Warped Scales

So many ways things can foul up, huh?! Even when you do everything right. Reminds me of the action of a bi-metallic strip, as in the old fan controls on furnaces, or in house thermostats, or even even the old flat coils in the automatic choke on your carbureted car (anyone remember those?)

They relied on the different expansion rates of the two metals to cause a curving motion. So anyway, it seems that either the G-10 expanded or the wooden scales shrank. But, why? i would have expected the opposite- for maybe the wood (even though its stabilzed) to absorb some solvent from the epoxy and then expand, but if anything it shrank (or the G-10 expanded).

I've never had that happen. But I too, glue up the liners, scales, and tang all at once. But, if that's what's keeping mine straight, then why didn't the angle iron you clamped to keep yours straight. This is smoking my pea-brain.
 
Bill Hubbell said:
So many ways things can foul up, huh?! Even when you do everything right. Reminds me of the action of a bi-metallic strip, as in the old fan controls on furnaces, or in house thermostats, or even even the old flat coils in the automatic choke on your carbureted car (anyone remember those?)

They relied on the different expansion rates of the two metals to cause a curving motion. So anyway, it seems that either the G-10 expanded or the wooden scales shrank. But, why? i would have expected the opposite- for maybe the wood (even though its stabilzed) to absorb some solvent from the epoxy and then expand, but if anything it shrank (or the G-10 expanded).

I've never had that happen. But I too, glue up the liners, scales, and tang all at once. But, if that's what's keeping mine straight, then why didn't the angle iron you clamped to keep yours straight. This is smoking my pea-brain.
Click to expand...

Great post, BIll. I'm at a loss, too. Especially since I just had a G10 scale / wood combo do exactly the same thing. Here's the interesting part: It was flat when I removed the clamps. I did the glue up with it clamped to my workbench top. It curved over the next couple of days. Riddle me that one, Batman!
 
I'm going to give my theory on this. After much reading on this, I think the reason is shrink during the cure. My googling has revealed that epoxy, especially unfilled epoxy, can shrink 2%-7% in volume during curing. This is also dependent on the cure rate which is in turn dependent on temperature. There are LOTS of variations of this curing cycle that can be "tuned" to an application. That may be critical in mitigating this effect in the future.

Anyway, my theory. G10 is much less porous than wood, even stabilized wood. The more open space at the surface equals more epoxy inclusion. The more epoxy there is "soaked" into the surface, the more measurable shrinkage there will be. The effect will be tension between the two parts, much like the bi-metallic spring @Bill Hubbell described above.
 
Hey Bill.
Yep many ways to foul up and I'm doing my darnedest to check them off as I go.
I remember the Bi-metallic coils and all the carb adjustments just to keep the engine from flooding. Preferred my pull choke !
From what West Systems chem engineer said, there are no solvents to migrate separately into the substrates.
Also told me that G10 has a release agent in its mix and so there's the need to scuff and clean well.
I think it is possible that the combination of things, the denatured alcohol, the exothermic heat (expanding the scales resin?)and perhaps the uneven clamping with the small spring clamps (3 per scale) combined to give me my results.
I believe I'll be able to sand these scales flat and have a thin line of yellow still. Maybe add another color afterward.
People still like pinstripes.
 
SS369 said:
Thank you Mark for your reply. I appreciate every little tidbit along this journey.
What you have written makes sense and I understand it. But, in this case I only hand sanded the materials to roughen them as they were acceptably flat.
In the past I have glued up sets to choose from, before completing the blades. This time I ran into a mystery.
I actually called West Systems and spoke with them to see what they could offer in the way of possible explanations.
They were great to talk with, very helpful and informative, but perplex as to the actual cause.
I went over my complete procedure with the chemical engineer and he pointed out a few things I could do differently. Shared some things about G10 I did not know. He also called me back later to tell me he had run it by some coworkers and got their input.
Long story short, he suggested to use isopropyl alcohol, not denatured, and use a caul to put between my clamps and the scales to spread the clamping force.
Now all I have to do is salvage these...
May choose to do liners to scales to blade in the future.
Click to expand...
Well, it sounds like you are getting to the root of the problem. I had assumed you were sanding under power. I see that mistake often in my new students and apprentices. You have an analytical mind and you will do good.

I'm sure the all-at-once construction will solve the problem.

Good luck but most of all, keep it fun.
 
tkroenlein, that was an interesting read, thank you.
Another thought came to mind while reading. The angle iron that I clamped to was most assuredly colder than anything else and perhaps that effected the epoxy crosslinking on the G10 side of the film. Drawing the exothermic heat out more so on that side.
I'm thinking I'll change my procedure a bit from now on . ;-)
 
tkroenlein said:
I'm going to give my theory on this. After much reading on this, I think the reason is shrink during the cure. My googling has revealed that epoxy, especially unfilled epoxy, can shrink 2%-7% in volume during curing. This is also dependent on the cure rate which is in turn dependent on temperature. There are LOTS of variations of this curing cycle that can be "tuned" to an application. That may be critical in mitigating this effect in the future.

