Clock Base - Silly Foam Project

dkemppai

H-M Supporter - Diamond Member
H-M Lifetime Diamond Member
Joined
Dec 20, 2021
Messages
1,048
This project starts back a ways, have a Synchronome master pendulum clock that needs to get hung up. They're fairly well known clocks, and were typically used to drive slave clocks in an office building, banks etc. As with any pendulum clock, they do better on a solid foundation. Solid being massive concrete foundation of a building or similar.

My workshops in the house have a good concrete foundation, but it's all buried behind an insulated stud wall and 1" of blue board. To tie into that, the wall needs an opening, and concrete extended out. So, why not over engineer it? Build a concrete base with legs extending off the wall. The idea is to insulate behind it, and the smaller surface area of the legs will reduce heat loss.

I've got a couple future projects that will be based on concrete poured into a form. So this seems like a good chance to test out the theory of milling a form, and casting concrete into it. A little cad work, an 8 legged base design with good mass and a flat surface were sketched up. The idea being the blue board will only be removed where the feet will contact the wall. The spray foam the rest of the area underneath.

ConcreteBase.jpg

Of course, the CAD is the easy part. The next step was to laminate up 3 layers of 2" blue board. That turned out harder than expected. Solvent based glues melt the foam. Water and alcohol based glues won't dry. Epoxy is too hard, and busts foam while milling. Eventually gorilla glue was settled on. The problem is that needs water to kick it off. So the trick was to wipe the foam board down with a damp rag before applying the glue. Then a little tape and a lot of full toolboxes for ballast to keep is pressed together. (No images)

DidThis.jpg

Cleaning the router table took three days! That was a royal pain! Everything that I didn't want to deal with for the last 3 years landed on it. It's mostly sorted out thrown out now. Also had to raise the gantry to the top position, it's adjustable for stuff like this.

Once that was done, built a frame. Didn't expect the foam would hold up to 150lbs of wet concrete without backing. Form was pretty simple, just a little scrap OSB and 2x4. (See that one stud, that looks twisted on the right side? That bugger has been kicking around the garage for over a decade. Finally flattened a side and put it to use!)

Ready.jpg

Then came the part that shouldn't have been hard. Getting a long reach large diameter ball end mill. Looked on line, $140! No way! Finally decide to mill foam it would be easy enough to just make one. Started with a ball end shape in cad, rotated it 5 degrees in two planes and projected it to a flat surface. Brought that back up, and added some lead in and out for the milling operation.
BallEnd.jpg


Next was taking some 4130 tube, milling off half, welding up the hole, grinding some clearance and TIG welding a few old drill bits onto it. Ground the hard edge back down to flat like a D bit. Mounted it in a collet block, and rotated the flat and the block to match the double 5 degree rotations in cad. Then milled the ball end shape.

Ball-1.jpgBall2.jpgBall3.jpg

A little hand clearance grinding and it cut foam.
TestRun.jpg

Thinking about it, became a little worried about running a home made bit at 8000-10,000 RPM. So epoxied an old carbide shank inside the tube near the collet end. Carbide being stiffer should take up some of the bending load, and stiffen the tube for the collet to grip it. Chucked it up, and tweaked the runout of the bit to be mostly zero against the cheap Chinese spindle and collet. Typed in 8000 RPM and hit go... sounded good.


GonnaBe.jpgMessy.jpgFoam1.jpg

The next Saturday revolved around chasing little blue dust bunnies with a shop vac, emptying the vac, and repeating. The last image was the start of the finishing pass. that was broken up into a waterline finish, and a few other zigzag passes to clean it up.



Foam2.jpg

After that, some quick sanding, then paint, then sand, then paint then... (No in process images, just wanted to get it done! ). To protect the foam and reinforce it, layered most of the outside with 3M packaging tape. Also taped over the bottom holes of the form where the legs end, and painted that too. This would keep the concrete from sticking to the OSB underneath. Quick layer of bondo below the feet on that board to help support the tape. The tape was mostly to prevent the form from wanting to split, and make cleanup and troweling the surface easier. That worked great! No damage to the top edge of the form.

