Giving the digital caliper a workout

[Chasper]

I've been trying to determine the appropriate size difference between the inside of tubes and the parts that get pressed into them. I have a generic digital caliper with three digits after the decimal and the number 5 pops up after the three digits if it measures between numbers on the third digit. I don't know how the rounding logic is set up that causes the 5 to appear. I realize that the caliper I am using is not terribly accurate. I tried to put the same amount of pressure on each part when I measured them. I believe that the large number of parts that I measured helps to smooth out variability that was introduced by tool and operator.

I measured 5 each of 5 different platings on slims, I believe they came from a variety of sources. After deburring the ends, all the tube IDs measured .243 to .244. All the transmissions measured .248 to .249. There was an outlier nib that measured .246, all the rest measured .2485 to .2505. The caps measured .2485 to .251. On average the parts being pressured into the tubes were .00555 or 2.28% larger than the tube.

Same type of measurements for 5 each of 5 platings of cigars; there are four parts to be pressed into the cigar tubes during assembly. On average the parts to be pressed in were .004 or 1.11% larger than the tubes.

Should I consider .004 to .0055 to be within the acceptable range? Or should I consider 1.1% to 2.28% to be the acceptable range. I would have expected the % difference to have been less on the smaller diameter tube, but it was the opposite.

The purpose of all of this is that I'm having something made and I've asked for the part to be pressed in to be .005 larger than the tube with a tolerance of ± .0005. The person making the item tells me that I'm asking for an impossibly close tolerance. I only asked for ±.0005 because that is the smallest my caliper will measure. What is the appropriate tolerance to ask for?

 

[kirkfranks]

A couple of thoughts.

Edit cause I missed a decimal point. .005 vs. .0005"

The company I work for buys and makes turned parts all the time. +/-.005" is standard tolerance for us and all the shops we buy from. Granted these will all be CNC or Screw machine parts but still that is the standard. We do regularly ask for tighter tolerance too, but that is an upcharge. +/-.0005 would be an upcharge and harder to do. I really don't see much tighter than +/-.001

For your part I think I would be more comfortable asking for the nominal size to be equal to the tube size with the +/-.005 so that if it is up to the max over then it is a nice press fit and if it is less then you can still use a drop of epoxy and have a good fit. If your supplier can do better than +/-.005 then get what he can do.

Good luck with your project.

 

[johnnycnc]

So you asked for a thousandth total tolerance? shouldn't be a problem from an experienced machinist..
I can hold that or less on my cheap garage metal lathe with slop in the gibs.

I would caution you on inside diameter measurements with calipers, it can be
misleading. without boring you to tears, remember there is a flat on those caliper jaws, and it never makes full diameter contact at the largest i.d.
Do these fit on your mandrel? and what does mandrel measure?
might give you a comparison, as your o.d. measurements may be very close.
.243-.244 seems rather small. I would question those numbers, with all due respect.
And, In my opinion, .005 is entirely too much press.
.001-.002 is generally appropriate (in my experience).

 

[JeffinWIS]

Gerry, I would ask for + or - .001 from the tube size. That will be MUCH easier for your supplier to actually give you. Much nice for assembly also. 1/4 drop of Loctite on each fitting = no cracked blanks and no problems, IMHO

 

[JeffinWIS]

Exactly, John!

 

[Sylvanite]

Gerry,

I don't think you are getting good measurements. First of all, calipers are not particularly good at measuring the ID (inside diameter) of a tube. Secondly, "generic" calipers like the one you describe are at best accurate to only +/- 0.001". Ignore the little "5" that sometimes appears - it is meaningless.

Slimline pen tubes are made to slip over "D" size drill rod (what they make the mandrels from). "D" size (which is 0.246"), plus 0.001 to 0.002" (for the slip fit) puts the ID of a slimline tube at 0.247-0.248".

In that diameter range, an interference fit (a.k.a. "press fit") for thin-wall brass should be 0.002-0.003", yielding nib/transmission/finial OD (outside diameter) somewhere around 0.249-0.250".

For press-fit pen parts, I'd aim for 0.002" interference with a tolerance of +/- 0.001".

I hope that helps,
Eric

 

[bitshird]

I wonder if a larger tube, would increase the acceptable limits, or the interference fit, it seems that the larger tubes are calling for a different percentage, but the actual amount of interference is close to the same, the percentage of interference is just smaller, also most of the pressed in components, nibs-transmission adapters, and cap finials have ridges that to my thinking are actually what's holding the part into the tube. some of the only parts I know of that don't have the ridges are the plastic couplers for the Jr. series and the full size gents and Statesman, undoubtedly there are more, just not familiar with them, .0005 isn't that hard to maintain on a good machine, a good second op lathe can hold .0001
But maybe I'm just not understanding it right.

 

[Sylvanite]

... most of the pressed in components, nibs-transmission adapters, and cap finials have ridges that to my thinking are actually what's holding the part into the tube.

The tenon on a kit nib or finial is typically stepped. That is, the very end is a slip-fit or close-fit. A little ways in, the diameter steps up to a force-fit. The end allows you to align the parts when pressing them together. The plastic couplers I've seen have a bevel on the end that accomplishes the same thing. 7mm transmissions also have a beveled end.

Regards,
Eric

 

[Chasper]

Thanks all, this is exactly the info I was looking for.