Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Dowels really useful? Just glue enough?

592 views
Skip to first unread message

PN1...@musica.mcgill.ca

unread,
Jun 28, 1994, 9:04:27 AM6/28/94
to
I have edge glued some pieces of oak recently, with dowels to
reinforce/align the glued pieces. But the dowels did not help that
much in aligning. I know many people use biscuits, but that is not
an option for me at this point. Why not just glue teh pieces together?
(with good clamping).

Marc Fortin
pn...@musica.mcgill.ca


Peter Williams

unread,
Jun 29, 1994, 8:47:32 AM6/29/94
to culist-...@cunews.carleton.ca
Just glue is definitely good enough for edge joining. when using
dowels/biscuits, they shouldn't have glue on them: otherwise, they can
help push the joint apart with humidity changes! they're for alignment
*only*

--
Peter Williams / Dan Riseborough : Geotechnical Science Laboratories
Carleton University Dept. of Geography
"The Geotechnical Flying Squad" Ottawa, Canada (pwil...@carleton.ca)

Kent M. Zickuhr,91-30 151,6-4889,

unread,
Jun 29, 1994, 10:03:49 AM6/29/94
to
In article 00...@VM1.MCGILL.CA, <PN1...@MUSICA.MCGILL.CA> () writes:
>I have edge glued some pieces of oak recently, with dowels to
>reinforce/align the glued pieces. But the dowels did not help that
>much in aligning. I know many people use biscuits, but that is not
>an option for me at this point. Why not just glue teh pieces together?
>(with good clamping).
>

Works for me. I do this when gluing panels for raised panel doors. If done
well and not clamped too tightly (you'll squeeze the glue out), they're
reasonably strong. When doing large panels, though, as for a table top
it certainly helps to have something to align your edges. For something
like that I use splines. I cut a slot in each piece with my router and a
slotting bit, usually 1/4". Then use 1/4" stock for the splines just a
little narrower that the depth of both slots. If the edges are going to
show, don't run the slots all the way to the end. If you have a thickness
planer, you can custom make your spline material for a perfect fit.


Kent Zickuhr

Dave VanEss

unread,
Jun 29, 1994, 1:14:43 PM6/29/94
to
In article <Cs5sv...@cunews.carleton.ca>,

Peter Williams <pwil...@SUPERIOR.CARLETON.CA> wrote:
>Just glue is definitely good enough for edge joining. when using
>dowels/biscuits, they shouldn't have glue on them: otherwise, they can
>help push the joint apart with humidity changes! they're for alignment
>*only*
>
When glueing wood together with the grain parrallel to the glue line you
do not need dowels bisquits, or ant thing else. The glue is stronger than
the wood. Every time I glue up a table top and cut out the extra on the ends.
I take the end piece, put it on blocks, and break it with a hammer. I have
never had a break at a glue line. Dowels or Bisquits can help with
alinement but can reduce the joint strength. Heres why.

The goal is to get to flat surfaces together with a thin layer of glue between.
Preasure is applied. However the surfaces are not square to the face of the
board. So the surfaces mate and the top becomes slightly zigzaged. This zig
zag can be removed with a planer. The problem with dowels and bisquets is
that they are not aligned with the glue surfaces but with the top or bottom
of the board. If the board is not square you got conflict. If you don't
believe it try the following. Prepare the glue edges that are 2 degrees off
square. When put together thet are 4 degree off. gule up one set with dowels
the other with out. Compare results.

Dave VanEss

George Uptain

unread,
Jun 29, 1994, 10:03:56 AM6/29/94
to

In article <Cs5sv...@cunews.carleton.ca>, Peter Williams <pwil...@SUPERIOR.CARLETON.CA> writes:
|>Just glue is definitely good enough for edge joining. when using
|>dowels/biscuits, they shouldn't have glue on them: otherwise, they can
|>help push the joint apart with humidity changes! they're for alignment
|>*only*

I hope to see more discuccion on the above comment. I have never heard that
you do not use glue on dowels and biscuits. If not then why do dowels
have cuts in them for glue to escape from and biscuits are advertised as
expanding when soaking up glue to tighten up the joint?
--
<><> George Uptain || Schlumberger Well Services
><>< Development Tech III || Austin, Texas
<><> 512.331.3183 || upt...@slb.com

Bramel, Jim

unread,
Jun 29, 1994, 2:24:54 PM6/29/94
to
Norm is going to be shocked out of his drawrs when he finds out
biscuits are not to be glued.

Bennett Leeds

unread,
Jun 29, 1994, 4:20:08 PM6/29/94
to
Dave VanEss writes

> Dowels or Bisquits can help with
> alinement but can reduce the joint strength. Heres why.

A new twist on an old argument.


> The goal is to get to flat surfaces together with a thin layer of glue
> between. Preasure is applied. However the surfaces are not square to the
> face of the board.

Just how out of square are your surface/edge joints? With my jointer fence
adjusted properly, mine come out with better than a degree tolerance.


