Help! Yet another rub joint question

[Chris Knowlton]

I've worked patiently on trying to glue up the top plates for more than a week, and though I am getting really nice seriously flat joints, I have not been able to get the glue joint correct. I see where I went wrong on some of the attempts, uneven pressure, glue too thick. I also tried clamping.

With the last attempt everything seemed to go really well. The surfaces were smooth and flat, the glue seemed to be right (see picture), I placed one plate in the vice, I heated the surfaces with a hair drier for about 4-minutes, I put the glue quickly onto one surface making sure that it was fully coated, I rubbed the plates back and forth until they stuck.

I ended up with what I believe is an acceptably thick (very small) even glue line which I assume would close except that there are areas that, when viewed with magnifiers, appear to have no glue in them. Even if the joint closes there can't be any strength to it.

My question is: How do you (in painful detail) correctly perform a rub joint assuming all of the preparatory work has been done correctly? For instance how much pressure should be applied?

I need some ammo for the 7th attempt.

[COB3]

With nothing more than your description and that one photo, I would have guessed that to be a good joint. You could try taking a few cross-grain shavings out of the middle, using a gouge, and bend them until they break-- do they break in the wood or in the glue-line?

I'd try a few shavings like that, (also look at the joint in the cut area where you carved it-- is it tight, and clean? ) then, go ahead and use it...though you have shown admirable tenacity, in re-working the joint until it is right.

[Tim McTigue]

Chris, with my limited experience, I would be seriously surprised if that is not a good glue joint. It looks like you used quite a bit more glue than I did, I didn't have any squeeze-out at all, you've got glue running down the face. If anything, I'd wonder if you may have used too much - understandable, given the circumstances.

What did you use to apply the glue? I used a small palette knife. I got some glue on the knife, then quickly ran it along the jointing surface, and used the knife to further spread it out and smooth it. Then I got some more glue on the knife, and repeated the process. I only did this twice, which covered the entire surface. Then I pressed the other half down on it, and the only rubbing I did was just very quick, to line up the joint on the marks. Then I left it to dry.

Take heart, the problems I reported last week turned out to be very minor, and easily fixed, at least that's what it looks like at the moment. I glued a couple of pillars on, then made up some glue and ran it along the joint where the problem was, and worked it up and down a bit to get the glue down into the joint. In the process, I discovered the defect was a bit longer than I had realized, so I put some more glue there and worked the joint again. Then I put a couple of small clamps on the pillars and left it to dry. When I checked it later (and again the next day), no light was showing through, and now the joint seems solid. If the problem occurs again, I know what I'll do now, and I fully expect this back will be fine once I get it graduated. I plan to put a couple of small cleats where the fix was, but they may not even be necessary.

[Andres Sender]

His squeezeout is textbook, just like in Johnson/Courtnall. I worry when I don't see that, although so far unnecessarily.

A brush is faster, and the back and forth rub is what gives a rubbed joint its name.

[Oded Kishony]

Where is the gap you mention?

~OK

[M_A_T_T]

Your glue squeeze out looks good, I hardly have any, it may almost appear like I have a starved joint, but once carved the joints appear fine. In the areas where you see no glue, there may by glue deeper along the depth of the joint, closer to where you finished piece is cut from. I have noticed this myself.

[Janito]

The best joints do not show any glue line.

You can only tell where the latter is by the different visual behaviour when the back or top is moved back and forth in a light.

I have had violin fronts where it has been impossible to locate the centre joint as a guide to bridge placement (and no, they were not one-piece - I glued them!!)

[NewNewbie]

This YouTube.com video by Brian Lisus, shows a rub joint being made

around 2 1/2 minutes into the video.  

I would suggest that heating the joint is not necessary if you are

in a warm room.

If you have perfect wood to wood contact, then heating the joint is

only going to distort the wood.

See in the video how fast the glue is put on with a wide brush, and

how little time is spent before the plates are brought

together.

Also I would not wait too long after getting a perfect wood to wood

joint before gluing.  The wood surface can only degrade after

obtaining a perfect surface from finger prints, oxidation, dust,

and other contaminants.

YouTube is getting to be a pretty good place to see violins made,

maybe we need to get MANFIO to film his next masterpiece.

[NewNewbie]

In the general woodworking community, which is foreign to violin

most makers, since they occupy a special niché, there is a

saying that it is almost next to impossible to get a starved long

grain joint.

