CutList Bridge
CutList Bridge
NESAW
New England School of Architectural Woodworking
CutList Plus fx
CutList Plus fx
Ads By Google


Lie-Nielsen Toolworks

CabWriter Home Page

« Previous PageNext Page »


Wood Expansion Calculator 1.0 Sample Input Page Wood Expansion Calculator 1.0 is now available. The picture at left shows some of the major changes.

When using the Relative Humidity input option the temperature can be specified in either Fahrenheit or Celsius. The other temperature scale will be calculated and also displayed.

When choosing a Calculation Mode input boxes appear for receiving user specified dimension inputs. These can be supplied in either Imperial (US) or Metric units.

Metric dimensions require one, and only one, unit; either m, cm or mm. The following are the only valid inputs:

     i m
     i cm
     i mm
     d m
     d cm
     d mm

Where i is an integer whose first digit cannot be a zero. Where n is a decimal number and the first digit cannot be a zero unless the decimal point is immediately to its right. In fact, a decimal number less than 1 must begin with a zero such as 0.967 cm. Note the space between the dimension and its unit. This is required.

Imperial dimensions may require more than one unit, for example 3′ 7 1/64" is a valid input. The following are the only valid inputs:

     i"
     n/d"
     i n/d"
     f’
     f’ i"
     f’ n/d"
     f’ i n/d"

Where i, n, d, and f must be non-zero integers who’s leading digit also is not a zero, e.g. 0123 is invalid. Note, unlike the Metric units, there is no space between the dimension and its unit; in fact a space will create an illegal entry.

Wood Expansion Calculator 1.0 Sample Output Page Consistent with supplying the alternate units results as was the case with temperature, the output will show results in both systems of measurement. The type supplied by the user will appear first and the alternate second. A sample results page can be seen at right.

A number of bugs were fixed in this release but do not materially change the functionality.

There are future changes planned for Revision 2.0 that include but are not limited to the following:

  1. Provide the capability for users to supply their own Regions and corresponding EMC values.
  2. Provide the capability for users to supply their own Species and corresponding shrinkage factors.
  3. Provide memory so that user settings and inputs are restored when next the tool is opened.
  4. Strengthen the help messages, especially for the dimension inputs, and add more help features.

If you have any suggestions please pass them along in a comment to this post (preferable because others can see them) or forward them in email.

Download Wood Expansion Calculator here.


Well, I have finally started crafting the trundle bed I wrote about in the Trundle Bed Design series. Many family and unrelated projects got in the way of this project for some time. But no more excuses. The show must go on.

Headboard And Footboard Panel Details I decided to begin with building the panels for the headboard and footboard. The headboard requires a panel 22 57/64” tall by 40 1/2” wide and two panels 8 3/4” tall by 40 1/2” wide. The footboard requires two panels 8 3/4” tall by 40 1/2” wide.

The final thickness of the panels is 5/8”, but I prepare my stock for 3/4” and bring it to final thickness on the drum sander after glue up has been completed. This will allow me to take out any slight mismatches in the glue up joints which are unavoidable. In addition the drum sander can bring the finish panel to precisely 5/8” with 220 grit paper. That way, after shaping the edges, I can immediately apply several coats of finish, which I always do before affixing panels in their frames (The headboard and footboard are essentially a frame and panel construction.). Subsequent shrinkage of the panels will not reveal unsightly voids of finish.

Edges Are Always Prepared With A Hand Plane Before Glue Up Preparing stock for glue up requires the standard jointer, planner, jointer and table saw sequence to face and edge the boards. But the final step for me is always preparing the edges by hand with a hand plane. This accomplishes several things. First it removes any oils on the edge that exist from handling or are naturally secreted by the wood. This is especially important if the time from wood preparation to glue up is hours or days. Second the edge is given a glass smooth surface void of machine marks and scratches. Third I get a better edge, i.e. perfectly straight and square.

All these add up to a better looking and stronger glue joint. One of the tests I use for a properly finished edge is that I can get  a continuous, very thin shaving, of equal width all the way to the end, and the length of the shaving is the full length of the board. Notice the shaving above right. A Lie-Nielsen smooth plane is the one I use for the final cuts. But I will start with a jointer plane if the edge is close to straight, or a block plane if I have to cut short local areas to correct for a bow for example.

Headboard Panel Glue Up When I have finished preparing the edges with a hand plane I immediately glue up. If I have a number of panels to do, as in this case where I have five panels, I’ll prepare all the stock on the power tools. But only the edges for one panel at a time is prepared on the hand plane so that the time from edge preparation to glue up is short, keeping the edges from getting soiled or dinged.

I have tested glued edge joints numerous times and always found that a properly prepared and executed  joint will always be stronger than the wood itself. How long a joint will last I will never know because I won’t live long enough to see its failure. But the accelerated life tests manufacturers perform indicate these joints will still be going strong hundreds of years from now (barring abuse such as prolonged exposure to water, high heat or direct sunlight).

One other idiosyncrasy I have is that I always leave joints clamped overnight. True, the manufacture says you can work the wood after only one hour of clamping provided there are no undue stresses placed on the joint. But I am not sure what an undue stress is. This is an analog world we live in. Stresses don’t magically become undue at 10 lbs of force but not 9.9 lbs. So I am conservative but feel much more secure this way.

The Performax Pro 22-44 Is Used To Final Thickness The Panel After curing for an evening the panel is ready for final thicknessing. I do this on my Performax Pro 22-44 drum sander. To gauge when a side has been entirely sanded and flat I mark the panel with red carpenter’s crayon in wide horizontal lines. When the marks are completely gone I have succeeded in flattening the side. See the picture at right. I use 220 grit paper for this final step. I will sand it one more time just before applying finish with 320 grit and an oscillating rotary sander.

