Monday, March 30, 2009

Etching ceramics

As I mentioned in the previous post, I am working on a number of sabinos right now. I thought I would show some pictures of how that kind of color is done in ceramic underglaze.

White hair detail like the kind seen on sabino roans are done with a process similar to what cold painters call etching - though ceramic artists call it scritching. Until the underglazes are fired, ceramic paint is chalky and easily marred. Since the base material (the earthenware clay) is white, ceramic artists often take advantage of that fact when they need to add white markings or white hairs.

Most of the time the favored tool is a very sharp #11 Xacto blade. With my own horses I use a wide variety of blades (many sharp, but some dulled) and erasers to get just the right effect. I will sometimes use an initial layer (or layers) of white underglaze to get the right effect.

That is what I did with the guy pictured in the previous post. There is a light layer of opaque white underglaze under his gray areas. Without this, the transparent colors I have used for his body color are difficult to remove. I've only used a light layer, though, because I want a softer effect. In general, the thicker the layer of white the crisper the edges will be.

In the previous picture I had started adding the irregular roaning of his sabino pattern. This side, however, has not yet been detailed. This is how he looked right after his base color was sprayed and the latex masking was removed. The latex is too gloppy to mask the actual pattern, so only the whitest areas are covered. The real outline of the pattern, as well as all the hair detailing, will be done with the blades and erasers.

This picture also shows just how flat and chalky the underglazes are. Underglazes begin to look opaque long before they really are, so the body color often looks quite flat until after the horse has been through the final glaze firing. That's why most glazers paint by memory, rather than by sight. Remembering how many times you hit an area, and how hard you hit it, is the most reliable way to predict the end coloring.

Here is the belly showing one side that has been detailed using scritching, while the other side is largely untouched. I tend to work one side at a time, which was a habit I developed long ago when I was a cold painter. I wish I had the discipline to paint one part (like the whole head, or all the legs) at a time, but I always get impatient to see one finished side.

In the other post I also mentioned "scritching claw". It is true that making precise ticks with a tiny blade is hard on the hand, but it's the left hand that truly suffers. I have mentioned before that unfired underglaze cannot be touched, and that even fired underglaze has to be handled very gently. One of the biggest challenges in working with underglazes is plotting your hand holds. Here with this guy I had already painted his mane, tail and feet and fired them so that I could have a three-point grip on him. (Notice that my middle finger is not actually touching his back, which has raw underglaze at the moment.)

I have to hold him like this, without rubbing those areas overly much and without exerting any sideways pressure on his legs. (Finn is pretty sturdy in that regard, but for finer legs this can be a real issue.)

That's one reason why I tend to gravitate towards clean legged colors for my first glazed piece in a run. Tobianos, with their white legs, have four obvious hand holds. Toveros have those hand holds, and allow me to experiment with scritching without a huge time investment. All of this allows me to get aquainted with the mold a bit, learning the logical handholds and also where I am likely to run into trouble. (Will it be hard to reach between the front legs to etch the lower chest without hitting the knee with the back of the blade, for instance?)

And here is my sabino guy with one side fully scritched, sitting next to the sheet protector holding the set of reference pictures I was working from with him. I don't usually work this closely from a specific horse, but I wanted to address a certain set of issues with this piece so I've been staying pretty close to the picture.

Since taking these pictures, his second side is half-finished. His firing buddy, a sabino Voltage, is not quite so far along. She has a lot more area to scritch, and a considerably larger amount of white. I've been taking pictures of her in-progress, too, but I want to post hers as a group once she is done. I have high hopes that at least some of these guys will be in their final, shiny form by the end of the week. Maybe by then I will have made a decision what color to paint my Vixen!

Thursday, March 26, 2009

One step closer

Here is the first bisque set of Vixen and Imp. I've learned the hard way that one bisque only tells me that it is possible to make the horse. It will take a few more before I know if the edition is feasible. But it's one step closer, and she presented enough challenges that there were times when I wondered if I would make it this far.

I usually keep the first glazed piece from each edition, but I must admit I cannot decide her color. I looked over my "first piece" collection, and realized that they are almost all tobianos and toveros. Those types of patterns take fewer sessions (either scritching or firing) to create and they present far fewer technical problems. Usually by the time I'm holding that first bisque, I'm impatient to see color on the horse and the bisques seem too precious to do a color that might seem a little dicey. I'm on the fence whether to try something crazy and time-consuming with this first one, or avoid tempting fate.

