An "appropriate" finish for a Stickley-style oak table.
Todd Knoblock
For the past six months, I have been working on a coffee table in the
style of Gustav Stickley. For the past twelve months, I have been
working on finishing it. I have almost always been dissatisfied with
my finishing attempts, and was especially concerned in this instance
that I find a historically and esthetically appropriate finish for the
piece, and that it be applied to a high standard. I started reading
about finishing even before the project was drawn or the wood
acquired. This note summarizes what I learned and eventually the
finish that was used. While I wanted a traditional looking finish, I
ended up departing from the traditional technique at almost every step
as will be explained. What follows is the list of steps used in
finishing the table along with a list of the considered, but rejected,
alternatives at each step.
Step I. Ground coloring.
I used an aniline water dye to provide the basic ground coloring. I
first wet the wood with water to roughly the same extent that I would
during the dying, and allowed it to dry thoroughly (two days). I then
lightly sanded the fuzz with 400 sandpaper (dry). The dye was applied
in two coats of half dilution (1/2 oz. per quart of water) to allow
unevenness in the color of the wood to be compensated for. It turns
out that two coats at half dilution is somewhat darker than one coat
at full dilution. It took some experimenting to find a good color. I
wanted a dark, rich brown that was similar to several craftsman style
pieces that I had seen. I ended up using J.E. Moser's English Brown
Oak. I tried their medium and dark "fumed" oak, but found those
colors unacceptable. The medium fumed oak was distinctly green, and
the dark had too much black in it.
Alternatives considered.
* Fuming. Traditionally, the ground color would have been provided by
fuming the piece in an air-tight tent with 28% aqueous ammonia, with
a touch-up with alcohol based aniline dye. I actually tried fuming
sample pieces off the table, but was dismayed with the variation in
color and with the quality of the color on the sample pieces. The
ammonia reacts with the tannin in the wood, and different pieces of
wood have different levels of tannin. Also, the sap wood has much
less tannin than the heart wood and so is almost unaffected in
fuming. I was aware of these problems before starting and had tried
to select wood that was similar in color and to carefully cut away
all the sap wood. But, the color after fuming ranged from light gray
to gray-black and it turns out that some sap wood had made it into
the table despite my best efforts to eliminate it (it stays white).
Old books recommend using wood all from the same tree, but this was
not an option at my local lumber store.
I was after a brown or brown-gray color, but did not like the
gray-gray that resulted from fuming. It did not match the kinds of
colors I had seen on old fumed-finished pieces. There are a few
species that are all marketed as "white oak", and it is possible I
had one that was a different species (burr oak?) than the traditional
"white oak" fumed pieces I had seen. It is also possible that the
color I have seen is only formed over many years of mellowing or
developing of the finish. Because of these problems, and the
logistics of the fuming process itself (goggles, respirator,
air-tight tents, and dangerous chemicals), I decided not to use
fuming. Stickley claimed that only fuming would darken the rays, but
I found the water based dye to darken the rays almost as much as the
fuming process.
* Alcohol aniline dye. Gustav Stickley warns in no uncertain terms
not to use water based dyes, and recommends the alcohol based ones.
He was concerned, rightly so, with the grain-raising effects of
water on naked wood. However, I have found that I can mostly
alleviate this problem by carefully wetting the wood with distilled
water prior to dying. It is important to allow the wood to dry very
thoroughly, though, in order to avoid creating hollows by sanding
before it is dry. I was concerned that alcohol based aniline dye is
not as light fast nor as deep penetrating as water based dye. I
have seen several old pieces that have faded places in the coloring,
for example, in sap wood that would have been dyed more originally
than the rest of the piece.
* NGR dye. Everything I read said that this lacked the clarity and
light-fastness of water based dye.
Stage II: Coloring the pores.