> So the surfaces mate and the top becomes slightly zigzaged. This zig
> zag can be removed with a planer.

We can argue this one too: depending on the planer and the wood, its
possible that the planer will bend the wood, take some wood off, and when
the wood comes out the planer it will spring back to its old shape.


> The problem with dowels and bisquets is
> that they are not aligned with the glue surfaces but with the top or bottom
> of the board. If the board is not square you got conflict.

Not true, for a whole variety of reasons:

1) There's at least a few degrees of play of a biscuit in the slot. After all,
the biscuit is designed to swell to fit the slot, so it doesn't have to be a
tight fit when dry.

2) The way many people clamp boards, alignment is performed on the faces,
which are usually pressed against the clamp bar. Thus, you have the same
misaligned edges whether you're using biscuits or not. If you let the
board buckle out some (the "zigzag" you refer to), then you'll find that
the whole assembly just pops out of the clamps.

3) The biscuit itself produces a stronger joint than the edge glue line it
replaces. Try it: Make an edge joint with a line of glue under 1/4" wide
and less than 2.5" long every 8" or so. Now make an edge joint with #20
biscuits spaced every 8" or so. You'll quickly see which is stronger.


> If you don't
> believe it try the following. Prepare the glue edges that are 2 degrees off
> square.

If your edges are 2 degrees out of square, then you've other problems than
worry about whether biscuits are adding strength of not. If you're using
a power jointer, one technique you can use even if you can't get your jointer
fence just right is to press the good face of one board against the fence,
and one the next board press the bad face against the fence. Thus, any error
is automatically cancelled out.


BTW, have you tried your test? What were the results?

- Bennett Leeds
ben...@mv.us.adobe.com

Lisa Lapidus

unread,
Jun 29, 1994, 6:34:32 PM6/29/94
to
HELP!

My fiancee and I are putting a set of shelves up and are looking for the
best way to make sure that we put the screws in the lumber. (Obviously
neither of us has a great deal of experience hanging shelves...)

The "StudSensor" and its competitors appear to do essentially what we
want. Do they work? Is there another solution if not?

--
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
# Shawn M. Winnie win...@umich.edu #
# "It's not my fault!" (617) 776-2963 #
# -Meano Culpa (617) 589-4497 (Work) #
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #

Bob Meyer

unread,
Jun 29, 1994, 8:25:56 PM6/29/94
to
In article <2ussto$5...@scunix2.harvard.edu>, lapi...@husc8.harvard.edu (Lisa Lapidus) writes:
|> HELP!
|>
|> My fiancee and I are putting a set of shelves up and are looking for the
|> best way to make sure that we put the screws in the lumber. (Obviously
|> neither of us has a great deal of experience hanging shelves...)
|>
|> The "StudSensor" and its competitors appear to do essentially what we
|> want. Do they work? Is there another solution if not?
|>
Hi folks,
The one I've had for years works just fine. I just used it when I put up about
75 feet of chair rail around a room. It only had trouble in a couple of weird
spots that must have had extra studs.

Good Luck,
Bob Meyer

Peter Williams

unread,
Jun 29, 1994, 10:54:43 PM6/29/94
to culist-...@cunews.carleton.ca
George Uptain (upt...@WILDCATAUSTIN.WIRELINE.NAM.SLB.COM) wrote:
: In article <Cs5sv...@cunews.carleton.ca>, Peter Williams

: <pwil...@SUPERIOR.CARLETON.CA> writes:
: |>Just glue is definitely good enough for edge joining. when using
: |>dowels/biscuits, they shouldn't have glue on them: otherwise, they can
: |>help push the joint apart with humidity changes! they're for alignment
: |>*only*

: I hope to see more discuccion on the above comment. I have never heard that
: you do not use glue on dowels and biscuits. If not then why do dowels
: have cuts in them for glue to escape from and biscuits are advertised as
: expanding when soaking up glue to tighten up the joint?

In applications where the dowel/biscuit is doing more than just align
things, then glue is certainly nessesary, and the flutes work as
described. For alignment *in solid wood* where the dowel is
cross-grain, then it should be unglued, or if glued it should be as
short as possible. Imagine the "worst case": a long (say 3 inch) dowel
across the grain. With seasonal wood movement, the wood across the
dowel will want to expand/contract by um... er... a significant amount
(my copy of understanding wood is at the other end of the house...),
but will be constrained by the glued-in-place dowel. In the worst case
where the dowel bottoms out as the wood shrinks, it will split the panel.

Lee Valley Tools used to (maybe still does) sell special plastic
dowels designed *not* to adhere.

Dan R.

Stuart Gresley Staniford-Chen

unread,
Jun 29, 1994, 11:47:52 PM6/29/94
to
Dave VanEss writes
> Dowels or Bisquits can help with
> alinement but can reduce the joint strength. Heres why.

