Adhesion: Making Handle Scales Stick, and Stay Stuck, on Blade Tangs

Somewhere on the paperwork that comes with Gflex, I remember reading that it was recommended to actually wet sand the epoxy into the surface. Makes sense to me though I have yet to try it.
 
jwccustom57 said:
Me too, hollowing out an epoxy reservoir has been widely recommended around here, as well as numerous epoxy pin holes, will be considering this point very closely. Question: peel forces on very thin flexible blades like fillet and kitchen slicer knives with or without bolsters. How can we minimize the risk of failure in these cases? More mechanical pins maybe?

Also it sounds like the recommendation of cross hatching is an important key correct? How do we go about creating the undercuts?

BTW thanks very much for this, very interesting.
Click to expand...

Peel forces are just plain bad. Maybe it did not come through clearly in the third post ... but I am thinking that if you use a metal bolster, soldered on to the blade, and make sure the handle is glued to the back of the bolster (which is flat and perpendicular to the blade), then when the blade flexes, the forces between the handle and the bolster are shear (not peel) forces ... and a good joint is much stronger in shear than peel. Make sense? If not I can post a picture that will make more sense....

The undercuts are mostly a statistical thing. You can not say "here is an undercut" ... but the more you roughen the surface the more undercuts will be likely to be created. Also - not "cross-hatching" ... but circular sanding motions by hand. If you think about it, the circular motion tends to push the cutting particle "sideways" against the channel it is gouging, thus increasing the likelihood of creating an undercut. the more you do that - the more undercuts might be created - sorry, it is just a numbers game all around :-(

re. using mechanical pins on flexible blades: without a bolster, the pins themselves will not likely stop the delamination of the front of the handle, but will stop the handle from coming completely off. You will, however, be left with an unsightly gap at the front of the handle, which can also accumulate grime. maybe if you used several small pins very close to the front of the handle, but then you run the risk of splitting the handle from the force of the pins???

You are very welcome - I do hope that this ends up being understandable and helpful to folks :-)
 
The Traditional Bowyers Bible had a chapter on glue and the results they found were similar results though they were just gluing wood and other natural materials. Great information. I hope it goes into the stickys.
 
Cushing H. said:
I would be honored :)
Click to expand...
Thanks for this topic , you answered us lot of question :thumbsup:
I have another question ...What about other type of glue , is epoxy always best choice ?
On pictures are my small Parang which I use on my fishing trip .handle are rubber , pins are rubber/for more comfort/ and glue is / I think you call it contact cement ?/ ordinary glue for leather ,rubber and stuff like that .It has yellowish color and need to dry before we join parts ..I m right ? Is it that what you in USA call contact cement ?
Now , this summer will be the fourth season how I use this Parang , and I was surprised how good hold that glue .Parang is very hard and often used on my fishing trip ....to clean new fishing spot , to turn dry wood in water .....
After all this years I can't see single point where glue loose
J7X7bIQ.jpg

P7QyOFT.jpg

krGseBz.jpg

U2CWJ1X.jpg


One more thing ...for my hidden tang knife I use epoxy mixed with steel sawdust from hand filing with file .I think that sawdust make epoxy many , many times stronger ? What do you think about that ? I save epoxy and have stronger glue inside ?

M08ManC.jpg

4jujhRx.jpg
 
Hi Natlek. sorry for the delay in responding to your earlier questions: I have been out on vacation and did not have great connectivity. Contact Cement (and I believe you are correct to call it that) is a simple glue that "cures" by evaporating its solvents. The remaining filler material is "sticky" until it completely dries - and hence the ability to put it on both sides of a bond, then stick the two sides together. It does NOT create a cross-linked "polymeric" system when it cures. The result is that it is likely not as strong as epoxy, AND will degrade/become brittle faster. A theme you are likely starting to hear from me is "is what you are doing good enough for your purposes", instead of "make it the best possible that you can". My guess is that your Parang will have some failure of the handle bond to the tang before one made with epoxy would (and it will not survive high temperatures as well as epoxy would, BUT it has lasted 3-4 years. Is that "good enough" for you? If so, then the contact cement might be worth for you (and others) to consider for a "quick and dirty, short time use" knife. Interesting thing to try!.