The table is made from quarter sawn white oak with a very prominent
tiger pattern, and a more subtle quilted pattern on some boards. The
quarter sawing process tends to longitudinally slice the capillaries
in the wood leaving a prominent (and interesting) pattern of pores on
the surface of the wood. Unfortunately, water-based aniline dye has
almost no effect on the walls of the capillaries. This leaves an
overly prominent pattern of white dots and lines on the face of some
boards (think of lots of tiny white PVC pipes on the surface of the
boards). By the way, this was also true of the fumed samples. A
couple of books mentioned that this was a potential problem, but the
suggested solutions did not work for me.
I used three coats of dark walnut danish oil finish (watco). This is
a light tung oil varnish/linseed oil mixture with asphaltium as a
colorant. I applied it, let it set for about 30 minutes, and then
removed as much as possible from the surface but leaving it in the
pores. This muted the white pores by filling them with a dark oil.
The asphaltium in oil is a nice colorant because it enhances rather
than obscuring the subtle quilted pattern in the wood grain. It also
added nice black highlights for the ground color
Alternatives considered.
* Colored silex grain filler. Colored filler is often used on open
grained wood like oak. However, it results in a uniform and
unnaturally glass-like surface. I wanted an "open grain" look in
the final piece where you could still feel the pores of the wood.
* Pigmented stain. This was also recommended to correct this problem.
However, I was not able to find a stain that worked. Either the
stain had too little effect on the pores, or it darkened the pores,
but also obscured the wood grain everywhere.
* Black wax. This is also a traditional solution. It worked well on
some samples because it is relatively easy to apply black wax
(furniture wax darkened with lamp black) and fill the pores without
changing the ground color. Unfortunately, the coffee table is likely
to encounter hard service, and a wax finish was not considered
sufficiently durable.
Stage III: Varnish.
I let the oil cure for several days to avoid interactions with the
varnish. I used a high quality natural resin varnish (Behlen's
table-top varnish). I applied two very light coats with a good bush
and lightly sanded with 400 paper between coats. I prepared the space
to be as dust-free as a garage can be, and was careful not to raise
dust while the coats dried. I used thin coats of varnish to avoid
excessive varnish film (that would fill the wood pores), but was
carefully to tip-off the varnish and not leave holidays in it.
Alternatives considered.
* Shellac. This is the appropriate traditional finish.
Unfortunately, shellac is susceptible to damage from water and
alcohol. This would not be a good choice for a table top.
* Lacquer. This is the standard professional choice. It is applied with a
spray gun in a spray booth with explosion-proof ventilation fans.
As an amateur, my measure of an acceptable finishing method is
different than the professionals. In particular, just as I am
willing to cut mortises and dovetails by hand, I am willing to spend
lots of time and labor to achieve a good finish. The professional
is constrained by the costs of labor and free to use much more
expensive equipment than I could justify for my hobby projects. I
was also somewhat surprised to learn that varnish is actually a
superior finishing product to lacquer, just too labor intensive to
use in a professional shop. By the way, Gustav Stickley keeps
mentioning "German Lacquer" in his notes on finishing. It turns out
that this is an old name for amyl acetate (a shellac thinner), and
has nothing in common with what we currently call lacquer.
* Polyurethane. Polyurethane, in addition to looking like plastic,
does not rub-out well.
Stage IV: Rubbing out.
I let the varnish cure for 8 days. Next, I hand-rubbed the finish to
produce a satin finish for the eyes and the fingers. I began by
carefully sanding with 600 paper wet with water. I then rubbed with
2F pumice followed by 4F pumice lubricated with rubbing oil (light
mineral oil). I was carefully, but not careful enough, not to rub
through the finish at the edges. I stopped with 4F pumice rather than
rottenstone since I wanted a satin finish.
Stage V: Waxing.
I coated the table with one coat of black wax followed by one coat of
natural (brown) wax. The black wax filled some of the most prominent
remaining white pores, and the natural wax sealed in the colored wax.
