The Armour Archive

I've just had a delightful time experimenting with soldering caps on rivets. Dave Rylak (Cet) dropped by and we employed our version of the scientific method, till we came up with a tentative protocol. We also had lunch at the local Indian place, and that was good too.

We found that the biggest problem was getting the solder to really wet to the steel. Our early tests failed easily, and although the solder had taken on the surface texture of the steel, it has not apparently really bonded with it. For these, we had cleaned the parts with HCl, and used 60/40 rosin core solder. The solder wire was wrapped around a mandrel and cut into rings to give supply a uniform quantity or solder.

Later, we switched to using Amco Blitz https://www.panthereast.com/item_brand/ ... amco_blitz on surfaces which we'd cleaned by grinding. This made a big improvement in strength. We also switched over to using a fluxless 50/50 solder. This eliminated the entrapment of burnt flux as an obstacle , but it made it harder to control the amount of solder we used.

We did most of the testing by making up a test assembly of a 14ga steel plate, a piece of 4oz leather, and a burr. These were riveted together lightly, as one might do for a lining strip. We than struck the test piece against the heel of the anvil in a sort of chopping motion so that the cap would catch on the edge of the anvil. In the worst cases, this knocked the cap off the rivet. If the solder survived that, we placed the cap against the heel of the anvil and hammered the edge of the plate until something failed. In our best examples the 1/8" rivet shank failed. So, it's clear that the solder joint can be strong enough, but it has to be good solder joint.

When we knocked off for the afternoon, it seemed like the important points were...
---to have the surfaces mechanically clean
---to use a good flux
---to pre-tin the surfaces
---to have enough solder that is gets everywhere it must

The problems we were still having were...
---bringing the rivet up to temperature, and getting the whole head wetted before it had a chance to oxidize
---positioning the rivet perpendicularly to the cap

Unlike soldering on the usual brass caps, the surface tension of the solder was not enough to get the rivet to self center and stand up. On one of our early tests, we cut some grooves in the head of the rivet to match the flutes in the cap. This showed some promise, and further tests need to be made.

Mac

I am not very familiar with soldering parts together, so I wonder about these images.
Where did the solder go under the corroded through caps?
They all seem hollow, not filled with solder.




I have seen numerous examples of soldered on brass caps where the brass has worn away.
Every example I can think of is still filled with solder.

Tom,

The solder forms a thin layer between the head of the rivet and the inside of the very point of the cones. If you filled the cones entirely with solder, they would be strong, but heavy.

I'll see if I can make one up this evening and cut in in half for a photo.

The small brass caps are a different beast. They are pretty much filled with solder, in which is embedded a rivet head of indifferent and irregular form.

Mac

This is a great thread! Filled with good info. ( AND fantastic pics of the various types of decorative rivets) I read something years ago about "lead filled brass rivets" And have seen examples where the brass wore through showing the lead. Someone told me they didn't fill rivets with lead, and I was seeing iron under the brass, and that made perfect sense at the time. ( why make armour heavier?) but now I see why, they just used lead tin alloy to solder the caps on. Sure it added a little weight, but was needed in construction! And they used the small amount needed for the job. There was a reason for the "madness" LOL.

Mac wrote:Tom,

The solder forms a thin layer between the head of the rivet and the inside of the very point of the cones. If you filled the cones entirely with solder, they would be strong, but heavy.

I'll see if I can make one up this evening and cut in in half for a photo.

The small brass caps are a different beast. They are pretty much filled with solder, in which is embedded a rivet head of indifferent and irregular form.

Mac

Thanks Mac,
I knew there was something I was missing.

Getting the rivet hot enough - do you put the cap on after peining the rivet?

If so, does enough heat transfer to the helm (like a heat sink)? Does it "cook" the leather inside, or does it not get hot enough?


I'm showing my lack of soldering knowledge

G--

I took some pics this afternoon to illustrate some things about the soldered cones.

Here is cone sitting in the special holder. That ring is rosin core solder. This is a pretty convenient way to get a standardized quantity of solder into the cones. I use this with "regular" brass capped rivets as well.



