epoxy

[Lorien]

Yep more surface area for the glue top adhere to.

I get that, but do you think there is a greater risk of air pockets, particularly if a pin goes through the reservoir?

 

[Stacy E. Apelt - Bladesmith]

Pins and Corby bolts won't cause any problems. You won't get an "air pocket. The hollow grind is very shallow. I just lightly run a 80 grit belt down the center of the tang. On most wooden handles I don't bother to do the same on the handle. On Micarta I do.

The main reason for a glue reservoir is it prevents a glue starved joint. If the two surfaces are truly parallel and smooth there will be very little epoxy between them once clamped.

The layer of epoxy needs to be a certain thickness to cure and form a strong bond. For gluing smooth flat wood to smooth flat metal that layer thickness is .006".
Part of this thickness requirement is a physical strength issue and part is a chemical issue of the ability to catalyze and polymerize in curing. A very thin layer may not allow a complete reaction.

Think of the layer in 3-D. If the space between the tang and handle has plenty of X and Y direction and almost no Z direction, it will polymerize along the X/Y plane. The direction you want it to be strongest is the Z plane between the tang and handle. Thus, in a tightly clamped situation, you get a very weak joint that is effectively just a .001" spacer between the tang and scale. Allowing five or six thousandths of an inch of Z axis is enough to get the polymer chains to go from tang to scale and form a strong bond in the direction most important.

 

[Lorien]

I get your point that spec calls for .006", ( I didn't look that up, but seems like it's in the ball park and you'd know better than me) and you say that thinner than that is too thin.

Do you know how thick the epoxy layer is within the reservoir?

To me, it seems like removing material for the hollow section and replacing it with epoxy is what's happening, and that's putting a lot of trust in the epoxy.

I don't necessarily think that's a bad thing- modern epoxies are incredibly strong. They hold satellites, spaceships, military vehicles, ships and submarines together.

The point here that stands out to me is that introducing this hollow geometry into the bonding area presents at least some potential for risk, ie; air pockets, too thick an epoxy layer etc. and why risk it when it's known that the additional surface area and interlocking and undercut structures from low grit abrading is enough to hold two flat planes together?

Either way, being careful is ultimately what leads to a strong bong, so it's probably splitting hairs trying to figure out the real world difference. Still, I'd need to see data to convince me that, all things being equal, hollowing out a portion of the bonding area is better.

For a tapered tang, which is also hollow ground, it makes sense for a number of additional reasons to use fasteners, and I think you could argue that fasteners in that case negate some if not all the concerns that I brought up for a pinned type construction.

 

[Natlek]

Can someone find some date/proof where OUTSIDE of knife world in ANY other application where epoxy was used they make *glue reservoir*

 

[Natlek]

Epoxy hold just fine on most wood and other material we use .Weak link is always steel ! Glue two piece of wood or G10 or Micarta and try to separate them ! Good luck with that ! Glue two piece of steel and try to separate them .......Easy task !! So only thing you can do to strengthen that connection wood/steel is to add mechanical connection ........................ and that are scratches IN steel !!!! what holds Teflon to the pan ?

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

My previous posts on another thread led to the conclusion that I should consolidate this discussion into a single thread. I have had time to type out more on this subject - and so am posting this. I am also re-posting the earlier material so that it is all in one place eventually. Apologies...

www.bladeforums.com

 

[Natlek]

polish steel to a mirror finish and glue them with epoxy and see how they will hold !

 

[Richard338]

A 2021 study seems pretty relevant.
https://www.sciencedirect.com/science/article/abs/pii/S0300944021000059
The link is an excerpt, but you can click on PDF to access a full open access version.
They tested the toughness of bonding between 2 pieces of mild steel, prepared with 4 different levels of surface roughness (400 grit, 60 grit, 36 grit, and 20 grit). For each of the 4 roughnesses, they tested 3 different thicknesses of epoxy (1mm, 0.5mm, and 0.25mm). 0.25mm is about 0.01".
For a given thickness of epoxy, the bond was improved by roughening the surface. For a given roughness, the bond improves strongly as the epoxy layer gets thinner. Unfortunately they didn't test even thinner layers, but the data are still rising steeply at 0.01" (0.25mm).

My takeaway is to prepare the surfaces flat and sanded with 60 or coarser grit, and apply a thin layer of epoxy.

 

[Lorien]

Thanks for posting that, Richard, it's a pretty stark data set

I mix my epoxy and apply it with a palette knife. Every surface gets epoxy and I apply it under a magnifier to make sure there are no bare spots.