I am careful during glue up to put the good side of the panel up, i.e. away from the clamp’s bars. This allows me to clean the entire surface unimpeded by the bars of the the clamps. See picture above left. I clean the other side too, but the bars always obscure some glue. When dried the backside will have little glue hills which I level with a putty knife. Still, there is remaining glue to be removed. So the backside is the one I drum sand first. Then I turn it over, mark the good side and continue drum sanding until I reach final thickness.

On a panel this wide each pass actually requires two passes. As you can see in the picture above right, the panel is wider than my drum sander. The 22-44 in the name implies you can sand a 22” wide panel in one pass, or one as wide as 44” in two passes. One note of caution about drum sanders; you must not let the work piece stop while going through the drum sander. If you do the sander will sand a horizontal valley into your piece deep enough that you may not have enough thickness left to remove it.

Squaring The Panel On My Large Panel Cutter Once the panel is thicknessed I use a hand plane to create a square and straight reference edge. I then use that edge in my large panel cutter to square the panel to finished length. This panel cutter has been a life saver and workhorse for me. If you don’t have one I strongly suggest you make one soon. With it I can cut large panels (wider than a kitchen cabinet end panel) perfectly square every time, and with ease. The panel shown is 24” wide and 40 1/2” long. This panel cutter uses both table saw slots, has a high fence to keep your hands away from the blade and has a block that completely covers the saw blade as the fence passes it.

Inspecting The Panel With Mineral Spirits (Paint Thinner) When the panel is cut to size I wet it down with mineral spirits to inspect for any remaining glue spots. Hopefully there are none. This step also gives you an idea of what the panel will look like when finish is applied.

Of course, this being cherry, it will darken considerably with sunlight and age. Most of the darkening takes place in the first few months of exposure to strong light, but it continues for a long time. In the picture at right the wood came from two piles, one which had not been subjected to light and one which had (it was on top of the drying stack). These pieces will darken to the same color in a few weeks time.

However, you will notice some sapwood in this panel. Purists argue that you should remove all sapwood when crafting fine furniture. I respectfully disagree. I have always felt that nature does a better job of designing wood than we do. I like to expose all “imperfections” in the wood, including dark pitch pockets in cherry, or cats paw markings. I feel they add to the piece. I am sure that the Shakers didn’t throw out pieces with these imperfections, and if its good enough for the Shakers, it’s good enough for me.

A Full Scale Print Out Is Used As A Template One of the really neat features of SketchUp is that you can print drawings to scale. I printed out the headboard to full scale (1:1). It took about 23 sheets of 8 1/2” by 11” paper, though most of them were blank and I put them right back in the paper stack. I taped one side of the swan neck together and then encapsulated it with self sticking clear plastic and made a template, which I then traced on the panel. Only one side is needed for a template because the curves are mirror images and you can flip the template.

The Delta BOSS Is Used To Remove Jig Saw Machine Marks After rough cutting the swan neck curves, I used my Delta BOSS with course paper to sand away the machine marks left by the jig saw. I usually use my band saw to cut shapes like this, but a 1 1/4” re-saw blade was mounted on it and I didn’t want to take the time to change to a smaller blade. The BOSS oscillating sander does a good job, however, in the end I had to finish the job with lots of hand sanding.

This panel is rather large for the BOSS table so I used adjustable roller supports to carry most of the weight while still making it possible to easily manipulate the panel. Note that the circle in the top middle of the panel is not cut out at this point. If I would have cut it out at this point, the shaper, which will be used to shape the edges, would likely destroy the delicate points that are formed by the circle (see the first picture).

Shaping The Edge With A Large Cutter I Am Especially Alert The next task was to shape the edges. During design of the bed I chose to do this on a shaper because I could get a cutter that would form a wider shape than possible on the router. But the cutter has a rather large 5 1/2” outer diameter. Plus the shape of the swan neck is such that I had to expose most of the cutter to be able to manipulate the panel during shaping. This makes for a somewhat risky and dangerous cut. In situations like this I am always super alert, especially during the start of a cut when the shaper can grab the piece and throw it, or throw sharp pieces at you. Also, I am conscience of where my hands are at all times.

The Circle Is Cut Out With A Jig Saw Finally I cut the circle with the jig saw and repeat the BOSS and hand sanding process. When cutting pieces like this where the panel has to hang over the edge of the table, I make the cut in sections, and support the cut-off by clamping it to the panel. That way it will not unexpectedly fall an split a piece out ruining the panel. These little extra steps can save a lot of work and material and pay for themselves many times over.

The Completed Panel Ready For Finish When the panel is completed I wet it down with mineral spirits again. This time I am looking for scratches or dings. This sometimes happens due to the hard surface of the shaper and BOSS tables. If I find a mark I remove it now. If I were to skip this step the imperfection would surely show up after finish is applied and would be much more difficult to repair at that point.

This concludes Part 1 of Trundle Bed Crafting. In Part 2 I will make the swan neck frames the will encapsulate the panel. Stay tuned.


I have written a number of blog posts dealing with techniques for handling the seasonal change of wood movement and I am sure to write some more in the future. If you have crafted much in the way of furniture, or other pieces constructed from rough wood, you know you have to take wood movement into account in design. Failing to do so will ensure cracks and structural failure. If you do not believe this let me point you to two very good articles outlining the problems and solutions. First, an article that appeared in American Woodworker, 1993 issue number 34. The title is Coping With Wood Movement – How to Build Furniture That Won’t Crack or Split, pages 38 – 43 and written by Jim Cummins. If you don’t have this issue you can find it on the internet by clicking here.

