Glueing with non-matching grooves

Hi Kviljo

Nowadays I rarely have failures in glueing and matching horn and core. I groove horn,s and core,s with http://www.mehr-als-werkzeug.de/product ... -Griff.htm

I coat the horn,s and core,s with 5 to 7 thin layers of issinglass glue, mix then issinglass with hide glue for another 8 to 10 thin coats - in the end horn and wood looks gloosy.

I am using the all wrapped in ... technique by Jaap Koppendrayer. I prepare the horn,s and core,s by boiling water in a kettle (using the the steam), heat the horn,s over the oven (back side) so it is getting hot, at the same time I am shifting back and for the core over the steaming/boiling water then I put on a last thick layer of glue on both and then the wedding starts. No hurry at all is necessary - the room temparature should be > 22 degrees.

Easy, fast and no air bubbles between horn and core.

Good luck and give it a try.

Reto







Old topic - some of my proposal to read (beside many others):

http://198.170.107.188/phpBB2/viewtopic ... ght=#18820

http://198.170.107.188/phpBB2/viewtopic ... c&start=30

http://atarn.net/phpBB2/viewtopic.php?t ... sc&start=0

Yes, the bladder glue should work better for you, since you can use even higher concentration for the final coat to remain liquid long enough. I also believe the dry bladder glue is more elastic in colder climates, so the separation is less likely. I have not had enough palate glue to use for horn, perhaps this glue was even better than bladder. In old bows the glue is thick and fills the grooves, which calls for the least-gelling types, used at the highest concentration. It also possible the thick glue was allowed to dry partially, then heated to melt immediately before clamping, so the grooves would be flooded with the melt.

I notice your bows are rather ambitious and high stressed with short, curved sals and long siyahs. I am not sure what is the poundage, but I would go with the matching grooves for bows like that above 60-70lb weight.

The other thing is the thickness of horn - with thick horn in sals, the glue line is closer to the neutral plane of the limb. This way there is only shear stress there and less compression. I found the stressed bows do not like thin horn, which tends to come off the core.

I have some doubts on the Japanese method of clamping that Reto likes. I think the pressure is less (so one needs the horn to be more flexible/thin/hotter) and it is applied at the wedges only - perhaps not as close as in the case of the clamps. If the curvature of the horns is very similar to the core's and the horn thinner, it should work well I think.

Anyway, you need to be very careful and have your methods well perfected for the heavier flight bows. My heaviest so far was 170-180lb @ 26in, 41in ntn bow - survived well enough, but horns were very thick, sals not too curved and the grooves matched.

Adam

For my first bow I used non matching grooves; during tillering the horn delaminated and broke apart. In examining the glue joint you could see visible gaps along with general poor adhesion due to a lack of proper sizing coats.

In talking with Denis Betoulle he suggested the following:
You must to apply several glue layers.
The first are apply with light glue (10%) in four to five layers
And you make again the operation with more glue each time (15%, 20%, 25%, 30% of glue)
Between two operations don?t late dry the glue completely and apply a layer of light glue before
At the end of operation the grooves are full and you haven?t voids when the horn and the core are glued you applied 20 to 30 layers of glue or more.

I would guess that if you have non matching grooves you would have to follow the above advice and have most of the grooves filled before final glue up; or use a very thick glue during the final glue-up. On all of my subsequent bows I have used matching grooves. I then apply 20 thin coats of fish bladder glue at 2% and then for the last coat I use 20% glue, I use C clamps for the glue-up. Note I first tried using higher percentages for applying sizing coats but ran into problems with air bubbles.

Most of my latest bows have failed but I am pleased that the horn joint has always held strong. I also have cut some of these bows apart to further examine the glue joints.

[external image]

To groove or not to groove?

I have (very belatedly) resumed my Magyar bow project, this being my very first attempt at a horn bow. After my horn arrives I will be able to continue.

The core is completed, and I went through great pains to make the bare belly as flat as possible, with the intention of having as little void space between it and the horn.

I have Adam Karpowicz's book and have again reviewed his discussion on grooving (pp 84 - 88), so I think I have a fairly good understanding of the purpose of creating matching grooves in the horn and core. However, I have some reluctance to attempt matching grooves because it seems obvious to me that unless it is done with great precision, it is would be easy to wind up with serious discontinuities along the interface of the horn and core.

