pneumatic tolerances

[roxcreek]

hey everyone,

I'm designing a valve that can be made with a lathe and drill press, but I do not know how thick I need to make the walls of the areas that are under pressure in order to make them safe for holding.... let's say 300 psi?

[xero28]

Well, consider say, a Palmer's Rock low pressure regulator, or most any other low pressure regulators for that matter. They are frequently used with PneuMag set-ups and they are not regulated down to 200-400 psi like on a cocker, but they are taking the full 800-900 psi from the high pressure tank. I'm not sure of the exact body thickness, but the walls of the regs can't be more than 1/16" or so.

[roxcreek]

yeah, now that you mention that, the microrock has a max input if like what? 1000 psi?

[Frizzle Fry]

yeah, now that you mention that, the microrock has a max input if like what? 1000 psi?

1600psi max input... That said, 1600 isn't ideal.

[roxcreek]

what kind of metal do you suppose Palmer uses for his regs?

[Frizzle Fry]

what kind of metal do you suppose Palmer uses for his regs?

Brass for the old ones, 7075 aluminum alloy for the newer ones.

[Pneumagger]

It all depends on your dimensoins and the yield strength of the material you are using.
Try to design to hold at least 2X more pressure than you ever expect to encounter.
The approximate stresses developed around p[resurized chambers is...

Hoop stress:


Axial Stress:

[roxcreek]

I really appreciate those formulas, I'll have to bring them to one of my physics teachers to figure out the variables. Where should I go to find the data of the material I am going to be using to enter into the formula?

[roxcreek]

here another question too:

when it comes to the actual construction of the parts, what kinds of restriction should I be placing upon myself when drilling the actualy holes? As in, what are the largest sizes for a drill bit that I could use on metal if I were to just be using a drill press?

[Pneumagger]

Depends on desired surface finish and the power of the drill, honestly.
Given a good small drill press and assuming you are center punching or center drilling to start the holes...

For aluminum the largest single hole I'd drill would be with a 3/8" jobber and for steel a 1/4" jobber. From there you can step up the diameter by 1/4" or 1/8", respectively, until you reach the desired diameter. I would not just drill a 7/8" hole in aluminum if that was my desired end-result... A smaller press probably would not handle that and the hole would not be too clean. I would start around a 1/4" bit and then use a 1/2" >> 3/4" >> 7/8". Make sure you choose the correct bits with proper coatings (or cutting lubricant) for your material.

Do you have a sketch of what you are trying to do? Perhaps that would help people's suggestions?

-----

Hoop stress (sigma sub-theta) equals pressue times the inside tube radius divided by the tube wall thickness
Hoop stress is what the strength of a tube needs to hold around it's curcumference to avoid splitting open like a hot dog.

Axial stress (sigma sub-z) equals the pressure times the inside tube diameter divided by 4X the tube wall thickness
If your pressure chamber is tubelike and is enclosed, this is the strength of the tube needed to keep the tube from pulling apart lengthwise.

The stress you calculate must be less than your selected material's "yield strength". If it is not, increse the wall thickness and/or make the diameter smaller... or choose a stronger material. It is considered best practice to choose a material at least twice as stong as your stress requires. Just google your material to find its yield strength.

[roxcreek]

Do you have a sketch of what you are trying to do? Perhaps that would help people's suggestions?

I don't know what is wrong, but when I try to upload the pdf file it says that there are erors. Any idea what's up with that?

[athomas]

Unfortunately, you can't post files to the AO site. You need to post the pdf to another site and then reference it from here.

[roxcreek]

Alright, I'll have it up tomorrow

[roxcreek]

Here is my first draft design

and, Here is my second draft version, without the rest of the mechanism (the DC and the rammer housing)

I think that for now I should just try to get the second draft sketch done with a lathe, and then try to add on the DC and the Rammer later. I've already got the materials and the lathe ready to go.

I can re-dimension it or email copies of the design to those who want to review it.

Another question: Does anyone know where to get springs made at custom lengths and strengths?

[Pneumagger]

Please tell me that's in centimeters or something.
That's pretty big. Are you making a potato cannon?

Google for spring suppliers. You would be surprised at the kind of springs that already exist.
McMaster is ALWAYS a good place to start for ANY project you encounter in life.

----------------------------------------

EDIT: (I can't resist)

Try to avoid using wording or process specification in callouts unless unavoidable. Remove the word "threading". Whoever is making that "1/8 NPT" callout in the shop should know what drill and tap depth will be for any NPT threaded feature. They should have a go/nogo gauge for that process. You may want to specify a small chamfer @ the entrance to threaded holes. About those holes...

Don't call out holes for screw sizes. Pick your screw and then pick a loose or tight fit hole. Google "tap drill chart"... lookup through hole sizes for various thread sizes. Go with "loose fit" and a washer. Loose fit will be more forgiving tolerance-wise since you are using a drill press. Which bring us to...

Geometric tolerances... where are they?
At least use dimensional tolerances or include a general drawing tolerance block...
Speaking of dimensions...

You are trying to imply too much or all components are under-defined. Flange thicknesses... are all .500? Is three decimal accuracy required on the flanges? For ~.5 consider .XX+/-.050

For the 5 radial holes, specify some axial centerlines on the end views. Call out the diameter of the pattern and use basic dimensioning to call out "5X 72*" wrt one of the axial centerlines.

Middle view should be called "SECTION VIEW" or the features should be shown as hidden lines. Avoid dimensioning to hidden lines. If you do want to show sections, also show a centerline... section arrows on the parent view help with intent too.

Oring glands. You'll have to seal all those flanges somehow. Google: Parker Hannifin Handbook.

You did use third angle projection, though.