The second article is titled The Shrinking and Swelling of Wood and Its Effect on Furniture by Carl A. Eckelman. I will make this and other references available via Adobe PDF download at the end of this blog. Both of these articles outline specific construction situations, the effects of moisture and resulting wood movement on them, and techniques to avoid the problems that can arise. In both articles the author points out the importance of knowing how much movement to expect and that is the focus of this post.

Wood Movement – The Basics

SEM View Of Poplar - Photo Courtesy Of N.C. Brown Center, SUNY Let’s start with a short explanation of why wood moves. (For a comprehensive coverage of this topic I have listed a number of articles and books at the end of this post.) When wood is alive and green it is made up of soda straw like vessels that tend to run end to end along the length of the tree. These vessels are made up of thin walls consisting of specialized cells. (See the scanning electron microscope photo of Poplar at left.) Both the vessels themselves and the cells that form their walls contain water. When the tree is cut and sawn the vessels begin to lose water and continue to do so until all the water is out of the vessels. At that point the cells that make up the walls still contain moisture. This point is called the Fiber Saturation Point or FSP. It is the point where the tree has given up all its “free water” but the cells are still saturated with “bound water”. This point is very similar in most trees and represents a point at which the tree (lumber at this point) still has a moisture content of approximately 30%. (Moisture content is the ratio of water weight contained in the specimen to its oven dry weight, i.e. no water weight. Many trees, when live, have moisture contents greater than 100% which means that water accounts for more the half its live weight.)

The drying process from green moisture content to FSP, while making the lumber less dense or heavy, does not cause any shrinkage and in fact the lumber is still in its weakest state when it reaches FSP. As the tree (lumber) continues to dry from FSP to 0% (called oven dry) it gives up bound water from the cells, shrinks and becomes stronger. The shrinkage is quite linear from FSP to oven dry and the total shrinkage, expressed as a percentage, can be measured in three dimensions: radially, tangentially and longitudinally. Longitudinal shrinkage, which occurs along the length of the lumber in the direction of the grain, is very minimal and is generally neglected.

Flatsawn (Plain Sawn) Versus QuartersawnTangential shrinkage occurs along a line tangential to the tree’s growth rings and radial shrinkage occurs along a radial from the center of the tree. These two factors are called tangential shrinkage factor (expressed in percent) and radial shrinkage factor and can be related to two common types of lumber: flatsawn and quartersawn respectively. Tangential shrinkage is usually about twice as large as radial shrinkage but they each vary widely from species to species. In real life no lumber is cut exactly quartersawn or exactly flatsawn (also called plain sawn). As a craftsman you have to make a judgment call. If in doubt you might want to consider treating your lumber as flatsawn because that would be worse case.

Map Of Regional EMCs For JanuaryOK, so we know that wood shrinks as it loses water from FSP to oven dry. But the reverse is also true; it expands as it takes on moisture from oven dry to FSP. So what makes wood give up and take on moisture? Its environment. Relative humidity and temperature, and both change throughout the year, usually from dry air in the winter to moist air in the summer. If lumber sits long enough in a given environment it will reach equilibrium with that environment. Suppose for instance that the average temperature and relative humidity have been 80 degrees Fahrenheit and 70% relative humidity for the past two months. It is likely that lumber stored in this environment will come to equilibrium at a moisture content of 12.9%. Clearly higher than one would like if one intended to build a piece that would reside in Arizona. This is why we bring lumber into our shop and let it sit for a few weeks; to let it reach equilibrium.

Map Of Regional EMCs For JulyMost homes do not have humidity controlled environments and are subject to seasonal moisture changes. There have been a number of studies that have measured the outdoor monthly averages over the course of many years. Through the use of computer models, these studies produced maps and tables of indoor seasonal ranges by region of the country. The Department of Wood & Paper Science at North Carolina State University produced such maps in 2003 which you will find by clicking here. These values are called Equilibrium Moisture Content or EMC and represent the moisture environment a piece of furniture would be exposed to, on average, for each month and by region.

Wood Expansion Calculator

An Opening Centric Analysis Of Cherry In The Massachusetts Interior This brings me to the real topic of this post – a SketchUp Plugin tool called Wood Expansion Calculator. I created Wood Expansion Calculator to estimate just how much expansion and shrinkage you might expect in a given structural situation. I developed this tool because I have for years used a stand alone application called Wood Movement Master by Kite Hill Software which does precisely this. You have probably read some of my articles where I used this tool. Unfortunately, it is no longer available or supported by Kite Hill Software or its developer. I decided to pick up the gauntlet and provide the same capability in a SketchUp tool. So here it is: Wood Expansion Calculator ZIP File.

One note of caution. In my professional life (I am now retired) I was an electrical engineer and engineering executive. I started my career when CAD tools were nonexistent and you had to figure things out with paper, pencil and a slide rule. The advantage of this was that you developed a close feel for what you were designing. You knew the expected behavior intimately. Years later, when I was an engineering manager and CAD tools were ubiquitous, it drove me nuts to see a young engineer model a circuit, run the tool and take the results as Gospel. He/she had no feel for, or personal intimacy of the expected results. Don’t use this tool that way. It is meant to give you a quick, efficient, and yes accurate result so long as the input and assumptions are correct. If, for example, the temperature is 100 degrees Fahrenheit today at 90 percent relative humidity, and expected to be 30 degrees Fahrenheit and 10% relative humidity tomorrow, don’t expect wood stored in this environment to respond to that change no matter what this tool might indicate. Moisture takes time to enter or leave a specimen; that’s what equilibrium accounts for. So use judgment as well as the tool.

