iDRY Vacuum Kilns

Sponsors:

Joist question... drop-in or dovetail

Started by danreed76, August 11, 2012, 06:27:13 AM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

danreed76

with just a few more details to sort out before the PE throws the stamp on our plans, we were discussing the value of the dovetails on the floor joists.  The PE seems to think we'll be okay with drop ins, but is okay with the dovetails too.  He just wants to save me some unnecessary work (hmmm... this I should have considered).  What's the consensus on floor joist joinery?  Dovetails?  Drop-ins? Tusks? 

-Dan
Woodmizer LT40 Hydraulic with resaw attachment |  Kubota MX5200  | (late)1947 8N that I can't seem to let go.

Jim_Rogers

I recently had a discussion with another timber framer about a drop in dovetail joint where the purlin meets the rafter, in a principal rafter, purlin roof system.

I personally thought that the industry has gone away from using dovetail tenons on purlins. The reason why is that the dovetail timber shrinks and there is a possibility of the dove tail pulling out:



What I mean is that this distance of the "red" dimension gets shorter when the width of the purlin/joist shrinks.
To offset that some timber framers pound in hardwood wedges at the blue location to squeeze the dovetail tenon and take up any gap that is created when the timbers shrink, to hopefully prevent them from pulling out.

Personally I don't like this idea, method or joint.

Now if you need to secure your frame from one bent to another then your frame design should be carefully reviewed to find ways to do that. You can do almost all the joists as regular square ended drop in joists and do one in the middle called a "tying joist" which will tie the bents together.

We use tying joists all the time, to tie sills together in shed designs.

Jack Sobon used them in house designs on second floor systems as well.
You just have to plan on it when you raise bents to have one end in and the other end supported while you raise the next bent up to it.

Raising sequence has to be considered when designing frames. You can draw stuff you can't put together. So be careful and think it through completely.

Jim Rogers
Whatever you do, have fun doing it!
Woodmizer 1994 LT30HDG24 with 6' Bed Extension

routestep

Most often a drop in, but as Jim stated above I use one or two joists to help lock the bents together on the second floor.

On this barn I have two captured floor joist in two bays and one in the middle bay. I drilled a peg hole through the cross tie top, into the joist and out the bottom of the cross tie. I didn't bother with draw boring. Then drive in the pegs.

shelbycharger400

mount up like a T 
Steel plates and bolts are WAYYY easier!
1/4 to 1/2 in steel plates  with 8  1/2 in dia bolts
thats how my uncles house was.

logman

I always use drop ins and put simpson straps across them. 
LT40HD, 12' ext, 5105 JD tractor, Genie GTH5519 telehandler
M&K Timber Works

nas

So if you are using a drop in joist, would the mortice be the full size of the joist, or housed, or how would it keep from showing a gap when the beam dries?

Nick
Better to sit in silence and have everyone think me a fool, than to open my mouth and remove all doubt - Napoleon.

Indecision is the key to flexibility.
2002 WM LT40HDG25
stihl 066
Husky 365
1 wife
6 Kids

Jim_Rogers

Quote from: nas on August 17, 2012, 10:40:47 AM
So if you are using a drop in joist, would the mortice be the full size of the joist, or housed, or how would it keep from showing a gap when the beam dries?

Nick

Nic,
That is a question that doesn't have an easy quick answer.
It all depends.

It depends on the span and load and type of wood, it depends on a lot of things.

I can't or no one can or should just say, "at 2" mortise is good enough...."

You need to have enough wood under the joist pocket to hold up the joist.

It not an easy thing to figure out either.

Jim Rogers
Whatever you do, have fun doing it!
Woodmizer 1994 LT30HDG24 with 6' Bed Extension

logman

I usually cut 1 1/4" deep housings.  If I'm using a 6x8 joist or purlin I cut full 6" wide by 5" depth.  I cut the joist or purlin 5" and angle back the bottom.

 
LT40HD, 12' ext, 5105 JD tractor, Genie GTH5519 telehandler
M&K Timber Works

routestep

The Timber Frame Guild puts out a good book on joinery and design. It's not a substitute for a PE going over your plans, but it can help you in the initial stages of planning your building.

The red book is entitled "Timber Frame Joinery and Design Workbook"

The green book is volume two of same title

These two books are useful tools to have.

Agricultural buildings here in VA don't have to have a PE stamp, but its wise in my opinion to have a design looked over.

routestep

Another way to lock in a drop-in joist instead of using a dovetail is to move the peg over to the side. Drill the peg hole where the side of the joist and the pocket wall of the tie meet. Then hammer in your peg, making an interference fit.

