Just a quick update..

After the past year of not having much time to do smithing and other things in my life.  I finally took a leap of faith and left my current job for a new one.  This job promises to be a lot less stressful which will hopefully translate to more forge time.  The reason I haven't posted over the past month is that I was going through the motions of changing jobs.

Anyway, I have a weekend of smithing planned...

Will post pictures tomorrow...

Just a quick update.

  Today I made a new twisting fork.  It is used to help bend the circular shapes of the clock pieces.  Also started to make the outer rim of the crown wheel.  I welded it closed but it got dark before I could really examine it to see how it came out.  I took a few photos but need to pull them off the camera.  I am planning on do some more work tomorrow and will post the pictures then.

A good set of books...

  I just purchased a batch of books from the York Archaeological Trust.  I haven't had time to go through all the books yet but one of them was well worth the cost.  The book "Anglo-Scandinavian Ironwork from Coppergate" is definitely a book worth owning for anyone interested in Medieval blacksmithing.  They have quite a few tools documented.

  One batch of tools is identified as punches.  The punches (they are a very basic design) look like something that could have come from my own toolbox.  There are even some that look like the punches were distorted during use.  As a punch is used, it gets hot, hammered against the anvil and it either mushrooms or bends.  A few of the punches in the collection appear to have done just that.  So a medieval smith, like me, was constantly having to retouch and reshape his punches.

  A bonus from this book that I was not expecting was a section on locks including one that closely resembles the Mastermyr lock I reproduced a few years back.  I think I may have to try some of the locks sometime in the future.

  Also in the book was too micro fiche.  I was not expecting that.  They contain additional information including metallurgical information.  I will have to see if the local libraries have a fiche reader so I can see what's on them.  I would like to get a better idea of common levels of carbon content on the tools.

Good day

  Started the day with doing more work on the remake of the crown wheel hub.  Basically this just involved straightening the metal out a bit more, balancing the center, etc...  After that, I punched the hole for the arbor.  If you look carefully on the anvil, you might see a white line.  This was used to ensure the spokes were square.





  It is not important that the arms are of different lengths as they will be trimmed later on.

After looking at the state of things, I decided that it was necessary to add a washer between the crown wheel and the smaller gear that contacts the main wheel.  If I didn't do this, I would have no end of problems down the road as the crown wheel and main wheel would contact and possible stop the clock.  If I recall on the original there was a similar washer.

Unfortunately there was not enough room on the arbor for the washer.  So I had to lengthen the portion of the arbor that holds the crown wheel.  These pictures are before and after.  Notice on the left that there is an additional 1/2" of the square portion of the arbor.







You may also notice that reworking the arbor meant that it grew an extra 1/4" to 1/2" on the right side.  That will be corrected on another day.

Then it was time to make the washer.  This was simply a matter of punching an arbor hole in a piece of metal then making it round.  This picture shows trimming the corners to round it up a bit.  After that, I did a bit of grinding and filing to round it up more.



After finishing the washer, I needed to punch a slot in the arbor for the crown wheel.  This allows to insert a pin for to hold everything on.  I forgot to take some pictures until I was almost done.  This slightly out of focus picture shows the pin going through the arbor.  To the right of pin there is the hub for the crown wheel, then the washer, then the smaller gear wheel (no teeth on it yet.)

Contacting me...

Over the past few weeks/months, I have noticed an increase in traffic on this site.  Because I know Xanga kind of locks things down as far as replying if you are not a member, I've decided to give out my email.  Just make sure that if you contact me, you put something in the subject that will let me know you are not spam.  A reference to the Xanga site or clocks will get my attention.

My email is (remove the spaces and do the appropriate replacements)    medieval smith at gmail dot com.  If you have any questions about anything I've done or you have suggestions, I do encourage you to contact me.

We've been here before...

Today I decided to start working on the replacement crown wheel.  I started the work on the new hub.  This is something I've documented here before.  Essentially a straight piece of metal is split on both ends and opened into an "X" shape.  Here are some pictures that were taken during the making of the new part.

The work on the hub is not complete.  The arms need to be evened out and the center hole punched.  My arm was getting tired (I haven't been doing enough smithing recently) and the afternoon rain showers were threatening.