The tool itself is quite simple to use, and has Help notes for each input. To install the tool use WinZip to extract the one Ruby file and one folder to the SketchUp Plugins folder on your system. When you open SketchUp you can access the tool via Tools/Wood Expansion Calculator or go to View/Toolbars and check Calculators for a toolbar and icon (a tree). What you need to know to use this tools is the stocks current moisture content (best to use a moisture meter to measure this), the region of the country where you expect the final furniture piece to reside, the wood species, the stock type (flatsawn or quartersawn) and the construction situation you want to analyze (board centric, opening centric or breadboard end).

A Printer Friendly Results Of An Analysis If you don’t know the stocks current moisture content you can use a Relative Humidity and Temperature mode to calculate it. But be careful. Consistent with two paragraphs ago be sure to use average temperature and average relative humidity of the environment the wood has resided in for a period of time long enough to reach equilibrium. Do not use seasonal or daily maximums. The wood’s moisture content will not respond to fast or temporary changes.

One other note; board centric, opening centric and breadboard end are not a specific construction method but a representative method. For example, opening centric includes drawer fronts that fit flush in an opening, but could also mean a panel that is framed to make a door. Breadboard end could represent any cross grain situation such as a mortise and tenon. Don’t let the labels limit your use of the tool. Understand what is being analyzed.

This tool currently works only with Imperial or US measurements.

For example 3′ 7 1/64" is a valid input. More precisely, the following are the only valid inputs:
i"
n/d"
i n/d"
f’
f’ i"
f’ n/d"
f’ i n/d"
where i, n, d, and f must be non-zero integers who’s leading
digit also is not a zero, e.g. 0123 is invalid.

In the future I will expand it to accept metric measurements. The region table is the contiguous US only. In the future I hope to get data for other regions of the world. But you can still use this data by selecting regions that you know are close to environmental conditions in your area. Also, in the future are plans to let the user build their own environmental tables or to add their regions to the existing table. Lastly, I plan to add persistence to the tool so that it remembers the last set of conditions analyzed and uses them as the starting point when the tool opens.

Please report all bugs directly to me, and include the conditions that resulted in the bug. Thank you in advance, and I hope this tool serves you as much as Wood Movement Master has served me in the past.

Related Reading

Centennial edition of the Wood handbook : Wood as an Engineering Material. This is an excellent reference book. You can download individual chapters in PDF format. Chapter 04: Moisture Relations and Physical Properties of Wood is particularly pertinent to this blog post.

Understanding Wood – A craftsman’s Guide To Wood Technology by R. Bruce Hoadley is a book that ought to be on the bookshelf of any serious woodworker. Chapter 6: Water & Wood is particularly appropriate to this subject, but the entire book is directed at the woodworker.

The Shrinking and Swelling of Wood and Its Effect on Furniture by Carl A. Eckelman was mentioned at the beginning of this post. It is an excellent article on designing for wood movement.

Download Wood Expansion Calculator Here


On my most recent project, the American Chippendale Mirrors discussed in my last post, I was forced to choose between using a table mounted router or a shaper. Specifically, was how to shape the picture frame molding which was complicated by the use of tiger maple hardwood.

Tiger maple is notorious for tear out whether hand planing, jointing. thickness planing or shaping. I have described in this blog numerous times how I thickness plane the final 1/8” to 1/16” of tiger maple using my Performax Pro 22-44 for just this reason.

CMT 855.902.11 Traditional 1/2” Shank Router Bit The bit used to shape the molding in this project was a CMT 855.902.11 Traditional 1/2” shank bit. Its overall cutting length is 1 5/8” and its overall diameter is 1 1/16”, rather small for even considering a shaper.

The primary decision making criteria between using a router and shaper is the bit diameter. Bit RPM being equal, large diameter bits have a higher tangential velocity compared to small diameter bits. Large diameter bits remove more material requiring more horse power. This is where routers and shapers differentiate themselves (portability is another but not applicable when comparing table mounted routers to shapers).

Routers generally spin at higher RPM, typically 10,000 to 21,000 RPM and range from fractional horsepower to 3 1/2 horsepower. Hence they are useful mostly for small diameter bits. Shapers generally have two or three speeds to select from, usually 8,000 and 10,000 RPM and start at 2 horsepower and range to in excess of 5 horsepower. Hence applicable to large diameter bits.

RPM are directly comparable, but not all horses are equal; a 3 1/4 HP router is not equal to a 3 HP shaper, the latter being much more powerful. I should also mention that shapers tend to be much more hazardous than routers, so special attention to safety is required.

Molding Picture Frames On The Shaper This all being true one would normally mount this bit in a table mounted router and shape away. I started that way and quickly realized I needed to consider the shaper. Even though I was using a 3 1/4 HP variable speed router with speeds selectable from 10,000 to 21,000 RPM I couldn’t achieve a tear out free finish, no matter how many light passes I made. I put the same bit in my shaper and selected 10,000 RPM and discovered I could make tear out free finishes if I cut the molding in five light passes and proceeded slowly and smoothly on each pass. I can’t explain this rationally but I can demonstrate it quite clearly. If someone has a technical explanation I would certainly like to hear it.

This is not the first time I discovered this about tiger maple. In fact, on almost every project requiring tiger maple molding I end up on the shaper. I always try to avoid the shaper because of the long setup time required, but in the end I succumb. So this time I committed to design a fence for my shaper that will allow fast setups, flexibility, and safety. Designing such a fence may take a while but will pay large dividends in the end.