Anyway, my theory. G10 is much less porous than wood, even stabilized wood. The more open space at the surface equals more epoxy inclusion. The more epoxy there is "soaked" into the surface, the more measurable shrinkage there will be. The effect will be tension between the two parts, much like the bi-metallic spring @Bill Hubbell described above.
Click to expand...

Now this makes a ton of sense to me. I don't get the curling when I use a fiber liner as a backer for a two-part scale. I also know the fiber liner "wets out" and takes up a fair amount of epoxy just like the scale does. There probably is equal shrinkage and therefore equalized tension, whereas just as you said, the G10 takes up zero and then you have a definite stress variance. Whether there's more to it or not, your explanation is perfectly consistent with what I have experienced.
 
Well its not just me then :) ….I had this problem only this week.

I stabilised some Elm burl and G flexed G10 scales to them. Three days later warped, they were all super flat before gluing.

My thoughts were the same as above with regard the bi-metallic strips as G10 and wood are dissimilar materials.

From experience I have found if the scales are warping at the ends ie outwards, you will have more chance of them failing. If the scales are bowed so the ends are flexing inwards ie a gap in the middle and the ends pushing onto the blade then you will have a much better chance of them being ok.

I also use tinted glues on every knife as a belt n brace approach. If there is a micro gap I use Zap it glue and clamp as that stuff fill gaps by capillary actions superbly.
 
tkroenlein said:
This was an interesting and brief page I found whilst googling.
Click to expand...

Just reading this and reckon the PID oven I built might be worth using for glue ups in future as depending on the temp/oven loading I am getting 1-3 degrees accuracy. I am wondering if the epoxy companies will have information on the best cure temps in such a situation and I could 'ramp' the cure times. It would possibly mean the problem of bowing would be resolved.
 
Sent a question out to an epoxy manufacture and here is their reply regarding warping of thin wooden material, which if I am correct mirrors the previous statements.

Whenever you add a liquid to a bare timber surface the timbers nature is to wick up that moisture this will have the effect of swelling the wood fibres. Therefore if you have a thin piece of timber that it epoxy coated on one side the swelling of the wood fibres may be enough to expand that surface causing the bowing. This would be exaggerated for timber with a very, very low moisture content.

Also for bonding applications the epoxy should be allowed to cure to a solid state prior to post curing particularly when bonding dissimilar materials.

We mean that when you make up the bonds they should be left to gel at room temperature before being heated.
 
Which epoxy manufacturer?
By chance, did you inquire about stabilized materials?
 
Interesting...I haven't done much with stabilized woods, but thought that after stabilizing, they wouldn't soak up much of anything. But, I have been wrong before... at least once, maybe twice. :-0
 
SS369 said:
Which epoxy manufacturer?
By chance, did you inquire about stabilized materials?
Click to expand...


I knew someone would ask ;) ...The info came from Gougeon Brothers who are GFlex agents in the UK.

Yes I told them in my email the woods are stabilised and even included a link to this thread.:)
 
Glad you inquired, thanks. The more info the better.

I fully understand the movement/warping regarding unstabilized wood, but stabilized wood, well, it is stabilized.
Perhaps when roughing the surface the material takes on the attributes of raw wood and the epoxy can migrate into the surface and cause some expansion.
Just a guess...
 
Bill Hubbell said:
So many ways things can foul up, huh?! Even when you do everything right. Reminds me of the action of a bi-metallic strip, as in the old fan controls on furnaces, or in house thermostats, or even even the old flat coils in the automatic choke on your carbureted car (anyone remember those?)

They relied on the different expansion rates of the two metals to cause a curving motion. So anyway, it seems that either the G-10 expanded or the wooden scales shrank. But, why? i would have expected the opposite- for maybe the wood (even though its stabilzed) to absorb some solvent from the epoxy and then expand, but if anything it shrank (or the G-10 expanded).

I've never had that happen. But I too, glue up the liners, scales, and tang all at once. But, if that's what's keeping mine straight, then why didn't the angle iron you clamped to keep yours straight. This is smoking my pea-brain.
Click to expand...
I used to have that problem too. It happend with stabilised wood, all kinds of plastics, etc,...
I tried everything to fix the scales, they just warp again. I made sure to keep everything cold during the scale prep.
This summer I noticed that my scales didn't warp at all,...
Now I keep my raw materials inside the house where it is warm and I heat my workshop to normal room temperature before I start working on the scales. I make sure my scales don't drop below room temperature before they are fixed to the blade,... no more warping. I spoke to some of my suppliers and they all work at room temp conditions. When I cut a block in scales in a 10°C workshop they warp.
 
Wondering if your introducing some tension between the wood and g10.. maybe when clamping?
Easiest solution would probably be just gluing directly to tang, either g10 then wood, or all at once. Then the tang itself will be stabilizing everything.
 
I usually layer my g10 and scales between two pieces of bar steel and then clamp the steel bars together . I then feel the steel can’t flex when everything is drying. I don’t know if it helps. Or use my bench vise which is about 4 inches long. I clamp the bottom of the scales with the vise then put clamps on the top.
 
You must log in or register to reply here.
Back
Top