The metal shavings mixed with epoxy is like what we in the US buy as "JB Weld" (which is basically a metal powder impregnated epoxy). Its claim to fame is that it withstands high temperatures better than ordinary epoxy. Why, I do not know - if the epoxy it uses is just standard epoxy, then that portion of the adhesive should degrade at high temperatures regardless of whether the metal is present. maybe they use a special epoxy that is more resistant to high temperatures? But the portion of the adhesive that "makes the bond" by flowing into scratches and undercuts is still the epoxy base. I also know you can drill and tap JB weld, and the resulting threads are stronger than if just made with epoxy. In this application (knife handles) I just do know (one way or another) if it has an advantage over epoxy....
 
What I have been doing recently is making sure the tang is flat, roughed up with 80 grit paper, scratch pattern all in one direction (tip to tang), but I also will scribe lines with a dremel or needle file in the same direction. I guess the better practice would be do rough up the tang in a circular motion with the 80 grit, and forget about the scribe lines? As far as the underside of the scales, I also make sure they are as flat as I can get them, rough them up with 80 grit, and drill small shallow epoxy "bridge" holes. Maybe I should skip the holes? And back to the tang, I drill holes in the tang strictly for weight reduction. However, when epoxying scales to the tang, I make sure that these holes are filled with epoxy to act as "bridges" as well. Maybe the "best" bond would be to just rough up the tang and drill no holes (except pin holes of course). And that the best way for weight reduction, if you're after the strongest possible bond you can get, would be to taper the tang. But then like you mentioned in your next thread, rhetorically, when is good enough good enough? I have had 2 non stabilized wood handle issues in hundreds of knives. One was left in the sink covered with water...user error. The other, luckily, was my sister's chef's knife. The highly figured (non stabilized) walnut peeled away from the tang at the front, but she didn't even notice it until I showed her. The rest of the handle was tightly bonded. I am assuming it had something to do with the fact that the wood was extremely highly figured (almost burl like), and not stabilized.
 
Cushing H. said:
The metal shavings mixed with epoxy is like what we in the US buy as "JB Weld" (which is basically a metal powder impregnated epoxy). Its claim to fame is that it withstands high temperatures better than ordinary epoxy. Why, I do not know - if the epoxy it uses is just standard epoxy, then that portion of the adhesive should degrade at high temperatures regardless of whether the metal is present. maybe they use a special epoxy that is more resistant to high temperatures?
Click to expand...

I believe it has something to do with making the epoxy more thermally conductive and possibly making its coefficient of thermal expansion closer to to that of metal.
 
I would think they would help ... unless you have shrinkage of the handle... which could cause shear as the handle shrinks in either the long direction of the knife, or towards the top/bottom of the handle.
 
Cushing H. said:
I would think they would help ... unless you have shrinkage of the handle... which could cause shear as the handle shrinks in either the long direction of the knife, or towards the top/bottom of the handle.
Click to expand...
Sir, first I would like to express my respect for your work and sharing your knowledge and by no means am I an expert or do I pretend to know better. I am quite new here and only made about 50 knives so far, some failed, luckily most still are in use.
But I seem to disagree with some details of the points you make, and that might be because of not fully understanding the factors involved.
Surfaces should be clean and roughened up, I fully agree. But in my experience glue has also a minimal thickness. CA being very thin, epoxy just thin and PU and polymers a little thicker.
That thickness allows for some 'play' especially with wooden handles Vs steel.
PU is generally more elastic than epoxy which allows it to compensate for the working of wood Vs steel better.
The other thing is that if you have defects, air, dust... whatever in your glue surface it will get less in absolute terms but probably will take up a higher percentage of the volume and thus make it weaker. As an analogy I would take the minimum thickness of concrete that has to be X times the size of the included granite to be strong. Squeeze the concrete out and you are left with mostly defects.
Of course we should avoid dust or air or whatever inclusion but somehow I tend to believe in the oversqueezing myth. But again, maybe I'm wrong.
As for all aspects of making knives, still a lot to learn
 