[Melvin Goldsmith]

quote:

---

Originally posted by:

Chris Knowlton

I've worked

patiently on trying to glue up the top plates for more than a week,

and though I am getting really nice seriously flat joints, I have

not been able to get the glue joint correct. I see where I went

wrong on some of the attempts, uneven pressure, glue too thick. I

also tried clamping. With the last attempt everything seemed to go

really well. The surfaces were smooth and flat, the glue seemed to

be right (see picture), I placed one plate in the vice, I heated

the surfaces with a hair drier for about 4-minutes, I put the glue

quickly onto one surface making sure that it was fully coated, I

rubbed the plates back and forth until they stuck. I ended up with

what I believe is an acceptably thick (very small) even glue line

which I assume would close except that there are areas that, when

viewed with magnifiers, appear to have no glue in them. Even if the

joint closes there can't be any strength to it. My question is: How

do you (in painful detail) correctly perform a rub joint assuming

all of the preparatory work has been done correctly? For instance

how much pressure should be applied? I need some ammo for the 7th

attempt.

---

................................................................................

...........

Hi Chris

I think you will find the rub join method in detail on MN if you do

a search......but there are some things you mention which I'll

comment on.

As Janito says there should be no visible glue line.

As Andres says a brush is best for glue application. Glue

application must be very fast so the rub is done while the glue is

still hot and liquid at pre gel stage. I notice you only applied

glue to one surface. This is incorrect. You MUST apply to both

surfaces for a strong join.

Also you mentioned using a hairdryer to heat the join for 4

minutes. This is a mistake....you should not use one at all. the

heat will distort your gluing surface....wood is moving all the

time with humidity changes.....an extreme local intervention like

this is not helpful.... What you should do is make sure your gluing

room is a suitable temperature fr gluing....20C plus for all the

planing and gluing process.

Your method of holding the lower plate looks OK...but always

remember that holding wood in a vise can distort it. always make

sure that your holding method is not forcing the join out of

alignment.... at all times.

[Janito]

One small comment about the Lisus YouTube movie where he is planing a plate for jointing. I noticed that his plane slowed down (and possibly stopped).

If the plane stops in midflow, there is a high probability of getting a small (micro) step when it starts moving again. This makes a perfect apposition of the joints more difficult.

So, 1 sweep along the board is best...if possible.

[David Burgess]

A lot of great information so far, particularly in Melvin's post. Glue should be applied to both surfaces.

The only thing I do differently is to apply the glue from a plastic squeeze bottle and spread with a finger, because it's faster than a brush. It may be next to impossible to get a glue-starved long grain joint, but sometimes we don't exactly have a long grain joint because of runout or curl.

[Chris Knowlton]

Thanks again everyone. You've given me the ammunition I needed to attack it again. I'll put the glue on both surfaces this time and no extra heat except for a warm room. Both of these items could have contributed to the results I am seeing.

Like most I've read a lot about plate jointing and gluing and I did searched the MN for information. I feel I could get a good joint with clamps and extra wood strips glued on the thin end of the plates, because it is most like what I would have done on a woodworking project in the past, but I really want to learn to get the rub joint right.

The information I was looking for is what you guys shared with me today. Thanks a lot. I should be jointing and gluing tomorrow night. I let you know how it goes.

I'll also try to get a picture of the gaps just for curiosities sake, and post it later this evening.

[Chris Knowlton]

Here are a couple of pictures of the glue joint. The files are a little large but I thought you'd see the detail a little better. The clean one is the back side of the top plate with the glue removed. The joint looks good. The other picture (with the glue residue from countless attempts) is enlarged quit a bit and shows the slight gap on the top side. You can see the glue bridging between the plates especially on the right side of the photo indicating little or no glue in between. I guess thats an indication of what happens if you put glue on only one plate.

I'm pump up for the next attempt.

Thank you all again.

[Chris Knowlton]

That did it! Thanks eveyone. It went really well. I didn't pre-heat, I applied glue to both surfaces, and I got a nice tight glue joint. Now I can move ahead.

[edi malinaric]

Fabulous feeling isn't it?

cheers edi

[violins88]

Chris,

I am glad you finally got a good joint. I guess the hair dryer was not needed. Also, for future reference, I have found that the rubbed joint method seems to be able to cause distortion. That's why I went to the clamping method.

[Brad Dorsey]

"I have found that the clamping method seems to be able to cause distortion. That's why I went to the clamping method."

Because you wanted distortion?

[JohnCockburn]

quote:

---

Originally posted by:

Brad_Dorsey

"I have found that the clamping method seems to be able to cause distortion. That's why I went to the clamping method."

Because you wanted distortion?

---

Senior Member having a Senior Moment?

[violins88]

JohnCee and Brad,

Ho, ho. Actually you are right -- senior member having a senior moment.

John

[jowl]

Have a look at Francis Kuttner's method of doing a clamped joint in the Jan 2013 issue of the Strad. Makes the whole process a cinch. Really good!