Finally, I could not help but take a photo of all the parts on the clock.  It is important to remember that most of the moving parts in this photo will be remade.  But they do give me a good idea of where I've made mistakes and how to correct for them.  If you look carefully, you can see how the main wheel will contact with the smaller gear that I started earlier this week.

Things got moving again.

  Today I did take off of work so that I could get something done on the clock.  The day started with some basic tool maintenance such as cleaning, sharpening and reshaping.  Then I turned my attention to the crown wheel.  After looking at things I decided that the best thing to work on was the small gear that is on the same arbor as the Crown Wheel.  This gear contacts the main drive gear.  The reason I decided that I should work on this piece is that the width of this gear directly impacts the placement of the crown wheel.

  I needed this gear to be a little thicker, around 3/4" thick.  Unfortunately I did have the right sized stock for it.  So I started by folding a piece of stock to the correct thickness and welding this together.



  After that, a hole was punched, drifted a little larger round and then drifted square to match the arbor.

 



  Next, I marked a rough circle to cut away some of the waste.



This shows the piece close to a circle shape.  The tool to the right is a hold down tool that held the piece stable while I was cutting it. 




  Finally the piece on the arbor to show its placement.  I used the old crown wheel to show how they relate.




  I will hold off shaping too much of the small gear.  Since it directly relies on the size of the drive gear I will not make it into a circle or cut the gears until the main wheel is done.  It doesn't make sense to put a lot of work into something that may need to change anyways.

  I did some test filing and there was no evidence of the welding.  If the welding had been done improperly, there would have been a seam line that could have potentially failed.

  I would have gotten more done but a typical Florida summer storm came along in the afternoon and cut the day short.  It felt really good to work on the clock again.

 

Workday tomorrow

Well, with work slowing down I have gotten anxious to get back to the clock.  So I am taking the day off tomorrow and going to try to get a full day of blacksmithing in.  My goal is start and maybe getting mostly done a new crown wheel.

Some more thoughts on the Crown Wheel

This evening I was going over some photos a friend had taken of the Salisbury Cathedral Clock  for me about a year ago.  In looking at the photos, there was a technique used for attaching the outer ring of the Crown Wheel to the spokes.  I think I may adopt this techinque because it makes a lot of sense and is a lot easier.  Essentially, on the back side of the Crown Wheel, four squares are cut out and then the spokes and hub are just dropped in from the back.  Then, the spokes are riveted over.  These two pictures show this.  Click on them to see a larger view.

This makes a lot more sense then what I was trying to do.  I wish some better pictures of the specific clock that I am modeling.  The only photographs I have are photo copies of a BW photo.

There is one minor problem with using this Crown Wheel as a source.  The original clock's Crown Wheel is lost to history and this one is actually a reproduction.  However, it makes sense and I am going to go this route.

 

5 years ago...

  I just happened to notice today that I have now been using this site for 5 years (joined September 14, 2003 and first post on October 31st, 2003).  And what that means is that I've been working on this project off and on over the past 5 years.  I've gone through periods of a lot of progress and learning and I've also gone through periods of inactivity that can last close to 4 or 5 months.  It has been a very challenging, frustrating but yet rewarding project.  Having said that, I really hope it doesn't take another 5 years.

Finally

Well work has finally slowed down that I can start looking at putting some serious work in on the clock.  Also it is starting to be blacksmithing season.  During the months of May-September, things get very uncomfortable in Florida.  Between daily rain storms and 90+ temperatures, it is a steam bath.  Well things are starting to cool off.

I was going to do some work today but I noticed that my tools needed some attention.  Because of the demos I do on the road and my erratic work schedule, they needed some attention.  So I decided rather than jumping right back into it to take the time and do some basic maintenance and repair.

The bad:
  About 1/3 of my files were exposed to too much moisture at some point.  They will be recycled into some other project, perhaps knives or fire starters.

  The tool boxes I made about 2 years ago for demos just are not cutting it.  I made them quickly (3 of them in 2 hours) and because of a simple mistake, they always left the tools slight exposed to the elements.  Also, they are not specific to the tools they are carrying.  I hope to remake them to have boxes specific to the drifts, chisels, punches, hammers, files, tongs, etc...

  Several of the chisels, punches, etc... need some attention.  I may do some work on those yet this afternoon.

The good:
  I had a lot more punches, chisels and drifts then I thought and in a lot more variety.

  I was able to clean up a lot of my files and get them wrapped up in an oil cloth so that will hopefully stop any rusting.