The rivet is set in place and the whole thing gently brought up to heat over a torch flame.



A soldered assembly sitting next to the holder.



One way we tested these earlier today was to assemble the completed rivet up as if it were securing a lining strip in a helmet....



...and then bonked the test assembly thus.



The one on the left shows a "bad" failure. The one on the right is a "good" failure.



Here's another view of good failure. The cone stayed soldered on, even though it was struck hard enough to shear the rivet shank. Ideally, we want a soldering procedure that will produce strong joints like this every time.



This is what the assemblies look like when sectioned right down the middle. That lighter colored line is the solder. This section was through the "valleys" of the fluted cone.



In this view, I made another cut to show the thicker solder line through one of the "hills" of the fluted cone.



Mac

Mac, those cross-sections are so instructive! A picture is truly worth a thousand words. Thanks for taking the time to section those tests and show them here.

Mac,

I see from your cross sections of your experiment that your rivets mate up pretty closely with your jello mold rivet caps. What is your plan to shape the rivets for the spirally fluted caps?

Anyway, last evening I attempted to drift my die larger to give clearance for the thickness of the cap.

First, my new dapping set was waiting for me when I got home:

http://i186.photobucket.com/albums/x171 ... xm9evr.jpg

So I dapped up some blanks from .032" 1050. I should have stopped right here and made some fresh discs to dap up. The punchings from the ironworker had a small tear across the punch nipple. You can see it in this picture:

http://i186.photobucket.com/albums/x171 ... dv2vdi.jpg

Next I tried to flute the cap by driving it with my spiral punch into a block of pewter. The results were not what I was hoping for. The flutes are not very defined, but I thought that when I drove the whole mess into a hot steel block, things would even out.

http://i186.photobucket.com/albums/x171 ... aveccp.jpg


This was not the case. Despite taking the sharp edges off of my punch's flutes as per Mac's advice, this test ended in failure:

http://i186.photobucket.com/albums/x171 ... ctxzzb.jpg

So for the next experiment I will make the following changes. First, I will make fresh blanks with no tears in the middle. Then, I will forge up a couple of more bottom blocks, the current one is getting a bit deformed from repeated attempts to drift it larger while hot. Last, I need a way to dap up a cap and get it to fit closely to the punch before I attempt to drift the whole mess into a block of hot steel.

I'll keep posting my progress here if there is interest.

Rob

Zanetto wrote:Mac,

I see from your cross sections of your experiment that your rivets mate up pretty closely with your jello mold rivet caps. What is your plan to shape the rivets for the spirally fluted caps?

I ground that one down a bit around the edge. It's a better fit than I expected. I think that for the spiral flute caps I will make up a punch to reshape the rivet heads to look like those ones on Churburg S30.

Zanetto wrote:First, my new dapping set was waiting for me when I got home:

http://i186.photobucket.com/albums/x171 ... xm9evr.jpg

Nice! I have punch envy.

Zanetto wrote:This was not the case. Despite taking the sharp edges off of my punch's flutes as per Mac's advice, this test ended in failure:

http://i186.photobucket.com/albums/x171 ... ctxzzb.jpg

I think I would try making up a dome out of mild steel for use as your clearance gaining punch condom. It will be easier to anneal, and you should be able to get it shaped up with a pewter block. It will probably take multiple punchings into the pewter before the valleys lie down properly.

Zanetto wrote:I'll keep posting my progress here if there is interest.

Please do!

Mac

Yesterday evening, I continued the soldering experiments. Instead of using the fluted domes, (which are now in rather short supply), I turned to the simpler model of soldering a flathead rivet onto a strip of sheet steel. The joint is then tested by striking the rivet with a crosspene hammer.



The rivet either breaks off more or less immediately , or hangs on for a while before giving up.



The desirable result is for the rivet to stay attached until the shank can be knocked all the way over.