Like a keyed shaft.

Morewood

Has anyone used these dovetail systems " Arunda or Ligna " ? Is it worth the money ? It looks like a nice system and could speed things up.
2008 Cooks AC3651 perkins diesel , JD 310C w/forks and thumb , International 8000 Dump truck , 16 ton trailer , and all the accessories .

Jim_Rogers

The whole problem with this type of connection system is that when the timber shrinks the dovetail gets smaller. And can then pull out of the dovetail pocket. Or slide down more. They have tried to counter this by making the dovetail on an angle instead of straight so that it would hold in two directions.

I have seen these tools at trade shows and I have seen them used face to face at the trade show cutting the joints.

But I have never seen them in a finished structure.

Jim Rogers
Whatever you do, have fun doing it!
Woodmizer 1994 LT30HDG24 with 6' Bed Extension

Aikenback

I have a ligna tool. It works great.One thing I noticed is with the v shape you can pound in a slightly twisted joist, and it will straighten it. As far as shrinking goes, a full housing will show gaps on the sides of the pocketed timber. The ligna joint does remain tight, however any gap will show light where the dovetail tenon is shorter in height than the timber. The joist/purlin also uses the entire pocket and sides of the dovetail for bearing because of the v shape. Like I said, the joint remains tight, and I use it all the time, but that being said, I like the look of a housed joint.

I hate any kind of shrinkage. Especially when swimming.
no whining.

Jay C. White Cloud

I have some real concerns with the direction I see many of these "dovetail," type of tenon joints evolving and the jigs that are making them.  Jim has, on several occasions, pointed out their potential flaws, and I can't begin to stress how absolutely true his assessments are.  There has been some rather serious failures of this joint in the past decade, and for litigious reasons, not much is out there about it. 

Whenever you bastardize the traditional design and use of a joint, you can quickly run afoul with using it in a frame.  I see way too many frames being built with these modified "dovetails," and the applied parameters are too far outside the scope of the joints intended use.  That has been happening to the "dovetail," since it's reappearance in the mid 70's.  This is a family of lap and scarf joint that exists wherever timber framing is practiced, but no place has a greater variation and application of the joint than Asian timber frames, especially Japan.  The new "jig" assemblies that are out there being sold, are almost exact copies of what has been in Japan since just after World War II, when there housing boom started.  Unfortunately, western timber framers are not applying the joint as originally designed!  "Dovetail," joints are, (99.9% of the time,) a "housed" lap joint or toggled/lapped scarf.  Unless you are building furniture, you would never rely on the tenon of a joint to take any kind of significant load.

When shrinkage is brought up for the reason of the modification, it only further demonstrates the lack of knowledge behind the application of this joint.  When properly housed, and the timber is "kerfed," (as is often done in Middle Eastern and Asian timber frames,) "gaping" of the joint does not occur or, at a minimum, is highly mitigated.
"To posses an open mind, is to hold a key to many doors, and the ability to created doors where there were none before."

"When it is all said and done, they will have said they did it themselves."-teams response under a good leader.

Aikenback

I'll let the engineers who stamp the work worry about it. Regardless of the joint, size, etc of any member or group of members in a frame. They are generally a conservative group of people.
no whining.

Jay C. White Cloud

It is very true that PE's are, as a group, very conservative, (sometimes to a distraction.)  That does not mitigate that sometimes things are overlooked, like the "re-purposed scarf joint."  My comment, (I do apologize,) was not directed at you or your work individually.  If you have a PE signing off on your frames, they take full responsibility for any failures, in most (but not all,) cases.

My concerns are with the overly generic application of the corrupted "dovetail" joint.  After speaking with, (no names for litigation reasons,) PE's that will no longer allow the use of this joint in frames they are part of, I can only agree with their reasoning.  This joint, as modified, can function very well, but only in very narrow parameters of application; which many are now being seen employed outside of.  The other key reason for apprehension: if a joint of most timber frames fail from some destabilizing event, (e.g. meteorological, seismic, etc.) they tend to hold the structure together even after failure, so occupants can escape safely.  The joints of the frame may even be repairable.  The original derivation of the "dovetail" joint, as found throughout Asian architecture, is more than capable of withstanding the rigors of these events, as history has proven empirically.  This new version of the "dovetail," even when done correctly, fails catastrophically during these events, every time.  Because of the lack of housing, (an improper redesign of the original,) the joint can only work one very narrow way and when that way is compromised they come apart completely, i.e. catastrophically.