  I put new handles on two of my hammers and have them soaking in a linseed oil bath.  I've talked to a few smiths online and they recommended soaking any hammers in linseed oil after putting a new hammer handle on them.  The linseed oil fills the voids, swells the wood (tight fit to the hammer head) and prevents them from drying out.  I have them in a mixture of 75% linseed and 25% mineral spirits.  The mineral spirits thins the linseed and helps it to penetrate the wood.

  I was able to clean up my work setup to make things quicker and nicer.  Hopefully during this week I will get some work in.

********

  Finally a note on the actual clock!!!  When I last left the clock, I had done a fair amount of work on the crown wheel.



  Although it may look okay, there are some balance, symmetry issues.  I think I will try to make this in a different way.

  In the first technique, I took a straight piece of metal and cut the teeth.  I then wrapped that into a circle.  Next, I made the spokes and hub to insert into the crown wheel.  This was very difficult to get inserted properly.  Part of the problem is that in the process the wheel got a little out of whack.  It is difficult to get it back into shape without damaging the teeth.

  I think this time I will start with just a straight piece of metal.  This will be wrapped around the spokes and hub then welded closed.  After that, I will true up the circle on the axle.  Once I have that trued up, i will then cut the teeth.  In other words, I going to get the circle true before I cut the teeth.  And that really makes sense now that I think about it.  Any gear on the clock needs to be a true circle prior to the gears getting cut.  It is only common sense but that is part of the learning process.

Some good forge time

  I spent a fair amount of time on the forge this weekend.  Nothing to do with the clock.  It was mostly work related to a primitive fire starting class I taught several times over the weekend.  Still, it felt good to get back to work and hopefully over the next few weeks I can get back into doing some more work.  At the very least, I expect over the next few months for work to really settle down.

Finally, some forge time

  The past few months have not been fun.  Way too much work, way too little time.  But, I finally got back out to the forge.

  It really wasn't that exciting.  This weekend I am teaching some classes including a fire starting class.  As part of that class, I am giving out fire starting kits.  So this afternoon I started forge some basic fire starters.

  I am really glad that I didn't work on the clock.  Because I haven't been doing any work on the forge, my muscles were out of shape.  It didn't take much for me to get tired.

  This weekend I will be doing some smithing as well but probably won't be able to get to the clock.

I promise, something new soon...

  Work sucks...  I can't go into too much detail here because I don't want to get into trouble but my company has been preparing for a possible strike and I've been in the middle of those strike preparations.  It hasn't been fun and it has kept me busy.  But it looks like things are very close to be resolved so hopefully I will be doing more here soon.

  Anyway, the only real news to post it that I have begun to assemble materials to make a set of bellows.  Right now for my forge I use a hand crank that is about 100 years old.  I've been wanting to make a set of bellows for some time.  I have mostly wanted them for when I do historical demos.  It is hard to look like you are doing something medieval when you have a 20th century hand crank.

  So I picked up some duck cloth for the bellows yesterday.  The canvas will take the place of leather.  I'm using canvas instead of leather for cost, maintenance, simplicity, etc...  The bellows that I am making will be about 5 feet long by 3 feet wide.  They will be a style of bellows known as double lung.  Although this style is more colonial than medieval, the average observer won't know this.  I am using this style because the true medieval style works best when you have a second person working the bellows and they tend to "puff" more.  In other words they alternate puffing and not puffing much like a set of fireplace bellows.  The double lung bellows tend to give a more constant stream of air.

  I hope to start providing real updates soon.  This has been a very frustrating stretch for me.

Hopefully some work soon...

Still alive...  Unfortunately over the past few months, work has been extremely busy.  I have had very, very little personal time and what little time I do have usually coincides with the afternoon rainstorms.  For those not from Florida, during the summertime we get rain almost everyday at about 4-5PM.  Since I work outside, this prevents me from working.

Anyway, I hope over the next few weeks that there will be a bit of a slow down and that I can get back to some work on the forge.

Sunday work

  Sunday started with some more work on the hub.  Essentially I had to draw out the arms of the spokes, straighten them in a perfect "X" and punch the hole for the arbor.  Finally, I had to create small tenons to connect into the outer part of the Crown Wheel.  Not much here to show that I haven't shown before.  This picture is just prior to forming the tenons.

