New WoodChat!
Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Good Hope Hardwoods - Walnut Slabs & Specialty Hardwood Lumber

Glueing up.

Posts

Glueing up.

#1

Glueing up.

Derek Cohen (in Perth, Australia)

This may be a little simplistic for some.

It occurred to me to keep a record of glueing up the two Military cabinets. In part, I always look for feedback to improve methods. And in part, it may help someone what to do - or what not to do! :) - as glueing together dovetailed cabinets can be a little unnerving.

I left off last weekend at this stage - a dry fit of the two cabinets ...


Here I present the right hand cabinet, which will have three drawers and three shelves ...


Glue? I am using Titebond III. It has about 15 minutes of open time, and is required to be clamped for 30 minutes for the initial bond.

The internal "seen" surfaces (around the shelves) have been scraped. All other internal surfaces are planes flat and smoothed but scratches and scrapes are left be. They will never be seen.

The clamps are ready, the glue bottle is clean. I have made up a fresh bunch of spatulas.

I make glue spatulas from an old, broken, cheap tape measure ...


I start with the sides:


I only add glue to one side, not both. I find it easier to place the glue on the pin board as there is a wide area, and it can be plastered on evenly and completely.

First one side ...


Then the other ...


And lastly the centre (even if it is end grain) ...


The clamps do double duty. As the boards warp a little and are no longer as straight as they were when first built, the clamps pull the join into alignment. Secondly, by angling the clamps you are able to alter the pressure on the sides and square up the angle.


Next the centre divider is glued in. Glue runs the full length of the stopped dado, and I add a little more for the undercut edge. I am never sure whether one should leave an unglued section for expansion here?


I clean up with tepid water as I go.

Now the top can be added ...



I had worked hard on the dados and the runners of the drawer supports and the shelves to fit well. I did not want to be surprised at this stage with sticking parts, drying glue, and panic. Happily the shelves slid in smoothly, the lower one first ...


I only glued the first quarter of the dado. This would allow the solid shelves to expand and contract with changes in the weather.


I did the same with the drawer supports althogh these will expand differently, probably hardly at all (as they are all mortice-and-tenon joints).


Clamps ensure that there are no gaps at the front ...


And then diagonal braces are used to fine tune the squareness ...


That came out just fine:


Next up - smooth planing the outside and building the bun feet.

Regards from Perth

Derek

Re: Glueing up.

#2

Re: Glueing up.

Stuart Hough

Thanks, Derek. That lays it out in a very logical, organised (note the English infuence!) fashion that is extremely easy for us neophytes to understand. We all appreciate the guidance you veterans of the craft give us here on the WC. By the way...the wood looks gorgeous, and the pieces will no doubt be outstanding when completed. Kudos!

Re: Glueing up.

#3

some comments

Bill Tindall

I am enjoying watching the progress.

You are using way to much glue. It takes a film of glue only two molecules thick to form the bond. Try brushing it on with a small brush. Before assembly scrap all that extra glue off with the spreader, or don't put it on to begin with. Trivia: the bond strength is 100,000 psi for a glue of infinite toughness. But in fact the glue cracks under stress reducing the bond strength of a perfect joint to only 10,000 psi (about) but still good enough.

For glue to work it has to wet both surfaces. If one puts it on both surfaces there is no doubt it has wet both surfaces. I think it a waste of time to glue the end grain.

If you put glue only on one surface(not recommended) then put it on the surface where the excess will be pushed to an outside joint, not an inside corner. In this case glue the tails. I have no problem applying glue to the tails with either a small brush or a flat spatula. But again, I apply glue to both surfaces. And I rarely have more than an occasional tiny bead of glue squeeze out.

I don't risk a partial case glue up and I have never seen a professional shop that does either. Somewhere down the road you will get one of these joints off a bit and not know it till the rest of the parts get added. Glue gets applied one joint at a time and that joint assembled until the whole of the case is done. It is unimaginable that more than 20 min would be necessary to do the assembly. I do often get help applying glue in a complicated assembly. I use liquid hide glue for dovetails. It has slow tack so assembly is aided.

Diagonal clamping is risky. When the case is dry fit I check for square in the unclamped state. If it is not square then I fix the issue. If clamps are needed to square a case then spring back is likely and the result is unsquare.