I did quite a few of these, varying the surface preparation, the flux, the solder, and the procedure. (there are more of them in the trash, and scattered across the shop floor)



The results were surprising. It turns out that within broad limits, the surface preparation is not critical, so long as there was no obvious rust, and the flux was Amco Blitz. Neither does it appear to matter if the solder is the solid 50/50, or the rosin cored 60/40. Either of those will produce specimens that satisfy the test procedure. What does matter is whether or not the flux has had a chance to finish boiling and out-gasing before the solder freezes. If there is any chance that gas from the flux can get trapped under the rivet head, the resulting solder joint is weak.

A corollary to this study is that the protocol I have been using for the normal brass capped rivets produces the weakest sort of joint. What I had been doing there is to "butter" both surfaces with a Zinc Chloride paste flux (Nokorode) and then assemble a sort of sandwich of cap/solder ring/ rivet. It turns out that some of the paste flux will get trapped against the face of the rivet under all circumstances. What's more, as the ring of solder melts, it creates a seal around the edge of the rivet that keeps gases from escaping. I had known that this procedure sometimes lead to cap failures and air voids, but now I understand how and why. Next time I need brass capped rivets I will have to do it differently.

Mac

Very neat stuff Mac.

It does make me question my own process when making the capped rivets on my Gorget though. I had problems with the rivet floating in the liquid pool of solder instead of staying at the bottom as yours appear to have done. Ended up using a pair of needle nose pliers to hold the rivet in place while the solder cooled to make it work. Which worked but was more fiddly.

Only thing different I did was not use any flux. Is that what made the rivet behave differently? Change in surface tension as well as deoxidant? Or am I completely off base?

Here was my process. There was some good discussion from folks here. Middle of page 3viewtopic.php?f=1&t=157286&hilit=Augsbu ... t&start=70

Mac
Just for reference, here is what you wrote in June 2009 in Jiri's brass caps on the rivets thread

Mac wrote:Here is the short version of the procedure I use to make brass capped rivets. There are probably as many ways to do this as there are armorers, but this will give you a start upon which you can experiment.

-Start with a convenient piece of brass shim stock. I usually use .012" (.3mm). It comes in 6" wide (150mm) rolls.

-Cut off a convenient length. I usually take about 6 or 8 inches (150-170mm)

-Anneal the brass.

-Remove the oxide from one surface, so that your solder will adhere. I usually use an 80grit greaseless compound on a sewn cotton buff. This will require a backing board to keep the, now soft, brass flat while you clean it. Take care not to catch the edges in your buffing wheel!

-Cut the brass into strips about 5/8" (15mm) wide. I use a scissors.

-Cut the strips into squares.

-Using a punch and a tin block; stamp a dome in each square of brass. Be sure that the clean surface will be the inside of the dome.

-Take 1/16" (.158mm) rosin core electrical solder, and wrap it around a 3/32" (2.5mm) rod.

-cut the solder into rings.

Note this gives a measured amount of solder which is right for my caps. You may need more or less solder.

-Remove the oxide from the heads of your rivets. I use an 80grit greaseless compound on a sewn cotton buff. I hold the rivets by their shanks with a big pin vice.

-Wipe a little zinc chloride soldering flux over each of the rivet heads. This step is optional.

-Pick up a brass dome by one of its corners, using a needle nosed pliers for which you care little.

-Put a solder ring into the dome.

-Put a rivet (head down, of course) into the dome. You will now see why I make rings of the solder!

-Hold the assembly carefully over a gentle torch flame.

-Heat until the solder appears around the rivet head. Capillary action will probably pull the rivet into an upright position. If not, gently poke at the rivet with a bit of wire until it is straight.

-As soon as the solder is melted, gently remove the assembly from the flame.

-Carefully set the assembly down to cool.

-Repeat until all your rivets are soldered.

-Take each rivet ( now with a cap) and trim the excess brass away.

-Polish the brass caps by whatever method you will.

-Electro-gild if required

I will be pleased to answer any questions about this.

Mac

I'm switching gears again today to address a problem we had at the fitting. It seems that I have the cuirasse sitting too heavily on the hips, and what was mild discomfort early in the day was bruises by evening.