The reason I've heard for not gluing dowels when using them in edge to
edge gluing is that they can cause the joint to fail due to moisture
cycling. This would only apply to longish dowels. The idea is that
if the moisture content of the table top (or whatever) increases, it
will be unable to swell because it's restrained by the glued dowels
which have their long-grain parallel to the direction of swelling.
This puts the table top material in compression, and if the moisture
content rises high enough the wood can be compressed beyond it's
elastic limit. Now, if the table top dries out and needs to shrink,
it's already smaller than it was and it will have a tendency to split.

I can't see any reason not to glue biscuits though - they wouldn't
seem to be wide enough to cause any such problem.

On the other hand, there isn't any very compelling reason why you have
to glue the biscuits or dowels in an edge to edge joint. If the glueline
alone isn't strong enough, something is wrong. Usually I glue my
biscuits, but once or twice when I had a big assembly to do, I didn't
in order to reduce my stress level a bit while trying to spread glue
and get everything together.

Stuart Staniford-Chen
stan...@cs.ucdavis.edu

PS - Of course, all of the above only applies to edge-to-edge joints.
It would obviously be bad not to glue the biscuits in an end-grain
joint.



Bennett Leeds

unread,
Jun 30, 1994, 12:08:04 AM6/30/94
to
Peter William writes

> George Uptain (upt...@WILDCATAUSTIN.WIRELINE.NAM.SLB.COM) wrote:
> : In article <Cs5sv...@cunews.carleton.ca>, Peter Williams
> : <pwil...@SUPERIOR.CARLETON.CA> writes:
> : |>Just glue is definitely good enough for edge joining. when using
> : |>dowels/biscuits, they shouldn't have glue on them: otherwise, they can
> : |>help push the joint apart with humidity changes! they're for alignment
> : |>*only*
>
> : I hope to see more discuccion on the above comment. I have never heard
that
> : you do not use glue on dowels and biscuits. If not then why do dowels
> : have cuts in them for glue to escape from and biscuits are advertised as
> : expanding when soaking up glue to tighten up the joint?
>
> Imagine the "worst case": a long (say 3 inch) dowel
> across the grain. With seasonal wood movement, the wood across the
> dowel will want to expand/contract by um... er... a significant amount
> (my copy of understanding wood is at the other end of the house...),

I think you better make the trip across the house. Even a 3" long dowel
only goes 1.5" into each board (and most dowels are shorter than this to
being with). 1.5" or even 3" just isn't enough to make the total change in
width significant, even if you choose a relatively unstable wood in a
relatively unstable environment with a relatively permeable finish.

If this were really a problem, we'd be seeing problems with mortise and tenon
joints, many of which have cross-grain areas this large and larger. As the
neanderthals point out, there are M&T joints that have survived hundreds of
years.

Finally, considering a biscuit only goes into each side a 1/2" at most (and
is not perfectly cross-grain), there's even less problem.

- Bennett Leeds
ben...@mv.us.adobe.com

Michael Winnett

unread,
Jun 29, 1994, 11:52:49 AM6/29/94
to
Shawn M. Winnie writes.........

>
> My fiancee and I are putting a set of shelves up and are looking for the
> best way to make sure that we put the screws in the lumber. (Obviously
> neither of us has a great deal of experience hanging shelves...)
>
> The "StudSensor" and its competitors appear to do essentially what we
> want. Do they work? Is there another solution if not?

I've never had a lot of luck with these sensors, possibly I'm too
ham-fisted with them. There are always cables in the wall and
other stuff that can fool the sensor into a false reading.

I generally use a long thin nail and a hammer. The studs are
most often fitted at regular intervals, so once you find a couple,
it normally gets easier to find the others. You can bash the wall
with a hammer and listen to the sound, it should sound a bit more
solid where there is a stud. You then tap a nail in to see
whether the stud is indeed where you think. Repeat this process
until you've found them all.

If any of the holes you've made are going to be visible, you will
need to fill them afterwards with decoraters filler.

You could save the money on the gadget and put it towards a lovely
Estwing hammer, which IMO will be money well spent.

Mick.

John Cornett

unread,
Jun 30, 1994, 9:25:38 AM6/30/94
to
In article <2ussto$5...@scunix2.harvard.edu> lapi...@husc8.harvard.edu (Lisa Lapidus) writes:
> HELP!
>
> My fiancee and I are putting a set of shelves up and are looking for the
> best way to make sure that we put the screws in the lumber. (Obviously
> neither of us has a great deal of experience hanging shelves...)
>
> The "StudSensor" and its competitors appear to do essentially what we
> want. Do they work? Is there another solution if not?
--------------------------------------------------------

The other solution is the old tried and true method which all carpenters
I ever knew use. The sensor is the hammer. Left to right, across the area
to be fastened. Listen carefully. The tone and feel will tell where the
stud is. You might drive one nail part way in to verify.

