Controlling the ID of a drawn tube

[germain87]

Hi all,

Looking for a little guidance/advice on how to properly call out the ID of a tube. The intent of the design is to provide a light pressfit (0.000-0.004") with an end cap/cover prior to laser welding the ends shut.

I've just joined the company and we have previously just sent our vendors a list of dimensional specifications instead of providing them with a properly toleranced drawing. As a result, the tubes that we get frequently need to be widened or squeezed in order to bring the ID back to the correct dimension. I am working to create the drawings but am getting a little hung up on how to control the ID. We need a nominal ID of 1.811" with a WT of 0.036". Can anybody tell me the best way to call out the ID such that we can create the necessary interference with a Ø1.811" +0.000 -0.002" cap?

For what it's worth - I'm assuming that the roundness of the tube is not terribly critical and that it will get properly rounded when it conforms to the cap.

 

[desertfox]

Hi germain87

Buy a tube with an I.d. That always needs machining to get the bore correct, I am guessing you only need a small length of accurate ID at each end of the tube, does it matter if the wall thickness is bigger than you want in the unmachined part of the tube?[pre][/pre]

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[BrianE22]

I believe DOM tubing is the most accurate "stock" tubing you can buy. If you could live with 1.875" OD x 1.812" ID you might find something. For that ID, the standard tolerance on the ID is .000/-.006. Some tubing suppliers also offer honed ID tubing. That would control the ID even closer.

 

[EdStainless]

germain87,
If you are cutting these off in a lathe then buy the tube a little heavy and turn a land on the ID as Deserfox suggests.
You will want these in a collet so that they are wound when machining.
We will make tube to ID-wall or ID-OD specs but it isn't standard product.
And yes you can get tighter tolerances, maybe +0/-004" but you will pay for it.
You are correct about roundness, this is light wall tube and it will easily deform into shape.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[pylfrm]

germain87,

forum1103 may have been a more appropriate place for this question, although you'll probably get more insights on the design and manufacturing aspects here.
[ol 1]
[li]What tolerancing standard do you want to use on the drawing? Some version of ASME Y14.5 perhaps?[/li]
[li]What is the length of the surface you want to control?[/li]
[li]What is the length of the interference fit?[/li]
[li]To what extent can you tolerate localized gaps in the interference fit?[/li]
[li]What quantity do you expect to buy?[/li]
[/ol]

pylfrm

 

[3DDave]

You'll need the AVG modifier on the diameter dimension and a circularity callout. This will limit the out-of-round condition while providing control of fit. See 5.5.3 and Figure 5-13 of ASME Y14.5-1994.

 

[dgallup]

We press fit millions of parts inside deep drawn tubes every year. We use a note that says the ID dimension is computed form the average of the minimum and maximum values going around the bore but the minimum measurement can not go below the drawing lower limit. Sort of a poor mans roundness spec. True circularity is very expensive to measure.

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

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[EdStainless]

This is not a dimensioning issue, this is about understanding the tubular product specifications and how they are manufactured.
You can put anything that you want on a print, but tubing does not have a center it only has an OD (or ID) and a wall thickness.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[3DDave]

Circularity has no center requirement, though runout does.

 

[germain87]

Hi all,

I need to keep this as low-cost of an item as possible so honing out the ID likely won't be an option. I do think 3DDave's and dgallup's suggestions of just calling out the ID as Ø1.811 AVG of calling holding to an average w/o violating the low limit will suffice as it should give us the ability to ensure the proper diameter to create the press fit we need and without having to worry so much about controlling the circularity. I appreciate the feedback from everybody - this has been extremely helpful!

 

[racookpe1978]

How deep inside the tube does the press fit need accurate to work properly?

I'd make a stiff cone of hardened steel, starting just under the ID you want, expanding to the ID you desire for the length you need + a small margin, then abruptly expanding to the max OD of the tubing + a margin.
Then just ram the cone into the end of each received tube until the seat hits the end of the tube.

Sledge hammer ramming is not recommended, but a true press is probably overkill.

 

[Compositepro]

Usually balls are used rather than cones, and the process is known as "ballizing".

http://www.itiball.com/products/tungsten-carbide/ballizing-ball-burnishing/

 

[hendersdc]

What are the reasons for needing such a tight tolerance on the ID? Just for a press fit to the end cap or something else as well?

If it is only for the end cap, and assuming you are working with plastic, you could consider updating the design of the end cap with crush ribs or flexible ribs that can work with stock tubing tolerances.