I am also somewhat leery of non-matching grooves. And quite frankly, I don't quite see the benefit of non-matching grooves, as it appears even more likely to get discontinuities due to high points sitting on top of each other and/or low points lining up. True, these are supposed to be filled in with multiple layers of glue and thus create a sort of transition zone of thin bands of glue alternating with wood, but I have doubts. While I have no experience with hide glue (yet), I have done a great deal of wood gluing using a variety of different adhesives, and the idea of non-matching grooves is contrary to the concept that to have a maximum strength glue joint you need to have two surfaces as flat as possible pressed together with a minimum of void space. But obviously, this feature of non-matching grooving has been used for centuries, no doubt for good reason.

Now, here's my question: for a bow of modest weight (say approx 65 lbs) will I be better off with non-matching grooves, for which some people here have experienced problems, or with just making sure that the core and horn are both as flat and uniform as possible and for the joint at the interface to have a minimum of void space?

As this is my first attempt at a horn bow, and since there are so many things that can go wrong, I need all the help I can get. So, what the thinking out there on 'to groove or not to groove'?

I would make matching grooves. It isn't hard to make them match properly. Aferall you are using the same scraper on both core and horn so you can't go wrong. Obviously the first guide groove must, must, must be perfectly straight. I actually use a very fine tenon saw up against a clamped straight edge as a guide. If you struggle to find a good straight edge (as I did) goto a general steel fabricators. I got some 1.5mm thick sheet off them and had it cut into a strip.
I had delamination on my first 2 bows but since then i've just followed Adam's methods to the letter and have had no problems since (I may have just cursed myself there!).
Good luck Kilvo.

[external image]
I cannot help finding it a bit contradictory when you make these non-matching grooves in order to create a surface that is not smooth and then you want to fill up the grooves again with glue before putting it together. Now you have created a smooth surface again with lots of glue in the bottom of the grooves and shall we say a normal amount on the tops. Right or wrong I cannot help thinking like that.
I am enclosing a picture that should give some food for thought when it comes to gluing up wood to wood. Many still claims that you should sand the workpieces before gluing with a non gap filling adeshive. If you ask them why the answer is often that they have always done so.
If the workpieces to be glued have been laying around for some time and have reacted with the air you could give them a light sanding with a new sharp paper.
If we only knew how the ancients did it. Markus

I see that the text below the pic is difficult to read so I'll write it here; (A)
A cross section of a portion of a bonded joint between two undamaged Douglas-fir surfaces prepared by planing with a sharp knife (magnification 120x). The dark area at the center of the photomicrograph is the adhesive bondline. The wood cells are open, and their Walls are destinct without evidence of damage.
(B) A cross section of a portion of a bonded joint between two damaged Douglas-fir surfaces prepared by abrasive planing with 36 grit ( magnification 120x) The cells in and adjacent to the bondline are crushed and their walls indistinct. At higher magnifications ( not shown) cracks in the cell walls can be seen.
It should of course be said that the knives of the plane or planer device should be sharp, otherwise it might do more damage than good.

Marcus,

Ah yes, that photo and accompanying discussion is the sort of thing I was looking for with regard to the merits of smooth surface bonding versus non-matching groove bonding. I am inclined to interpret this as suggesting that while non-matching grooves will put a lot of glue between the wooden core and the horn, it might not create the best bond.

This could be especially true if in the process of making the grooves you create a lot of tears, abrasions, and other forms of damaged fibers. In which case, you might be partially bonding loose or damaged fibers to other loose or damaged fibers.

So then, I suppose my next question would be: can one make a horn bow of moderate draw weight (say less than 65 lbs) that will not be prone to delamination by avoiding grooving altogether but making two very carefully prepared flat smooth surfaces, thus creating a glue line of minimum thickness?

Or is something else at work here?

Edmund Dohnert wrote:

Marcus,

Ah yes, that photo and accompanying discussion is the sort of thing I was looking for with regard to the merits of smooth surface bonding versus non-matching groove bonding. I am inclined to interpret this as suggesting that while non-matching grooves will put a lot of glue between the wooden core and the horn, it might not create the best bond.

This could be especially true if in the process of making the grooves you create a lot of tears, abrasions, and other forms of damaged fibers. In which case, you might be partially bonding loose or damaged fibers to other loose or damaged fibers.

