Making Round Thin-Wall Cylinders 4

[ThomasUtley]

All,

How would you suggest fabricating brass cylinders that are very close to true round? Would likely be starting with brass extruded tube, 1" OD with wall thickness of about 0.070". Finished length approximately 8". Smooth enough to contain compressed air well enough for damping purposes without the use of o-rings or lubricants.

Lathe + reamer? Boring bar only? How to avoid the lathe jaws deform the walls?

Maybe some kind of ball swaging technique like musical instruments?

Thanks,

Tom

 

[Cockroach]

I would turn it, cut the OD first and leave the ID to solid to mitigate deflection during the cutting operation. I would also leave about two inches outside of the jaws as a solid ring.

Then I would do the ID taking heavy cuts, easing up on the depth of cut as you get closer to the finishing ID. The last two cuts would be super light, say 1/256 inch maximum. As you get closer to the thin wall, consider wrapping the OD in surgical tubing to keep the "singing" down from the vibration.

When finished, part the cylinder from the solid ring just right of the jaws.

I would also double cut the fluid, have a decent flow right over the nose of the cutting tool.

Good luck with it, not easy at all.

Regards,
Cockroach

 

[btrueblood]

Even with the precautions Cockroach mentions, I'm not certain you could maintain the tube round after removing whatever restraints it had during cutting.

FWIW, I've found as-drawn seamless brass and copper tubing to be pretty darn round as they come. A quick pass with a brush hone down the i.d. to get the surface finish where I wanted it, and they made lovely low-pressure air cylinders.

 

[ThomasUtley]

Thanks, Cockroach--I like the idea of the surgical tubing to minimize harmonic vibes. Nice.

Btrueblood, can you elaborate a little more on the low pressure air cylinders you made? The whole reason I need the cylinders is to fabricate some air damping cylinders. I'm limited to max 1" OD for the cylinders due to packaging space constraints. However, I need to control motion of an assembly that's being pulled upward by spring tension. Net force (upward force of the springs minus gravity on the mass being lifted) is a max of about 75 lbf over a travel of 6". With two such dampers, I need to contain (translation: slow down to a crawl) roughly 40 lbf per damper. If I'm limited to a 1" OD, and a "thin" wall of, say, 0.065", that gives me a piston face of about 0.59 in^2. 40 lbf/0.59 in ^2 works out to something like 67 psi without including any dynamic transients, which there will be. Call it a design spec of 100 psi maybe?

More detail on what I'm trying to accomplish is provided under the posting entitled "One-Way Linear Escapement?" on this same forum.

Would love to hear about how you made your cylinders and any associated valving that went with them. Especially interested in how you capped the end(s)...soldered? Threaded? Press fits?

Thanks!

Tom

 

[KENAT]

thread404-321039 if it helps.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[berkshire]

Thomas Utley,
Could you not adapt the barrel of a bicyle pump to do what you want?

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[btrueblood]

These were one-way cylinders for a test, no pressure on rod side. Soldered the rod cap end, and had a soldered threaded fitting on the opposite (non-rod) end, this allowed loadin. Piston (acetal) fits were probably .010 on the diameter (i.e., sloppy) with o-ring grooves. Grooves were sized to put about 10% compression into the o-ring section, and the cylinders were used on shop air (90-100 psig). No internal valving, just port on the cylinder ends. They held up surprisingly well, 100k cycles with no leakage.

 

[ThomasUtley]

Berkshire,

I do like the idea of the bike pump, if for no other reason than I can likely salvage pistons and valving. However, for cosmetic appearance reasons, a brass cylinder would be a home run with the client. It's intended to be an heirloom piece and having a "Made in China" powder-coated pump barrel showing is not what he's paying for.

I think with all the great input received on this post for the cylinders and the more general post for the mechanism referenced above, I have what I need to proceed to trial some hardware. I'm definitely going to start with an off-the-shelf ball pump to make sure I'm not off the mark with forces and pressures. If all OK, then I'll move forward to fabbing some nice-looking brass cylinders to complete the system.

