What tolerances can I expect from a local machine shop?

[metalman8357]

Hi all,

I'm having simple fixtures made for some material testing. They are essentially plates with through holes to insert precision dowel pins. The center to center spacing for these holes must be accurate, and my design allows for no more than a +/-0.001" tolerance. My question is, is it safe to assume that a normal machine shop can hit these this tolerance no problem? Also, what type of tolerance can I typically expect for the overall width of the plate? (both faces will be end milled). From what I remember working at my uncle's machine shop as a kid, it wasn't uncommon for us to hit +/-0.0005. I just want to make sure that I'm not making some absurd request.

Thanks,
M

 

[ajack1]

I am finding it hard to believe that there are engineering companies out there that need to be told to inspect parts at 20 degrees C or would measure something in imperial that was dimensioned in metric without knowing how to convert between the two.

I guess I must have just been lucky over the years and nothing should be taken for granted, but it does leave me wondering what exactly you could take as a given.

 

[mfgenggear]

Kenat

your post are always very helpful. & I agree.

Bill

The OP as I interpreted it, is for a small testing fixture for the lab. but size & complexity does make a difference.
I to worked with large equipment & parts. up to 10' diameters, & longerons that the starting material was aluminum plate.
that specially fabricated, it was approximately 20-30 ft lengths. I am guessing now. because it has been years.
These were machined on plank mills. the 10 ft dia was machined on & with a 20 ft bed VTL.
it was all verified optically. except the dia was verified with a pie tape, I believe the tolerance was +/- .010"
in the restrained condition.
it may have been closer.




Mfgenggear
if it can be built it can be calculated.
if it can be calculated it can be built.

 

[KENAT]

ajack, at least here in the US a significant part in the drawing standard development was the legal aspect making sure all kinds of things like this are explicitly defined so that in the event of a 'you made my part wrong' type court case there is minimal ambiguity of requirements - if the drawing standards were followed. Essentially the engineering drawing defines the part you will accept effectively as part of a contract.

My recollection was that the philosophy in the UK was similar, though not quite as hardcore on the legalese aspect.

Posting guidelines faq731-376

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

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

 

[automatic2]

My question is, is it safe to assume that a normal machine shop can hit these this tolerance no problem?

CNC, yes, manual, not likely. What normal?

 

[geesamand]

Hello dead thread, welcome back.

I would argue (based on several years of working with local machine shops and +/- .0005" tolerance, is that your past experience in a shop is missing the point. The point is that shop machines routinely read out to the .0005" or better, and will repeat to .0005", but rarely are *accurate* to .0005". In my experience, unless the shop has a robust CMM system they simply have never learned the difference between repeatability and accuracy.

It gets much worse when you include the fact that the z axis of travel is usually not exactly at 90 degrees to x and y, so that if you machine features at different depths, the perpendicularity error will throw off locations by up to a .0005 to .001" per inch of z.

Then don't forget clamped condition vs. as-used and as-measured condition.

Certainly in the environment of local job shops, you may succeed with this part but IME you won't have a chance of success unless that vendor or your company has appropriate CMM equipment to prove it.