So then, I suppose my next question would be: can one make a horn bow of moderate draw weight (say less than 65 lbs) that will not be prone to delamination by avoiding grooving altogether but making two very carefully prepared flat smooth surfaces, thus creating a glue line of minimum thickness?

Or is something else at work here?

I'd say yes there certainly is something else at work here.
For a start the grooves aren't sanded in - they are made with a cutting tool like a scraper. They are not made with 36 grit paper. Very few people use 36 grit paper to sand surfaces before gluing. Also it makes a huge difference what glue you are using. Coolgen glues prefer perfectly (or as close as possible) flat, smooth surfaces.
I don't think the picture (whilst interesting) shows anything useful for hornbow making.
Adam explains the thought process behind grooving in his book.
I think i'm correct in saying that all the old bows had non matching grooves - they didn't do it for no reason! We are really only playing at making these bows. I think at the time when hornbows where being made /used/fought with regularly then their production methods would have been at the height of perfection....they had a lot of brains thinking about the same problems/processes and we simply don't :)

Mikekeswick -

Thanks much for the comments. I have Adam Karpowicz's book and have read his discussion on non-matching grooving (p 85), but am still a little confused as to the merits of non-matching grooving. As best I can tell, the rationale is that the glue has mechanical properties not unlike horn and that by filling the void spaces created by the non-matching grooves with glue, you are essentially 'injecting' horn into the wood. Perhaps this occurs with hide glue, but with wood gluing in general, it is considered bad practice to rely on the mechanical strength of thick layers of glue alone.

I know that the grooves are made with scrapers rather than sanding. I have no problem with that, as I have used homemade flat scrapers before and have used such a scraper to create the flat surface on the belly of the maple core. However, you can still do bad things with a scraper, especially if it doesn't have a sufficiently sharp edge.

As you said, there must be a good reason that non-matching grooving was in such widespread use for so many centuries. So, as far as the question, 'To groove or not to groove', I am now inclined to conclude that the answer is Yes. Regardless of whether I attempt matching grooves or go with non-matching, once I make my scraper I had better do some practicing on scrap wood before I attempt the real thing.

Yes, regarding bow making, I'm sure that there's still a wealth of lost arcane knowledge that we have so far incompletely recreated. I guess that's one of the things that fascinates be about this stuff.

The video that Thomas Duvernay shot shows the korean master making non-matching grooves at 1:23:00. He just pulls his toothed file forcefully along the material just a few times.

Now, if you've seen how he handles the adzes, I wouldn't be surprised if the grooves matched perfectly like being cut with a router, but being serious again, I'd doubt that.

I too don't think that the picture tells us much. We're not glueing wood on wood in this case. Since the mechanical properties of the glue are very similar to horn, think of the bond of making the "horn" expand into the enlarged surface of the wood.

It's simply no accident that we're always looking for a surface that's as large as possible.

Thanks for the input, guys! I think there should be enough to play with now. I will try to thicken the final layer of glue as much as I can, and perhaps build up more glue during the sizing too. Also, I will have to continue working on the design of my bows, as I think they have all failed at places where the shape have not been optimal. I guess I could go for matching grooves too, but it does not appeal to me somehow. If it was possible to make heavy short bows with non-matching grooves back then, then it should be possible today also - somehow.

Reto, are you using matching grooves?

I think these pictures of this cut up bow is the only direct source I have seen with regards to what the grooves looked like (except for the picture of the damaged bow in Adams book). These ones are properly non-matching, and are also not very deep. Curving the glue line probably help prevent failure though, as the vulnerable edges are closer to the neutral plane.

http://kviljo.no/bue/atarn/a.jpg



And on a side-note: Adam, will we see speed-tests of your 170#++ flight bow? 8)

This is not about bowmaking, but it explains the basics of laminated bending structures.

http://www.oneoceankayaks.com/Sandcore.htm

The glue line will hold better when it does not have to do work. This means it has to placed in the neutral bending axis. What Adam says with a thick horn layer makes perfect sense because it "moves" the glue line closer to the neutral axis.
I am not into making composite bows, but to me, looking from outside, a successful glue-up depends on getting the apropriate thickness of the materials based on their particular properties. This means for different quality of horn and wood, different thicknesses must be used. How to evaluate this based on the material you are working with is the greatest challenge. I admire and respect the work you guys are doing. Keep it up!

this is how ancient nomads did it

bow of Duguj Cachir, Mongolia
10/11th century (Kitan era)

see book: Steppenkrieger

Uli -

Interesting picture.