Foam2.jpg

3D Printed a few parts. Washers and spacers to hold some conduit (bolt clearance) and a steel washer, and a form for bolt head relief. The head relief form was printed with 1/4"-20 threads, and a bolt came through the form bottom and bolted up the whole thing.


Foam3.jpg

Mixed up some mud, and poured it. A bit after midnight it was finally set up enough to try to finish the surface. Turned out OK. Left it in the form for another two days.

Foam4.jpg
Then finally popped it out. Came out easy, the 5 degree draft angle did the trick. Pulled most of the bolt clearance forms (Two are stiff, gonna leave them until the mud it set more). Set some cardboard down, 5 mill poly sheet, and set the concrete on that. Wrapped it up, poured a gallon of water on it, and sealed it up. I'll let it sit like that for a few weeks, keeping it wet the whole while. Then let it dry out, and start finishing it up.
 
@dkemppai

Thank You!
I hope you keep sharing the clock build.

Brian
 
Feel free to also post back in the POTD, but this shows us to subscribe and follow this particular build. Thanks! This is awesome!

Sent from my SM-T500 using Tapatalk
 
This next post is a long time in the making.

Picked up a import "Long Throw" polisher/grinder and some cheap import diamond discs. Really surprised how well they worked to polish up the concrete surface. Started with 50 grit and worked down to 400, lots of water, wet dirty pants lead to a baby bottom smooth concrete surface. Plenty good enough for who it's for!
SillyConcrete.jpg

After baking dry again in the hot garage for a few weeks, the underside of the base was coated with a bunch of coats of 'Epoxy' floor paint from walmart. It actually seems like some pretty good stuff. I was surprised how well it sticks, and how hard it is when cured. Next the top was coated/sanded with a few coats of water based poly. That didn't seem to want to dry smooth no matter what I tried. So, after letting it cure, moved to an oil base satin poly. Right now the process is to sand/coat/sand/coat/sand/coat until it's smooth enough (The rough surface from the water base poly goes away).

Here's a picture with wet poly on this afternoon. Just blocking it out with auto body sanding blocks and 220 grit. It might need one or two more coats to be done. Might spray the last few coats, if brushing thinned down poly doesn't leave a nice finish.
NearlyFinished.jpg


Beyond that, I had a buddy help me run an FEA on a 'constrained' 3D model of the part made with concrete material. That constrained model assumes the feet are rigidly attached to something massive. It should mimic this being bolted to something like the massive wall it will get bolted to.

I'm by no means an FEA or NVH guy. But I work with them on a daily basis, and record a lot of data for these types of measurements that is used to verify FEA results. For anyone who doesn't know what FEA (Finite Element Analysis) is or does, it's basically a 3D model 'built' to match a real part. The part can be run through various analysis packages that simulate various properties of the part when subject to loads, strains, or vibration. In this case, I wanted to see what the modes of vibration of this complex part might be. Since this is a clock base, the wish is that the vibration modes would be significantly different in frequency than the clock (1 second beat pendulum).

RestrainedModes.jpg
Well, wish granted.

Although, I hadn't run the FEA before building the base, based on the size and stiffness of the concrete my guess was the modes were not going to be an issue. As long as nothing in the clock actually excites these modes, the base should be pretty 'still' compared to be pendulum. We also spit out some of the modes as animations, to physically see how the vibrations would look.

Tapping the concrete with a nylon hammer unconstrained (Not bolted to a wall) with a spectrum analyzer/microphone nearby indicates the modes are indeed all pretty high. Even without being constrained to the wall, nothing really happens below ~200Hz. This is pretty close to the modes shown in the free body FEA (although not exactly).
UnconstrainedModes.jpg
This is the 'loudest' resonance, as sitting on the sawhorses seen above. (That support may be changing the resonances from the completely free body FEA data).

Here's the animated GIF's for a few of the modes. Note, for all of these actual amplitude is amplified. That is, it's exaggerated by a huge amount so that the movement can be visualized. Hopefully your web browser animates the images properly. (Edit, it looks like if you click on the image, it will select that image and show the animation! This is good to know...)

Mode1-642Hz.gifMode2-662Hz.gifMode3-713Hz.gifMode4-851Hz.gif

Mode5-889Hz.gifMode6-991Hz.gifMode8-1323Hz.gif
 
Back
Top