Mirror Clock Project Part 2 - Gilding

Gilding

Gilding is an important detail on early New Hampshire mirror clocks. Some have sadly been painted-over in gold paint over the years, while others have survived in brilliantly shiny condition even after 200 years. Here's an example of a wonderful clock that's still in excellent original condition:



When it comes to gilding, you have two basic methods that are used: oil gilding, and water gilding. Often the two techniques are used on the same piece for contrast. Oil gilding simply involves a quick drying linseed oil varnish (size), and the application of the gold leaf. In general, it produces a somewhat brilliant finish, but is slightly more matte than water gilding. Water gilding involves the use of animal glues, and clay boles, and the gold is applied with water. The water reactivates the animal glue in the prepared surface, which bonds the gold to the surface. It produces the most brilliant finish, and it can subsequently be brought to a highly reflective mirror finish (using agate burnishing tools).

If you are interested in a quick tutorial on water gilding, I would suggest watching this episode of the New Yankee Workshop:

How to make a Giltwood Mirror (24 min)



For this particular reproduction, I wanted a not-too-glossy gold finish, and I wanted the gilding to be antiqued and faded. I had also planned to use imitation leaf, which works best with oil gilding. That said, I still had to prepare the wood base in nearly the same way, using layers of sanded gesso, and a base layer of deep red.

Leaf

When it comes to gold leaf, you have several options and brands available. Imitation leaf is also known as Dutch metal, and it comes in several different shades (Gold, Silver, Copper, etc). Most of the imitation leaf comes in sheets 5" to 5 1/2" square. Real gold leaf comes in various karats and comes in small sheets about 3" to 3 1/2" square. Both types of leaf usually come in "books" of gold leaf contain 25 leaves. In addition to "loose leaf" (which is my preference) there is also transfer leaf, which has the leaf attached to a backing paper, and it must be pressed in place. Price-wise, the current costs for imitation leaf (depending where you buy it) is between 10-20$, and the average price of a book of genuine gold leaf is around 50$ or more.

Shown below is my current inventory of gold and silver leaf. Two types of imitation gold, one package of imitation silver, and two types of genuine gold leaf (the top one is loose leaf, and the bottom one is transfer leaf).



Another thing to keep in mind is that apparently imitation leaf can go bad. I wasn't really aware of this, but since it's basically a metal blend that is close to brass, it can become tarnished, and even develop spots, as was the case in one or two sheets from the book I used below. These had been in a plastic bag (the original packaging) and stored in the basement for several years, so be aware of this.



Only the first two sheets were affected, and only on one side. They were used for testing and samples. The rest of the book seemed fine.



For this project, I decided to use imitation leaf, since I did not want to invest the extra money in the specialized clay bole, and rabbit skin glue which are needed for water gilding. I was also aiming for a more antiqued and distressed look, so the imitation leaf would work just as well.

Gesso

Gesso is basically paintable plaster mixture which seals and smooths a surface in preparation for painting or gilding. Traditional gesso is a mixture of rabbit skin glue, whiting (chalk), and water. Recipes for gesso can be found easily online. I chose to use a pre-mixed acrylic gesso. You need to apply enough gesso to build-up a thick enough layer that you can then sand it smooth. I believe I applied about 8-10 coats of gesso on my turnings.

In this photo you can see one section of turning before and after sanding.



The sanding is NOT fun, and it's very time consuming. You want the gesso to be as smooth as possible, since the leaf will show all the imperfections. Since the door frame was fairly smooth, I did fewer coats (around 4).





In lieu of a red clay bole, I used a burgundy acrylic paint. The reason for the dark red paint is to give the gold a richer appearance, and as a background for any visible gaps and voids in the leaf. Other traditional ground colours for gold leaf include grey and ochre (mustard yellow). This was coat number 1:



The paint was a bit too transparent, and I had to do a lot of coats to get a nice deep burgundy. This was coat number 5:



I ended up doing 7 coats of burgundy to get a deep enough red. If I were doing this again, I would skip all this acrylic paint and just use an oil based burgundy spray paint. The spray paint would only need one or two thin coats, and it would give a smoother finish.



The paint was then polished down and smoothed as much as possible with #0000 steel wool.



The areas that won't be gilded are taped off with painter's tape, and the oil gilding size is applied. I made the mistake of applying ALL the size in one shot, since it was supposed to take 3 hours to dry, with at least 1 or 2 hours working time, however, the gilding took much longer than I had thought, and near the end, the size was nearly dry. The size I was using also came up to tack much faster than the 3 hours it should have taken. It was ready to use after only 50 minutes!



