Hand Tool Headlines
The Woodworking Blogs Aggregator
The Barn on White Run
In Roubo’s description of the finishing processes and materials included in L’art du Menuisier (and thus our To Make As Perfectly As Possible translation) he used the word “juncus” when referring to the fibrous plant from which the polissoirs were made. At the time we had competing dictionary definitions and identifiers, “rush,” “grass,” and “straw” all showed up in one dictionary or another, and in the end I decided to simply use the word “grass” if I recall correctly.
Yannick Chastang, like my Roubo Project collaborator Philippe Lafargue, was trained in the full multi-year program at Ecole Boulle in Paris, chided me that Roubo chose the word “juncus” on purpose and I should have as well, at least in concert with the word “rush.” Fair enough, but in retrospect since the word Juncus refers to a genus with over 300 species of grassy rushes I cannot beat myself up too much for that editorial decision.
I was talking about this to Mrs. Barn one day, she being a botanist/mycologist by training, and she said something like, “Well, you are in luck since we have Juncus effusus growing around the pond.” She took me outside and sure enough, we have a number of fairly immature clumps at the shore of our pond. When Daniel the stonemason was here building the hand-knapped dry-stack wall a few months ago he mentioned that he had loads of juncus growing around his pound and I was welcome to harvest as much as I wanted.
As a break from our activities during ManWeek John and I took the morning an went to Daniel’s place to harvest soft rush, or juncus. We first spent a minute ogling his greenhouse. Mrs. Barn will be most impressed with it when we visit again.
Then we headed to the pond and there was indeed a multitude of soft rushes ringing one end of it. In less than an hour of harvesting we had the back seat of the Envoy completely filled.
Back at the barn we sorted and arranged the rushes to dry in the sun before moving them inside a few days later.
Yannick avers that polissoirs made from these fibers have a very different feel and performance than the ones I get made from sorghum broom straw. After this material gets fully dry and I make some Juncus polissoirs I will be able to make my own determination on that.
While I was occupied with the Roubo bench slab in the center hall of the barn John was a dozen feet away in the classroom tinkering with the Winterthur ripple molding cutter. When we gathered earlier as a group we identified a number of modifications that might serve to transform it into a reliable, precision machine. I ordered all the materials and supplies we thought we needed for this undertaking so everything was ready to go for John to dive in to making these modifications a reality.
As a moment of review, the ripple molding machine is simply a contoured scraper being drawn across a length of wood, with either the scraper or the workpiece being undulated by some sort of linear pattern. In short, a ripple molding is the result of controlled chatter.
In the case of this machine it is the cutter that remains fixed relative to the length of the frame, but which undulates up-and-down via a horizontal “follower” rod affixed to the cutterhead frame, pressing down on the pattern running the length of the machine frame. We found in our earlier efforts that either the pattern or the follower ere being degraded and even destroyed by the very process of creating the moldings.
I do not know how this problem was dealt with historically, but for our applications we decided to replace the extant follower rod with a new rod and tiny roller bearings to instead ride along the pattern, transferring the up-and-down impulse without friction to the cutterhead. John spent extensive time retrofitting the cutterhead to accommodate this modification without damaging or changing irrevocably the machine as it was presented to me.
After installing the new follower system John reported to me with a grand smile that it as a perfect solution to the problem, and would guide our design considerations as we move forward with new machines in both our futures.
When I last left the oak Roubo bench 4+ years ago it was still quite ways from being done (one of the great benefits of building a bench a la David Baron is that it can get done in a week). The leg tenons were all cut, but only two of the dovetailed mortises and none of the rectangular mortises, so clearly a lot of drilling and chopping was in store. There was nothing exceptional about the task or process other than it required flipping the top a couple of times to get the job done. The last two dovetailed open mortises took about an hour to knock out.
Drilling and chopping the closed mortises went smoothly. For three of the four. And the fourth? Grrrrr! For some inexplicable reason I switched from a Forstner-style bit to a long auger bit for my drill, and it went astray. Not just astray but bound tighter than a drum and would not move forward or backward (a theme that was not yet fully played out). After a lot of fussing and fuming I was eventually forced to drive it through the other face using my sledge hammer. Sheer brute force. I was reminded of my late friend Mel Wachowiak’s quip, “With enough force you can pull he tail off a living cow.” Or drive a 7/8 auger bit through an inch of solid oak.