The feel of this proves the stud is there. It will meet resistance.
Once you have found one stud, the others should be at 16" intervals.
You can use a level to mark vertically, your rule to mark horizontally.
Just to be sure, test with tapping hammer again to see if it falls in
ballpark (that they do not have 24" centers). That should do it.

John Cornett

George Uptain

unread,
Jun 30, 1994, 9:24:21 AM6/30/94
to

In article <2ussto$5...@scunix2.harvard.edu>, lapi...@husc8.harvard.edu (Lisa Lapidus) writes:
|> The "StudSensor" and its competitors appear to do essentially what we
|> want. Do they work? Is there another solution if not?

Shawn, I own a Craftsman studsensor and it works as advertised. It works
by sensing changes in density. You hold the sensor against the wall
and calibrate it by holding the switch down until the led's go off. Now
without letting go of the switch move it across the wall slowly. As it
approaches a stud (higher density) the led's start lighting up. As it passes
beyond the stud they go off. The center of this zone is the center of
the stud. While the instructions do not say to do it, I like to go both left
to right and right to left to average out any error.
On rough textured walls or ceilings move the sensor along a piece of cardboard
held against the wall.
If you happen to calibrate on a stud it will work in reverse of above.

Kent M. Zickuhr

unread,
Jun 30, 1994, 10:12:39 AM6/30/94
to
In article 5...@scunix2.harvard.edu, lapi...@husc8.harvard.edu (Lisa Lapidus) writes:
> HELP!
>
>snip>

> The "StudSensor" and its competitors appear to do essentially what we
> want. Do they work? Is there another solution if not?
>

They work great. My wife gave me one and it's been a very useful tool. The one
thing that will throw it off is wires or pipes in the wall. This makes it a
little troublesome in bathrooms. You generally only find wires above light
switches and receptacles so you can be aware of those before you start. These
things will find the edge of the studs so you can tell if you have a double
stud. It certainly beats the old magnetic finders that looked for drywall
nails.

Kent Zickuhr


gary preckshot

unread,
Jun 30, 1994, 11:59:54 AM6/30/94
to
In article <4...@satchmo.win-uk.net>, Michael Winnett
<mwin...@SATCHMO.WIN-UK.NET> wrote:
>
> Shawn M. Winnie writes.........
> >
[stuff]

> > The "StudSensor" and its competitors appear to do essentially what we
> > want. Do they work? Is there another solution if not?
>
> I've never had a lot of luck with these sensors, possibly I'm too
> ham-fisted with them.

They work fine for me. Studs are vertical. Wiring usually runs
horizontally. Likewise firestops. You can usually tell whether you have a
stud by the fact that it shows consistently over the entire stretch from
floor to ceiling.

[stuff about using a hammer and a nail...]

I used this method before I got a StudSensor. I also used a lot of spackle
fixing up afterwards. Those little holes look tacky.

Peter Williams

unread,
Jun 30, 1994, 10:47:52 AM6/30/94
to culist-...@cunews.carleton.ca
Bennett Leeds (ben...@ADOBE.COM) wrote:

: If this were really a problem, we'd be seeing problems with mortise and tenon


: joints, many of which have cross-grain areas this large and larger. As the
: neanderthals point out, there are M&T joints that have survived hundreds of
: years.

A mortise and tenon joint is different from an edge glued panel, in
part because it has inherent mechanical strength. And the true
neanderthal pins the tenon in the mortise.


: Finally, considering a biscuit only goes into each side a 1/2" at most (and


: is not perfectly cross-grain), there's even less problem.

I agree that it's not a serious problem with short dowels or biscuits,
but the glue is not nessesary, which was the main point I was
makeing. For long dowels, though.....

I'll take a trip across the house.

D.R.

Myron Hammond

unread,
Jun 30, 1994, 12:38:44 PM6/30/94
to

>
>> The "StudSensor" and its competitors appear to do
>essentially what we
>> want. Do they work? Is there another solution if
>not?
>>
>
They basically work OK, but you still need to use some common
sense. If you just take a single quick reading over a small
area, you could easily be fooled! It is best to check a larger
area, making small marks as you go, and verify your readings by
checking the spacing between the marks. If the measurement
corresponds to your stud spacing, you're OK. These sensors
sometimes have problems in small areas, such as between the
tops of doors and the ceiling, and they aren't much help if you
encounter metal studding.

Bill Currier

Lawrence M Lidsky

unread,
Jun 30, 1994, 12:53:54 PM6/30/94
to

--
I've found that studsensors work quite well on drywall and moderately well
even on plaster over wood lath. The trick is to map out the hidden structure
in detail, using knowledge of standard spacing, support structure location
and whatever other metaknowledge you can bring to bear. This enables you to
dismiss false positives and, in many cases, to pinpoint the centerline of
the studs to better than 1/2" accuracy.

In my opinion, a studsensor is $20 well spent. You can still buy yourself an
Estwing hammer to make the job more fun.