The results were quite satisfactory. The color is good, dark brown
with black markings. The clarity is excellent: the rays are luminous
and the grain pattern and color changes in a complex mosaic as the
viewing angle changes. The pore pattern is subtle and interesting,
and still apparent to the touch. The finish has a pleasant satin look
and feel. There is minimal streaking caused by the wax. If I were
finishing a piece that was not subject to the wear of a table top, I
would use shellac instead of varnish since it would result in a
thinner finish film. Aside from that, I would use this method again.
--
Todd Knoblock
to...@microsoft.com
http://www.research.microsoft.com/research/graphics/knoblock/
Jeff Mayhew
Hi Todd:
After applying aniline stain and a sealer coat of shellac (sometimes
done in several stages to bring out the ray-flake), I find that a
"filler" coat of BLACK PAINT, which is wiped off so as to fill the
pores and nothing else, does a good job of simulating a popular
Mission finish of the period. I've discussed this with several others
who have all conceded that the "filler" steps alluded to in the period
literature are really nothing more than an attempt to overcome the
shortcomings of the stain (i.e. a failure to reach the pores), not
to truly fill.
I share your feelings about fuming.... I did the experiments, and
got results ranging from dark green to sick brown to black.
I don't like wax much though... too fragile.
Regards,
Jeff Mayhew
jma...@cris.com
Bob
Great experiment Todd! Only thing I can add is to treat the wood with
tannic acid rather than with water before any stain is applied. Using
tannic acid increases the reaction with fuming and gets a more uniform
distribution of tannin across the grain of the wood.
Bob
You can't miss what you ain't got,
You can't lose what you ain't never had.
- Muddy Waters
P. Rose
In article <01bb8ad3$d9430040$3a8a389d@toddk>, "Todd Knoblock"
<to...@microsoft.com> wrote:
>For the past six months, I have been working on a coffee table in the
>style of Gustav Stickley. For the past twelve months, I have been
>working on finishing it.
[snip]
What follows is the list of steps used in
>finishing the table along with a list of the considered, but rejected,
>alternatives at each step.
[snip]
>* Fuming. Traditionally, the ground color would have been provided by
> fuming the piece in an air-tight tent with 28% aqueous ammonia, with
> a touch-up with alcohol based aniline dye. I actually tried fuming
> sample pieces off the table, but was dismayed with the variation in
> color and with the quality of the color on the sample pieces. The
> ammonia reacts with the tannin in the wood, and different pieces of
> wood have different levels of tannin. Also, the sap wood has much
> less tannin than the heart wood and so is almost unaffected in
> fuming. I was aware of these problems before starting and had tried
> to select wood that was similar in color and to carefully cut away
> all the sap wood. But, the color after fuming ranged from light gray
> to gray-black and it turns out that some sap wood had made it into
> the table despite my best efforts to eliminate it (it stays white).
> Old books recommend using wood all from the same tree, but this was
> not an option at my local lumber store.
> I was after a brown or brown-gray color, but did not like the
> gray-gray that resulted from fuming. It did not match the kinds of
> colors I had seen on old fumed-finished pieces. There are a few
> species that are all marketed as "white oak", and it is possible I
> had one that was a different species (burr oak?) than the traditional
> "white oak" fumed pieces I had seen. It is also possible that the
> color I have seen is only formed over many years of mellowing or
> developing of the finish.
[lotsa good stuff snipped]
Todd,
Thanks for a very interesting and informative post! This is a "saver" for
me, and I'll be taking a while to digest it all.
However, being a G. Stickley fan who early-on embraced the ammonia fuming
technique, I find part of the fuming results you described to be much at
odds with my experience. In particular, the fact that you obtained a light
gray-gray to gray-black color (rather than brown) leads me to question
whether or not you applied any finish (i.e., oil or varnish) *after* doing
the fuming. The color of fumed (but *unfinished*) oak tends to be pretty
light and puky. I guess "dull, light grayish/brown" is an approximation to
it. But when a finish such as Watco natural (or others) is applied, it
(the fumed color) immediately changes to a dark chocolate brown or
brown/black, depending on wood and fuming time.