Next, on the crown wheel I had to punch the holes for those tenons on the spokes.  I messed one of these (made it too big) but was able recover from it.  The really difficult part here was to force the spokes into wheel.  This really took a lot of muscle, hammering, clamps, etc...  to get it into place.  In the process, the wheel became slightly distorted.  I think I can do some corrective action.  Hopefully the wheel is still in good shape.  This picture shows the spokes in the wheel.



Finally, some pictures of the nearly complete wheel in the frame.





The main problem with the wheel is that the arbor is not perfectly perpendicular to the wheel.  This causes a wobble when the wheel is spun.  With my experience of the great gear, I know I can correct this to some degree.  Hopefully it will be enough.

Even if this wheel does not work out (I hope that it does!), the important thing for me to remember is that I created this in under two days.  If you look back in my blog, when I first attempted the great wheel, that was over several months.  My experience level is getting to the point where I am much faster, more confident, blah, blah, blah...  :)  I will beat this project yet...

Some more work on Sunday...

  I did get in some more work on Sunday.  However, I have not had a chance to upload photos yet.  I will do a longer entry this evening.

Good progress day...

Today I wanted to get started on the crown wheel and I was able to make good progress.

I started with the hub of the wheel.  This is just like the hub I did for the test great wheel I did months ago.  The metal is split into a "Y" shape and stretched out.  Then it is flipped around and repeated.

This is the last photo of the hub that I took.  I wanted to hold off on finishing it to make sure it would work with the crown wheel.

Next I turned to the crown wheel.  The crown wheel has a diameter of approximately 8" with 33 teeth.  Something of an interesting note.  An article that I have used for a lot of my source information had some interesting theories on how you would lay out 33 teeth.  He believed that since the number could not be easily split into halves that it would be difficult to layout with a divider.  He had some interesting theories on how you could layout the teeth.

In reality, it was quite easy with the divider.  I divided by 11 then divided those by 3.  I started by laying this out on a straight bar.  It was about 1/8" thick by 1 1/2" (I think).



I then removed the waste with a hacksaw.  Because of the thinness of the metal, this was fairly easy.  I am still a little bothered that I don't have more evidence on using hacksaws but it seems perfectly logical.  Medieval hacksaws do exists and they could easily be made to be hard enough it is just that I do not know of any woodcuts, manuscripts, etc...  that show a smith using a hacksaw.



You may notice a little tab on the end of the stock.  I was originally going to rivet the wheel together but I decided to braze it instead so the tab was cut off.  After cutting the metal, I began to curve it into shape.  This shows the crown wheel in the fire.



Finally, after curving the wheel, I took a photo of the crown wheel with the verge and foliot just to see them together for the first time.  The tabs are not yet on the verge and foliot and I haven't placed the hub in the crown wheel.  Also at the bottom of the photo, you can see that the crown wheel is not yet welded together.  I want to hold that off until I get the hub in the crown wheel to make it safe.



Finally, a picture of the crown wheel on its own.  I did double check and when it is brazed together, the teeth are perfectly spaced  (probably no more than 1/32" inch variation.)

Clockmaking weekend...

  I have tried to clear my schedule for this weekend.  I am going to make the serious effort to do the crown wheel this weekend and get the clock ticking.  At this moment, I don't have any commitments either personally or professionally so I am really hopeing to get a solid 2 days of blacksmithing in.

  The key right now for me to work on the clock is simply time at the forge.  I truly feel that I have all of the skills (although I am sure I will make plenty more mistakes.)  I've also noticed that over the past 6 months in particular that when I do work at the forge, I have a lot more confidence and that I am working much faster.

  I may try to get some evening work in as well this week.  With summer coming on, there are a few hours after I get home that I can work.

More of a note to myself...

  I am hoping to get started on the Crown Wheel next week and I was just mulling over an issue I had with the wooden version of the clock.  Before that, a reminder of how the crown wheel and verge and foliot work together...  This is taken from the Wikipedia entry for Verge and Foliot

  In my clock the Crown Wheel will be turned 90 degrees from what is depicted here.  In other words, the arbor will be horizontal rather than vertical.  It will also have 33 teeth.  In addition, the verge and foliot (the bar in the top of the diagram) will be vertical rather than horizontal.  The way this works is that as a weight turns the crown wheel, it catches one of the pallets (letter q in this picture) and that rotates the verge and foliot.  Then, just as the Crown Wheel rotates one of the pallets out of the way, it catches the other pallet and it has to rotate it the opposite direction.  This repeats over and over.  So the Crown Wheel is always trying to rotate the same direction but to do so, it has to rotate the verge and foliot one direction, then the other.  This oscillation is the "tic toc" of a medieval clock.