The issue about thinness really does presuppose that you have enough roughness on each side of the bond to supply presence of adhesive in the areas where those scratches on opposite sides cross each other ... with enough roughness you will not be able to squeeze that out (it is trapped).

the issue about defects is not so much the number of defects,but the worst defect - failure will occur at that worst defect. If you reduce the number of defects (cleaning surface, reducing the amount of adhesive present) you are reducing the likelihood of a single defect being present that is bad enough to fail given the stress applied ... it is just a numbers game.

the point you make about the adhesive layer being a “buffer” between materials that may expand differently is reasonable ... but the problem is that as you increase thickness you increase the probability of having present one of those fatal defects. Pick your poison I guess.

this is an old thread, so I will stop there, less a moderator yells at me. :-). PM me if you want to discuss more...
 
Cushing H. said:
The issue about thinness really does presuppose that you have enough roughness on each side of the bond to supply presence of adhesive in the areas where those scratches on opposite sides cross each other ... with enough roughness you will not be able to squeeze that out (it is trapped).

the issue about defects is not so much the number of defects,but the worst defect - failure will occur at that worst defect. If you reduce the number of defects (cleaning surface, reducing the amount of adhesive present) you are reducing the likelihood of a single defect being present that is bad enough to fail given the stress applied ... it is just a numbers game.

the point you make about the adhesive layer being a “buffer” between materials that may expand differently is reasonable ... but the problem is that as you increase thickness you increase the probability of having present one of those fatal defects. Pick your poison I guess.

this is an old thread, so I will stop there, less a moderator yells at me. :). PM me if you want to discuss more...
Click to expand...
Thanks for the answer. And yes, It is an old thread, and I was actually trying to find answers on a completely different topic but as so often I got lost in all the info available on the forum.
And forgive my arrogance but its the first time I had the feeling my experience/opinion was worth posting.
We all use epoxy, I do to, it works if done properly and I mostly agree with your guidelines but I have been thinking a lot about that flex issue and I don't know if epoxy is the best answer. It sure isn't always outside knifemaking.
But maybe one day that will be another topic.
 
Cushing H. said:
this is an old thread, so I will stop there, less a moderator yells at me. :). PM me if you want to discuss more...
Click to expand...

I think we (the internet community in general) really need to move past the "don't post in an old thread just because it's an old thread" mentality, because it is not rational. When a thread is found by someone doing a an internet or forum search on the thread's topic, and that person has a new perspective, or information to provide that can generate a fresh discussion, it is more sensible for that discussion to take place in the same thread. Why? Because now people who search the topic in the future will be able to benefit from the added perspective/discussion without needing to track down additional threads and get lost in all the info available on the forum. All the information is in one place.

The "bad" form of necro posting is, in my opinion, posting in an old thread to offer nothing of value to the discussion.

And Gorilla Glue is fantastic for knife handles. Never had it fail - except when I ignored the cardinal rules (surface prep) and destruction tested the knife.
 
Maybe if it works it just works. I didn't mean to discredit his advice because it's clearly good and balanced.
I just have had failures with glueing brass on brass with epoxy (clean and roughed up).
I'm just trying to find out the absolute best way to glue up scales.
Probably just need to do more research on glues, especially their behaviour after 10 or more years.
 