There are two times important to glue. One time involves how long the joint can be left exposed to air. This time is typically in the range of 10 min and it of course depends on temperature and humidity. Glue can be diluted with 5% water to increase this time. The second time is how long after assembly can the joint be adjusted. 10-15 min is typical and it depends on moisture content of wood and variety of glue.

Re: Glueing up.

#4

I find myself in the strange place of....

roger lance

agreeing with much of what Bill said. A joint involving end grain doesn't have much glue strength. Too much glue and Bill is also right about glueing in such a way as to push the glue to the outside of the carcass....easier to clean up....save the inside corners from having to be cleaned which is great because I know they were ready for finish when you started the assembly....also the wet rag thing to clean up the squeeze out, its tempting, but let it dry to a rubbery consistency which is easy to clean off....I do partial glue ups of large carcasses but then also dry fit the other sides/pieces so that all is drying square and true....this takes some of the pressure off the process of glueing. I still hope that you will adopt me.

Re: Glueing up.

#5

Expansion

Bill Houghton, Sebastopol, CA

Derek,

You said in reference to gluing in the shelves, "I am never sure whether one should leave an unglued section for expansion here?"

The shelves appear to be of the same species of wood as the case, with the grain running in the same direction. Wouldn't the case and the shelf expand at the same rate at the same time? I would think you could glue the whole length.

You then said in reference to gluing in the shelf supports (sometimes known as "web frames"), "I did the same [glued just the front of the dado] with the drawer supports although these will expand differently, probably hardly at all (as they are all mortice-and-tenon joints)."

It's true that the shelf supports won't expand much. But they're being glued into a solid case, which will change sizes with the seasons, so gluing just the front of the dado makes a lot of sense in this case. I don't recall whether you've discussed how you're handing the back of these cases, but the back will presumably hold the sides from spreading, and thus keep the shelf supports firmly in place. Had you been raised in my family, you'd have cut some slotted holes in the shelf support toward the back to drive a screw into the side of the case and center support for added strength (we tended a lot to what we called "cast iron engineering," meaning that, if doing A and B made it strong enough for a century or so, then doing C and D must make it even better).

I couldn't tell from the pictures whether you'd run the shelf supports all the way to the back of the case (less the space for the back) or left some room for the case to expand and contract relative to the supports.

I am here, foolishly, offering lessons to a respected colleague who doesn't need them; but I'm pleased that you're doing this discussion of glue-up technique, and offer these comments with hopes that they'll improve the conversation.

Re: Glueing up.

#6

wood moving

Bill Tindall

It is a perilous assumption to think that all wood of a species responds to moisture changes equally. In addition to the well known "rule of thumb" that radial changes are only half tangential changes, other factors can vary this response. I can't lay hands on the reference(filing system is hopelessly flawed) so I won't risk quoting the value I think I remember, but the standard deviation within a species is substantial.

Re: Glueing up.

#7

Re: I find myself in the strange place of....

Bill Tindall

It is not at all surprising that once the discussion gets around to actually making something the divergence lessens among people experienced at making stuff .

Re: Glueing up.

#8

damm....I agree again !!

roger lance

Re: Glueing up.

#9

Re: Glueing up.

Ron in Sterling, CT

Derek,

Over the years I have done a lot of "glueing up", but it is always good to see how, others with much more experience than I, approach the subject. To that end your excellent write up, and pics, has gone a long way to explain your process, and show me some new teachniques. I thank you for that.

Your thread has also brought forth excellent comments and experience from others, and this is important to anyone who wants to learn more. I thank you all.

Ron

Re: Glueing up.

#10

Thanks.  Very useful information

Bill Houghton, Sebastopol, CA

Re: Glueing up.

#11

Re: Glueing up.

Warren in Lancaster, PA

I use hot hide glue to put together a carcass like this. I don't have a camera, but probably wouldn't have time to take a picture while gluing. I use a brush and put a thin coat coat of glue on both pieces. Even though hide glue doesn't stain like Titebond you really don't want a lot squeezing out. I often put dividers, shelves and such in after the carcass is dried; I think this is the traditional method.

I generally refuse restoration work if the piece is put together with yellow glue. It just is not worth the aggravation of trying to get the thing apart and cleaning off the old yellow glue. Nobody wants to repair that class of furniture. And most other guys with restoration experience use hide glue for new construction as well.