While those Mongolia bows were obviously quite effective, as far as grooving goes, those are pretty poor looking grooves. They aren't even all parallel to the limb and some ppear to run off one side.

I suppose this should be somewhat encouraging to someone like myself who hasn't done any grooving yet, because it suggests that non-matching grooves do not have to be all that perfect for the bow to succeed.

... and I d like to add:
we should keep in mind the distinction between

a) making the perfect (composite) bow
b) making the exact replica

both is desirable but aim is different

I do not think those look like grooves. I think they look more like the grain of the wood. Or it may just be the picture.

Well, I think I'm beginning to understand a little bit more about non-matching grooves. I recently made myself a homemade scraper and guide fixture to ensure that the grooves are all parallel to the centerline of the limbs.

My grooves are a bit on the shallow side, as I purposely didn't want to overdo it.

I have just completed applying two coats of thin (10-percent) hide glue to both the grooved horn plates and the grooved limbs. The wood appears to soak it up quite readily, the horn somewhat less so. I think I need several more coats of the thin glue before I start with the thicker glue (maybe 20 -30-percent?) leading up to the final application before clamping. (I've made a clamping strip similar to the one shown in Adam Karpowicz's book.)

It would appear to me that for non-matching grooves, the grooves on both the horn and wood should be almost totally filled with glue before the final gluing, so that there are no void spaces when the whole thing is done. However, it would also seem that there might be some danger in filling the grooves beyond the high points of the grooves, as that would lead to a thick glue line and potential bonding problems.

I was wondering if it might not be a good idea to give both the horn and the wood a VERY light final scraping with a flat scraper (such as a razor blade) to remove any excess glue from the high point before the final gluing with the heavy glue, so as to ensure two very flat surfaces and to avoid any lumps due to excess glue in one spot or another. Or would this be looking for trouble?

Any thoughts out there would be most appreciated.

IMHO the collagen glues contract so enormously during drying that any voids including the grooving should be filled. your idea is interesting, because with much coating the surface will not be flat but rounded at the edges. in a very thick glue line you will notice how the glue that initially is squeezed out or bulges from the joint will later retract toward the center.

Thanks Uli -

That pretty much summarizes my thinking. Non-matching grooves would appear to not accomplish much if you do not have perfectly flat horn and wood surfaces prior to doing the final gluing. But you really can't achieve that if you have layers of glue built up on top of the 'peaks' of the grooves. Thus, I would think that a very light scraping with a straight scraper before final gluing would go a long way in ensuring a nice flat glue line.

So, I think that once I get all these shallow grooves filled with dried glue, I will give both the horn and wood a very light one or two passes with the flat scraper to make the adjoining surfaces more flat. Then I'll apply a very thick coat of glue and hurriedly do the clamping.

Right now I am very patiently filling grooves with multiple coats of thinned glue (10%). It is slow going.

Yes but just before clamping the surfaces are made really quite hot. This heat allows the horn to conform to whatever shape the core is any way. Plus you have got a thick piece of wood to act as a form. Clamp with the rope method and you will be getting a very strong clamping force that will easily make the two surfaces mate perfectly.

mikekeswick -

Thanks for the advice.

I made both the core and horn strips nice and toastie before gluing, and I am certain I had fully liquid glue even well after everything was clamped in place. As I don't trust myself with the rope/cinching method of clamping, I made a pressure pad (per Adam Karpowicz's book) consisting of six wood blocks attached to a strip of poster board and a strip of felt over that. I clamped the horn strips in place with six big C-clamps.

The time and effort I took to make the horn strips and core belly as flat and matching as possible (with nice parallel shallow grooves) appears to have paid of. After filing the edges of the limb to get them closer to shape, I appear to have a very straight and practically invisible glue line. Of course, the proof of whether this was successful will come much later.

I will now spend some time CAREFULLY adjusting the cross-sectional shape of the horn/core, as both are still rather thick for the draw weight I am going for (no more than 60 -65 lbs). In its present form, even without any sinew attached, the bow is VERY stiff.