Making a stupidly simple part. 2

[bnmorgan]

I am the part designer, not manufacturer. I have a very simple cylindrical part, with particular importance to the inner and outer surfaces. It is basically a sleeve bearing. Our manufacturer is, for some unknown reason, having trouble meeting the requirement for the surface finish of the ID. Could you suggest proper tooling or method for them to accomplish this. If not, could you suggest where I might find such information. I intend to specify exact tooling for them to buy and use to give me the parts I have designed. The part is produced from C.R.S. and plated afterward. I have included a portion of the print, for reference.

If there is further information required for you to make a suggestion, please feel free to ask.

Much obliged,
Byron Morgan


Byron Morgan - Tupelo, Mississippi
1947 Mayline
SolidWorks 2008
NX4

 

[BobM3]

Grinding or honing. That would be more than additional tools. That would be an additional operation. Have you considered buying tubing honed to size?

 

[MikeHalloran]

To what feature does datum flag A refer?

To what feature does Note 1 apply?

Are you asking them to tumble finish the INSIDE diameter of a cylinder?



Mike Halloran
Pembroke Pines, FL, USA

 

[KENAT]

http://en.wikipedia.org/wiki/File:Surface_Finish_Tolerances_In_Manfacturing.png

You're only calling up 32 so I'm a little surprised they're having that much trouble. Maybe look at a different supplier?

I'd be hesitant to specify exact tooling, at least on the drawing, you should just specify your desired end result. There is often more than one way to skin a cat, and by specifying specific tooling you may end up disqualifying perfectly adequte techniques. Even if you're stuck with the current vendor now, this could change in the future, do you want to have to rev the drawing just for this?

Also, an off topic general comment, much of your symbology does not comply with recognized industry standards, at least the ASME version used in the US. I don't think this is causing your current problem but FYI.

Posting guidelines faq731-376

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

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

 

[bnmorgan]

@BobM3, we considered it, but it was decided against (over my head)

@MikeHalloran:
d.A refers to the axis of the part.
note 1 applies to the entire part (this surface being the surface of the plating)

@KENAT
We are locked into this supplier, although I think they are contracting this part out. We aren't going to call for exact tooling on the print, nor likely are we calling for a specific technique. We will tell them ideas how they can accomplish our part with their tools, or what sort of tools they should be using.

on the ASME, we are aware. It is an in-house system, for various reasons. I would prefer ASME, but we have noone here versed enough (myself included) to apply it properly, so we continue with the system that was already in practice within the company.

Byron Morgan - Tupelo, Mississippi
1947 Mayline
SolidWorks 2008
NX4

 

[KENAT]

Well then, you're stuck.

For example datums symbols should be applied to actual features not centerlines or axis, hence Mikes question. Having datum A applying to the 'axis of the part' is open to interpretation. You should select either the OD or ID to be your datum features. The other diameter would then be related to that. To get the basics right you don't have to be that well versed, just spend a little time reading the relevant standard, I'd suggest ASME Y14.5M-1994. If you have questions on this aspect I'm sure the good folks over in forum1103 would be willing to help.

Also I agree with Mike on 'To what feature does Note 1 apply?' it's really not clear, even after you've explained what it's meant to mean I dont think most readers of the drawing would interpret it that way. The part has multiple surfaces, OD, ID, both ends and chamfer surfaces.

On the surface finish issue:

http://www.finishing.com/430/60.shtml

http://www.mfrtech.com/articles/2528.html#

However, given that you are subbing to someone who is then subbing, I'm not sure you have any chance in hell to really control the process. Did the supplier ask for tips on how to meet the requirement? Otherwise I'd be tempted just reject the parts and tell them to make some that meet the drawing requirement. One of the joys of subbing out work is, so long as what you've asked for is fundamentally possible, it's up to them to work out the details. In this case I'd expect the shop to know more about this than design guys.

Posting guidelines faq731-376

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

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

 

[BobM3]

Has your machinist tried a "wiper" insert? They have a flat on them that rubs the workpiece which gives better surface finishes:

http://www.kennametal.com/images/pd...s/TechTip_131_WiperInsertBenefits_Turning.pdf

 

[BobM3]

I've not used it but maybe roller burnishing:

http://www.cogsdill.com/PDF/USCatal...shingminipdfs/CogsdillBurnishing_Overview.pdf

 

[halfmark]

I have used bearingizing tools in the manufacture of sintered bearings. This probably suit your requirements.

 

[mfgenggear]

The type of machinery & tooling that your supplier has is important.

the small tubing would be a perfect fit for a screw machine
shop that can hold this tolerance & finsh day in & out.

32 micro & .005 tolerance is easy for them to hold.
& they can do it cheap.

they will most likely run 3 foot lenghts & turn & bore with
no problem.Plating may also be the culprit. flash plate if not done correctly will look unpleasent.
look else where for your parts.