--larry

------------------------------------------------------------------------------
room 38-174 phone 617-253-3808
M.I.T. fax 617-253-0458
Cambridge, MA 02139 net lmli...@athena.mit.edu
-------------------------------------------------------------------------------

John Cornett

unread,
Jun 30, 1994, 12:45:41 PM6/30/94
to
In article <940630151...@atc.boeing.com> Dave Adams writes:

>I'VE HAD ONE FOR ABOUT 3-5 YEARS, AND USE IT ALOT, IT'S THE OLD TYPE AND HAS
>NOT GIVEN ME ANY PROBLEMS AT ALL. ........................................
---------------------------------

I have no knowledge of stud finders, but can give some insight into work
without finder. Using old method, finding stud with first probe is about
75% (my estimate); with third probe (1" to left, 1" to right) perhaps
90-95%; the remaining 5% can be difficult. Tapping method does not
find edge of stud.

I did not like the old magnetic finders and am not up on these new ones.
What is the cost?

John

Chris Knoell -- BCSS

unread,
Jun 30, 1994, 2:50:54 PM6/30/94
to


I hate to even get into this again, but George is right, bisquit
slots are cut to allow a glue pocket and expansion, and I still
hold to my position that glued bisquits on "edge joints" are not
necessarily the only way to go but I highly recommend it. I have
used them for years making cabinet doors and table tops and numerous
other things, and have never had anything come apart as a result of
a glued bisquit, or at all for that matter.

Chris ( A glutton for punishment)

Dave VanEss

unread,
Jun 30, 1994, 2:46:01 PM6/30/94
to
Dowels or bisquets can be used as alignment tools but are frequently installed
in reference to the wrong surfaces. Bennett (I am using your first name to
keep this friendly, this is a difference of opinion) implied that perhaps I
have not done this test. I have, when I was glueing up a table top out of 4 8ft
8/4 10 wide boards. I used dowels for alignment. We are talking about 4 boards
each weighing about 40 pounds. I had to undersize the dowels to keep them for
interferring with the surfaces I was trying to glue. A 1 degree error on
each board causes a 1/16 in gap before preassure is applied. True I could
have put both pieces through the jointer together, but I'm not physically
able to manage 80 lbs of wood through my joiner. I had to undersize my dowels
(sander) so they were not snug. With the time I spend drilling dowel holes
I could have aligned the boards while in the clamps while under slight pressure.

I do not put glue on the dowel (or for that matter the bisquit) The strength
should come from the faces properly mated with a thin layer a glue applied
under presure. This makes for a joint that is stronger than the wood that will
not fail with the expansion and contraction of the wood. The same can't be
said for dowels and bisquits. I know that chairs built with dowels or mortise
and tennons must be occasionally reglued. I know the stresses a chair takes
are no where those of a table. However, I try to make my stuff to last 400
years.

I will summarize this with the following.

Dowels and Bisquits are tools that can help with the alignment of boards
during glue up. They are not needed for the strength and can reduce the
strength of joints and cause gaps in the joints if applied incorrectly.
It is desireable to install them normal to the glue surface instead of parallel
to the board face.

Dave VanEss

ps People who respond with spelling corrections should be required to
answer if "anal regressive" should have a hyphen.

Peter Williams

unread,
Jun 30, 1994, 2:09:02 PM6/30/94
to culist-...@cunews.carleton.ca
Bennett Leeds (ben...@ADOBE.COM) wrote:
[stuff deleted]

I made my walk across the house to get Understanding Wood (1980 Bruce
Hoadley, Taunton Press), and I find that it was there that I first
read about this phenomenon. The defence calls Dr. Hoadley to the stand
as expert witness:

Page 171:
"Despite the obvious success of the draw-bored mortise and tenon,
dowels are most often misused as gluing accesories to hold parts in
alignment. For example, in making a tabletop, boards might be edge
glued and held with a series of bar clamps. To ensure alignment, of
board surfaces at the joints, dowels are sometimes used. However, if
gluing is correctly done, full wood strength can be developed by a
plain side-grain to side-grain joint - no reinforcement is necessary.
Because they do not provide strength, the pins therefore need only be
long enough and numerous enough to ensure alignment. For edge gluing 1
inch lumber, 3/8 or 1/4 inch dowels 1 inch long are plenty. Dowels
should fit snugly into accurately positioned holes. When the joints
are clamped, no attempt need be made to glue dowels into the holes in
the mating edges. The loss of glueline due to the dowels is
negligible. For example, in edge-gluing 3/4 inch lumber, a 3/8 in.
dowel placed every 8 in. along the joint reduces glueline area less
than 2%. Although it might seem advantageous to make the dowels "Good
and long" and glue them in "good and tight", a negative effect can
actually result. The restraint to normal shinkage and swelling may
cause the wood to fail at or near the glue joint. If gluelines fail at
edge joints, the problem should be rectified by troubleshooting the
gluing procedure rather than by pinning a bad joint with dowels in an
attempt to bring it up to standard. If the gluelines are properly
made, there is little to gain in trying to reinforce the joints, since
the strength of the wood on either side of the joint is still the
limiting factor."