The Supreme Drill Press Table Mounted On My Delta Drill Press In March of this year I found myself wishing I had a drill press table with a fence to aid in accurately drilling a series of holes. My first thought was to build one; then my long standing rule of “using my time and efforts to build furniture and not jigs or fixtures” kicked in. So with the help of my large collection of woodworking catalogs and the internet I researched drill press tables available on the market. I settled on the Supreme Drill Press Table from Peachtree Woodworking Supply, Inc. (http://www.ptreeusa.com) shown left attached to my Delta drill press. The table is 15” deep by 24” wide and 1 3/8” thick. As shown there are two 22” fences which are closed to produce a 44” fence. These can be fully extended to form a 72” fence. It comes with two hold downs and two UHMW stop blocks. There are also two inserts to plug the hole in the center of the table that is provided for drill through.

Drilling A Series Of Holes Aligned By The Fence Since March I have used this table on numerous occasions leaving me to wonder how I ever worked without it. T-tracks on the bottom allow you to fasten the table to the drill press and provide plenty of travel front to back. The hold downs are secured in T-tracks that run front to back and are great for securing single thickness boards or boards with backing as shown at right. The star knobs allow for quick adjustments between drillings while providing plenty of clamping power. This is particularly necessary when drilling large wholes with a drill or Forstner bit.

Shaker Clock On Drill Press Table With Supporting Rollers I recently completed a wall hanging Shaker clock. The clock doors are held closed with magnetic catches which are secured in the sides by recessing them in shallow holes. I didn’t want to drill them by hand for fear of drilling them off vertical alignment. Further, I wanted to control the depth of the holes very accurately. After pondering this for a few minutes I wondered if my new drill press table could do the job. In order to get the sides under the Forstner bit I had to bring the table completely forward which caused me to think the setup might be unstable. Also the clock is about 4 feet long and holes had to be drilled close to one end, creating another potentially unstable situation. The former was no problem at all and the latter was solved with the use of adjustable roller supports shown left above.

Bessey Bar Clamps Are Used To Hold The Clock In Place For this operation I removed the fence and centered the holes in the side by eye. I adjusted the depth of the Forstner bit and locked it in place. Then I simply slid the clock along between drillings. Though it may not have been necessary, given the weight of the clock, I used Bessey Bar clamps to it in place while drilling.

I probably could have completed this operation without the use of the drill press and table. However, it sure made me feel at ease knowing I wouldn’t screw up this last step, which surely could have ruined my whole day. My brother-in-law, Winter Bargeron, calls these critical steps, with their potentially disastrous consequences, the “money cut”. Well, this table costs about $250 and is worth every penny.


Two days ago I installed all my blast gates and hook ups for my major pieces of equipment. Yesterday morning I sent my dust collection system on its maiden voyage. Using wide cherry boards I thickness planed and finish sanded them on my 15” Jet Thickness Planer and my Performax Pro 22-44 Drum Sander respectively.

Black Plastic Blast Gates & Drops For Major Equipment

Much to my pleasure not a micron of dust or a chip of wood escaped the dust collection system. I still have drops to connect, but they are for much less demanding stations. I gave each of these the hand test; I opened the gate, one at a time, and felt the air flow. Admittedly not a scientific test, but I am convinced they work.

Jet 15" Thickness Planer & Wide Cherry Board

The thickness planer is the most demanding machine in my shop. It is approximately at the end of a 35’ 4” PVC run. A 3’ plastic hose connects the machine and the drop. At the other end of the PVC run a 10’ plastic hose connects the PVC to the Powermatic PM1900 3hp Dust Collector. This hose will be shortened, but I intentionally left it long for this experiment.

Performax Pro 22-44 Drum Sander & Wide Cherry Board My second most demanding machine it the Performax Pro 22-44 Drum Sander. The dust generated from this machine is spread over a drum 24” wide and is funneled by way of a plastic housing to a 4” hose 36” long that connect to the drop. It is easy to inspect the board as it come out of the sander for missed dust: none was visible. I ran the board through positioned far left, center and far right. No problem.

The Powermatic PM1900 3hp Dust Collector specifications read:

Air Flow @ 8” Port – 1891 CFM
Velocity @ 8” Port – 5393.7 FPM
Static Pressure – 12.15” of Water

The 8” port is reduced to three 4” ports. At the moment I am using only two ports. One goes only to my Grizzly 8” Jointer through a very short 6’ plastic hose across the floor. The jointer is right next to the dust collector and there is no foot traffic where the hose lies. The unused port is reserved for my table saw, router and upstairs expansion.

Note The 8" Port Reduction To Three 4" Ports & The Dust Collection "Swirl"

The easiest way to check for reasonable air flow is to observe the cyclone “swirl” in the collection bags. This is subjective and you need to be familiar with the system, but I can tell by the healthy “swirl” that the air flow is strong. The picture at left doesn’t capture this as clearly as I would like, but trust me, it is obvious in real life. Also in the picture at left you can see the 8” port reduction to three 4” ports and the short hose to the jointer. In addition you can make out the types of wood I have been using in my projects: cherry, maple and walnut.

Being an engineer I will complete this testing with actual air flow measurements using calibrated equipment. I need to borrow this equipment from my brother who lives two hours away. So it will take me a week or two to complete. I’ll post the results here on my blog.


A wood shop is not a static object. It has a life of its own and it evolves just like a living organism. Fortunately, though its evolution can be slow, as mine has been, it also tends to change dramatically in one’s lifetime. Mine has taken eight years to evolve from a completed two story empty building to the shop I will discuss today.

My Brother Ron Standing On A Ladder Assembling A DropThis latest evolutionary step is a big one. It was originally planned to be completed before I even moved in; and it’s one I would not have completed even now if it weren’t for the help of my older brother Ron.