 

[AVigil]

Thanks for posting that, Richard, it's a pretty stark data set

I mix my epoxy and apply it with a palette knife. Every surface gets epoxy and I apply it under a magnifier to make sure there are no bare spots.

If you are using mechanical fasteners, it would not really matter if you sanded the surface or not. The epoxy is simply being used as a barrier from the elements.

That being said the properly prepped, sanded, surface will always make a stronger bond then a glass smooth surface.

 

[Stacy E. Apelt - Bladesmith]

Note that the minimum I am talking about is half that test data's thinnest test. I agree that there is a sweet spot between .005-.010" where the bond is strongest. My point was you don't want to go below that point in thickness.

I'll recap my statements on installing scales in case there is a translation difficulty:
The rougher the surface the better the bond.
The "reservoir" is only a minor additional thickness to get a sufficient bond. We aren't talking about a lake, just a shallow puddle.
The glue line should not be less than .005" thick.
With proper sanding at low grits and minimal clamping pressure the reservoir is probably unnecessary, but won't hurt on a properly installed knife handle.
Mechanical fasteners should always be used on knife scales.
Corby bolts are the strongest mechanical fastener for knife scales. Once you learn how to install them properly and drill the counterbore seats they are also very easy to install.
The best epoxy for scales is slow cure type resin. 24-hour cure is what I recommend.
Clean everything well and wipe off with alcohol.
Alcohol is the solvent for un-cured epoxies. Acetone is the solvent for cured epoxies.
All glued surfaces should be reasonably flat and sanded to no higher than 120 grit. A lower grit is even better.
DO NOT clamp hard. This can squeeze out most or all the resin from the joint, creating a "glue Starved" joint.
Kant-twist clamps are one of the best for clamping scales. Other choices are the weakest spring clamps you can find. Screw clamps are not a good choice.
With Corby bolts you don't use clamps at all. Just tighten to snug, wipe off the squeeze out, snug again a tad, clean up, let cure. Again, don't torque them down hard, just comfortably snug. Another plus with Corby bolts is the scales can be put on and taken off while fitting the scales to the tang. Clamping the scales together with the bolts will also guarantee that the scales will be identical at the ricasso when finishing that area prior to final assembly.

That should provide the basic info I was trying to convey.

I'm going to start dropping out of these discussions when Natlek starts in. I have better use of my time than replying to a troll. Natlek can post whatever he wants, I'm not going to respond. You folks deal with him.
However, as a moderator, if any thread takes a downward turn or becomes argumentative, I or one of the other mods will step in.

 

[pafka]

A 2021 study seems pretty relevant.
https://www.sciencedirect.com/science/article/abs/pii/S0300944021000059
The link is an excerpt, but you can click on PDF to access a full open access version.
They tested the toughness of bonding between 2 pieces of mild steel, prepared with 4 different levels of surface roughness (400 grit, 60 grit, 36 grit, and 20 grit). For each of the 4 roughnesses, they tested 3 different thicknesses of epoxy (1mm, 0.5mm, and 0.25mm). 0.25mm is about 0.01".
For a given thickness of epoxy, the bond was improved by roughening the surface. For a given roughness, the bond improves strongly as the epoxy layer gets thinner. Unfortunately they didn't test even thinner layers, but the data are still rising steeply at 0.01" (0.25mm).
View attachment 2462347
My takeaway is to prepare the surfaces flat and sanded with 60 or coarser grit, and apply a thin layer of epoxy.

this is obviously very interesting paper.
I have read some of it because I was interested in the testing procedure.
I remember my dad saying to me that epoxy has a good strength in tension, but does not cope well with shear. I am not saying it is true, it's just a background.
I looked in the paper and they say they use "single lap shear (SLS) test". Which appears to be a test where they pull the joined pieces "along" the plane in which the adhesive was applied. Ok.
But for knifemakers, would even a simple hidden pin (i.e. a mechanical connection) increase strength in this type of load significantly? I suppose yes.
Then the next question is whether the glue strength is impacted by the surface roughness and thickness similarly in pure tension load (i.e. pulling the pieces apart perpendicularly to the plane in which the adhesive was applied).
Do you (by you I mean anyone) know if there is a difference between shear strength and tension strength in these modern epoxies? Just thinking out loud.

 

[Cushing H.]

I'd love to hear your results from that test, please!! And, if it comes out good stuff, please post a photo of the can. Thanks and good luck with it. Hope it's clean.
Brake cleaner is a good choice, too!

I would also like to see the results from this. It would also be my guess that chemicals bought at the hardware store have really minimal purity controls in place, and will likely leave crud on the surface.