This blew out a chunk of the face adjacent to the mortise, leaving me less cheery than you might expect, my anger being tempered only by the fact that all this damage took place on the underside of the slab. An hour later I had knitted together all the splintered wood and glued it back in place to leave overnight. In the end it was a patience-expanding experience.
The good news is that the repaired place (epoxy and shavings filled) held up perfectly when chopping the mortise in that area. The repair felt just like the adjacent wood and held a nice crisp corner with no chipping or fracture.
So now the mortises were all done and seemed to provide a nice snug fit, and I was looking forward to driving the legs home in the morning.
Oh, about that…
My acquaintance Bill Robertson, maker of astonishing miniatures, is featured in a new TED Talk. Watch, and prepare to be astounded.
With a little logistical cogitation John and I, both 60-somethings and neither of us mesomorphs, managed to maneuver the 300+ pound top of the French Oak Roubo Project workbench out into the light. Immediately I was struck by both the magnificence of the 240(?) year old white oak slab, and the waney void adjacent to a glue line on the underside of it. I suppose at one time I was just going to leave it as-is, an admittedly foggy memory going back four years, but given that one of the leg mortises needed to go right through the flawed region I decided instead to fill it. I could have grafted in another piece of oak but instead fell back on a tried-and-true method of repair that I have employed several times in the past as it was especially well suited for a repair of this size.
I first sized (primed) the margins of the effected area with standard West System epoxy, thinned about 25% with acetone to get deep penetration. One of the reasons for any potential epoxy failures, whether in adhesion, consolidation or filling, is that the epoxy does not penetrate adequately to knit the entire construct together nicely. What then often happens also is that the density differences between the high density inelastic epoxy and the less dense, much more elastic wood, may result in a fracture at their margin when they are intimately bound together in a cyclic stressful environment. The diluted epoxy addresses the first of these problems, the filling of epoxy with large wood flakes addresses the second.
In this case I ran a scrap of oak through the power planer to yield the typically large shavings you would expect from the machine. I took handfuls of these shavings and packed them down into the void that had been previously primed with the thinned epoxy.
I then drizzled un-thinned epoxy on top of the wood flakes, then sprinkled on more shavings and packed them again through some wax paper. I let the entire fill to harden overnight.
An additional feature of fills like this is that when the volume is large enough, the exothermic reaction of the epoxy hardening causes the adhesive to actually boil in place, aerating the fluid as it hardens, reducing further the density of the hardened fill. This is a very good thing.
The resulting repair is much closer in density to the wood, thus reducing the risk of a system fracture at their interface, and yields a repair that can be easily smoothed with a rasp or Surform tool.
The success of the repair can be clearly seen in the edges of the mortises I drilled and pounded through the slab and the repair (next blog post). It held together wonderfully and had working properties nearly identical to the adjacent oak.
With our ambitious agenda awaiting us for the Man Week at the barn, our first task was to begin the Tetris game that always seems to be on tap for any type of reorganizing the shop. The ripple molding machine was easily accessible for John but I had to move a ton of stuff to get to the FORP I workbench parts that were behind the parts for all the other workbenches that are not yet finished, and a large pile of old oak salvaged from the shack deconstruction that we were working on when I crossed paths with the cranky wheelbarrow that put me out of commission for the better part of a year.
The first thing I noticed from the pile of salvaged oak was the presence of frass in between each piece of the stack. It might have been old frass from a no-longer-active infestation or it might not. It was not an extreme amount but I was not going to take a chance as I was going to be making furniture for the cabin from it.
I mixed up my typical batch of insecticide with a gallon of marine anti-freeze and two 8 oz. cups of dry borate-complex powder (Disodium Octaborate Tetrahydrate), then mixed with a paint stirrer in my cordless drill.
I used a cheap garden sprayer to saturate the boards and stacked them under plastic to let it all soak in thoroughly. After 36 hours I set them against the barn to dry, and two days later moved them inside and will put them to use when that project moves to the top of the pile.
If you’re paying attention you might know by now I will be one of the presenters at the upcoming Colonial Williamsburg annual conference Working Wood in the 18th Century, January 25-28, 2018.
