Replies

1. 
   ErnieConover | Mar 22, 2006 12:39am | #1

   The skirts are actually called "aprons" and they would normally be mortise and tenoned into the legs in what is called frame construction. This would be the proper construction for a table, of which a workbench is just a study example. As you surmise wood movement in the top is going to be a problem. Wood shrinks and expands across the grain but not along it. Therefore the frame will always be about the same size but the top will be wider or narrower depending on the time of the year. Various attachment methods have been devised over the years to allow for this movement. One is with screws and oversize holes and the other is with right angle blocks screwed to the underside of the top that fit into mortises in the aprons. There is metal hardware available to accomplish this same end. The attached drawing will explain all.

   Actually, I do not think you want to build a table as a workbench. I favor a European style bench but even a standard American bench is usually just a heavy top (1-1/1" to 2" thick) attached to legs. The legs can be a traditional frame with aprons but allowing considerable overhang around for clamping and vices. I would advise you to get a copy of The Workbench Book published by Taunton Press. It will give you all the options with a number of great designs you can build.

   Regards,
   Ernie Conover

   
2. 
   ErnieConover | Mar 22, 2006 03:56am | #2

   Double checking this on line (to late, why do it?) I see that I labeled the screws FHWS (flat head wood screw) when I meant Round Head Wood Screw (RHWS). Sorry!

   Regards,
   Ernie Conover

1. 
   HoustonHeights | Mar 22, 2006 05:57pm | #3

   I'm not explaining myself very clearly.  Take a look at the attached file.  My question is how this joint accounts for expansion/contraction. 

    

   Edited 3/22/2006 11:04 am ET by HoustonHeights

   
2. 
   HoustonHeights | Mar 22, 2006 06:07pm | #4

   Here is the attachment.  -- Is expansion minimized by laminating quartersawn pieces together?

   
1. 
   ErnieConover | Mar 22, 2006 06:25pm | #5

   I cannot open your attachment.Regards,
   Ernie Conover

1. 
   HoustonHeights | Mar 22, 2006 06:54pm | #6

   Try this one.

   
1. 
   ErnieConover | Mar 22, 2006 07:49pm | #7

   Although it is hard to tell from the photograph it looks like you are trying to make the tail vice of a European style workbench which does indeed often have a dovetail in it. They just about have to be hand cut and it is a simple enough matter to do so. It looks from the photo that the end apron is mechanically attached to the bench so that the top will float and the dovetail will not cause a problem. Without more information such as the plans you are using, etc. I could not say for sure.With best regards,
   Ernie Conover

2. 
   HoustonHeights | Mar 22, 2006 09:15pm | #8

   I'm obviously not communicating my question very clearly.  Ignore the tail vise.  I posted the pictures to illustrate the kind of "apron" I am referring to.  Notice that the bench has a solid core about 2" thick.  The core is surrounded by a 5" wide apron, flush with the top.  The long ends of the apron (front and rear of bench) are joined to the short ends of the apron with half blind dovetails.  I understand that there will be seasonal movement in the width of the bench, hence my question: how should I account for this expansion in the core of the top with respect to the apron?  It seems like if the top expands, it will pop apart the dovetail joint.  Does that make sense?

    

3. 
   ErnieConover | Mar 22, 2006 11:06pm | #9

   Traditional benches of the type you are showing me have a heavy front apron which also contains the mortises for the dogs. Behind this is a thinner top section (usually about 1") followed by the tool well. The ends are bound by heavy sections equal to the front apron. These sections have grooves which rides on a tenon on the end of the thinner part of the top. These end sections are indeed dovetailed, often with one large tail and two half pins to the front apron. This leaves the rest of the bench to float on the tenon and these ends prevent any warpage. The ends are actually what are called bread board ends which are used on bread boards, chest lids and such to prevent warpage. The back of the tool well is often dovetailed to the end pieces in the same fashion as above. If I can be of further help do not hesitate to ask.With best regards,
   Ernie Conover

4. 
   HoustonHeights | Mar 23, 2006 02:04am | #10

   Thanks for your patience, Ernie.  I now understand that the function of the breadboard end, but my original question still remains.  How do I deal with expansion of the top with respect to the longer side aprons -- the one with the bench dogs in it?  Do you see what I mean? The 1" thick top will expand and contract seasonally, causing the long side apron to move with it.  The floating bread board end, however, will not expand and contract longitudinally at nearly the same rate, thus putting seasonal stress on the dovetail joint.   How do I account for that stress with respect to the side apron?  Do I attach it to the top with a floating tenon as well?  Kind of like a frame and panel door?  Again, I appreciate your patience.  The answer probably seems obvious to you, but it isn't to me!

5. 
   ErnieConover | Mar 23, 2006 03:53am | #11

   In this case the bread board ends are dovetailed solid into the front apron and it is the top that floats in the ends. The front edge of the bench stays put and the back edge moves. We have really beat around the bench on this one!With best regards,
   Ernie Conover

6. 
   HoustonHeights | Mar 23, 2006 05:41pm | #12

   I think I understand now.  The breadboard ends float on the tenon in the benchtop.  The front apron is fixed to the top and dovetailed into the breadboard end.  The back apron (which is also dovetailed into the end) is not fixed to the top to allow for seasonal movement.  Is that right?

7. 
   ErnieConover | Mar 23, 2006 06:47pm | #13

   Precisely! Send me a photo when it is done.With best regards,
   Ernie Conover