Though perhaps with all the sabinos I have currently in the works, I think my "scritching claw" might never go away long enough to paint her!

Wednesday, March 25, 2009

The quest for the perfect nose pink

It never fails that if I mix a useful color, I will forget to write the formula down. Or the company will discontinue one of the key components. With the mix I was using for pinking, it was both of these things. I couldn't remember all the ingredients, but the one that I knew was in there was discontinued some time ago. Since that time I've experimented with a variety of pinkish colors, but I never took the time to develop a fool-proof color.

Pinks and reds are unpredictable during firing, especially in small kilns like the ones I use, but some are worse than others. I wanted to find a fairly durable one. I also wanted something that wouldn't edge towards bright, true pink - a color that my friend Joan refers to as "boiled baby possum" pink - or something too orange or brown. I created a few mixes, sprayed a gradient on some 1.75" tiles, and then glazed them.

I was pretty sure I wouldn't get the right color this time around, but I wanted some clues as to what direction to take. That was exactly what happened, so I'll need to do another mix or two before I have my perfect hue. Once I have that, I plan to work on the mix until I have the actual shade I need. In the past I've used lighter coats of a color a bit darker than I needed. This works of course, but it involves a fair bit of guesswork. Unlike acrylics and other color painting methods, underglazes don't really show layers of shade accurately until after their final fire. That means that the nose might be much paler than I thought, or much darker. My hope is that by matching the shade of the glaze at full saturation to the deepest pink I need (and no darker), I can at least eliminate the possibility of the nose ending up too pink.

This is all important because so many of the horses I am working on have a lot of pink skin. Now that I am back to glazing, I have been revisiting the problem of how to best tackle the indistinct, soft colors like sabino roan.

Here is a sneek peek at one of them where I just started adding the pattern. I have a group of these at the moment, all some type of ticked or roaned pattern. I'm finding it helpful to work on them as a group so I can put one down as soon as I start to fall into regimentation with the pattern, and pick up another. I think it's even more helpful that one of these horses is a different scale (an old Voltage, which has been cool to work on!).

And with luck I'll have a bisque Vixen to add to their little group today!

Tuesday, March 24, 2009

Off with her head!

After a handful of unsuccessful castings from a (now) dry mold, I had to give up on my hope of casting Vixen in one piece. Her head is turned too far to the left to work, so it's going to have to come off.

What I discovered, however, is that I think I can take it off inside the mold. My problem isn't that the head will not cast properly. It's just that the angle necessary to free the body and the one needed to free the head are incompatible. The design of the mold means the body has to pull first, and that pull breaks the left ear every time. But if I do the head and neck cut in the greenware, I can pull the body out and leave the head. Then there is room to pull the head at the angle it needs to come out. The two pieces can then be reassembled just as if they were cast separately.

I have one piece assembled and ready to fire. She looks great so far. If this works, it will save having to make another set of master molds (one for the body and one for the head). She'll still have to be assembled, which will make the bisques more time-consuming, but I think it's still better than dealing with two molds. And right now I'd do most anything to avoid having to make new rubber masters again!

(In the picture above, this is the stage in demolding when I make the neck cut.)

Monday, March 16, 2009


Okay, so I admit this doesn't look like success. But I do believe the mold design does work and that I will be able to cast Vixen whole.

I did try to cast from the mold a little too early. Castings tend to stick when the mold is damp, which is what caused her legs and ear to break. There are some minor alterations to the mold that I will need to make, like adding a small inset piece to the mane around her withers, and I suspect she'll be a bit of a fidgety mold. Still, it's a level of fussiness that still falls short of what we be involved in a multi-part casting. I am going to give the mold a little more time to dry and try again, but I am more hopeful than before that this mold design will work.


I am anxiously awaiting the delivery (any day now!) of my copy of the third edition of Equine Color Genetics. This edition is supposed to have more information on some of the "new" colors and patterns not discussed in previous editions. I know there is a section on the manchado pattern because Dr. Sponenberg contacted me some time last year about it. (I have a small book on the color that the author self-published in Argentina.)

I had heard that there would also be information about the black dilution found in Arabians (sometimes referred to as the Muhaira dilution), the "mushroom" color in British Shetlands and "calico" tobianos. Those are the buckskin and palomino tobianos that have patches that have reverted back to an undiluted red. Lots of neat stuff!