I have two time slots, the first being a discussion of the acouterments of a Parisian woodworking atelier in the late 18th century, including Roubo workbenches and ripple molding machines. If all goes well we will be demonstrating these machines, making ripple molding right there on stage. My second session will be the concluding presentation of the conference IIRC, reviewing and demonstrating the practice of woodfinishing of the era.
I hope to see you there. Say “Hi” if you make it.
Varnishers of The World, Unite!
This post is in great part my celebration of a grand circle of friends who provided me with the wood I needed for the project.
In correspondence with the client for interpreting the c.1820 writing desk, it was clear that he wanted something made in the manner of craft technology of the period, and if at all possible, using wood of the period, or at least very old wood. Since the task of acquiring verifiable 200 year-old mahogany was at best an iffy proposition I simply determined to find the oldest, best wood I could find. In fact I already owned about half the wood necessary for the project thanks to my own acquisitional proclivities.
Among my inventory was a superb piece of dense, lightly figured mahogany I needed for the veneers that would wrap the box of the desk. It was one of the three critical pieces I needed.
The desk writing surface was a second vital component and I sent out requests to everyone I knew who might be able to supply my needs. Before long a UPS truck bearing the piece I needed showed up in the driveway. Then a second. And a third. And a fourth. Sean, Ben, and Alf all contributed spectacular pieces to the venture.
One last look through my inventory uncovered the final piece of this particular puzzle, a wildly figure slab of flame crotch that was needed for the veneers on the outer leg elements.
But that was not the end of it. My friend John brought a small pile of vintage mahogany with him to the next MWTCA gathering, and I took it off his hands. Josh emailed me about a stash he had, and delivered it to me.
Then my orthopedic surgeon told me he had a storage unit full of pre-WWI era lumber including some prized mahogany. I loaded all that was there and headed for home.
In the end I would up with enough vintage, unused dense swietenia to make at least two additional desks and, thanks to the willingness to part with some of their holdings by my circle of friends, I probably will.
Plans done? Check. Wood in-hand? Check. Ready to dive in? Uh-h-h-h-h.
My recently scheduled barn workshop, “Make A Traditional Workbench,” was mercifully “cancelled” due to the fact that all four of the scheduled registrants notified me they were not coming. No students, no workshop. I say “mercifully” because it would have started the day after Barndaughter’s wedding weekend, and I was already worn to a nub. Nevertheless, my friend John, who participated in the workshop last year and was scheduled to be my teaching assistant for the week, decided to join me anyway for a grand week of man-time in the man cave, a/k/a The Barn.
We had a delightful week of fellowship and working on projects; John concentrated on modifying and tuning up the Moxon-style ripple molding cutting machine while I emphasized bringing my FORP workbench from many years ago closer to completion. In addition, John being a trained theologian and well-engaged citizen of The Republic, our conversations were vibrant and varied, and by the end of the week we were almost sentimental about our shared experiences.
The success of the week can be summarized in the observation that by Friday afternoon it looked like a tool-and-shavings bomb had been detonated there. I’ll recount our adventures in greater detail in coming posts. Stay tuned.
Recently I was a presenter at the SAPFM Blue Ridge Chapter on the topic of saw sharpening. I would not call myself an accomplished saw sharpener mostly because my results are inconsistent, generally due to the lack of hours at the task. But there are times when the result is excellent, for example my favorite old back saw that I last sharpened sometime in the 1980s and has cut hundreds of joints since, and remains sharp and the cuts crisp and clean.
Using some oversized props I reviewed the notions of tooth spacing and shape (rake, and fleam), and how these come into play for crosscutting and ripping at varying degrees of scale, precision and effectiveness.
I the moved through the nearly unlimited options for holding the saw during sharpening, and finally set up to actually doing some sharpening under less-than-ideal conditions of a large lumber warehouse with diffuse illumination. I find that getting the lighting correct is perhaps the most important thing when sharpening a saw, and this setting wasn’t it.
My explanation of the process was certainly better than the actual sharpening during the demo, but I think the attendees got the idea.
As an aside, I was delighted I had my petite Roubo bench with me and realize that it has become a treasured part of my traveling side-show kit, as it fits neatly into the back of my S-10, is moved easily with a hand truck, and performs most excellently.
Recently when I was visiting plane maker extraordinaire Steve Voigt I had the chance to use his Sterling plane setting hammer, and I liked it and said to myself, “Self, you gotta have one of those. Right now.” Since even in the era of the interwebs, on-line purchasing does not provide instantaneous delivery so I got up the next morning and made a plane setting hammer for myself, using scrap from my inventory of stuff.