The imitation leaf is much thicker than real gold, and you can actually pick it up and handle it with your fingers without it falling apart or tearing. If you are using genuine gold, you must be very careful and gentle when lifting each piece using a gilder's tip. To cut the leaf, you are supposed to use a gilder's knife and cushion, but I simply used a sharp/thin palette knife, and a scrap of leather (suede side up). I don't plan to do a lot of gilding, so I'm not ready to invest a lot of money for the specialty tools that are usually needed.



Gilding finished, with the black portions of the columns freshly painted (one coat of black acrylic paint).





It was at this point that I decided to do a gold door frame, instead of a black door frame. I felt that the clock would be too dark, and most of the examples I found during all my research use a gold door frame.

Here is a close-up showing the gilding. I later went back and did some touch-ups on all of the large missed spots (especially in the cracks). You can see that a lot of the brush strokes from the burgundy paint show through the gilding (the surface ends up looking a bit wrinkled), and that it would have needed a much more thorough sanding for a smoother surface. Unfortunately Acrylic paint doesn't sand that well, and I would have been much better off with a red (oil based) spray paint as mentioned earlier.



To save time and to get a slightly different shade to the gold, I used black paint for the base colour on the door frame.



To make the least amount of mess, and to avoid scratching or damaging your leaf while you work, start with all the awkward areas first. I applied the gold on the door in multiple sessions. I did all the interior corners of the corner blocks first, followed by the interior lips (where the glass will sit), and lastly the exterior portions in 2 "L shaped" sections.



I waited until the very end to brush off all the excess gold overhanging edges and joints.



The centre of the door frames won't show under the columns, so you don't need to leaf the entire frame.



You WILL end up with gold flakes EVERYWHERE. Try to clean up as you go, and avoid any drafts (walk slowly, close furnace or AC vents, etc).



The tops of the corner blocks can be either black or gold. I chose black.



Topcoat your leaf, to prevent it from tarnishing, and to protect it. The topcoat will also serve as a barrier for the antiquing and glazes, and it will deepen the ebonized sections.





You can see much fewer imperfections in the final (sealed) leaf (compared to the previous detail photo earlier). I did want SOME imperfections to replicate crazing, scratches, etc. but not too much.



The next part will cover reverse painting the tablet.

Mirror Clock Project Part 1 - Building The Case & Door

The original inspiration for this clock was largely based on the following three clocks:



I originally really loved the overall black look of the first clock (by Abiel Chandler), but I later changed my mind on the colour scheme. I preferred the shape of the columns on the third clock (by J. Chadwick), and the centre clock (by Leonard W Noyes) was just nice in general (as far as the colouring and aging).

A lot of research was done while building this clock. I needed as many source images as I could find, and by the end of the project, I had stockpiled close to 200 photos. These include construction details, corner rosettes, case sides (which are rarely photographed), antique mirror and paint textures, movements, and a plethora of painted tablet images.

I can't stress enough that if you plan to build a reproduction, you need to do as much research as possible to get the best results.

To get the most convincing effect possible on the case interior, I built this clock using old scrap pine boards with existing old surfaces (as much as possible). A lot of this wood was just junk boards that were found around the house while cleaning out the garage, and during renovations.



One of the drawbacks of using old wood, however, is that you will rarely have large usable (clear) pieces, you will also often have nail and screw holes in your boards. You must be very careful when running boards through machinery, since any missed nails can damage your tools, or become dangerous projectiles. Another disadvantage is that your old boards may be warped or twisted. If these issues are too difficult to work around, or you simply prefer to work with new kiln dried furniture grade wood, by all means, use what you're comfortable with.

I find that clocks in particular make good use of old boards because the majority of the pieces you need are fairly small compared to other woodworking projects. Other useful sources of wood are salvaged tongue and groove, barn boards, or old damaged furniture.

The Case

When building any clock, I tend to build around the movement that I plan to use. I had originally planned to use an ogee clock movement, striking on a cup bell for this clock, however, this too was changed later on (as you'll see later). For this reason, the case was built smaller than a traditional New Hampshire mirror clock to fit around an ogee clock movement. A standard NH mirror clock ranges slightly in size, but is generally 30" tall, and around 13.5" to 15" wide. This clock was scaled down to 26" high x 13.5" wide, and 2 1/2" deep (not including the backboard or the door). A standard clock also uses slightly thicker wood for the case body (3/4" thick). My clock uses 5/8" thick stock.