Friday, March 13, 2009


I gave the wrong model number for the Peak airbrush. What I have is an X-5, not a C-5. The "C" brushes are gravity feeds and the "X" models are bottom feeds. I did get the gravity-fed Richpen since I knew if it worked it would just be for detailing, but for general use I much prefer a bottom-fed airbrush.

Thursday, March 12, 2009

New toys!

I have to confess that I have been intimidated about glazing my Imps. His small size presents a number of technical issues, not the least of which will be the need to hit some pretty small targets with an airbrush.

I started out glazing with rather marginal airbrushing skills. Although many believed my older acrylic work was the result of airbrushing, the fact was that all I ever did was block in the basic tonal shifts. All the real shading was done by hand with water-thin layers of acrylic. Switching to ceramic underglaze has forced me to improve my airbrushing skills. I have come a long way with my ability to target my spray to a specific spot - and to avoid other areas. (Joan's ability to leave strands of mane flaxen while darkly shading the body around them never fails to impress me!) Still, as my wrapped-up Imp shows, I am heavily dependant on masking.

For something like Imp's feet I can use masking. (Even if removing the latex from his spindly legs is nerve-wracking.) The problem was that with a head less than a half-inch long, even the most basic facial shading was going to take some serious precision. I decided it was worth it to try out some new tools.

For years the standard airbrush used for underglaze has been the Iwata Eclipse (top). Many of us had tried finer brushes only to see that the pigment would not pass through the brush. It wasn't that they splattered; the paint simply did not come out. But I wondered if some of the newer intermediate brushes might work, especially now that so much of my own underglaze work involved translucents. Translucent underglazes like the Duncan E-Z Strokes have a much lower clay content that traditional underglazes.

Underglazes are hard on airbrushes anyway. They are deceptive, because unlike acrylics they never dry completely so they don't clog the brush. They clean with water no matter how long they sit in the brush. But the clay content puts a lot of wear on the internal workings. It's a bit like slow-motion sandblasting of all the internal workings. I'm used to replacing the cones, and eventually the entire brush. It didn't seem sensible to invest in an especially fancy airbrush, even if it would work.

But my favorite supplier, BearAir, was running a buy-one-get-one special for the month of March. They had stopped carrying Iwata products a while back, and I had already considered trying one of their replacement brands. I thought I would use the opportunity to try the new brands as well as some of the finer brushes, for not that much more than I would have paid for a single Eclipse.

After speaking to the salesman at BearAir, I decided to try the Peak C-5 (middle) and the Richpen Apollo 113C (bottom). The fellow I spoke to felt that I stood a better chance with the Peak C-3, which was not quite as fine as the C-5, but that one was not included in the special so I took the chance. We were both pretty sure the Apollo wasn't suitable, but I've wanted to try some of my improved airbrushing skills on a micron-like brush and it seemed like a good compromise. (This brush is one step below "micron technology".) I figured the worst that could happen was that neither might work, and it might force me to revisit cold-painting some day just so I could get some use from the brushes.

What has surprised me is that both brushes - which arrived this morning - worked just fine with my watered-down translucents. Even more surprising was that the Peak gave a far finer line than the Richpen. It may be that the pigments were too coarse for that brush to perform properly. Eventually I'll have to test it with airbrush paints to see. But it really didn't matter, because the improvement from the Peak was more than I had ever hoped for! Even new my Eclipse brushes have never been able to give that fine a line. It was also really easy to maintain that fine line for long periods, which is something I've never been able to do.

What's nice is that both the Peak and the Richpen use the same fittings as the Eclipse. That was part of what made the decision for me. I've become really fond of the all-in-one "Fast Blast" bottle caps. They don't have any metal parts, which is great since I used to worry about corrosion contaminating the underglaze. They also come in sizes that fit both the cheap flip-top paint bottles (24mm) and the manufacturer's bottles for the translucents (20mm). With so much time invested in getting all my paints and bottles just the way I like, I wasn't keen to try anything that couldn't just plug in to the existing system. Fortunately both pens not only fit the bottles, but they use the same hose fittings.