My first step was to take a piece of brass and turn one end of the head on my wood lathe, which is easy enough to do when using turning chisels set up as scrapers rather than turning gouges (virtually all of my turning is with beefy scrapers with very rare use of gouges; it’s an old habit from my early years in the pattern shop). I then turned a wooden end of the head from a scrap of lignum vitae, then drilled and tapped both sections and screwed and epoxied them together.
Then I grabbed a piece of exotic wood from the waste bin (probably bubinga) and made a handle in about ten minutes. I drilled the hole in the head through which the handle passed, then used material from an ivory piano key as the wedges for the handle in the head. All told I spend maybe 90 minutes on this hammer.
The result was immensely gratifying and its weight and proportions and performance have made it my “go to” tool for this purpose, and several folks I have shown it to have expressed interest in purchasing one. Who knew? I guess I will have to get set up to do it. I might have to actually order some supplies.
Many, many months ago I was commissioned by a client who asked me to create an interpretation of an early 19th Century desk. I approached the original artifact caretakers, requesting a set of the drawings I knew had been made for that artifact. My request was declined, so my first task was to derive a working set of designs based mostly on images from the web.
About the time I was set to begin work on this project I crossed paths with an angry wheelbarrow, and the resultant broken hip left me out of action for many months. One thing I could do was sit at my laptop and noodle up some templates. I started with the images from the web and the handful of measurements that were also on-line and got to work. My importing the pictures into Photoshop and distorting them I got something resembling “face on” images for the critical elevations. Still, some was spitballing at this point with details to be resolved at a later time.
By importing these manipulated Photoshop images into a vector drawing program, in my case CorelDraw, I was able to ascertain the various measurements and contours I needed for the construction templates. If I was either younger or more computerily cognizant I would have use SketchUp, which I believe can do most of this processing almost automatically, but at this point in my life I am trying to forget computer applications, not learn new ones.
Should you be in a place to need construction details, measurements and proportions based solely on photographs it is best to have images where the camera is square to the desired face of the furniture, at point zero on both X and Y axes, with the longest possible distance from the object . From there it is a piece of cake to get the details darned near perfect, provided you have at least one or two firm dimensions known. At some point upcoming I will write bout the best way to capture the images with an eye towards creating drawings, but I have not written that missive yet.
For this project I was able to derive all the dimensional and profile details I needed, so soon enough I was off to the bench. Working in the manner to which I was accustomed from my time in the pattern shop I drew out the detailed drawing at full scale on a sheet of clean plywood. Once I was satisfied with the results it was time to get started with the building.
But first I needed to gather the necessary lumber. Stay tuned.
One of my interests for some time has been “Every Day Carry” practices and even forums on-line discussing the stuff we have on us every day, with a special emphasis on emergency situations. I find the ingenious creativity in manifesting the ideas to be captivating sometimes, and over-the-top zombie apocalypse silly at other times. The current issue of Backwoods Home magazine, one of the two or three periodicals I take these days, had a feature article on the subject that prompted me to reflect on my E.D.C. in the shop. Since pretty much everything I need is within reach or a few steps at most, the inventory is much, much smaller than when I worked in Mordor and my tactical vest was packed to the gills.
This is what I carry virtually every day, all day long when in the shop.
First off is my Victorinox Spirit multi-tool, which I carry any time I have pants on, whether in the shop or not. Over the years I have owned and used a couple dozen multi-tools and this one is the best I’ve owned, hands down. Certainly pricier than the $10 knock-offs at the Dollar General, but I use mine hard every day with nary a complaint from me or it.
Next is my DelVe square from Woodpeckers, invented by my friend Tom Delvechio. Simply the perfect layout tool for the hip pocket. I bought an extra one just in case this one gets lost or stolen.
An antique folding two-foot boxwood rule is my newest addition to the ensemble, and I just love its utility and compactness. I picked it up for not much money at a tailgating session at MJD Tools one summer and it has been part of the kit ever since.
A 6″ Starrett machinist’s rule has been in my carry tool kit for as long as I can remember. They never go bad nor out of fashion.