Others might have similar questions or observations, so I think your post was fine, as it added to the discussion. It might have been nice to see you pop into the introduce yourself thread first, though. ;)
 
olivier coen said:
Maybe if it works it just works. I didn't mean to discredit his advice because it's clearly good and balanced.
I just have had failures with glueing brass on brass with epoxy (clean and roughed up).
I'm just trying to find out the absolute best way to glue up scales.
Probably just need to do more research on glues, especially their behaviour after 10 or more years.
Click to expand...
If you had failure with a metal to metal bond... was it after grinding? Excess heat will kill epoxy, independent of how you prepped or squeezed the surfaces (don’t ask me how I know... :-(. )
 
olivier coen said:
The other thing is that if you have defects, air, dust... whatever in your glue surface it will get less in absolute terms but probably will take up a higher percentage of the volume and thus make it weaker. As an analogy I would take the minimum thickness of concrete that has to be X times the size of the included granite to be strong. Squeeze the concrete out and you are left with mostly defects.
Click to expand...
Oliver - first, thank you for your question and comments - it is just that we recently had a discussion about posting to old threads - and the jury is still out with regard to whether this forum want to allow or discourage that activity.... In this case we will let the moderators decide.....

On the quote above and your analogy regarding concrete as an example of included defects .... though reasonable, I am not sure if the analogy really translates to the points I was making. First, we need to distinguish between the surface and the bulk of the adhesive. Lack of cleanliness of the surface (grease, or particles) creates a weak spot in the bond that is strictly located in the surface and can cause a bond failure to initiate at that point. for now though, lets assume that we have a perfectly cleaned surface, so "surface defects" are not present.

We are still left with the bulk of the adhesive. When I talk about defects in the bulk of the adhesive, I am talking about bubbles (of various sizes), or dust or other particles (again, of various sizes) that are UNIFORMLY distributed within the adhesive bulk at a certain concentration. Hopefully it is intuitive that if you have a really, really high concentration of defects in the bulk material, the bond will eventually have problems. So you clearly want to keep the bulk of the adhesive as "clean" as possible. But even then - when it comes to bulk defects, it is not the concentration of the defects, it is the ABSOLUTE NUMBER of defects that are present that presents a problem .... because it only takes **one** defect above some threshold to **initiate* a failure in the bulk (that will then propagate through the bulk of the adhesive as a propagating crack, ultimately causing total failure of the bond). I still find the analogy of a metal chain with a "weak link" in the chain very appropriate. It takes one and only one weak link above a threshold for the whole thing to suffer a catastrophic failure. Because the links will always vary from one to another in terms of their strength, the shorter the chain, the fewer the links, and the lower the probability that a catastrophically weak link is included in the chain.

When you have defects (bubbles or particles) uniformly distributed in the adhesive, when you make the adhesive layer thinner, you squeeze out both the adhesive and the potential defects equally, thus reducing the total number of potential defect sites (thus "shortening the chain"). (this is where your analogy to concrete and imbedded granite, I think, does not follow - with that mixture you are not going to be squeezing out concrete and stones in an equal ratio. Also, the fact that you do not want too many stones versus concrete follows the same thinking made above that "too high a concentration of defects is bad" - and so I think is saying exactly the same thing I am saying?

The other thing with concrete (which is admittedly a pretty nerdy point) is that concrete is a "composite material" ... that is a material with two or more materials each of which contribute something to the final behavior of the material that is better than either of the component materials. But there is always a "sweet spot" in terms of the mixture to achieve that final better performance. Add rebar to the mixture and you get even better performance - but rebar just by itself would not have anywhere near the properties of concrete/stones/rebar combined. Another example is fiberglass (which is strands of glass embedded in a resin). If you use just the loose strands of glass, you have just a pile of fibers that will blow away in the wind. If you use just resin, it will be really brittle, and crack under small stresses. The two together have incredible strength and durability. In an adhesive, something that can be called a defect (bubbles or particles) are just plain bad - you want both the smallest concentration and the smallest number of them you can achieve....

But again, all this really does require that you have a sufficiently rough set of surfaces that will hold the adhesive in those grooves, regardless of how hard you squeeze. If you do not have that, and you have two basically flat surfaces pressed against each other, then of course you are going to squeeze the adhesive out - especially in those areas where you have a relative high area opposite a relatively high area on the other side of the bond.

Kind of long winded - but hopefully makes sense??
 
You must log in or register to reply here.
Back
Top