The defence rests :-).

Dan.

Chris Knoell -- BCSS

unread,
Jun 30, 1994, 2:56:35 PM6/30/94
to


This may sound crazy but, I always knock on the wall with one
knuckle and listen for a dense sound, then I stick a straight
pin in the spot I like and see if I meet the stud. If I miss,
(which I do on occasion) you can't see the pin hole unless your
nose is against the wall. Has worked for me for years. Good
luck....Chris

Bennett Leeds

unread,
Jun 30, 1994, 3:29:10 PM6/30/94
to
Dave VanEss writes
> Bennett implied that perhaps I have not done this test.

I just asked if you had (you hadn't said), and what the results were. No
offense intended.


> I have, when I was glueing up a table top out of 4 8ft
> 8/4 10 wide boards. I used dowels for alignment. We are talking about 4
boards

> each weighing about 40 pounds.... A 1 degree error on


> each board causes a 1/16 in gap before preassure is applied.

That makes sense if you didn't have any play at all of the dowel in
the hole, and the hole were drilled perfectly parallel to the faces. Both
are rare in my experience, but then I never had luck with dowels in any
use.


> True I could
> have put both pieces through the jointer together, but I'm not physically
> able to manage 80 lbs of wood through my joiner.

You don't need to do both boards at the same time. One at a time works
just as well on the power jointer. BTW, the hand tool equivalent is to clamp
the two boards face to face on your workbench and handplane them at the same
time. I've done this with 4/4 stock, but not 8/4 stock.


> With the time I spend drilling dowel holes
> I could have aligned the boards while in the clamps while under slight
pressure.

I agree that dowels are not usually useful.


> Dowels and Bisquits are tools that can help with the alignment of boards
> during glue up. They are not needed for the strength and can reduce the
> strength of joints and cause gaps in the joints if applied incorrectly.

Whoa, I don't see how you've made the leap from 3" long, tight-fitting dowels
to 1/2" wide, loose-fitting, expanding after assembly biscuits. A biscuit
joint is much stronger than a dowel joint in my experience.

- Bennett Leeds
ben...@mv.us.adobe.com

Bennett Leeds

unread,
Jun 30, 1994, 4:01:54 PM6/30/94
to
Peter Williams writes

> The defence calls Dr. Hoadley to the stand as expert witness:
> Page 171:
> "Despite the obvious success of the draw-bored mortise and tenon,
> dowels are most often misused as gluing accesories to hold parts in
> alignment.
> ...

> For edge gluing 1 inch lumber, 3/8 or 1/4 inch dowels 1 inch long are plenty.
>...

> Although it might seem advantageous to make the dowels "Good and long" and
> glue them in "good and tight", a negative effect can actually result.
>...
> The defence rests :-).

OK, I'll no longer argue that long and tight dowels won't cause problems,
except to say that using Hoadley's own wood movement calculations (earlier
in the book), I get about .01" of expansion on 2" of cherry in the north-
eastern US.

It's too bad Hoadley isn't actually available for questioning, though:

Since 1" long dowels would seem to be fine accordingly to Hoadley's statements,
(see requotation, above) one would think that a 1" wide biscuit (whose
grain isn't as cross-grain to the board as the dowel's) would also be fine.
And, the biscuit does not require the same critical milling as dowel holes do,
allowing a few degrees of play before they expand and set, so would not have
any of the alignment or movement problems atrributed to long dowels.

Hoadley's book was apparently written before the use of biscuits became
popular in this country, as it doesn't discuss them, except perhaps in
passing. It'll be interesting to see what he has to say about them in the
next edition (if there is a next edition) of his book.

> I will summarize this with the following.
>

> Dowels and Bisquits...

My summary is that there are enough differences between dowels and biscuits
to make conclusions and experiences of one not really apply to the other.

- Bennett Leeds
ben...@mv.us.adobe.com

Bennett Leeds

unread,
Jun 30, 1994, 4:54:33 PM6/30/94
to
Dave VanEss writes
> ...I was glueing up a table top out of 4 8ft 8/4 10 wide boards....
> A 1 degree error on each board causes a 1/16 in gap ...

> I had to undersize the dowels to keep them for interferring with the
> surfaces I was trying to glue.

How did you plane the 11/32" off of each face of the board to get it flat
afterwards?

Let's look at this: With two 8/4 boards each out 1 degree, you've got a 2
degree wedge 1.75" long, which is about a 1/16" gap at the wide end. The same
trig yields a peak to valley distance on each face of 11/32" if you instead
mate the glued surfaces. Your tabletop must have ended up at most 1 1/16"
thick instead of 1 3/4" thick, not to mention the difficulties you had in
planing a 40" wide by 8' long top flat. I hate to think of how thin the
top ended up being if the wedge orientations didn't alternate....

As I said previously, you've got problems other than the dowels in this
situation.

