This is the fourth and final installment of the (longer-than-War-and-Peace) tutorial on making a pate de verre panel. In the first post, Carla, Shelby and I designed and made our models. In the second, we “invested” them, i.e., poured refactory plaster over them and made molds, and in the third we filled and fired the molds.
This time around, we’re seeing what comes out of the kiln and investigating some options for finishing them. All told, this project took the better part of three Sunday afternoons. The finish step–the one we’re talking about here–probably took the least time of the three. (We made the models and molds in one very FULL day).
- Protective gear. You don’t want to breathe spent, dusty plaster molds, and they are extremely drying on the hands. Wear old clothes (this stuff is MESSY), make sure you have on closed-toe shoes so you done ruin yourself from shards of glass or dropping plaster on your foot. Above all, WEAR A MASK when you’re breaking open the molds, cleaning up the plaster, etc.
- Lined garbage can. Drag this right up to the kiln, and drop the spent mold fragments into it as you clean. If you use drawstring can liners, you’ll save a lot of time and mess.
- Awl/icepick and bamboo skewers. One or the other, or both. In a pinch, an old phillips screwdriver will work, too. You work the points into the spent mold material and it usually comes apart easily.
Rubber hammer or mallet. You probably won’t need this, but if the mold material turns stubborn you can use this to knock the mold apart. A metal hammer can break the glass, so it’s NOT a substitute. If you don’t have one, a rubber-soled shoe works, too (not kidding).
- Pressure washer. Beg, steal or borrow one; they are hands-down the best way of removing plaster from glass castings that I’ve seen. If you don’t have a pressure-washer, a waterpik or garden hose with a hard spray are (less effective) options. Mine is a 1700PSI unit I picked up from Costco for a bit more than $100, and it’s worth every penny. You can sometimes find these at Goodwill or on Craigslist. You can also rent them, but you can buy your own with the price of three or four rentals.
- Tile saw. If you want to square off the edges of your panel you’ll need access to a tile saw. My (distant) second choice would be grinding them down on a flat lap or wet belt sander. A glass cutter won’t work, but if you want to trim by hand, get hold of Paul Tarlow’s Coldworking Glass without Machines. It’s a very useful book for beginning handworkers.
- Sponge and wet-dry sandpaper for finishing the panel. You’ll want at least 200-grit, 400-grit, 600-grit and, if you can find it, 1000 or 1200-grit. How far you go depends on the finish you want–in some cases you may not need to do much finishing at all. The sponge can be any cheap kitchen sponge; cut a piece that fits easily into your hand.
- FLEXIBLE diamond hand-pads (optional). I love these things, they make hand-finishing quite easy. HIS Glassworks sells a full set (Grinding pads in 60, 120, 200, 400, 800, and 1800 grit, smoothing pads in 100, 220, 325, 600 and cerium) for about $77, but you can get the 200, 400 grinding and 600, cerium smoothing pads for $34 and that’s all you need if you’re also using wet-dry sandpaper. The nice thing about these pads is that they flex (right) with the contours of the piece and preserve them; hard diamond pads knock the tops off detail and flatten the relief.
- Sandblaster (very optional). If you want a matte finish on your panel (or if your panel’s surface needs to be evened out before you finish it), a sandblaster is a great option. It’s wonderful for quieting a wildly colorful, too-busy piece and can dramatically change the character of your glass. I sometimes sandblast a fresh-from-the-kiln piece before finishing, just to see what will happen. If you’ve never used one before, ask for help.
- Flat surface covered with an old towel. This is your work surface; I like to cover a plywood scrap with a couple of layers of old towel. They help secure and cushion the work, give you a little extra “bite” when you’re bearing down on the piece, and also absorb stray drops of water. Other people use a cookie sheet from the Goodwill, lined with a piece of industrial flooring. Both work.
- Water dish. A cup or small bowl wide enough to fit your hand and the sponge.
