OK guys, we need to be very careful here. Getting very close to the ad hominem arguments, and taking things too personally. Why doesn’t everyone take a break for the evening, and come back after a night’s rest and some time to chill. I’d hate to have to find someone with the ability to lock the thread…
1 Like
Sure. Why not?
I apologize if I didn’t write clearly enough. Questioning is what we scientists do. However, something important is lost in translation if we’re not justifying our questioning with thoughts on where the prior theory was weak. That Asimov link from @skidd is a great read because it highlights that questioning is more about refining past understanding than broad rejections of prior art.
To your specific question about whether anyone had done wood in steel on a MPCNC, I think you’re right that it hasn’t been done. However, nautical architects have been using wood as a core for a composite construction for almost a century, so you’re following in good footsteps. You’ll find many modern boats which are made similarly to what you propose, only they use wood (or urethane foam) in a fiberglass or carbon fiber tube instead of steel. It’s surprising how awesomely useful wood is. I’ve been very happy with how the custom booms and bowsprits I’ve made have turned out.
The challenge for your application is that steel and wood aren’t a particularly complementary fit. Wood provides crush and impact resistance, which is essential for fiberglass and carbon fiber. However, steel outperforms wood on this front. Perhaps a novel approach is a super thin-walled section of steel where the wood keeps it from deforming under the bearing load, while the steel provides a hard wearing surface. But I think you might have more mileage out of pour foam, though, since it’s a lot easier to apply in this circumstance and it generally has similar performance to wood when serving as a composite core.
I was actually thinking about that spray insulation foam as a binder for an insert. I just don’t know how well it would bond to the steel…
Degrease maybe a wire brush and it will stick but multiple compressions would break it free but also destroy the tube
So basically gorilla glue.
1 Like
Sure, but that stuff really only works well on porous materials… i.e. wood.
You know… everything in this thread could be simulated in Fusion360 in about 10 minutes. But then I guess we’d be relying on science and math and assuming a calculation is correct.
3 Likes
It sticks to metal pretty good. Especially in the amounts this would use.
I’ve never had good luck with it long term. But maybe it was my application.
I wouldn’t rely on it to hold me in the air, but the inside of the emt should be rough enough. We’re also not bending these things to failure.
3 Likes
The biggest bottleneck with stiffness on the MPCNC, in my experience at least, has been the belts and cable ties (and this has nothing to do with tension, belts are inherently designed to flex a little bit). My machine is relatively small and uses stainless tubing, so other people may have different experiences. But all this talk about conduit stiffness is a bit of a red herring to me.
The only exception to this might be something that improves the dynamic properties of the machine i.e. the dynamic stiffness and vibration damping. But I feel like the 3D printed parts do a fair job of this already and I’m not sure any of the ‘improvements’ that people offer would do much for this (I would be happy to be proven wrong with numbers).
What’s the saying? ‘Premature optimization is the root of all evil.’
2 Likes
If the belts are tensioned correctly they shouldn’t flex too much. One side tight against the corner the other side has the loop, with a second tie holding the loop flat.
1 Like
Are you a software engineer? I thought this was a software saying. I interpret it as, you need to know where the biggest problem is before you go fixing stuff. If you need to save money, you need to make a budget to decide if you need to give fewer gifts or cut out trips to the coffee shop.
1 Like
Barry, they don’t flex a lot really, it’s hardly noticeable (and yes one side is tight against the corner – I did it the wrong way at first) I’m just pointing out that, for me at least, the flex in the belts seems to be much more than any flex in the rails.
1 Like
Man, Christmas is going to be sparse this year!
I was trying to think of larger items, but rarer vs. a smaller, more frequent expense.
Conduit or belts, wood or concrete 
, one good aspect of this is the attitude toward measurement as opposed to assumption. That’s essential. Disregarding and perhaps slightly disparaging the expertise of others is not nice, but the orientation toward seeking ground truth is still a good thing.
4 Likes
How did this turn out?
2 Likes
TLDR: I am guessing best case you gained 10% stiffness at a 15% running weight increase. I don’t think running weight will be much more than the spindle and carriage weight so negligible. I wouldn’t recommend it since that is a lot of work to get 10%. I would just buy thicker conduit. Also sorry to 1 month necro but again waiting on parts to print and looking for what people said about the Primo and saw this was brought up to the front and I did not see any maths which made me sad face… see excel in zip.
Full:
Waiting for my Primo parts to print and I saw this thread and thought i could give a quick tip for calculating this yourself since I didn’t see anyone put up the maths. I am assuming you have perfect adhesion between your core and shell which is another can of worms. I also used a very stiff pine value and a light weight just to be as generous as possible for the effects. Don’t use concrete. ^^
The stiffness value is the relationship of the EI.
I, which as someone mentioned up above I is the second moment of inertia and is related to the stress fiber equation of Mc/I. (List of second moments of area - Wikipedia)
E, is the Modulus of elasticity or the stretch-i-ness or stiffness of a material. For E a high value is steel and a low value is rubber.
If you modify the stress max fiber equation to Mc/EI that is the strain in the fiber direction under a bending load. The strain is a useful property because it relates to how much something stretched under load vs its original length. Since these rods are used as a carriage holder they feel a bending load from the trucks on the rail similar to cars on a bridge so EI is an appropriate way to approximate total stiffness of your structure. (If it was a strut you would use EA [area stiffness])
Most of the time you add a core to stabilize the skins or shell of your structure because they are so thin they buckle without out of plane stiffness. Not because the inside is a good place to stuff more stiffness. The opposite is true better stiffness comes from adding thickness to the outside, better to add more steel further away from the center. Why the mechanic will just start welding bar-stock to the outside of a pipe if it starts sagging, I saw you in this thread crazy mechanic. 
Tried attaching excel I made to hide it in zip file to show maths. Should be virus and error free but you get what you pay for… no warranties expressed or implied.
MPCNC_CoreAdd.zip (10.3 KB)
Cheers,
Lucid
2 Likes
A fin of bar stock off the bottom of the rails, between the bearings, would probably be the best bang for the buck, then. You lose the ability to rotate your rails, but you get perfect adhesion along the length of the stabilizer (full access to the weld joint), and a vertical fin will give you the best resistance to flexing under load. Which is all and good for the outside rails, but you might have to settle for more of a short square bar on the gantry rails (more like a key than a stabilizer), as you have far less open space going through the core. I don’t know if the gains (stiffness) outweigh the costs (weight) at that point. I suppose, if you have the time and energy for it, you could slice a gantry rail along one side, and weld a fin into it, so that the bulk, if not the entirety, is inside the rail. Of course, maintaining a circular rail while slicing it and welding it back together might be… less simple than it sounds. 