  Anyway, in the wooden clock that I made, I had a lot of trouble getting this to work.  The problem was that sometimes it would jam or skip teeth in the Crown Wheel.  Sometimes the pallet would not swing into place in time or it would swing into place too early.  Fine adjustments with sand paper was required to get it to work properly.  Even then, variations in humidity affected this.

  I think when I made the wooden clocks, I was trying to have the pallet engage the Crown Wheel as soon as the other pallet released the Crown Wheel.  In other words, it would already almost be touching.  I think the solution is quite simple.  I am going to try to design it so that there will be a significant gap (maybe about 1/8 to 1/4 inch) when one pallet disengages to where the next one engages.  This should cause a suddenly slip in the Crown Wheel as it rotates that gap to engage the other pallet.  It will also lead to a more distinctive ticking.

  If you don't understand this, don't worry.  I just made this note because I've been mulling this over and I wanted to get it down on paper and make myself think it through.
 

Good to be working again.

Well, work and health issues kept me from the forge but I finally got a full day in today...

Last time that I did serious work on the clock, I had been playing around with the crown wheel arbor.  The crown wheel together with the verge and foliot perform the medieval version of the pendulum.  They help regulate the running of the clock.

Well last time I work on the clock, I noticed that one of the frame pieces had a few problems.  I had punched the crown wheel arbor a little off center (not that big of a deal but visual distracting) and I had also placed it a little too high.  Also, the mortises for the piece were a bit too far apart which meant that the piece had to be convinced to get on the frame by using a hammer.  Luckily this particular piece is not permanently attached so I decided to go ahead and remake it.

I started by punching the lower mortise on the piece.  This was done using a simple rectangular punch.  Next I needed to drift the hole for the great wheel arbor.  I took this picture because I wanted to emphasis one of the benefits of traditional blacksmithing over metal fabrication.  The frame piece is about 1 1/4" wide while the hole needed was 1".  If I simply drilled the hole, there would not be enough metal.  If you look at the following picture, you will notice the circle drawn on the piece.  This shows how little metal would be left.  The top piece is the new frame piece while the bottom piece is the old, bad piece.

 

To start, I make a small slit in the piece that is the same as the diameter of the hole.



This is drifted using progressively larger drifts until the desired size is reached.


After the great wheel arbor is complete, I move to the crown wheel arbor.  In the top part of the photo is the piece of the frame that is in the middle of the clock.  The lower piece is the new piece and it goes on the end of the clock.  On the top piece, the arbor hole is actually in the "U" shaped piece of metal.  The reason for this will be obvious in the last photo for this entry.

Note that the great wheel arbor holes (right side of photo) are different sized.  This is intentional because of how the great wheel arbor is constructed.


The crown wheel arbor is punched then drifted.  This just shows the punch.  I forgot to take a photo after drifting.



This picture shows the old frame piece.  Notice the white lines.  I marked it to show how much the mortise needed to be shifted.  I then used those measurements to finish up the new piece.


After completing the frame piece, I then did some cleanup on the crown wheel arbor.  I had worked on it some in a previous session but it need some cleanup and a little reshaping.

In this final photo, I wanted to show the placement of the crown wheel arbor and the use of that "U" shaped piece of metal.  The crown wheel arbor is located just above the gear.  Notice that the left side of the crown wheel arbor rests inside that "U" shaped piece of metal.  The verge and foliot extends down through the clock and it can be seen between the "U" shaped metal.  So that "U" shape allows for the correct placement of the verge and foliot while also giving the arbor somewhere to rest.

I am still going to try to get the clock "ticking" by the end of the month.  It is important to me to feel like I've made progress.  Essentially this means I need to make the crown wheel and put the pallets on the verge and foliot.  I hope that this can be done with a few evenings worth of work.

Time to get precise...

Work is still keeping me busy (for about the next 2 weeks) then things should get much better.  In the meantime, as a mental exercise, I decide to come up with the basic math for gear rations and size.