Re: Glueing up.

#12

Re: some comments - replies

Derek Cohen (in Perth, Australia)

Hi Bill

Many thanks for your thoughtful reply. You raise many of the issues that are often in one's mind, and the choices that we have to consider. I will respond from my experience here, mainly to clarify what went on in my head and what was unseen in the pictures (not that cool under pressure as suggested by having time to take all the photos!).

You are using way to much glue. It takes a film of glue only two molecules thick to form the bond. Try brushing it on with a small brush. Before assembly scrap all that extra glue off with the spreader, or don't put it on to begin with.

Although it looks a lot of glue in the pictures, the fact is that there was very little squeeze out - the tiniest bead only. Squeeze out was immediately cleaned up with clean paper towel soaked in tepid water. I'd rather that there was a little squeeze out - the merest suggestion thereof - than none as then I am reassured that I put on enough glue.

For glue to work it has to wet both surfaces. If one puts it on both surfaces there is no doubt it has wet both surfaces. I think it a waste of time to glue the end grain.

There are two schools of thought when it comes to one- versus two sides. I have done both. In this situation, with as many dovetails to glue, the open time of the glue would effectively have been halved - that would not have been sufficient (and THEN I really would have panicked!).

With regard end grain, I do agree with you .. intellectually. Emotionally it is another thing :\ ... my fear of leaving some join unsupported gets the better of me.

If you put glue only on one surface(not recommended) then put it on the surface where the excess will be pushed to an outside joint, not an inside corner. In this case glue the tails.

Yes to ensuring that the surfaces coming together do not created a starved joint. I ensured that the glue was a little thicker at the top of the pins. Pushing one end over another does not favour gluing the tails - that does not make sense. I chose the pins because it is easier to spead glue fully over the surface.

I don't risk a partial case glue up and I have never seen a professional shop that does either. Somewhere down the road you will get one of these joints off a bit and not know it till the rest of the parts get added. Glue gets applied one joint at a time and that joint assembled until the whole of the case is done.

I think that if I did this more frequently (2 or 3 large projects a year is my quota, with a dozen small ones thrown in that do not require this kind of pressure), then I would be more efficient and more confident. I have better control working with smaller sections. I wonder how others feel? In any event, your point about having a joint move is taken - which is why I am fanatical about measuring and monitoring the squareness as I go along (you can see several squares being used in the pictures). As I mentioned, clamps are used to maintain the angles, not simply to add pressure on a join.

It is unimaginable that more than 20 min would be necessary to do the assembly. I do often get help applying glue in a complicated assembly. I use liquid hide glue for dovetails. It has slow tack so assembly is aided.

I am in a double-bind with hide glue - I lack experience in using it, and therefore I avoid it. I recognise the advantages regarding repairs (as Warren mentioned below). Help? Rufus (the Golden Retriever) would rather chew any Jarrah than help me. And Aura (my son's shih tzu cross maltese) just bats her eyes at me, then lies down to rest) ..


Diagonal clamping is risky. When the case is dry fit I check for square in the unclamped state. If it is not square then I fix the issue. If clamps are needed to square a case then spring back is likely and the result is unsquare.

I will find out whether there was any springback later - I am in your camp here. I do not rely on these braces to square the carcase. They were just fine tuning - there was a fraction of a mm that I wanted to correct.

There are two times important to glue. One time involves how long the joint can be left exposed to air. This time is typically in the range of 10 min and it of course depends on temperature and humidity. Glue can be diluted with 5% water to increase this time. The second time is how long after assembly can the joint be adjusted. 10-15 min is typical and it depends on moisture content of wood and variety of glue.

Thanks. Good info.

For reference, it is Winter here and the weather is cool (18-21 Centigrade/about 65 degrees F).

Thanks again.

Regards from Perth

Derek

Re: Glueing up.

#13

Re: Glueing up.

Derek Cohen (in Perth, Australia)

Thanks for replying Warren.

See my reply above about the hide glue. I shall be including it in future pieces. To be fair, however, I have not seen PVC fail in 20 years. On the other hand I have not thought of my furniture being used for more than my lifetime - never considered it good enough that someone else would want to own it, nevermind repair it, in generations to come.