Rather than mess with the dowel holes, the easier fix would have been to
flip one or more of the boards (so the errors cancel out, but you may not
get the best grain up), or replane some of the edges. But, you whether
you're using dowels, biscuits, or just glue, you really want your glue-ups
to have the board faces parallel to one another.

- Bennett Leeds
ben...@mv.us.adobe.com

Lisa Lapidus

unread,
Jun 30, 1994, 9:53:41 PM6/30/94
to
Thanks one and all for the (rapid) help. Both sets of parents are
visiting for the fourth and we appreciate the volume, quality and
promptness of the reponse. (We weren't exactly sure how such an
obviously neophytic question would be received... :)

We've saved some of the responses; if there's an FAQ we'd be happy to
submit or summarize.

Bob Groschen

unread,
Jul 1, 1994, 2:49:11 PM7/1/94
to
writes:


I have two of these that are the ultrasonic type, and they work very well. I
also have one of the magnetic type and these are a joke as they require that
you're looking for a magnetic (steel or iron) fastener and that you know just
about where it is. The last part is sort of contradictory to the point of
having a stud sensor............

Regards,

Bob Groschen (rpgro...@mmm.com)

Data Cartridge Systems Laboratory
Memory Technologies Division
3M Co.

Standard Disclaimers apply here -->[X]

-------------------------------------------------
Why does a _Slight_ tax increase cost you $200 and
a _Substantial_ tax cut save you $0.30?

David Haile

unread,
Jul 1, 1994, 4:52:36 PM7/1/94
to
In article <2v1hsh$8...@dawn.mmm.com>, Bob Groschen <rpgro...@mmm.com> wrote:
>
>In article <2ussto$5...@scunix2.harvard.edu>, <lapi...@husc8.harvard.edu>
>writes:
>>
>> My fiancee and I are putting a set of shelves up and are looking for the
>> best way to make sure that we put the screws in the lumber. (Obviously
>> neither of us has a great deal of experience hanging shelves...)
>>
>> The "StudSensor" and its competitors appear to do essentially what we
>> want. Do they work? Is there another solution if not?
>

I agree with the last guy, whose comments I zapped. The ultrasonic types
work _really_ well, but the little magnetic gizmos are a waste of money.
I purchased the extra-depth version, but I think I would have been fine
with the regular version. I can find studs that exist behind double
thickness wallboard with a 1" foam insulator behind that! The actual
stud is about 1.75" behind the surface of the wall. Quite amazing.
You can use them on your kids' heads as brain density meters and estimate
how much you're going to have to save for their college funds. You can
use them in the garden to determine where the moles live. If I was blind,
I would use one to help me navigate. With slight circuitry modifications,
they could be used to locate where to dig a well.


--
/-------------------------------------------------------------------------/
/ David W. Haile - Ft. Collins, Colorado - dha...@csn.org /
/-------------------------------------------------------------------------/

Steve Shapland

unread,
Jul 1, 1994, 6:36:55 PM7/1/94
to
Dave VanEss (da...@u.washington.edu) wrote:
: ... True I could

: have put both pieces through the jointer together, but I'm not physically
: able to manage 80 lbs of wood through my joiner. ...

Use the IO-IO method. (WARNING - ASCII GRAPHICS )
TOP VIEW END VIEW
+-----------------------------------------------+ +---+
| | | |
| | | |
| I | I| |O
+-----------------------------------------------/ /////
| O | O| |I
| | | |
| I | I| |O
+-----------------------------------------------/ /////
| O | O| |I
| | | |
| | | |
+-----------------------------------------------+ +---+

For your FINAL pass when you joint the edge, always place the "I" face against
the fence. Mark the boards with a chauk during layout. Then the minor angular
errors will cancel each other. This final pass needs to be a very thin pass
to avoid tear-out in the "against-the-grain" cut.

Works for neandrethal jointers too. My #7 won't handle a 4" wide cut.
Steve Shapland, some-time wood butcher

Chita_...@f1010.n202.z1.fidonet.org

unread,
Jul 2, 1994, 2:00:50 AM7/2/94
to
I've had good luck with the StudSensor by Zircon. I have the model ich
gradually lights up as you approach the stud, finally lighting up
completely when you're over the center. About $15 at Harbor Freight.

--- ProBoard v2.01 [Reg]
* Origin: The Alien Biker Kat, ZOOMing..! (1:202/1010)

Paul Houtz

unread,
Jul 1, 1994, 1:35:37 PM7/1/94
to

Regarding edge-gluing boards that are whose edges are not 90 degrees to
the faces:

Here is an interesting production tidbit:

Higher volumne cabinet shops *don't* joint the edges of their panel
stock to 90 degrees, on purpose.

They joint it at an angle that is easy to see, like 80 degrees.

Then, they reverse alternate stock when edge gluing up the panels.
The non-90 degree angle is very easy to see so they don't make
mistakes, and reversing alternate pieces of panel stock GUARANTEES
that the faces will be dead flat.