The case for a typical New Hampshire clock involves dovetails, or simple butt joints at the corners. A very select few use a half-blind dovetail (with the sides of the case showing no visible joinery). You may use whichever joint you like.

There is no need to be fancy or extremely accurate with the dovetailed corners. The dovetails on my case were loosely traced on a piece of carboard, and quickly cut and trimmed in about 10 minutes. It is, however, important to note that the tails should be on the top and bottom of the case, and not on the sides. If any of your joints are less than perfect, you can fill any gaps with thin wooden shims or slices of veneer. It will all be sanded, puttied, gessoed, and painted, later.



You can see here how the joinery looks, and how I usually use a full-scale drawing when building projects like this. You can also see that I was able to keep 2 old edges, while the other two had to be cut to get usable wood. All the small nail holes on the case sides and bottom get filled with wood putty and sanded smooth. In most other places (like the backboard) they stay exposed as "character".



The Door

When it comes to the construction of the doors on mirror clocks, it's harder to tell how some are put together. A lot of them (the majority) appear to have mitered corners, while a few have lap joints. I'm not sure if the mitered corners are held together with a hidden dowel, a slotted joint, or simply nails plus the corner blocks on the front, but I decided to use simple (and strong) lap joints for my door. As with the frame, the door is also 26" x 13.5", and built from 1 1/2" x 5/8" stock. If you are building a full sized copy, use 3/4" thick stock, and wider frames (1 3/4" to 2" wide). The doors on these clocks are QUITE thick compared to a lot of other clock doors I've come across, but they are all made this way, with some doors being nearly 2" with the corner blocks attached.

Build the rectangular frame first, and the centre divider will be made afterwards. The door frame was built from old wood, but I milled it flat (down to "fresh" wood) on both surfaces for better accuracy with my lap joints. For something like a door frame, try to find the nicest, clearest and flattest stock.



Once your rectangular door frame is glued, sand it flush on the corners (edges), and on both surfaces, then run a deep rabbet along the interior edge to fit your glass. I chose to do this with a router, and to cut out the rounded corners by hand with a sharp chisel. You want the rabbet deep enough to leave you with about 1/8" lip to hold the glass at the front of the door. Do not worry about the centre divider yet.

Once your frame is ready, prepare the piece that will be your centre divider. This is by far the most difficult, and potentially dangerous piece to manufacture, and I would suggest making a piece double or triple the length needed simply for safety. I made mine on the table saw, and I left a very small sliver on each side of the centre (forming essentially an M shape). The extra slivers were cut off afterwards with a chisel. My centre bar was 5/8" x 5/8", and I would suggest a 3/4" square bar for a full sized clock.



Next, you need to carefully trim off about an inch (or less depending on the width of your rabbet) of the top portion on each side. The centre portion needs to fit between the top part of the frame.



Once it's trimmed to size, you need to carefully cut out two mortises. I did this with a drill bit, and a 1/8" chisel.



The placement of your divider will depend on the proportions you want for the mirror, or around the dial. I used 7 1/2" down from the top of the opening (or around 9" down from the top of the door), which gave me a fairly rectangular dial glass. Some clocks have much more of a squarish dial glass. Use what looks good to you.

The completed door so far should resemble this:





Any small errors at this difficult-to-cut joint can just be filled-in with putty as I did here.



The routed inside corners get cut with a chisel.



The Turnings

Find a suitable closed-grain wood such as maple, clear pine, basswood, or poplar for the turnings, and prepare 2 blanks. The blanks should be the same width as your wood frames, and longer than the overall door to accommodate screws and lathe fittings. Glue your centre joint with a single sheet of newspaper, and secure both ends with wood plates with screws. These are "split turnings", which means you are making two halves together in one shot. Because they're glued together with paper, they risk flying apart, and you also can't mount them in the lathe normally, because your centre points will separate the two halves before you even get a chance to start. This is why you need the screwed "end caps".

I used maple, but poplar would have been a much better choice.





For the turnings, make a pattern to work from. There is a fairly wide variety of turnings used on these clocks, but most are along the lines of the three examples at the beginning. When it comes to the number of rings, most of the short ends have 2 or 3 (plus the tulip/blulb portion), but the centre sections can have as many as 9. Since my clock is slightly smaller, I decided to roughly divide the top turning in 3 sections, and from the size of those rings, (and keeping roughly the same proportions as most clocks) I ended up with 5 rings in the centre. I could have easily done 6 instead, but I preferred to have an odd number. This is another reason to have a full size drawing of your project. I drew one version (which didn't completely add up) on one side, and redrew the final version on the other side. After I had my pattern figured out, I cut it onto 1/8" scrap wood to bring with me to the lathe. If you are not good at figuring out patterns, you can simply blow up a photo of an existing clock and print it out (or just a section of it) at actual size.