In the past when working on the wooden versions of the clock, I did some basic math to make sure the gears would work.  However, the wood was much easier to correct if there was an issue.  It was really easy for me to redo a gear if I needed it.  With the metal gears, an awful lot of work goes into one of those gears.  I can't afford to fake it anymore.

The problem is that given a particular distance between two arbors (axles) and what ratio of movement was needed (for every turn of one gear, the other should turn six times), I came up with a fairly simple formula.

d = distance between arbors
r1 = radius of smaller gear
r2 = radius of larger gear
t = ratio between gears

This is what I came up with.
r1 = distance / (t + 1)
r2 = distance - r1

Anyone care to check me on this?  :)  My address is medievalsmith@gmail.com if you don't have a Xanga account.

I hope to get back to the forge soon.  When I do, I need to start being precise on my measurements, I am going to try to be very careful with this.  However, I also want to try and be precise using what they may have done in the middle ages.

I will use a divider to do the formula above.  For example, assuming a 6 inch distance and a need for ratio of 10 to 1, I will take the divider and attempt to divide the distance 11 times (remember the formula up above, ratio plus 1...).  The measurement of one division would be equal to the distance on the divider.  That will also be the radius of my smaller gear.  The other radius would be 10 of those divider distances.  It may sound complicated but it really is not that bad.  I wanted to come up with a simple technique that would not require the use of a ruler marked to very fine degrees.

Another nice thing about doing it this way is that what few examples I have seen of medieval clockmakers, they always have dividers but I've never seen one depicted with a ruler.

NOTE: When I refer to a radius up above, it is the distance if the gears were simple wheels with their outer edge touching.  For actually gear teeth I will have to add a bit to the size listed above to allow tooth contact.  In most cases that will a little under 1/4" added to both gears.

Another thing is that up until now when I have been working on the clock I have often been experimenting.  This means that I often rushed parts especially if I was doing it during a public demo.  From here on out, I am going to try and adopt the attitude of making the final part.  I don't want to have to redo many more parts...

Just a bit of an update.

  This past weekend I worked at a festival near Jacksonville.  I spent the entire day doing blacksmithing for the public.  I am surprising myself with my speed at producing half way decent looking items.

  Work is really keeping me busy.  :(  However, I am going to try to make time this weekend to get some time in on the clock.  I noticed a problem with the clock that I will show details of later.  Not a big issue (and to be honest, I'm glad that it happened) because I wanted to remake the one part anyway.

A bit of a helpful video.

  Found this on Youtube.  It is of the Salisbury cathedral clock.  Although it predates the clock I am working on, all the parts are basically the same.  At about the 30 second mark is some good video of the verge and foliot rotating as well as the crown wheel.

http://www.youtube.com/watch?v=5UKlVaGwJ78

More progress...

  Managed to spend more time on the forge today and made some decent progress.  Mainly I was working on the arbor for crown wheel.  Not really anything special about it.  The process was very similar to working on the arbor for the great wheel.  The arbor is mostly formed.  It just needs a bit of touching up.  I will post some pictures of it later.

Progress...

I am attempting to drive to getting the clock to actually ticking so I continued work on the escapement assembly.  Today I made two simple pieces.  The first was an "L" shape at the bottom of the verge and foliot.  This piece has two purposes.  The first is to hold the verge and foliot in place as it comes in contact with the crown wheel.  It is also a bit of a safety feature.  If the wire holding the verge and foliot breaks, this piece helps to keep it from falling through the clock.  It was a pretty simple piece and these pictures show it being made.





This final picture shows it riveted to the frame.  I have the verge and foliot resting in it.



  The next piece is mildly more complicated.  It is a "U" shaped piece of metal.  It is designed to hold one end of the crown wheel arbor.  Because of the placement of the verge and foliot, the end of the crown wheel arbor cannot hook into the frame.  This piece is designed to wrap around the verge and foliot and provide the arbor with someplace to rest.

  I started by rounding the end of a piece of metal.



The center part is then thinned out.



After some more thinning, a hole is punched and drifted to form the hole where the arbor will connect.



The other end is spread out and I punch two holes for rivets.



I give it the "U" shape.



The piece is riveted to the frame.  I used the old arbor from the great wheel to show how it works.  Notice that the end of the arbor goes into the piece rather than the frame.  This allows the verge and foliot to be in its proper positiong.