Going back and looking carefully at pics I made at the fitting show that while the shoulder straps are tight, they are not doing enough to bear the weight. This detail tells much of the story. The angle between the strap and the breastplate shows that the strap is pulling more "back" than "up". Likewise, the upper backplate does not seem to bearing on his back.



My plan is to bring the upper breast and upper back in somewhat so that whey may bear some of the weight. This should effectively shorten the height of the cuirasse from waist to shoulder without actually removing anything. By doing this, I hope to lift the armor up by at least 1/2". The other think I hope to achieve here is to bring the bevor in closer to his face. I have already reworked the bevor to make the upper plate sit closer and a bit lower as well. I did that "in secrete" because it required a lot of concentration, and I did not feel as thought I had the brains left over to use the camera... (sorry).

In preparation for this modification, I have adjusted the padding on the armor stand to get the armor to sit more accurately. Here is a pic taken from the same camera angle as the one above, and then cropped the same way so as to make a better comparison.



Since I plan to make the bevor sit more closely in this process, I need to have some record of where it currently is. I snipped up a cardboard template and then copied it in masonite. It's not perfect, but it will do what it has to.



The process will begin this morning with the neckline of the breastplate.

Mac

Mac, I think the reason your so good at what you do, is you always strive for absolute perfection. What would be "good" for most isn't good enough for you. I think that's both a blessing and a curse! LOL. I tip my hat to you again Sir! ( being the fine artist and a sculptor in steel that you are helps too!)

A fine mechanic and engineer as well, cant forget that. Your an artist and a thinker. And then the scientific experimentation you do, cant forget that! ( testing rivet assemblies, attempting to figure out exactly what works, the geometry of articulation ect. ) Your a true Renaissance man! Your an inspiration, thanks for sharing your knowledge, ideas and theories!

Mac,

I recall a discussion (maybe from one of Dave's threads, I will try to look it up later) about shoulder strap attachment to backplates. I posted pics from the KHM and Thun Sketchbook that seemed to indicate the buckles (one those examples) were probably were on short straps attached a few inches down from the top on the inside of the back plate.
We were theorizing that this might help tge strap engage / conform to the shoulder better.
Do you recall that discussion?
Any current thoughts or comments on the merrit of such a configuration?

Otto von Teich wrote:I think that's both a blessing and a curse! LOL.

Indeed, Otto. Sometimes I wonder if I would be happier if I were normal.

Mac

I began the work of making the cuirasse sit higher by marking where it sits on the armor stand padding....



....and making up another gauge that's the shape of the profile of that breast without the bevor.



After reshaping the shoulders and the neckline in general, I found that I had moved the neck in a bit. A bit more might be better, but at this stage, I think this is all I can ask for, and any little bit is to the good.



Having brought the shoulders of the breast in, and mounted the straps a bit lower on the back, we can see that things looks a bit smoother.

Before... ...after



In addition to moving the attachments lower on the back, I switched things around a bit. Having looked at the gothic breasts and backs in Goll's pics, it was clear that the most typical thing is to have the straps attached to the back. I made wider ones while I was at it.



I also took a few minutes to put rollers on the buckles before I stuck them on the breast. These really make a difference in a thing like this. When I make up the permanent buckles to replace these I will definitely put rollers on them, and that goes for the one on the waist belt as well.



Between bringing in the neck, and reshaping the gorget plate of the bevor to fit that, I managed to bring the "nose" back about 1/2" or so.



I loose a bit of that when the bolts are loosened, though. The buckles push up on the gorget plate. I will have to reshape the shoulders of the plate a bit to fix that.



Mac

Tom B. wrote:Mac,

I recall a discussion (maybe from one of Dave's threads, I will try to look it up later) about shoulder strap attachment to backplates. I posted pics from the KHM and Thun Sketchbook that seemed to indicate the buckles (one those examples) were probably were on short straps attached a few inches down from the top on the inside of the back plate.
We were theorizing that this might help tge strap engage / conform to the shoulder better.
Do you recall that discussion?
Any current thoughts or comments on the merrit of such a configuration?