This, of course, plays havoc with bookmatching grain, which is why
you virtually never see bookmatching in commercially made doors.
In fact, aesthetically they try for totally random grain matches as
it actually lends a more uniform look to the cabinetry than semi
book-matched doors would.

For my kitchen I used 90 degrees on my jointer and bookmatched
the grain and it turned out very nice. I didn't use biscuits
or dowels.

Michael D. Sullivan

unread,
Jul 4, 1994, 1:11:26 AM7/4/94
to
ckn...@ronjon.msfc.nasa.gov (Chris Knoell -- BCSS) writes:
>I hate to even get into this again, but George is right, bisquit
>slots are cut to allow a glue pocket and expansion, and I still
>hold to my position that glued bisquits on "edge joints" are not
>necessarily the only way to go but I highly recommend it. I have
>used them for years making cabinet doors and table tops and numerous
>other things, and have never had anything come apart as a result of
>a glued bisquit, or at all for that matter.

Maybe they haven't come apart, but if you use glued biscuits for edge
joining you had better allow *lots* of drying time before final surfacing
and finishing. The water in the glue causes the wood in the vicinity of
the biscuits to expand, and it takes a fair amount of time for this
moisture to go through the wood and evaporate through the surface. If
you wait long enough before scraping or sanding, fine. But if you don't,
you may find that when the biscuit joint finally dries, there is a
noticeable depression in the finished surface, because you sanded or
scraped away the expanded wood that has now contracted. This may take
months or years to occur if you put on a barrier-type finish
(polyurethane, for example).
--
--------------------------------------------------------------------------
| Michael D. Sullivan | INTERNET E-MAIL TO: |also: avog...@well.sf.ca.us |
| Washington, D.C. | m...@access.digex.net | 74160...@compuserve.com |
--------------------------------------------------------------------------

James Foster x2912

unread,
Jul 1, 1994, 1:28:09 PM7/1/94
to
`
I never liked the tapping/probing method either. I got one of the "automatic"
Zircon stud sensors about 10 years ago. It currently looks like it was in a battle
and is a little erratic, but still works. I used it a lot on lath and plaster
and some on drywall. In older construction with firestops in oddball places
it can be a real lifesaver (or at least patience saver). As another poster
recommended you can use one to "map" out an area - find the studs and stops
and lightly mark everything with pencil. I don't think I ever had any false readings
from electrical wires though water, vent stack, and radiator pipes will
be picked up. Usually you have a good idea where these are running so you
can be on the lookout for false readings. All in all I'd highly recommend them
for anyone doing anything from remodeling to hanging pictures. If I recall
they're less than $15 now, and certainly worth it.

Mark Lord

unread,
Jul 4, 1994, 11:34:59 AM7/4/94
to
> Shawn M. Winnie writes.........
>>
>> My fiancee and I are putting a set of shelves up and are looking for the
>> best way to make sure that we put the screws in the lumber. (Obviously
>> neither of us has a great deal of experience hanging shelves...)
>>
>> The "StudSensor" and its competitors appear to do essentially what we
>> want. Do they work? Is there another solution if not?
>

Assuming we are *not* talking about the old ones that use magnets,
then, yes, they work. Very well. Very little falsing.
--
ml...@bnr.ca Mark Lord BNR Ottawa,Canada 613-763-7482

Bennett Leeds

unread,
Jul 5, 1994, 3:14:58 PM7/5/94
to
Michael D. Sullivan writes

> ckn...@ronjon.msfc.nasa.gov (Chris Knoell -- BCSS) writes:
> >...I have

> >used them for years making cabinet doors and table tops and numerous
> >other things, and have never had anything come apart as a result of
> >a glued bisquit, or at all for that matter.
>
> Maybe they haven't come apart, but if you use glued biscuits for edge
> joining you had better allow *lots* of drying time before final surfacing
> and finishing. The water in the glue causes the wood in the vicinity of
> the biscuits to expand, and it takes a fair amount of time for this
> moisture to go through the wood and evaporate through the surface. If
> you wait long enough before scraping or sanding, fine. But if you don't,
> you may find that when the biscuit joint finally dries, there is a
> noticeable depression in the finished surface, because you sanded or
> scraped away the expanded wood that has now contracted.

My experience agrees with what Hugh Forster writes in his "Biscuit Joiner
Handbook," that "puckering" isn't a problem as long as the biscuits slots
aren't too close to the face of the board. Apparently some people felt the
need to put two rows of biscuits in 4/4 stock, which inevitably resulted
in slots too close to the board's face.

I've successfully biscuit joined pieces as thin as 1/2" walnut and 5/8"
maple without any evidence of puckering, and I scraped/sanded the day
after letting the glue dry overnight.


> This may take
> months or years to occur if you put on a barrier-type finish
> (polyurethane, for example).

Those pieces were finished with an oil finish. Both pieces are about
two years old.

- Bennett

0 new messages