- Good movie or friends who are also coldworking and like to chat. Handworking can be repetitive and tedious, so I treat it the way others treat knitting: As a way to occupy my hands while I’m engaging other parts of my brain. I’ve gotten so that I measure coldworking by the number of movies it takes to finish the piece. A 4-movie piece is pretty intense.
- Sealant (optional). You can bring the color up in your panel (and seal it from fingerprints) with the use of a sealant such as Unelko’s Clean Shield. It’s not very expensive, you need only a small dab, and a single tube lasts a very, very long time.
- Mounting hardware. These panels can be simply set on an easel, on a table, or they can be hung using something like the Hang Your Glass hangers. I’ve also mounted the panels inside shadowboxes for a very effective presentation (right). Whatever you choose, read the instructions carefully and add enough time for adhesives and whatnot to cure.
De-kiln (is that a word?) the mold
All told, our firing time was close to 60 hours (I fired these in a well-insulated firebrick kiln, which takes a bit of time to cool down). Sad experience has taught me not to rush castings out of the kiln; the firing finished around 1AM, but I let them continue to cool in the closed kiln until about 8:30AM.
The reason: The glass is encased in refractory, which insulates it and slows down its rate of cooling. Besides, glass is a very poor conductor of heat, which means it’s slow to cool anyway. So when the kiln hits room temperature, the surface of the glass probably will still feel warm…and inside the casting it may be warmer still, possibly warm enough to thermal shock.
These panels are thin enough that it really wasn’t necessary, but why take chances? Plan your firing so that it’ll finish around midnight, and wait to open it until morning (or, better still, in the afternoon after work).
Opening time! Put on your protective gear, open the kiln and see what you’ve got. The fired (spent) mold material should be dry and a bit crumbly. The glass should be solid; it won’t be see-through because of the bubbles. There may be holes in the surface because of bubbles rising and popping (there’s one in Shelby’s mold in the center of the photo)–these won’t show from the front.
50/50 plaster silica mixes come apart easily when they’ve been fired. If I’ve got a clay carving tool and an awl, I can generally take these molds apart with no trouble. If your mold mix has additional additives for strength, or you’re using something like hydroperm, you’ll have a harder time taking the mold apart
Clean the glass
Leave the molds in the kiln while you remove the glass; the kiln contains most of the crumbled mix and speeds cleanup, especially if a big trash can isright beside the kiln. WEARING PROTECTION, gently pull as much of the mold apart as possible by hand. If a mold is a bit stubborn (Carla’s, in the middle, was still pretty hard), very gently knock the rubber mallet (or tennis shoe) against the inside wall of the mold, breaking it outward.
Once your (gloved) fingers have taken as much mold material off as possible, go to work with the awl/icepick. Gently (GENTLY) work the awl into a crack in the mold and wiggle it SLOWLY, putting pressure on the side of the crack. (NEVER put pressure on the glass). Slide the point alongside, not against, the glass. Chunks of the mold should come away; if you’re lucky and you’ve done this right, you’ll be able to lift great sections of the mold off the glass.
Watch out for sharp glass edges; the frit that slides down into the mold can leave a trailing edge that’s as sharp as a surgeon’s scalpel. As you’re removing mold material and see these very thin shards sticking up, carefully knock them off.
Once you can see the rough outlines of your piece, put the awl away and get out the bamboo skewers. You’re too close to the glass to risk punching it with metal; the skewers are softer than the glass. Use the skewers as you did the awl, to remove as much as you can easily. When it’s getting too hard to remove more plaster, it’s time for the pressure washer.
Take the glass outside, and put it on a soft surface (I usually coil up the hose and set the glass inside). Hook up the hose to the pressure washer and turn it on. Hold the wand nozzle about 8-12 inches above the glass and spray. (If you’re not used to a pressure washer, practice on concrete or something first).