That means I can use the Eclipse (with it's cheaper cones!) for larger work, and then switch to the Peak when I need to paint really precise things. I am excited to see what might be possible with this set-up. But mostly now I'm more confident that I can shade Imp's eyes without giving him a dark facial mask. I was sure folks would start to wonder if all the foals were some kind of dun or silver!

Tuesday, March 10, 2009

Another use for the needles

This the watched pot that won't boil. That's the mold that will tell me if Vixen can be cast in one piece. I've placed it next to the kiln (which has been running pretty steadily for the last few days) hoping that it will dry faster. So far the only thing drying have been the rubber bands!

I almost forgot to mention that I found another use for the large gauge veterinary needles. One of the things I wanted to try with this test mold was pouring the head piece. Horses with turned heads are problematic in ceramic, and usually require a separate piece to fit over one side of the head and neck.

"Finn" has one of these. In the past I've made these by hand, much like I did with the tail piece for "Imp". I thought with "Vixen" I would try to include this piece in the master mold, and pour it just like the others.

The problem with pouring these kinds of pieces is that they wrap under the original. The poll piece sits on top of the ears and forehead, but the second piece has to include the underside of the head.

Since the plaster pours from the top (the area tinted blue), what often happens is that some amount of air - often quite a lot! - will get trapped under the jaw (the area tinted lavender).

What I found was that I could fill the needles with newly mixed plaster, tip the mold sideways like I have in this picture, and then insert the needle just past the tip of her nose. I measured that distance and marked the needle, since the last thing I wanted to do was inject plaster into the rubber! But doing it this way allowed me to fill the area under the jaw. Then I turned the mold back the normal way and continued filling the rest of the cavity. The finished plaster piece was completely bubble-free.

The nice thing was that plaster cures slowly, so the needle can be cleaned and used again. With the rubber the syringes have to be thrown away since it is impossible to clean them afterwards. So I have the marked needle ready to use again for future plasters, if this mold design works.

In the meantime, I've reclaimed my studio and have started glazing again. It feels good to get back to all the projects I left hanging when I started working on the molds. I've enjoyed pushing myself with some more complex molds, but the end results aren't nearly as charming as finished, shiny horses!

Monday, March 2, 2009

Indoor camping

That's what the Kathman family has been doing since yesterday, after we traded our electricity for five inches of snow. As you can see from the picture, my oldest son Brandon totally approves!

I can almost forgive him his enthusiasm; after all these years in Scouting, he's at least really handy with a camp stove.

I'm not sure this guy would agree. It's probably not visible in the picture, but the poor little fellow had icicles hanging from his tail feathers! And I can attest that the only thing more useless than an electric kiln during a power outage is an airbrush. My plan to return to glazing today was thwarted.

But I did promise to post how this was fixed.

Here are three of my legless Vixens. I have quite a collection of them now. Sarah suggested that I turn them into ornaments and hang them from my Christmas tree. I might have to add a small tree to the studio next holiday, just to do it. If course it's a lot easier to laugh at the idea of a tree festooned with mangled Vixens now that I have one with four legs!

My final solution was to open the mold, cement the hind gussett in place (thank goodness for good mold keys), and then pour the lifted leg first. Once the leg was filled and had set up enough that it would not spill out again, I closed the mold and poured the rest of the horse in the normal way.

This worked just fine, though if I had it to do again I'd cut a sprue level with the leg pour. That's because once the mold refills, air gets trapped where the first and second pour meet. A sprue would allow that to vent. It was deep in the hindquarter, though, so it was really just cosmetic. The bubbles did give me a chance to test out my new 16 gauge hypodermics, though, and they worked beautifully. I'm still skeptical about being able to insert rubber through the mold walls, but I suspect I'll get opportunities in the future to test it.

I also think there is a better tool for pouring into the open leg. I used a disposable plastic spoon, but that method lacks pressure. Pressure is part of what drives the viscuous rubber down the thin legs, and I think that had I not been using really fresh rubber I would have still encountered problems. Next time I'm going to try this angled syringe that my friend Elaine Lindelef suggested. That will let me pour the rubber with a little more force, and better avoid getting excess rubber on the rest of the mold.

So with a four-legged Vixen, I went ahead and cut my inside pieces apart so I could pour a plaster test mold. I'm not real confident that Vixen will cast as one piece, but there was one design that might work so I have to give it a try. If it does work, it will save me hours of assembly work.

If it doesn't, I'm afraid I'll have no choice but to lop her head off!