Thanks!

And when all that's done, I'll finish the woodworking, too. Ha.

 

[SnTMan]

ThomasUtley, how bout a made in China bike pump with an attractive brass outer sleeve?

Go standard guts if you can.

Regards,

Mike

 

[imcjoek]

How about

http://www.airpot.com/

?

 

[rb1957]

how do they make bicycle pumps ? (i'm they're not machined from bar stock)

probably a deep drawing process, that suits Al.

if Brass doesn't have the ductility for this, how about anodised Al ? whatever colour you like.

how true is the ID of a bicycle pump ??

 

[racookpe1978]

Final dimension of a 0.070 thin wall of a soft material?

I'd take care to make a mandrel of OD = ID of your final tube cylinder, then "draw" the cylinder over the mandrel. Seems that machining the OD of a solid mandrel is going to be more accurate than trying the "make" the hollow cylinder out of solid material.

Then, since you're drawing (slightly expanding the possibly out-of-round shape of the original purchased soft tube material, you can accept normal commercial variations in the original tube.


Go too far of course, and you're stretching the original tube and you will create a "too-thin" wall in the stretched tube. But at a "bicycle pump" pressure of 120 some-odd psig final pressure that won't be significant.

 

[ThomasUtley]

Racookpe1978,

That is an excellent idea. When I mentioned swaging up above, I was thinking of round ball bearings or something similar. Your concept of a cylindrical mandrel seems very reasonable. And it seems like it would also be self-straightening. Bonus!

Thanks,

Tom

 

[btrueblood]

Thin wall tubing is generally drawn the other way, i.e. decreasing o.d., over a mandrel. Going the other way you end up with your tube tightly shrink-fitted to a mandrel, and no good way to remove it; see

http://en.wikipedia.org/wiki/Tube_drawing

Also, drawing will tend to give you poorer surface finish than the material you started with.

Is there no size of off-the-shelf tubing that will work?

 

[btrueblood]

Somebody mentioned airpot cylinders, which are pretty neat little devices. Glass tubing is, or can be, very smooth and round on the i.d. in the as-drawn condition. With a graphite plug for a piston, the friction can be reduced to ridiculously low levels. The challenge is forming/retaining end caps for the cylinders, but there are ways.

 

[Pud]

.029 wall thickness brass tubing ?:

http://www.specialshapes.com/round029.html

http://www.tynevalleyplastics.co.uk

Why be happy when you can be normal?

 

[btrueblood]

Aluminum tubing on Pud's page is another option - extruded Al tubing also generally is pretty round. And you can have it anodized and dyed in a bewildering array of colors...

 

[ThomasUtley]

I was thinking of using this product from onlinemetals.com:

http://www.onlinemetals.com/merchant.cfm?pid=1558&step=4&showunits=inches&id=84&top_cat=79

1" OD, 0.065" wall, 0.870 ID.

As for the concept of drawing over a mandrel, seems as though, if necessary to do, shouldn't be all that hard if my mandrel isn't very much oversize. Say, a few tenths with a lot of lube and a rigid post over which the mandrel is mounted. Push the other end of the tube down over it using some kind of plug that spreads the open end rather than collapsing it.

When the tubing arrives from the vendor, we'll see if it's even necessary. Might be just round enough for what I'm trying to accomplish. After all, I'm not trying to stop movement, just slow it.

Thanks,

Tom

 

[btrueblood]

" Might be just round enough for what I'm trying to accomplish. After all, I'm not trying to stop movement, just slow it."

There ya go.

 

[hydtools]

Maybe find a grandfather clock weight shell that's close enough.

http://new.blackforestimports.com/AdvancedWebPage.aspx?cg=415&cd=4&SKUTYPE=202&SKUFLD=SKU&WEBID=2225

Ted