Sweep the washer slowly across the casting, and you should see plaster flying away. Keep the stream sweeping in at an angle, linger in crevices, and keep going until you’ve removed all visible plaster. Then give the casting a quarter-turn and repeat until the piece is back in its starting position.
We’re trimming these, so we didn’t worry about the edges, which can collect a fair amount of plaster crud. If you’re not going to trim the edges with the saw, you’ll also want to pressure-wash the edges. DO NOT hold the piece with your hand while you hit the edges. I have scars from trying that; the spray took a chunk of meat from my knuckle while leaving the glove intact. Instead, rig something to hold the glass upright and secure.
Examine your glass (and make a plan)
Unless you’re from the Planet of Unexcitable Art-Haters, you’ll have already done this. Who could put this much work and time into something and NOT get excited about seeing how it came out? There’s a reason that the kilnforming world is famously prone to clibanus interruptus,* after all.
*Clibanus interruptus: The tendency of a glassist to open the kiln before fully cooled, risking total glass destruction because s/he just can’t wait to see the result.
So look to your heart’s content; turn the panel over, view the sides, back, front. A still wet, just-pressure-washed casting is always gloriously gorgeous; the water hides flaws, and it’s just so cool to see your vitrified sculpture that you’ll forgive any little flaws. Once it dries, the flaws will pop out like zits on a teenager, and your goal will become getting that wonderful, drenched look back in your piece.
Take a picture or two of the wet panel, for reference, and then get serious: What needs to happen to finish this piece? Noted artist and instructor Richard Whitely once said, “It’s only half-finished when you take it out of the kiln,” and he’s right: Coldworking is often the difference between hobby craft and artwork.
So…what does yours need? On my bigger sculptures, I’m doing just as much carving on the glass as I did on the wax or clay. These panels, though, are pretty simple: A flat expanse of glass with some low-relief detail. In most cases, they won’t need much coldwork.
Carla’s, for example (left), had some slightly ragged edges. They were charming in a shabby-chic kind of way, but she thought they’d take attention away from the really lovely relief lily-of-the-valley pods and iris leaves. So she opted to trim the edges straight.
On the other hand, she very much liked the slight unevenness of the background, which showed some marks from too-fast plaster thickening and some unevenness in the clay slab of the model (if you click to enlarge the image, you’ll see what I mean). She could easily grind the background flat and smooth with a power grinder and some hand pads, but she wanted to retain the organic, natural look, with the soft gleam of old leather.
Mine (right) also had uneven edges and, as I’d predicted, a tiny bit of clay that had gotten stuck between the chestnut pod and the plaster, getting bisqued into the glass (one of these days I’m going to stop using inorganics and clay entirely!). My coldworking plan needed to include digging that out, with an awl if possible, diamond drill and awl if not.
As sometimes happens, the way you design it turns out not to be the best way to display the finished piece. I’d built the original model with the two bigger pods at the top. Once I started playing with the casting, however, I really liked it on its side. (A good lesson in staying flexible with what comes out of your mold)
Devit and incomplete fills: Soda lime glass (the fusible glass made by Bullseye, Uroboros, and Spectrum) is a bit harder and more viscous in a casting than lead crystal. It tends to need longer times at hotter process temperatures when filling small, insulated areas in a mold. Nothing wrong with that, but:
- Soda lime glass is more prone to devitrification than lead crystal
- Frit is more prone to devit than solid (sheet or billet) glass
- Powder is more prone to devit than frit
- Powder packed against wet plaster is more prone to devit than powder on a kilnshelf
So a powder-filled plaster mold has a fair amount of devit potential, which makes the firing schedule something of a balancing act. You want to keep the mold hot enough, long enough to completely fill…but the longer it processes, the more likely you are to encounter devit.