 

[tygerdawg]

ditto KENAT. Your drawing needs redimensioning, and you need to resolve Datums to physical features. For example you've got the straightness callout, now you need cylindricity to give your machine shop a good datum upon which to grip (and your standards lab a good datum from which to measure). What follows is concentricity of the ID, perpendicularity of the edge faces, and dimensional tolerances and finishes.

ditto MFGENGEAR. A good screw machine shop could do this (but only if dimensioned properly).

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[tomwalz]

I would just specify the parts I wanted and put it out for bid. As a guy, who both makes parts for others and buys parts, it seems to work better that way.

We have some suppliers who do much better work for much lower prices than others. Typically they are using something unique. If you specify a standard practice then you lose that advantage.

Example: We sell carbide saw tips among lots of other things. Carbide saw tips come 100, 250 etc. per package. Because the parts are powder, pressed and sintered, they are not quite identical and you cannot weigh count them exactly.

A few years ago I bought a pill counter, as used in pharmacies, which counts the parts to a minus zero / plus one standard. This means I can ship you the correct number of parts without having to over ship just to make sure. (The other option is to risk short shipping.)

If you were to order carbide saw tips from me and specify a counting procedure it is unlikely that you would specify a pharmaceutical pill counter. You would most likely specify a weigh counting scale. I have several of those and I would use one if you wished. However I would have to set them at plus 1% to make sure you got at least the right number and raise your prices at least 1% plus an additional amount for the extra labor.

This is an unusual case but the world is full of shops with unique and clever tricks and equipment.

Tom


Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[mfgenggear]

all though BN Morgans drawing is not to the ASME standards.

it is clear to me what the drawing requires.
& it's as stated "a simple part". & it is.
there should be no problem fabricating it.

 

[KENAT]

mfgenggear - there are several relatively tight tolerances which may be affected by datum selection or use of the wrong type of tolerance based on funtion etc. While I don't see an obvious link to the current problem it is a potential issue.

Posting guidelines faq731-376

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

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

 

[vc66]

Maybe the shop is so confused trying to read that drawing that they forgot about the surface finish requirements.

V

 

[bnmorgan]

Much appreciated all, well, most. We have solved the problem. The early production showed problems, but the latest samples have been exceptional in this aspect, as an improperly setup tool was found and corrected. Not sure how much of that can be attributed to a Chinese subcontractor to a Chinese contractor. At any rate, he surface finish is now well under the 32µin finish required.

I apologize to those of you who have been offended by the drawing itself, but I am in no position to change, or even suggest changing, the in-house standards used on our drawings. It was developed long before I was here, and likely will remain long after. I'm in the "low toad on the totem pole" situation. Sorry.

Thanks to those of you who left the constructive and helpful comments.


Byron Morgan - Tupelo, Mississippi
1947 Mayline
SolidWorks 2008
NX4

 

[dicer]

The drawing is fine, way better than some I have seen.
Maybe it has been covered here I didn't study all the posts.
Why is it plated? The finish should be easy to hold, not sure about tumbling though.

 

[bnmorgan]

the plating is because in the application the roller bears a 1.4mm nylon cord running over it, and in our application that cord cannot be appreciably discolored through its useful lifespan. Early prototypes I created used SS but cost prohibited its use for production. It was decided this was our best second choice.


Byron Morgan - Tupelo, Mississippi
1947 Mayline
SolidWorks 2008
NX4

 

[tomwalz]

Dear Mr. Morgan;

Nice post. I gave you a star. The initial question was valid and interesting but the process that developed was fascinating right down to the surprise ending.

In my opinion, which seems to be correct occasionally, answers to process problems are best solved by examining the process on the plant floor rather than discussing them. Discussions take on a life of their own which can be extremely valuable in a “brainstorming” session where you are looking for new conclusions. They don’t seem to work so well in this sort of application where you already have a defined conclusion.

I have to constantly watch myself when I get interested in a project. I tend to get sidetracked and lose sight of priorities. Many engineers seem to have the same mind set. We want to make everything perfect and correct all flaws no matter how unimportant.

I am glad you are getting good parts.

Tom


Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[bnmorgan]

"answers to process problems are best solved by examining the process on the plant floor rather than discussing them. "

And therein lies the fundamental problem with production being on the diametrical opposite side of the globe. We barely get any information about how they make our part, especially when it is a contractor to our contractor making them. Sometimes I think the difference in cost just isn't worth the trouble. Then again, those who are looking at the cost don't much care what I think. If it's cheap, but it doesn't work, then it's "my problem".

Such is life.



Byron Morgan - Tupelo, Mississippi
1947 Mayline
SolidWorks 2008
NX4