As to the problem of color variation after fuming, I agree that this
certainly can happen. Even with wood from the same board (let alone just
the same tree). Again, thanks for all the useful detail about your
experiences.
Phil Rose
Jeff Mayhew
pjr...@servtech.com (P. Rose) wrote:
[snip]
>However, being a G. Stickley fan who early-on embraced the ammonia fuming
>technique, I find part of the fuming results you described to be much at
>odds with my experience. In particular, the fact that you obtained a light
>gray-gray to gray-black color (rather than brown) leads me to question
>whether or not you applied any finish (i.e., oil or varnish) *after* doing
>the fuming. The color of fumed (but *unfinished*) oak tends to be pretty
>light and puky. I guess "dull, light grayish/brown" is an approximation to
>it. But when a finish such as Watco natural (or others) is applied, it
>(the fumed color) immediately changes to a dark chocolate brown or
>brown/black, depending on wood and fuming time.
>
>As to the problem of color variation after fuming, I agree that this
>certainly can happen. Even with wood from the same board (let alone just
>the same tree). Again, thanks for all the useful detail about your
>experiences.
Yikes!!! Why didn't anybody mention this before? I've seen all
sorts of references in the literature to fuming, and have never
heard this business about the true color coming out only after a
finish is applied. I mean, we all know that the depth or darkness
of a color will change (as with stain), but to go from grey to
brown??
For example, Gene Lehnert's Morris chair article in "Fine Woodworking"
has a color photo of the author peeking at his creation under the
fuming tent, and it's a glorious nut-brown color. What gives??
Guess it's time to do some more experiments with that gallon
of killer ammonia I bought for this purpose years ago......
Thanks, Phil... live and learn.
Jeff Mayhew
jma...@cris.com
P. Rose
>Great experiment Todd! Only thing I can add is to treat the wood with
>tannic acid rather than with water before any stain is applied. Using
>tannic acid increases the reaction with fuming and gets a more uniform
>distribution of tannin across the grain of the wood.
Bob, tannic acid treatment is something which has occurred to me in recent
months, but I haven't tried it or seen any info about it. Is this
something you've done, and can you give any specifics? Thanks.
Phil Rose
pand...@gmail.com
Thanks for all your research and for sharing. I'm going to follow your process on my Stickley coffee table.Planning on building another one for myself.
Paul
Brewster
> On Thursday, August 15, 1996 at 3:00:00 AM UTC-4, Todd Knoblock wrote:
>>
>>
>>
>> The results were quite satisfactory. The color is good, dark brown
>> with black markings. The clarity is excellent: the rays are luminous
>> and the grain pattern and color changes in a complex mosaic as the
>> viewing angle changes. The pore pattern is subtle and interesting,
>> and still apparent to the touch. The finish has a pleasant satin look
>> and feel. There is minimal streaking caused by the wax. If I were
>> finishing a piece that was not subject to the wear of a table top, I
>> would use shellac instead of varnish since it would result in a
>> thinner finish film. Aside from that, I would use this method again.
>>
>>
>> --
>> Todd Knoblock
>> to...@microsoft.com
>> http://www.research.microsoft.com/research/graphics/knoblock
>
> Hi Todd,
> Thanks for all your research and for sharing. I'm going to follow your
process on my Stickley coffee table.Planning on building another one for
myself.
> Paul
>
As to fuming, I do that a lot with my White Oak pieces. You really don't
get the correct (final) color until you top coat it, then you get the
golden brown (I top coat the raw fumed wood with Danish Oil (ringing out
the danishes for the oil can get tiring 8^)
I use hardware store ammonia (5-10%) with good results after 8-12 hours,
but you are correct about the sap wood and you need to monitor every few
hours in case of wood from different trees. Basically it comes down to
experience. One of the redeeming qualities of fuming is the browning
goes deep, no worries about leaving bare spots if you do a bit of finish
sanding or accidentally rub through a sharp corner.
-BR