Each year my family has a reunion which occurs on the first weekend in August (I am one of twelve siblings). People come from all over the country to attend and often stay for a week to a month; they visit family, old friends and high school classmates. Ron and my sister-in-law Wanda come from Tennessee and usually stay with me for at least a few weeks. Typical of my family we don’t sit around watching TV and chatting. We have to do something. This year Ron and I decided to add dust collection to my shop.

My Brother Clark With Red Auerbach During the eight years I have been planning this project I read every article I could get my hands on concerning dust collection issues. I was appropriately frightened by articles on the potential for static electricity fires if non-conducting piping were used. I read an article written by an MIT professor debunking that myth. I have no less than five books on building your own shop and/or dust collection system in my library. I debated flexible hose, metal piping, plastic piping with grounding wire and PVC. It wasn’t until my oldest brother Clark piped his basement shop with PVC recently that I settled on it as the pipe of choice. I must admit here and now, my extensive research played a minor role in my decision, which was based primarily on my brother’s choice, PVC’s availability, its relative inexpensive cost and the MIT professor’s article (I’m an electrical engineer and it made sense to me).

At this point the project is not complete. I need to add fourteen 4” blast gates, all the manual plastic variety, and flexible hose connecting the shop’s equipment. I also left a 3” drop near the Lie-Nielsen Workbench providing a vacuum system for my Random Orbital Sanders (ROS). The current state of this project is shown below. My Powermatic Dust Collector has three 4” ports. Two of them are used to connect to the two returns shown. Not shown is the third port which will connect directly to my 8” Jointer with hose only, no piping. We also made provisions for a future second floor line. Click on the picture below to enlarge it.

Dust Collection Piping Without Blast Gates & Connecting Hose

The project used approximately 100’ of 4” PVC and numerous wye’s and 45 degree fittings plus a few 4” to 3” fittings – all schedule 40. In addition I will use 14 4” blast gates. All told it cost me $615.14. I chose to buy the PVC pipe and fittings at a local plumbing supply because they had all the fittings, J hooks and ceiling hooks I needed. Home Depot had all but the ceiling hooks. To my surprise the plumbing supply charged $1.60 per foot of PVC while Home Depot charged $1.13, a difference of $47. The fittings were also less expensive at Home Depot so I spent about $100 more than I should have.

4" Black Plastic Blast Gate Fits Snugly Inside Schedule 40 PVC Pipe My brother Clark discovered that standard black plastic 4” blast gates used in many shops fit snugly inside schedule 40 PVC. All that is needed to secure them is two sheet metal screws and silicone sealer. I used PVC cement on all the piping and fittings because I am unlikely to make many changes. However, the blast gates may change in the future should I decide to use switching gates or add a wye to the drop for expansion. Removing two sheet metal screws and sealer is easy compared to sawing off cemented pipe.

Note the picture on the left shows a 4” to 3” reducer and drop. This will be used to connect to a vacuum hose with the other end connected to a Random Orbital Sander. I will need to add further reduction to match my two sizes of ROS hoses.

Ceiling Hangers Were Used Away From The Wall's One thing my research convinced me of is  loss due to 90 degree fittings. Ron and I chose to use two 45 degree fittings and a 6” connecting piece in lieu of a long sweep fitting to achieve the same effect (in some places a wye, 45 degree fitting and 6” connecting piece was used). Forty-five’s are more readily available and more flexible – the design changed numerous times during the project and this flexibility was very much welcomed. One significant change was to put most of the piping on one leg, leaving a second for future second floor expansion. The third leg is also lightly loaded; only the router and table saw are on it. I may expand this leg to provide more shop vacuum capability.

J Hook's Were Used Along The Wall'sSupporting this much PVC required both J Hook’s and Ceiling Hangers shown above right and left respectively. This allowed us to hang the PVC 5” on pipe center to the ceiling everywhere along the run. The piping, done this way, does not interfere with the fluorescent lighting. That is, it blocks no light and casts no shadows. I will complete this post when I have added the blast gates and hookups. Hopefully I will be able to report a positive experience using dust collection. Stay tuned.


Veritas Bevel-Up Smoother PlaneMuch has been written about the relatively new bevel-up (aka low angle) planes from Veritas and Lie-Nielsen – most of it complimentary. In particular, two woodworkers I admire, Chris Schwarz and Lonnie Bird, have been outspoken about the advantages of bevel-up planes. This month’s issue of Popular Woodworking (August 2009 #177) had a great article called The Case For Bevel-Up Planes by Lonnie Bird. Anyone trying to decide between bevel-up or bevel-down should read this article. My experience with bevel-up planes, however, has been quite different. I offer the following account to present a different view.

The Lie-Nielsen Low Angle LN-164 Smooth PlaneIn December I bought my first low angle plane, a Lie-Nielsen LN-62 jack plane, largely on the strength of an article by Chris Schwarz. In March I bought a Veritas bevel-up smoother plane for a class I was to take at Lonnie Bird’s school. This was my first non-Lie-Nielsen plane and I felt like I was cheating on my wife when I purchased it. Somewhere in between I bought the Lie-Nielsen LN-164 smooth plane. In the intervening months I spent a lot of time tuning and using these babies, all along comparing them to my bevel-down Bedrock planes.

Lie-Nielsen No. 4 1/2 Bedrock Smooth Plane As an engineer by education and profession I am no stranger to design analysis. When I analyze the design of bevel-up planes I am convinced they will perform better than Bedrock designs, cost less, set up easier and are more flexible. Yet, my side by side usage and comparison over the last six months have left me wanting to stick with the Bedrock design, what Lonnie calls the “antiquated” design.