Devit in an open-faced mold will typically manifest itself on the top, exposed surface of the glass–you’ll see some clouding, as if the glass looks dirty. If it progresses, the corners start to wrinkle a bit (if you skim down to the picture of the hanger on the back of a garden panel, you’ll see some mild devit). You can also get a sandpapery surface on the glass next to the mold which needs to be coldworked, but I’m never really sure if that’s devitrification, or simply the effect of abrading wet plaster with powdered glass.
If your mold REALLY devits, as it did once for me with a 2-inch mold (right, the model and master mold are shown above, for reference), it’ll look like one of those crystal-growing experiments you did in grade school. Looks cool…but not really what you want. 😉
My favorite process temp for smaller plaster pate de verre molds, 4 hours at 1485F, can give me devit issues sometimes. I wanted to keep that to a minimum. So I reduced process time to 2 hours, and accepted that taller, more difficult areas might not fill completely.
I, too, liked the idea of a soft, natural finish without a lot of over-smoothing, so I planned to just knock off the
The largest chestnut pod, top right, had an area of incomplete fill, where the glass didn’t completely move into the void left by the protruding chestnut pod. These are pretty easy to identify: The glass has a rounded, stubby appearance and is usually glossy, since it didn’t actually touch the mold.
Sometimes those can be blended into the piece with a little coldwork. More often, they don’t match the surface of the rest of the glass and need to be removed.
Shelby’s piece really didn’t need much at all, outside of trimming. It had a few holes, probably from the shorter process time. If she liked, she could choose to drill those out, even up the panel and coldwork to a soft sheen…or just leave them.
But again, she’d made it as a vertical, portrait-format panel. When it came out of the kiln, the vertical orientation was nice, but the horizontal, landscape-format orientation was really striking.
The only real concern we had was the proximity of the blossoms to the edge; it was going to be difficult to trim the panel without cutting them off.
Trim your panel
If you’re lucky, you have perfectly straight, smooth panel edges that need no coldwork. They’re beautifully rounded, and just slightly uneven enough to look hand-crafted. Do a little sanding, and you’re done.
Most of us aren’t that lucky, so we trim the edges. If you haven’t tried cutting glass with a tile saw, check out my post on using a tile saw. You want to make sure you’ve got a great blade on your saw; if you don’t, you might be better off grinding the edge down.
You just want to square off the edges and cut off any contaminated edges. What you don’t want to do is overtrim, or trim one side unevenly and then spend a lot of time trying to correct the trim. Take a little time to mark where you want your cuts to be made–you want to trim back far enough to get nice, square corners.
What you’ll probably notice is that you don’t have to cut as much off as you think–cutting off one side means that you don’t have to remove as much from the adjacent side to get to a 90-degree corner.
Now position the panel on the saw bed. The panel isn’t square, so don’t try to snug it up against the fence–you’ve got to cut one straight side first. Align the cut line you made with the blade in front, measure where it’s sitting with relation to the edge of the bed and then line it up the same way in the back.
Hold the glass firmly and make the first cut. (Some tile saws have an adjustable fence that you could use to secure the panel, but my saw is an el-cheapo eBay model without such luxuries–I just hold it down).
Now you’ve got a straight side you can put against the fence. Give the panel a quarter turn, so the straight side is on the fence, line the trim line up with the blade, and cut again. Repeat until you’ve trimmed all the edges.
This was Carla’s first time with a tile saw, so she practiced on some of my old casting offcuts until she was confident (left); then she went for it (right). Success!
Not only did her panel work, but she noticed that cross-sections of these panels make really nice panels. So she also cut them up for pendants!
Coldworking the panels
If you’re satisfied with the look and feel of your panel, all that’s left is displaying it. Skip this section.
If your panel has some rough spots, or if it feels like sandpaper, or you simply want to try a silky, hand-sanded texture, you need to do some coldworking.
If you’ve got the power tools to do it–Foredom or Dremel with diamond bits, bench grinder, wet belt sander, flat lap–great. You can carve, flatten, grind, polish…just be careful that you don’t overdo it, which is very easy with power tools.
