from the other end. Thus, you can turn those rows of dots into slots and the lame will function OK. That is to say, it will not "pull through", "crash", "pop out", or anything else you might like to call that particular failure mode. At the other end of its travel, ie. when the lame is pushed back in to the cop, it will be a bit loose, but that's OK.

This all really calls for diagrams, but I have not got a good idea how best to render them. I'm warming up to an essay on articulation, though. Sometime in the next year, I think.

Mac

[Aussie Yeoman]

Thanks Mac, very clear!

[Mac]

Ehm, nevermind. Its my late night brain playing tricks on me. The knees being bent the way they are with the top lame still inside the cop got me confused. I withdraw my previous ramblings.

I had a suspicion that it had something to do with the way the lames were flexed. I kept looking at it, and everything seemed like was pivoted more or less symmetrically, within the range of normal error.

Mac

[James Arlen Gillaspie]

A big part of riveted articulation of a joint is length expansion or retraction. An arm that has it wrong will have the cuff either move up your wrist or down as you bend your arm. Also, for knee cops, the S curve through the leg (viewed head - on) needs to be accounted for some how, and making the distance between the rivets on the inseam side longer than on the outside helps with that. Finally, the diameter of the cuisse, even at the bottom, is wider than the top of the greave, if you're normal.