Veritas Bevel-Up Smooth Plane ComponentsSo why the disparity between what technically seems true, what the leading woodworkers say, and what my experience tells me? Part of this I believe is due to my (perhaps too long) learning curve. Perhaps I haven’t used bevel-up planes enough yet. But I am convinced that the disparity is in large measure real and physical.

Components Of The Lie-Nielsen Low Angle  LN-164 Smooth PlaneI agree with Lonnie Bird that bevel-up is the way of the future for all planes; largely because they have fewer parts, hence cheaper to manufacture and easier to tune. Both the Veritas and the Lie-Nielsen bevel-up planes have only three simple components; the body, blade and cap-iron (the Lie-Nielsen actually has four components if you count a small spacer). Compare this to the Bedrock design which has a body, frog, blade, chipbreaker and cap-iron. Also, you can buy one bevel-up plane plus two or three additional blades ground to several bevel angles and essentially have the equivalent of three or four planes.

Lie-Nielsen No. 4 1/2 Smooth Plane Components For example, Veritas offers a 25, 38 and 50 degree blade. Their planes have a 12 degree bed. Hence a resulting 37 degree low angle configuration for end grain, a 50 degree York configuration for well behaved grain and a 62 degree high angle for highly figured hardwoods. Three planes in one. In the Bedrock design you have to purchase multiple frogs to achieve this. Even then it is difficult to get a 62 degree cutting angle, which appears optimal for highly figured woods.

Sharpening A Lie-Nielsen Bevel-Up Blade Requires A Jig Like The Veritas To Maintain A Right Angle All of that is true yet, call it feel, call it human engineering, call it whatever you want, but bevel-up designs are at least two generations away from getting it right in my opinion. While the Bedrock design has more parts and is more difficult to tune, like many woodworkers I have a number of planes tuned for specific purposes. This means I tune them once and avoid reconfiguring them so that I don’t have to tune them again for quite some time. What’s left after tune up is minor blade depth and lateral adjustments – and that is where the Bedrock design excels. The human engineering of the Bedrock design is such that you can make either a blade depth change or a lateral adjustment between strokes while returning the plane to the starting point of a pass. Both adjustments are right there at your finger tips. No hunting required, no need to turn the plane over or change hand position to make an adjustment. Also, both adjustments are silky smooth on the Lie-Nielsen Bedrock designs, and both are like vernier adjustments that give precise results.

Lie-Nielsen bedrock Tote (Front) & Veritas Bevel-Up Tote - Note Difference In Angles Contrast this to the Veritas and Lie-Nielsen bevel-up planes. The Veritas seems to be the most clumsy of the two manufacturers. The adjustment mechanism for both depth and lateral adjustment is a knob that is placed too low for the thumb or index finger to access without moving your hand from the handle. You can see this in the picture at right. Further, I find that if I tighten the lever cap to a degree required to keep the blade from shifting on its own the lateral adjustment is too tight to operate smoothly.

Lie-Nielsen Bedrock Tote (Front) & Lie-Nielsen Low Angle Tote - Note Difference In Angles The Lie-Nielsen low angle design doesn’t require a lateral adjustment relying instead on a tighter machining of the bed alignment to the edges of the blade. However this requires that the blade be ground and honed precisely at 90 degrees to the edges, a task that is easily performed, ironically, with the aid of a Veritas honing jig. The Lie-Nielson’s depth adjustment seems slightly better human engineered than the Veritas. Neither come close to the ease, smoothness and controllability of the Bedrock.

One of the features of the bevel-up design that is far superior to the Bedrock is mouth adjustment. Both the Veritas and the Lie-Nielsen planes make this easy. Simply loosen the front knobs, adjust the mouth and retighten. In the Bedrock design you have to move the frog position requiring the involvement of three screws. It is quite difficult. However, like I said, in practice this adjustment is rarely required if you dedicate your planes for specific configurations.

The Lie-Nielsen No. 4 Bronze Bedrock Smooth PlaneTuning a Bedrock is second nature to me since I have been doing it for so long. I can achieve wispy thin and wide shaving with very little effort. For some reason I can’t explain other than a learning curve, I can’t achieve a tuned state nearly as quickly with the bevel-up planes. This is puzzling to me but I can only hope it is not a permanent condition.

The last part of this comparison is a somewhat nebulous factor, let’s call it feel. The Bedrock feels right in my hand. It’s beefy (those extra components have a benefit), the grips (tote and knob) are just right and everything is in reach without moving my hand from the grip. I can easily get in a zone and plane away effortlessly even with the 55 degree middle pitch.

Mouth Adjustment Is Easy On A Bevel-Up PlaneThe bevel-up planes, on the other hand, seem too light; my hand and arm extension seems wrong; I find myself fighting the lower angle which is tiring. Look at the two pictures above comparing the totes of the Bedrock design versus the totes of the bevel-up planes. Notice that both Lie-Nielsen and Veritas bevel-up planes have a lower angle tote forcing your bodies forearm and shoulder lower. I suspect the designer’s rationale was that for steep pitch configurations you need to be pushing more behind the plane to compensate for the increased force required. That makes sense, but it’s unnatural; and you have to pay this price even for mid and shallow pitch configurations, which are the pitch configurations most often used in the shop.

Lastly, in this nebulous category called feel, the bevel-up controls require me to break my planing stride. I usually approach a smoothing task, especially for figured woods, with the blade slightly retracted. On each pass I extend the blade ever so slightly until I am taking that wispy thin cut I want. During this time I also make lateral adjustments. With the Bedrock I do this on the fly. I can’t do this with the bevel-up planes. It’s stop and go with each adjustment

Maybe I am just too biased to allow for a new feel, I don’t know. But I have been at this comparison for nearly six months and still haven’t achieved the advantages my engineering analysis, and the woodworkers I admire and aspire to be, say I should. So for now I’ll stick with my Bedrocks thank you.