It's useful to have two (or more) calipers, for a consistent column size, and turned ring size. I used the calipers for the rough sizing, and eyeballed everything else.



During the turning of the larger column, I ran into a problem with some wobble, and while adjusting the speed, the partially turned column got caught on the tool rest, and flew across the room.



I didn't want to completely start over, so I decided to see if I could reglue both halves and the broken chunk, re-mount it into the lathe, and finish it.



I was able to finish the turning with minimal visible damage, and any small nicks or glue joints would eventually be buried under gesso and gilding. Here are the split turnings trimmed to fit, along with the 4 corner blocks glued in place.



The Backboard

The backboard for the case was made of several old boards. Original clocks in most cases used a SINGLE WIDE BOARD for this, which is often around 14" wide. If you have access to wide old boards, use them, otherwise, try to have as few joints as possible. One joint down the centre would be the second best option. The backboard should be fairly thin, and it is sometimes feathered around the outside edges (on the back) with a plane. I made mine 3/8" thick. You can also go thinner, around 5/16", maybe even 1/4". Do not use plywood, no matter how tempting it may seem.



The back board is glued and nailed in place (preferably with square nails), with about 1/8" projecting from the sides, but flush at the top and bottom. No that is not a mistake, and yes that is contrary to modern building methods that take into account wood movement. Back then, they didn't seem to care as much about wood movement, and all the original clocks I've seen have had the backboards firmly glued and nailed across the grain (top, bottom, and edges). With this building method, there's a good chance that the backboard will split, but I don't see that as a big issue (just more "character"). If you really don't like this, you can install your back panel into a rabbet, and glue fake side mouldings to the edges of the case to look like the backboard "edges", but it won't look historic/correct. Some clocks have these two side projections cut square (basically just 1/8" reveal similar to what's seen on some Banjo clocks), while others are rounded over into a semi-circle profile. I chose to have mine rounded (see below).

This photo shows the case with the back attached, and with a few coats of gesso applied (more on gesso later).





Part 2 will cover the process of gilding the turned columns.

Auction

I went to an auction yesterday. The star attraction for me was a stunning French Comtoise longcase clock. I was really hoping I'd be able to take it home. Sadly, I was outbid on it, but I still wanted to share some of the auction items, and my two purchases.

First off, the star of the show, and the only reason that I dragged myself out of bed at 7am on a Saturday, the longcase. This is a French longcase clock along with its original grain-painted case. The clock dates from around 1860-1870, and it was in fairly decent condition.







The case on it was pretty warped, very flimsy, and thin, but I've been told that they are all made this way. It's sort of like a fragile house of cards holding up a 30Lb movement. The grain painting and surface decorations were absolutely wonderful, and in excellent shape. The dial and pendulum were also beautifully painted (in the original paint) and in great shape. These are just thin stamped brass, and they get damaged very easily.

Some of the problems with the clock, however, were not going to be easy fixes. One of the major problems was that the original hinges of the trunk door were broken away. This means that there are chunks of wood missing from the door, and that some of the original hardware is now missing. The way these hinges work was basically just an eye bolt, with "L" shaped bent wires sticking out of the case body. These are very simple hinges, and you can just lift-off the doors if you need to.

Other small issues were that some of the boards on it were severely warped, like the left upper panel shown above. There was also a deep crack through the right side of the case which would have been impossible to repair. The clock also has no key, which is usually a very large crank. You can buy new ones, but old original keys are always much better.

I was willing to pay quite a bit for this clock, but nothing crazy. I had told myself 600$ max, but when it came up for sale, I bid up to 650$, and lost to a bidder who got it for 700$. I was pretty bummed out after that.

The last detail I'll mention about the clock before I move on, is that it was much taller in person than I had expected. I'm 6' tall, and I think the dial bottom was at about my height. The clock was probably a little over 7 feet tall.

At the auction there was also another longcase clock, an Art Deco style German one probably from around 1900-1930.