Finally, the only thing I did not have in the picture was an LED flashlight, probably because I just forgot to pull it out of my pocket. My favorite value in this tool category is the Ozark Trail pocket flashlight that I buy in the camping section of Wally World. I have several, and they perform admirably and seem almost indestructible. I make use of a small flashlight usually several times a day.
That’s it. Even in my own workshop, I have tools in my pockets all the time.
With the walnuts raining down and the their leaves yellowing, and the sound of chain saws off in the distance, we are definitely moving from the cusp of autumn to the reality of it. Last week my dear friend Bob came over to bring down several dozen tons of trees for me to prepare, mostly for next winter and perhaps the one after that. We already have more than half of what we need for this coming winter but I really want to get way ahead of future demands. The local tradition is to enter every winter with two years’ worth of firewood in hand, and that is my goal as well. Our objective for this cut was to select several trees that were either damaged or in the wrong place (I am trying to establish a cleared path to the southwest of the barn so I will no longer lose winter sun at 3PM), get them on the ground for me to work with, and emerged unscathed ourselves. In two hours we accomplished all of the above.
Working with Bob is a great learning experience as he has been felling large timbers ever since he was a boy. I am fine with cutting it up once it hits the ground, but I’ve heard there is unending paperwork if you drop a twenty ton tree on yourself so I defer to him in this enterprise. He stands at the base of the tree looking at its trunk and crown, judging both the direction it would like to fall and the degree to which that trajectory can be altered. Then he sets to work, back notching then felling the tree. In every instance of the two dozen trees we (and by “we” I mean “he”) dropped it came down exactly where he wanted it to come down.
Now it is up to me to cut them into short bolts, process them with the hydraulic splitter, and stack them to season. Starting next week I will begin filling the firewood crib and the front porch with a mountain of BTUs.
The most beautiful sound in the depths of winter is when Mrs. Barn remarks, “Hmm, kinda warm in here, isn’t it?”
A couple weeks ago I ventured into the barbarous climes of Mordor to deliver the workbench to the Library of Congress Book Conservation group. If the traffic and multitude of high-dollar construction projects are any indication, the travails of the provinces are not being felt in the capital city. In fact it looks like a boom town that has four trillion of our dollars at its disposal every year. And since we apparently are not motivated enough to demand that they stop spending those four trillion dollars every year on us, that trend line will remain unchanged.
The logistics of getting into a secured facility (and in Mordor virtually every facility is secured) is a challenge. It turned out that the most efficient way to get the workbench into LC was for me to drop it off at the curb in front, with LC staff taking delivery of it there. Once I parked and rejoined them we were able to get through the security checkpoint and proceed to the conservation lab. Admittedly, I felt under dressed with my Victorinox Spirit muti-tool sitting in the van outside.
The path to the final home for the workbench was uneventful, and the crew there was delighted to get their new tool. Particularly pleased were the petite members of the staff, many of whom wrote me a “Thank You” note for taking their physiques into consideration when fabricating the variable height configuration of the bench.
The bench fit perfectly into the tiny Tool Room space they have, and after I spent a little time explaining its features it was given some tryouts almost immediately.
And then I escaped before the Dark Eye poisoned my heart any more.
The newest PopWood arrived int he mail recently and it contains my latest article for them. If the topic interests you, I hope you will join me at the Marc Adams School of Woodworking where my workshop on parquetry will revolve around making and using these jigs.
As we run-up this week to nuptials for Younger Daughter we were blessed with a visit from her last weekend. Much of the time she spent with Mrs. Barn doing wedding-y stuff, but she spent a few hours in the shop with me turning a bowl. The wood for this bowl came from a plum tree in the Maryland house yard that died of natural causes some years ago (she remembers climbing the tree as a tyke), and I harvested the wood and set it aside for something special. This definitely fits the description.
I had in recent months found the faceplate for the lathe and ordered a threaded insert from Woodcraft so it could be put to work. Before she arrived I mounted the piece on the faceplate and roughed it round (she is not yet experienced enough to bring a really rough piece to round comfortably). The lathe is a bit high for her, so in the early stages she was most comfortable with the scraper tucked in the armpit. I will be building a lower base in the coming weeks.
I gave her only a few pointers as she developed the outer shape she wanted.
Before long she had the outer surface defined and embarked on an initial sanding and polishing.
With the base established and the shape determined it was time to remove the faceplate in favor of the small bowl chuck and get started excavating the interior.