I often put dividers, shelves and such in after the carcass is dried; I think this is the traditional method.

How do you do that, the dividers I mean?

I was doing a constant dry fit - measuring when assembled, dissembling to create the dados (for example), assembling to test the fit, dissembling to tweek ... Not a efficient method I know, but I really did not want to try and do this work after the carcase was permanently assembled.

Regards from Perth

Derek

Re: Glueing up.

#14

Re: I find myself in the strange place of....

Derek Cohen (in Perth, Australia)

the wet rag thing to clean up the squeeze out, its tempting, but let it dry to a rubbery consistency which is easy to clean off

Hi Roger

Thanks for the reply.

I know that many, like yourself, prefer to scrape off the residue. However water works best with Jarrah. It is so hard and oily that it is not affected by water. Around the world Jarrah was used in the construction of piers and wharfs, and the sleepers sat in the water for a 100 years without erosion.

I recall that Norm, in the days I watched his show, also washed away the glue residue. It did not appear to affect his finish.

The other point is that I only use clean paper towel, never a cloth. Cloths tend to get used for many things and, if they pick up metal filings, this with leave black marks on the wood.

Regards from Perth

Derek

Re: Glueing up.

#15

Re: Glueing up.

Warren in Lancaster, PA

With some designs internal pieces need to be put in at the time of gluing. However, with your design everything could be fitted and installed from the rear after gluing. In restoration work it is nice if pieces (shelves or runners for example) can be removed from the rear without disturbing the carcass.

Re: Glueing up.

#16

Images of glue squeeze out.

Derek Cohen (in Perth, Australia)

Below are a couple of images from the glue up today. The first one was about average for the amount of squeeze out I generally experience, while the second is a little more than average. Both are, I believe, quite a small amount, and far less than I think that Bill was imagining. Yes/No?

First image ..


Second image ...


Regards from Perth

Derek

Re: Glueing up.

#17

Having done both...

John in NM

I too find the partially dried peeling to work better, though I usually give the visible area around the joint some tung oil before glue up to make sure the glue pops right off.

Both techniques work well, when the limitations of each are understood. My preference likely comes from habit as so many do.

Re: Glueing up.

#18

Re: Images of glue squeeze out.

Todd Nebel

Prior to applying the glue I will use blue tape on the inside corners. I run it tight up aginst the baseline of the tails and pins. I make tabs on the top of the tape so that it's easy to grab and peel off later. Several minutes after everything is clamped and the little bit of glue squeeze out is starting to harden I carefully pull off the tape and the squeeze out comes with it. If done right the inside corners are perfect, clean and ready for finish.

Re: Glueing up.

#19

Re: wood moving

Robin Corell in Atascadero CA

I think the actual potential movement is only a fraction of why Derek is concerned about if he should glue the entire length of the center divider. With the center divider almost entirely contained within the cabinet it will have very little exposure to the changes in humidity like the outside of the case will. I would be much more concerned with this differential in movement than the standard deviation or the difference between radial and tangential shrinkage.

I think the actual structural function of the support needs to come into consideration of how it should be glued. How much does the member contribute to the overall structural integrity of the cabinet? In this case it appears to be minimal and that it provides support for the shelves and drawer supports but everything is positively located mechanically by dados.

Of course no one has mentioned the fact that the entire joint in the center support is end grain to end grain. There is no side grain to side grain contact in the joint. I have had very good results in getting a strong end grain glue joint by "priming" the joint with hot hide glue and then scraping/sanding it smooth before glue up, but because of how yellow glue works this technique is futile.

Robin

Re: Glueing up.

#20

Re: I find myself in the strange place of....

roger lance

Derek....I'm not familiar with the properties of jarrah....it certainly is a beautiful wood....and your craftsmanship is outstanding....but the point should be made that the technique of wiping away glue with water does offer up the potential for glue contamination of surfaces especially on porous open grain woods which can lead to finishing issues....the concept of glue management is well worth discussing with the goal of directing where the excess goes to the benefit of the woodworker....Norm did use the wet sponge/rag method and he was criticized for this by many knowledgeable people....it was an area where he could have picked up his game but did not do so....and I think this is the idea of these discussions....improvement.

Re: Glueing up.