I know as a seasoned timber framer, I do not want my frames to ever fail that way, now or hundreds of years from now.  That is the legacy of this craft, and I hope to honor it the best ways I can.  Even though I may have a PE approve my design, does not take away my responsibility of do dillagence, not only for my work, but for what I have learned is a flaw, and perhaps a very dangerous flaw at that.
"To posses an open mind, is to hold a key to many doors, and the ability to created doors where there were none before."

"When it is all said and done, they will have said they did it themselves."-teams response under a good leader.

Aikenback

no whining.

Jay C. White Cloud

Aikenback,

That is good that you have no doubt, and you seem to present good intentions, as you use a PE. on all your timber frame designs.  However, this is a public forum, with many lay folk looking for guidance.  They should receive all the information available to them to make sound decisions. When all the information about a topic is available, folks can way how they may want to proceed.

Respectfully submitted,

Jay
"To posses an open mind, is to hold a key to many doors, and the ability to created doors where there were none before."

"When it is all said and done, they will have said they did it themselves."-teams response under a good leader.

Aikenback

Oops, darn phone.
What I meant to say was. There is no doubt in my mind that a tusk tenon is superior. Also, I would expect veteran timber framers to have valuable insight. No disrespect intended by defending. Engineer always requires a 5" GRK rss as a precaution. After reading some earlier posts by Jim, and looking at the traditional housed dovetail joint, I agree that it isn't a good joint. But for more reasons than withdrawal. If the joist/purlin shrinks in height, it could "float" in the housing, bearing only on the dovetail tenon. The tusk tenon could never get in a withdrawal situation, nor a bearing situation like the traditional dovetail. Also, the joist/purlin is restrained at the bottom bearing of the joist, resisting any twisting or withdrawal in the receiving timber. I make these observations as a long time journeyman carpenter, not as a veteran timber framer. Are these accurate assessments in your opinion?
Also, as far as the new cnc style dovetails, I would be interested in it's history. Who designed, when, etc. And why it persists if it is flawed.
I think it's a viable joint (better than a simple pocket with a screw) if it is used with a secondary restraint to help mitigate the failures you describe. For me, it has worked well, but nothing I have done is with wet material or in an earthquake zone. I think maybe with D.Fir it is working, because what I'm using is quite dry. Once again I wouldn't use this joint without a screw or strap. So much for traditional joinery.
I do respect your, and other veteran timber framers' opinions/knowledge. Particularly those that not only produce frames, but study historical methods. It obviously was not conveyed in my above arrogant post : )
I can absolutely see the limitations of these "cheating" techniques, especially in seismic zones.
B.T.
no whining.

Jay C. White Cloud

Hey Aikenback,

No worries, your post have been fine, and defending a position on a topic is good for the conversation.  If all sides are heard, we all become wiser for it.  ;) 

You have addressed well the proper way to use this modified dovetail.  D.Fir dry is really a good idea, though you could use it in some application in green wood, as long as you had some type of tying joint in proximity to it and/or metal fastener back up.

House dovetails, when cut in green wood are tricky.  You want the dovetail to function as the element that resist withdrawal, but it should actually be slightly under cut from beneath, so as when the timber shrinks the housing gets tighter, as does the bearing surface.  I've seen these joints perform well even after hundreds of years, but you really need to know how to cut them, understand/reading wood grain, and a dozen other subtleties.

If I ever did a project with you, (I consult and teach a lot,) believe it or not, I wouldn't stop you from using this joint, since it is part of your design style.  I would just make sure that all the counter measure had been put into place.  Like the fact you back them up with metal fasteners of some type, (i.e. strap, screw, etc.)  Now with this new age of timber framing, we have the luxury of seeing, after centuries or millenia, were the traditional joints begin to fail with age.  I design my frames as traditional frames, however, I will, especially in critical situations, add hidden metal elements for the once in a life time event that could weaken the joint.  I do not think there is anything wrong with that.  It is even considered standard practice in historical restoration and museum work.

Regards,

Jay

"To posses an open mind, is to hold a key to many doors, and the ability to created doors where there were none before."

"When it is all said and done, they will have said they did it themselves."-teams response under a good leader.