I wasn't really interested in this one, because I'm not a super big fan of Art Deco, and I don't think I'd like having this huge piece of furniture that I didn't absolutely love. It sold pretty cheap at 130$ to an antiques reseller (I spoke to him later on). It wasn't until after the auction was over that I truly realized how terrible the condition of this clock was. The man who bought it was trying to move it onto a dolly, and the entire top part of it was loose and broken from the rest. You could see screws into the wood that had been added (and that weren't holding it together anymore). I also noticed that the top door hinge had broken right through the wood side of the door and it had been very poorly reglued, and there was also a break in the wood case behind the hood door. On top of all this, I also saw that the two lock plates on the case looked like they were missing, so even though the locks and key were still there, you couldn't keep the doors shut. It also had just general problems visible from the start, like the chipped upper crown, and loose mouldings everywhere. Besides all this, the veneer on this clock was ABSOLUTELY gorgeous. I'm not even sure if it was burled walnut, or some other burled wood, but it was some of the nicest cuts I've seen on any similar clock (which is saying a lot). Even the sides were just as nice as the front (a lot of clocks and veneered furniture use cheaper or plainer veneers on the sides).

There were several other mantle clocks at the sale, but only two were worth photographing. This first one was a mint condition Seth Thomas "Black Mantle Clock". I don't know what model it was, but it had a Seth Thomas label on the back, and the plates were stamped ST, so it was all original. A lot of these clocks have been heavily refinished, so it's always nice to see original ones like this in good shape. I bid on it, but I wasn't willing to go over 100$ on it. Some are selling for much, much less on eBay these days. I think it sold for 120$ or 130$.



Another Art Deco piece. This one was a Perivale Westminster chime clock, from the UK. I'm told that it's not the best brand to look for if you want a Westminster clock. I think it sold for 75$. IT was in very good shape, and it ran, because it started up during the auction and kept ringing during the day. So did the Seth Thomas, actually.



If you're curious to see some of the other items from the auction, and what I DID buy, go visit http://my1923foursquare.blogspot.ca/2015/02/auction.html.

As the auction was starting two wrap-up, I got a chance to talk to the guy who bought the Comtoise. He's a local collector as well, and he seemed like a really nice guy. I helped him take the clock apart (we had a VERY hard time unhooking the pendulum, because the mounting piece had a bit of a curve to it in the opposite direction that was needed for easy unhooking). He tried for a good 5 minutes, then let another lady we were chatting with try, and then I offered to try and I managed to get it off safely. Once the bob was off, we noticed an old dark paper label on the back. It said something like "Balancier...??? fleurs" and I can't remember the city or town name. Basically it indicated that it was a floral model pendulum.

He was trying to get the clock case onto a dolly, but it really wasn't going to work, so I offered to help him carry it to his van. I think I would have been really pissed if he managed to ruin the case just by trying to leave the building with it. Plus, I wanted to be nice, in case he decides to come to me for repairs (I gave him my info, but I never even got his name).

***

Work continues on the mirror clock project. The case is not finished because my planer blades are gone out for sharpening, but the beautiful gilded door is finished, and so is the ornate reverse painted glass tablet for the dial, and I have a custom movement that a clockmaker friend is making for the clock, so stay tuned!

Here's a teaser photo for now.

A New Project

I'm currently working on a new project. It's a reproduction of a New Hampshire mirror clock. One of the reasons I've decided to work on this project (instead of finishing other ones) is to practice some of the restoration skills and techniques that I've learned over the past few years. One of the other reasons is that I'm in the mood to paint, and most of this clock will have painted surfaces. It's also winter, so I can't work on any large house projects, or do too much woodworking in the freezing cold shop.

This clock will be a fairly simple build (I've been working on it for a few days and the main case is nearly done), but the real challenge on this clock will be the treatment of surfaces to achieve a realistic antique look on what's basically a brand new clock. The look I'm aiming for is a period 1820s clock in good, but not perfect shape. This means that I want it to have a few blemishes, imperfections, and "character" but without going overboard. That said, I'm also making an effort to make the clock not so convincing that someone later on could pass it off as a fake. For this reason, I'm choosing to use one of my spare ogee clock movements in the clock. I'll go into more details about that later.

On this clock, I want to practice gilding on gesso (something I've never done before), antiquing new wood to match old wood (ie: getting that perfect shade of old reddish brown), making a convincing silvered mirror, and reverse painting in bronze powders.

I also want to print my own antique style label for the clock, paint a custom dial, and possibly even cast some custom lead weights.

There's still a lot to do on this clock, but I'll be doing my best to document the process as I go along. The next step will be the backboard, and the turnings.

Hopefully I'll have some photos to share soon.