#21

Indeed

Bill Tindall

Hot damn! someone finally built something and we are arguing about making furniture instead of how many angels can sit on the edge of a chisel.

A good building philosophy is not to plan, or need, to fix something later that can be readily avoided to begin with. Experienced builders just don't have glue squeezing out where it isn't suppose to be and its done with proper application of the right amount of glue, to the right places. In some cases providing glue a place to go where it is harmless is a strategy, for example the bottom of mortises.

It is rare that the right amount of glue can be applied by squeezing the glue bottle. Glue bottles make gushers. I buy small plastic cups from the restaurant supply (or swipe them from buffet lines in restaurants). A bit of glue is added to the cup and it is transferred from cup to wood with a spatula or small stiff brush (acid brushes work great). One can apply glue to something like dovetails very fast with this arrangement.

To the tapers, scrapers and washers..... there won't be any more glue in a M&T or dovetail joint than the gap between the mating parts. Ideally this amount is about 0.005" more or less. It doesn't take much of a layer of applied glue to achieve this thickness. In fact the yellow glues tack so fast that it would be near impossible to scrape off enough glue to have too little to bond properly, especially if glue is applied to both mating surfaces, something that just makes a lot of sense.

Perhaps it is worthwhile to explain how glue works. To work glue must wet the mating surfaces. "Wet" in this case is used in the surface tension definition, wet = spread and become one with and not bead up or be repelled from the surface. The glue must spread across the surface like water spreading on clean glass. Some physics mumbo jumbo results in the bond between what is spreading, and what it is spread upon, being stronger than the strength of the weakest material, in this case wood. If glue is applied to only one mating surface glue may not get properly transferred to the other surface and wet it. The yellow glues tack extremely rapidly, and, their surface dries fairly rapidly. In each case the high viscosity glue that remains will be reluctant to wet the dry mating surface. Or in a tight joint all the low viscosity glue will be scrapped off during assembly and the high viscosity remaining layer will not wet the mating surface. A starved joint will result in each case.

If anyone thinks it trivial to transfer glue from one surface to another try putting glue on one side of an edge joint, push the mating parts together and then pull them apart. In fact, it takes a fair amount of rubbing to get complete transfer from the wet surface to the dry surface. You only get one rub with assembly of M&T or dovetail and it's a dry rub at that.

Re: Glueing up.

#22

Moses Yoder

Re: wood moving

Moses Yoder

The joint in the center support is end grain to face grain. I have done a lot of joints like this, up to 26" wide, glued the full length, and never had problems with failure. The wood moves close enough to the same that there is no joint failure. The glue does have some holding power. It is possible for the earth to stop spinning as well, but I think I will worry about that when it actually happens.

Re: Glueing up.

#23

Thank you for sharing your glue-up approach

North of the Border

We all learned something from your post or from those of others who commented on your operation. The measuring tape cut-offs do make some good glue spreaders.

I always put glue to the tails only, no glue on any other parts (pins or end grains) because the glue squeeze is kept to the minimal and to the outside surfaces (where clean-up is a lot easier). TAILS win!

Also, for this kind of scale of glue-up job, I'd use commercial or shop-made assembly helpers like this product:

http://www.leevalley.com/en/wood/page.aspx?p=51101&cat=1,43838

Re: Glueing up.

#24

1 side vs 2 sides and Norm

North of the Border

I agree that putting glue to one side of a piece is good enough; I never have had a glue joint failure as a result of not putting glue to two mating sides or faces, provided enough glue is put on the first side. Dovetail joints are inherently strong and usually lack of glue is not the reason for a joint failure if it does fail.

However, I wouldn't trust what we see in NYW. Using a wet rag to clean up glue is OK if the surface with the glue smear is not to be stained. 10 out of 10 times the glue spots, if severe, will show after the staining process...I suspect if the camera in the NYW closed in on some of the project surfaces, we might see something we couldn't see from a distance. I agree with Norm on a lot of things he says or does, but not on how he cleans up the glue squeeze.

Re: Glueing up.

#25

Re: Expansion

Chris Friesen

An alternative with web frames is to leave the rear mortise/tenon joints unglued, but have the back rail dovetail to the rear edge of the carcase. This holds the carcase sides together but lets the web frame side members move relative to the carcase side.

This technique is used by Chris Becksvoort in FWW.