Aikenback

Hey JC,
I recently have downloaded Thomas Tannert's thesis on RDC from U of B.C. 2008. 256 pages of engineering testing, data, etc. Just for my own peace of mind. Talk about a dry read! One of the strongest recommendations is that the material be fully dry, and the joint created carefully with minimal gapping, and Definitely reinforced with a GRK or other self taping screw installed at 55 deg. Coming from a conventional carpentry background, I found there were things with timberframing that can be counter intuitive to anyone uneducated in timberframe and other historical building techniques. (as some of your other posts have shown). And I was wondering if my aversion to mortising deeply into a supporting beam midspan for a tenoned joist was justified.  The original reason for using the RDC method was for ease of use, but the main reason now for me is that it seems to be lest intrusive to the supporting beam, therefore weakening the supporting timber less than a row of tusk mortises through the middle of the beam. Especially where there are joists on both sides. Considering my short time (so far) in this type of construction, I would like to ask what has your (and other long timers') experience been with engineering beams for tusks over RDC's. Does the supporting beam  ( in general) need to be enlarged significantly due to the tenon? I'm already using some big stuff, don't want to go too crazy. I suppose joists dont HAVE to be mortised, they could sit atop the beam. I do like the look though, and on exterior work,  barn swallows love those joist spaces on the beam. However they do eat mosquitoes.
no whining.

Jim_Rogers

QuoteI would like to ask what has your (and other long timers') experience been with engineering beams for tusks over RDC's?

I guess I may have missed it. What is/are RDC's?

Jim Rogers
Whatever you do, have fun doing it!
Woodmizer 1994 LT30HDG24 with 6' Bed Extension

Aikenback

Sorry jim, didn't mean to leave you out. The engineering thesis I was reading (trying to read) calls these new joints Rounded Dovetail Connections. RDC's.
no whining.

Jay C. White Cloud

Hello Aikenback,

I can't believe you got through the read of Tannart's thesis, (can you say nose bleed!)  great info, but wow that is hard to all chew at once.  I haven't gotten through all of it, but have a copy I "gnaw at," when ever this subject comes up.  I think now you understand my concerns with this joint and have addressed them well in your frames.

To answer your question about "aversion to mortising deeply into a supporting beam mid-span for a tenoned joist," each case is different, as you well know. Perhaps though, your concerns aren't as justified as you may believe.  The primary evolution of the RDC is ease of execution of joint and assembly of frame.  Your concern for weakening the supporting beam can be alleviated by placing the "tusk" in the correct location within the receiving member.

Don't worry about your short time doing this kind of work, for it is refreshing to see how thorough your approach to the craft is.  If there is to be "modernization," in approach to cutting and executing joinery, it should be done with the kind of care and forethought you have put into it.  To address you question about the receiving beam I present the following for your consideration and questions:


For the most part, (but there are cases,) the supporting member does not have to be enlarged, or enlarged very little.  It is the placement of the "tusk," that is critical.  You need it to penetrate the neutral axis of the receiving member above or near the vertex of the "hour glass," center.  In this region of the timber you are between the "compression" and "tension" zones, thereby not "weakening" it, or very little.  Housing into the supporting member is another issue, where the "rules of 2s and 3s - 3s and 4s," should be applied.  I will explain this as the Amish "barn wrights," taught me.   On top of your receiving member, a housing should not get closer than 50 mm (2") preferably 75 mm (3") across to the other side.  On the side of the supporting member, there should be 75 mm (3") preferably 100 mm (4") of wood beneath the inserting timber. They also said this was more a guide than a "rule," but it has stuck with me all these years and has served me well.  There is, of course, exceptions and should you have concern consult with others.

I tend work in "folk styles," where receiving members don't receive but support, because the other member rest on top or is only partially housed.  This all depends on the style you work in and/or your own individual style/preference.  I like barn swallows (and other birds;)  giving them a place to nest and rest in architecture can be a good thing.  ;)
"To posses an open mind, is to hold a key to many doors, and the ability to created doors where there were none before."

"When it is all said and done, they will have said they did it themselves."-teams response under a good leader.

timberwrestler

I tend to do tenoned joists a little differently, like this:



I think that's generally called a soffit tenon, but I think that may be backwards (the tenon's not on the soffit after all).  I like it because it's a little less meat out of the receiving beam.  That's a 5x8 joist, reduced to 5.5" going into an 8x10.  The 8x10 has a nominal 1/2" housing.  With a lot of those dovetailed joist details, when the receiving timber shrinks, the joint looks like crap because the joist isn't housed.

I like the look of the smooth reductions, and engineers love em (even the NDS throws them some love).  Here's a row of drop in joists to give you look:

www.uncarvedblockinc.com
www.facebook.com/uncarvedblockinc

Thank You Sponsors!