The Backs And Doors Are Custom Fitted Spalted Soft Maple With Its Black Lines, Tan and Greenish ColoringThe backs of a custom piece serve a number of functions and they are far from simple pieces of wood. The upper back in this piece provides a mechanism for hanging the clock while it also serves to keep dust out of the clock’s works. The upper back is not, however, visible since it is hidden by the clock dial.

The bottom back is visible, just behind the weights and pendulum that drive the clock. It also serves to keep dust out of the case. However, because the swinging pendulum will draw all eyes to itself and the back, it is important that the back not look like a plain piece of wood, but rather adds to the beauty of the clock. For this clock spalted maple serves that purpose. The random black lines of the early fungus and the tan and greenish color of the wood provides the viewer with an artistic drawing that only nature could render.

Elongated Open Holes, Washer And Screw Allow Seasonal Movement The Backs Are Centered With Gaps On Either Side For Expansion Backs almost always require special treatment to allow for seasonal expansion and contraction. In large pieces I often use ship lapped boards that are spaced one from the other to allow for seasonal movement. Theses backs are not wide enough to accommodate this approach. Instead, after calculating the expected movement, I cut the backs narrow by 1/4” and fastened them with slotted open holes, washer and screws. I cut them narrow because expansion season has only barely begun, and at its peak, the backs will expand to close the gap. If this were peak expansion season I would have cut them to fit and let them shrink to their minimum size. The washer and screws hold the back flat but also lets it move under the washer. I am careful not to tighten too much. Notice that I center the backs so that the gap for expansion is equal on each side.

The hardware and glass are on order and as soon as they arrive I will attach them. Then it is a simple matter of applying finish. For this clock I am going to use Min-Wax Wipe-On Poly Satin Finish.


Thickness Planing With A Drum Sander - The Top Is Open For VisibilityThe doors of this clock are made of black walnut (aka American walnut) chosen to provide contrast to the lighter, and more red shade of cherry. Walnut, while an excellent furniture wood, is not one I like working with much. There are very few adverse health effects related to walnut though there have been documented reports of skin irritation, rhinitis and asthma. But the saw dust generated by walnut is very fine and highly noticeable even with the use of dust masks. I find its taste bitter and unpleasant. So, while it is a beautiful furniture wood I tend to use if for contrasting trim and doors and seldom build an entire piece out of it.

Tapering Legs With A Drum SanderMost of the walnut pieces in this clock are short, about 12”. So I have chosen to thickness plane them with my drum sander and avoid the problem of sniping. I load the drum sander with 80 grit paper. After joining and planing three sides with a power jointer followed by a hand jointer and smooth planes, I cut them to near length and thickness them on the drum sander.

I Use The Table Saw To Cut The Rabbets And Save The Off Cuts For Securing The GlassAfter bringing the pieces to within 1/16” of final thickness using 80 grit paper I switch to 220 grit for final thicknessing. I don’t have to run through all the grits in between because I am taking off more with the 220 grit paper than the depth of the groves left by the 80 grit paper. I would not suggest trying this with a random orbital sander though.

The drum sander has many uses not immediately obvious. For example, power planing a tiger or blistered maple board will often leave tear out because of the rapid grain changes. The drum sander is an excellent choice for final thicknessing in this case. Also, the safest way to taper legs is a drum sander. You might first rough cut the taper close to the line with a band saw and follow it up with a drum sander, or skip the drum sander altogether. Either way you avoid the dangerous step of either a table saw or a jointer.

The Tenoning Jig Makes Cutting The Open Mortises And Tenons Easy And SafeWhen thicknessing is complete I cut a 3/8” wide by 1/2” deep rabbet in all pieces. Normally I would do this with a set of dado blades. But if I do this on the table saw instead, the off cut pieces are exactly the size I need to secure the glass in the door.

For this Shaker wall clock I have chosen simple doors constructed with slip joints. This is consistent with many Shaker clocks in existence. More importantly, for a given rail and stile size, slip joints provide more glue area and are stronger. The rails and stiles on this clock are only 1 1/8”, so this added benefit is quite important.

Final Slip Joint Fitting Is Done With A Shooting Board And Shoulder PlaneSlip joints are basically a mortise and tenon with the mortises being open. The table saw and tenoning jig make cutting the open mortises and tenons easy and safe. I cut the open mortises first and then cut the tenons to fit. The jig has a fine vernier so that I can creep up on the correct tenon thickness.

Two of the stiles for the long door are 37” long, not a piece I would want to hold manually while guiding it through the table saw. This jig is designed to hold them secure, at perfect right angles, and hands safely clear of the blade. It is heavy and tightly fits the table saw groves so the cuts can be smooth and slow avoiding tear out.

Completed Doors - Notice That One Side Has A Rabbet To Provide For Securing The Glass I cut the tenons so they fit a little too tight in the open mortises. Then I final fit them with a shooting board and shoulder plane. This gives me a perfect fitting slip joint. The doors are crafted over sized, one quarter in wider on all sides. This leave me the ability to custom trim them to the carcass. A quarter inch may seem a little overkill, but it also allows for a little tear out on the ends of the stiles and mortises which will not remain after trimming.

The doors, after glue up but before custom fitting, are shown above right. Notice the rabbet shown on the back side of the long door. This rabbet provides and inset for the glass. Tuesday of this week I will be out to the wood yards picking out the figured wood for the back. After that only mounting hardware and applying the finish remains.

« Previous PageNext Page »


Back Issues of Chiefwoodworker's Newsletter