I took that to heart, and moved the attachments down another 3/4" in back. I may move them more later.

Mac

Making the cuirasse sit higher has two components. The first is to modify the top of the breast and adjust the strapping so that it can be pulled up by the shoulder straps. The second is to make the cuirasse wide enough that it can be worn in its new position.

The lower part of the cuirasse is a sort of a cone, and so is the human back from the waist to the armpits. It has been clear to me for a while that the cuirasse did not look wide enough or "round" enough between the waist and the armscyes. That needed to be addressed.

The first thing I did was to make a tracing of the sides of the breastplate to have as a reference.



Next, I needed to make a brace to keep the waist from spreading while I manipulated the rest of the breastplate. I made up a strap out of some scrap steel. That strap secures to the fauld pivot holes.



I then cut the ends of a piece of EMT so that I could wedge the upper edge of the plackart apart about as far as I thought I needed it to go. Playing a torch alternately on one side and then the other made the new shape permanent.



I then turned my attention to the underlap of the upper breast. It looked like it could loose at least an inch. That will help me make the sides of the breastplate a bit rounder.



So, I thought I would start by cutting off about an inch. I can cut more later if that seems like a good idea. I was surprised that that this old blade did the job in spite of the occasional shower of sparks when I cut through a harder spot.



With the breast reassembled, I used that same pipe and torch technique to widen the upper half of the sides.



A little spot heating and bossing out from the back made things look more lively.



Returning the breast to the earlier tracing shows the how much the sides were spread. Note that the waist stayed the same because of that temporary strap.



Looking at the sided of the assembled cuirass from the back, you can see how much breastplate width I gained in the process. The two gaps are not equal, and that will have to be addressed tomorrow.



The front view looks better now, although you can see that asymmetry. I will have to decide tomorrow which side needs to be corrected.



Mac

I did a bit more of that business with the length of EMT jammed into the breastplate to push out the right side. The trick is to only heat the side you want to move.

After that, the day was mostly about getting the backplate to match up better to the breastplate. That happened in two phases. The first was to clamp the parts together and heat the sides of the back to let the parts relax into their new shapes. I worked from one side to another, and moved the clamps around as they were needed. This is one of those processes where you can't have too many clamps.



After that was all done and cool, I disassembled everything and smoothed out the kinks and dents that the clamping an heating caused. The rule of thumb is to hammer on the outside if you want to part to treaighten, and on the inside if you want it to curl up. Since this was all about getting the sides of the backplate to open up, most of the planishing had to be from the outside. I will probably do another round of those two steps before I am done here, but it's starting to look OK.



Unfortunately, I did not take a pic from this angle before I pushed the sides out. I hope you will take my word for it when I say it has improved the look of the armor as well as (I trust) improving the fit. Giving it a bit more rib cage room was a win/win modification.



Mac

Mac wrote:

Tom B. wrote:Mac,

I recall a discussion (maybe from one of Dave's threads, I will try to look it up later) about shoulder strap attachment to backplates. I posted pics from the KHM and Thun Sketchbook that seemed to indicate the buckles (one those examples) were probably were on short straps attached a few inches down from the top on the inside of the back plate.
We were theorizing that this might help tge strap engage / conform to the shoulder better.
Do you recall that discussion?
Any current thoughts or comments on the merrit of such a configuration?

I took that to heart, and moved the attachments down another 3/4" in back. I may move them more later.