Soon she was in pretty deep.
We stopped for the night, but on returning the next day she refined the shape and surface.
To be sure the watchful papa bear was never far from the action. The working height was just plain awkward for her but she hung in there without complaint.
After the final shaping she moved to sanding and then polishing with beeswax melted into the surface, buffed with a linen rag while turning. She particularly liked my method of placing a dry sponge between the hand and the sandpaper, it allows greater vigor with less heat.
And here it is, an heirloom with a priceless memory attached. In all likelihood it was our final private time together with her as Miss Barndaughter until those moments just before I walk her down the aisle, and it was a precious treasure.
Doggone, something must’ve flown into my eye…
With the foundation laid for good finishing it was time to move on to undulating surfaces, the kind of finishing that gives many woodworkers fits and nightmares. Fortunately it is no more complicated or straightforward than finishing plain flat surfaces. It’s all about surface prep, varnish prep, and tool selection.
Switching to the “carver’s model” polissoir the surfaces were burnished in preparation for varnishing.
Then, on to applying the varnish. The true key to success is the right brush, a fine bristle watercolor “Filbert” with a rounded tip.
The Filbert allows for tremendously good “drape” of the bristles around the surface, not sqeegeing off varnish with the resulting runs like you might get with a square tip brush.
A few applications of the shellac varnish to these surfaces and they were ready to set aside, to be burnished with steel wool and waxed later on.
Next we revisited the luan panels we had started the day before, undertaking a light scraping with disposable razor blades followed by a brief but vigorous rubbing with 0000 steel wool. I have found scraping to be not only historically accurate (obviously not with modern disposable razor blades, but the concept and practice are still the same) but now to be an integral component in my finishing process.
Then another inning of shellac application, followed at the end of the day by the third and final inning. By then the surface was beginning to get some sparkle.
One last exercise was to finish a raised panel door. I do not recall where these came from but they have served me well in this regard for many moons. Again, a few applications of shellac followed by rubbing out with steel wool and paste wax yielded a luxuriant surface.
The large panels were rubbed out the third morning with steel wool and wax, and buffed with soft cloth. The result was, as one participant said, “The best looking piece of luan ever!”
By mid-day on Sunday the party started breaking up, but the students left with a new confidence and a sharpened set of skills. Folks may be reluctant to come to The Barn on White Run because of its remote location, but once here they always love it and go home with more knowledge and skill than they arrived with. That’s not a bad outcome.
I recently hosted and taught a “Historic Finishes” workshop at the barn, with five attendees from around the country and my long-time friend DaveR as a teaching collaborator. The objectives were to help the students overcome any hesitancy about finishing by learning new habits and techniques, and the results of the exercises indicate success.
Our first exercise was the one that was most time sensitive in that it required three inning of finishing over two days, which was pushing the technology a tad. Fortunately the weather was cooperative. The task at hand was to take an essentially unprepared 24″ x 48″ panel of luan from Lowes to see what could be done with it, some well-prepared shellac varnish, and good brush. After a brief scuff sanding with 220 they began to lay down the 1-1/2 pound shellac as I have taught multitudes before them. The purpose is of exercise to overcome the trepidation in applying shellac spirit varnish.
Next came the grain-filling of some solid mahogany panels with molten beeswax as the foundation for pad polishing. This was how they did it in the old days, and it is still my preferred technique. The wax was melted in using a tacking iron (I cannot believe I did not get any more of this on camera), then scraping off the excess and buffing it out with linen.
Even at this point the results are impressive and in some circumstances the finishing would be called complete.
DaveR came on stage next to introduce pad spirit-varnish polishing, sometimes known as “French” polishing,
All eyes were glued to Dave as he walked through the process of this technique which has garnered much (undeserved?) mystical reverence.
He demonstrated the process of making a good pad, or “rubber,” which can last a finisher for decades, and before long they all set to making their own.
And the padding began.
Before long we were seeing some mighty fine sheen.
It was time to introduce the newest tool in the contemporary finisher’s kit, the polissoir. Everyone got their own brand new one that needed to be tuned up on a piece of fine sandpaper.
And out to work, first over bare scraped wood, then in concert with beeswax that had been scrubbed on to the surface.
Again, the final results were immediate and gratifying.
Up next, brushing carvings and other undulating surfaces.