 

[Lorien]

it seems to me that fasteners, pins or otherwise, provide a level of shearing resistance many times greater than any bonding agent.

 

[Cushing H.]

this is obviously very interesting paper.
I have read some of it because I was interested in the testing procedure.
I remember my dad saying to me that epoxy has a good strength in tension, but does not cope well with shear. I am not saying it is true, it's just a background.
I looked in the paper and they say they use "single lap shear (SLS) test". Which appears to be a test where they pull the joined pieces "along" the plane in which the adhesive was applied. Ok.
But for knifemakers, would even a simple hidden pin (i.e. a mechanical connection) increase strength in this type of load significantly? I suppose yes.
Then the next question is whether the glue strength is impacted by the surface roughness and thickness similarly in pure tension load (i.e. pulling the pieces apart perpendicularly to the plane in which the adhesive was applied).
Do you (by you I mean anyone) know if there is a difference between shear strength and tension strength in these modern epoxies? Just thinking out loud.

This (thinner is better) is the result I would expect (read through my earlier thread on adhesion for the reasoning … focusing on the “weak chain link” analogy. *however* I do *not* discount stacy’s “epoxy reservoir” recommendation. Obviously experience says it helps… right now I can not reconcile the two perspectives.

the greater strength in tension is a physical thing (edit - ie geometry), not chemical (again, read through that adhesion thread) … I would be surprised if current epoxies are any different (other than increased absolute strengths). Pins will definitely help protect against shear forces. Hidden pins would protect against those shear forces without being visible … and if well roughened would also provide some additional strength against tensile forces (ie forces that would tend to lift the scales away from the tang)

 

[Natlek]

it seems to me that fasteners, pins or otherwise, provide a level of shearing resistance many times greater than any bonding agent.

If we drill holes in scale for pins and then we insert pins/tight fit/ in scales ...no glue ,ONLY /safe/ way to disassemble scale is to push pins out . Attempting to disassemble scale from the ends or from any other direction can only break the wood, we will not disassemble the scales . We can do it, but it will take a long time, raise a little one end and then a little on the other side and so on ..Personally, I don't see the difference between a normal pin, penned pin or fasteners. Properly done, they will all do the job for which they are there .

 

[Crag the Brewer]

In my opinion from what I've seen, epoxy is Only a seal between the scales and the steel. A barrier to help keep moisture out. Nothing more.

Why would you risk selling a knife that Will fail eventually. If it's a decorative knife only, that's fine, I guess..... But a real knife that is going to be Used NEEDS mechanical fasteners too. Straight pins/tubing is Not good enough. (My opinion)

This is not my knife, but another well known maker.....


We could probably have an entire post of just failed handle materials....

 

[Britt_Askew]

Testing things in your shop, to see what works for the way you do things is a good idea I think. When I started I would only drill holes for the pins and the rest of the blade was clean and smooth. When I did destructive testing using a hammer I was able to knock the scales off the blade. Now those first ones I used wood screw clamps and clamped down pretty hard on it and it was possible to break the bond with a hammer. In my mind I attributed that failure to glue squeeze out and changed the way I did things. I now drill holes and grind a slight hollow in all my blades to make squeezing all the epoxy out impossible ( and use alot lighter clamping pressure ) and my knife handles will hold when beating them with a hammer. Is it absolutely necessary to grind the hollow?? not sure but I do know that when I do I can beat it with a 20oz hammer without failure and in the end thats whats important to me.

 

[Britt_Askew]

But a real knife that is going to be Used NEEDS mechanical fasteners too. Straight pins/tubing is Not good enough. (My opinion)

( respectfully ) We can agree to disagree , IMO pins and good epoxy will make a trouble free knife handle.

 

[Lorien]

different methods for different design parameters is my credo. Every method is good, no single method is the best for every set of specifications

 

[Richard338]

This (thinner is better) is the result I would expect (read through my earlier thread on adhesion for the reasoning … focusing on the “weak chain link” analogy. *however* I do *not* discount stacy’s “epoxy reservoir” recommendation. Obviously experience says it helps… right now I can not reconcile the two perspectives.

the greater strength in tension is a physical thing (edit - ie geometry), not chemical (again, read through that adhesion thread) … I would be surprised if current epoxies are any different (other than increased absolute strengths). Pins will definitely help protect against shear forces. Hidden pins would protect against those shear forces without being visible … and if well roughened would also provide some additional strength against tensile forces (ie forces that would tend to lift the scales away from the tang)

The issue for us in our workshops is that we do not enforce a standoff distance to set the layer thickness as they do in the study. We simply adjust how tightly we clamp and hope that we don't squeeze out all the glue.