I prefer handworking these panels because I can precisely control the level of finish and keep it even. If you’re unfamiliar with handworking techniques, take a look at Paul Tarlow’s Coldworking Glass without Machines; excellent instruction.
The object of the game is to do as little coldwork as possible. Once you coldwork an area, you must coldwork the entire section, or it will look uneven.
Most often, I use a 200- or 400-grit flexible diamond pad on the entire background, knocking off the sandpapery texture, and making it feel smooth. And I’ll use a hard 400-grit block to cut a slight bevel on the sharp, trimmed edges.
Then I’ll swap over to a 600-grit wet-dry sandpaper and take the background to a silky finish. If I’m in the mood, I might go higher, but usually that’s good enough. I rarely do much work on the relief unless it has some damage that needs correction. (And that’s another post).
You can also sandblast or etch the panels, for a much softer, very matte finish.
Either way, it’s a good idea to seal the panels, to bring out the color and protect from fingerprints. The Unelko sealer I mentioned in the supplies list works very well for this purpose, although it takes about 3 weeks for it to finally lose its shiny gleam. Other people put a drop or two of mineral oil, lemon oil, or another clear oil on a soft, clean rag and rub it all over. It doesn’t last, so if you sell/give these panels to someone, you’ll have to advise them to periodically renew the finish.
You can also fire-seal the panel by re-firing it to slightly below whatever is a firepolish schedule in your kiln. It’s tricky, because fire-polishe levels can vary with the viscosity/color of the glass. And, personally, I don’t really like fire-polished pate de verre; it looks TOO shiny, for some reason.
Or…you can just leave your beautiful panel alone.
Displaying garden panels
You have several display options:
Let it stand by itself. If you made your panel thick enough (usually about a half inch or more), it will easily stand on its own. If you put it in a relatively protected place, on some museum gel, it will become its own 3D sculpture.
(Lola and Nikki will convert any museum-gelled panel into a nice pile of shards, so I don’t do this one myself)
Add a hanger on back. Hang Your Glass sells a nice kit of adhesive, prep, stand-off and wall hanger that make a sturdy, attractive display method. In 10 years of using them, I’ve only had one failure. These are small panels, so you simply glue one stand-off dead-center in the back of the panel and you’re done.
One caution: Any time you’re relying on the shear strength of an adhesive to keep glass hung on a wall, you want to really, really follow the manufacturer’s directions. Clean the glass and the stand-off (I like to slightly abrade the glass in that spot), prepare it with amino silane or whatever treatment is recommended, position it, and don’t try to rush the cure time.
Put it on an easel. A lot of places (including Michaels) sell little tabletop easels for 5×7 or 8×10 picture frames. Not my favorite display method, but they would work for a garden panel.
Mount it in a shadowbox. This is rapidly becoming my favorite presentation for these panels; it gives the glass a lot more presence, the hanger system is built in, and you can reinforce the mounting for heavier panels.
I keep an eye out for beautiful shadowboxes (or nice wooden boxes I can turn to that purpose), and stockpile them, removing the front glass because it isn’t necessary.
The glass panel is usually too heavy for a cardboard backing that most come with, so I’ll cut my own hardboard backing. Then I’ll cut a smaller piece of hardboard (or scrap plywood) about an inch smaller than the pate de verre panel.
I center the smaller board on the larger, and glue it on. Then I take my favorite fabric or hand-made paper, and stretch it across the board assembly, over the raised board and around to the back, securing it on the back with double-faced tape, hot glue, whatever works.
Then I slit open the fabric/paper in an X across the smaller board, carefully, pull it back and cut it away, exposing the board. The glass won’t stick well to the fabric/paper background, so I glue it to this board. If it’s heavier, I’ll sometimes use a mechanical fastener instead.
And that’s it. Apologies for such a long-winded series; I was trying to be as complete as possible. If you’ve got questions, please feel free to post a comment, and I’ll try to respond.