Mac

I finally dug up the images in question, I am adding them here for completeness:
Corey's Original Thread - Breastplate & Backplate Overlap


A58 at KHM (note: straps should be attached to the inside via the empty holes)


Thun Sketchbook detail

I don't know why I didn't chime in on that earlier thread. A 62's backplate is wider than the inner dimension across the breastplate, and it takes a bit of a trick to get them mated, as the spring tempered steel can throw back two men at once who don't know the trick. The difference in width means that the breastplate can take a substantial hit without coming free of the turn pin and spring pin, which it would have to break off anyway. I think it's a nice feature that makes the wearer much more difficult to crush; it is conceivable that a fellow wearing a cuirass that is only belted could be rendered unable to breath if his horse fell on him. I don't think that the backplate had been sprung a bit wider by a severe impact, as the breast is more likely to have this happen. It seemed very intentional, as the outside waist widths of breast and back are identical. Oh, and the Armamentarium Heroicum shows a very narrow waist belt. This does not guarantee that this was original, however, and Ferdinand II's people felt free to punch holes, add turn pins, and do whatever else they felt like to make things do what they wanted.

The photo shows A 62's current backplate buckle. There do not seem to be a pair of holes there. The backplate has a very rare feature, namely pivoted gussets at the armholes like those in the front.

Today's activity was mostly about dealing with some small matters of fit. I made use of my new clamps again to get the bevor to fit nice and snug to the breast.



The gorget plate is now flared enough that it will not foul the shoulder strap buckles.



The rest of the day was spent cleaning up the shop. Anyone who has ever seen my shop will recognize that this is about as clean and organized as it ever gets.



Mac

This post marks the end of the spring armor season. Tomorrow is the first day of the Billy and Charlie's Pewter Good's preparations for Pennsic. Between now and the end of July, we need to pour about 75 pounds of pewter, clean a few thousand objects. and make molds for another fifteen or twenty new items.

I will still check in here, but I expect that the armor will just sit like this till mid August.



Mac

Noooooooooooo! How can we now get our fixes?! Need.... more... armour... porn...



Good luck in the pewter world Mac! Dont burn yourself.

Love the amount of light in your shop, jealous of the amount of table space at your disposal. That knock-down table looks like it comes in handy.

Do you do anything to the armor, as far as preventing rust during the interim storage?

Xtracted wrote:Noooooooooooo! How can we now get our fixes?! Need.... more... armour... porn...



Good luck in the pewter world Mac! Dont burn yourself.

Thank you! I hope to keep the burning to the usual couple of small splashes.

Mac

Keegan Ingrassia wrote:Love the amount of light in your shop, jealous of the amount of table space at your disposal. That knock-down table looks like it comes in handy.

I boosted the exposure a bit in that pic, but in general it's bright enough in the shop. There are eight of those high output fixtures on the ceiling. Someday, I need to sheetrock that ceiling, but that will have to wait till the armor project is done.

Table space is a two edged thing. It's nice, but it invites clutter. That trestle table has been with me for a while now. It's our work surface at Pennsic.

Keegan Ingrassia wrote:Do you do anything to the armor, as far as preventing rust during the interim storage?

It will be OK like it is. I'm not too worried about rust at this stage.

Mac

I will echo the chorus of "Nooooo! I don't want to wait!" I have read this thread with excitement since its inception.

I am jealous of your patron, as despite his long and patient wait, for no matter what the fee was, he is paying far too little for such excellence.

Mac.

Please start a new thread for your casting projects. We'd all like to see the process and design work

We'll (im)patiently await your return!

Mac wrote:This post marks the end of the spring armor season. Tomorrow is the first day of the Billy and Charlie's Pewter Good's preparations for Pennsic. Between now and the end of July, we need to pour about 75 pounds of pewter, clean a few thousand objects. and make molds for another fifteen or twenty new items.

I will still check in here, but I expect that the armor will just sit like this till mid August.

Mac

Do you do anything as a rest preventative in the mean time? Spray with WD-40 or oil? Throw a tarp over it to prevent dust? etc?

Sean

Sean Powell wrote:
Do you do anything as a rest preventative in the mean time? Spray with WD-40 or oil? Throw a tarp over it to prevent dust? etc?

Sean

Nothing at all, Sean. When I come back to the project, I can either deal with a little light rust getting all over my hands and clothes, or I can deal with oily rust doing the same. I guess I'll take the light rust. Later in the project the choices will be different, but for now, it's OK.

Mac