I've previously had it suggested to me that around .014 dia position tolerance (corresponding to around +-.005) is a good guideline for various types of threaded inserts.
Doesn't mean tighter isn't achievable if required but if I need tighter I go to greater lengths to check with whoever will be making the part.
Thanks KENAT. That is helpful. I was given the same suggestion a few years ago by a design manager, but was told recently by a co-worker that my suggestion for a .010" position tolerance was TOO LOOSE. That co-worker told me to call out 0.005" positional tolerance on the threaded insert position.
At the end of the day the tolerance chosen has to be functional (i.e. mating hole patterns/fasteners fit) and manufacturable at acceptable cost.
Your initial question sounded more like addressing the manufacturability aspect.
From that point of view your colleagues comment makes no real sense.
So I assume they were talking about functional requirements?
Without knowing mating parts etc. no one here can really comment on that, ASME Y14.5M-1994 annex B has calculations for mating hole patterns.
Another trick is to avoid using threaded hole patterns and their fasteners for precision location, instead use features such as dowel pin pairs with a corresponding slot & hole. This means the threaded fasteners only have to 'clamp' the parts together so mating holes can be bigger and tolerances looser.
The tolerance in this case needs to be very tight because of the small size of the part and the minimization of it relative to the components it contains. Therefore, I need to know how tight I can tolerance the position of the threaded helical inserts. For rb1957, I mean: what can be achieved as far as the position of the pitch diameter of the insert with respect to the part datums without incurring a ridiculous increase in cost or a very high likelihood of non-conformance.
Part of the issue here is that I cannot make the corresponding through-hole diameters too large because of their minimized edge distance. As you probably guessed, that is why I need a tight tolerance on the threaded hole position.
I'd talk to the machine shop that will be making the parts if at all possible.
I generally avoid threaded holes below around .009 pos dia - I've been forced to put tighter and machine shops have claimed they can hit .003 on smallish parts but I don't know that our inspection ever verified that claim.
Assuming installing a helicoil will introduce a bit more tolerance then I'd suspect .005 pos would be pushing it but maybe the right machine shop will say it's easy.
Well match drilling is a whole other kettle of fish, generally to be avoided when possible but I know with aircraft skins is routine due to rivet tolerances, countsunk heads...
Thanks guys. KENAT, what you are saying agrees with what I have found from other sources. Sandia Labs conducted a test where they found they could achieve Ø positional tolerances of about 0.003 for threaded holes in aluminum. However, in steel they were able to achieve 0.0004! Like you said, on top of that you have to figure in the tolerance of the insert alone. On a final note, I have found that using a jig boring machine is the best way to achieve these tight tolerances.
Most often helical coil thread inserts are used in mechanical assemblies [forged/wrought/cast materials], such as gear boxes to accept cover-plates, secondary covers, seal plates, etc.
In these cases, the holes with helical coil thread inserts [all in one assemly or the other] recieve long-threaded bolts, full threaded screws or cap-screws to provide pure joint tension, without shear capacity, IE: optimum joint clamp-up... usually for gasketing/sealing purposes. In addition to these bolts, the same flange, both mating halves, will have several precision-located close-reamed or transition fit high strength shear-pins around/between the bolts [and on internal webs, etc][or sometines coiled or 'C' spring pins], to stabilize the component joint in shear/torque. Between the tension threaded fasteners and the shear-pins, very durable mechanical case/component joints are made.
Regards, Wil Taylor
Trust Me! I'm an engineer!
Trust - But Verify!
We believe to be true what we prefer to be true.
For those who believe, no proof is required; for those who cannot believe, no proof is possible.
Unfortunately, in science what You 'believe' is irrelevant û "Orion"
The use of a threaded insert usually guarantees that the shear plane between the two mating parts will pass through the threading of the bolt, not its shank. Meaning the bolt's minor diameter, not its full diameter is the shear load path. Shear-critical joints are not what these fasteners are intended for. The corresponding mating part should have enough "slop" in its hole sizes (oversize) to accommodate the variances in the tapped hole locations.
A mention of an "edge distance problem" by the OP implies that you do have a shear-critical joint. Or maybe you just have an eager designer trying to "tighten up" the tolerances as a general policy, mostly because it sounds good, not because it makes a better connection between parts. But I haven't seen the parts that you're trying to assemble... Is a 1/64" oversize really going to cause an edge distance problem?
There is no significant shear load to transmit through the screws, and micrometer-level slip between the fastener and the joined part is acceptable under CTE mis-match conditions. I just don't want to have a through-hole so close to the edge that it might break through the edge or leave it paper thin. I am working on a very small electronics package. The distance from the hole to the edge is only .022 in. So, 1/64 oversize plus positional tolerance is leaving it uncomfortably thin. I have implemented a solution that thickens the screw boss and lengthens my edge distance on the through hole. Yes, this has a minor impact on the mass, but we are still well under the baseline.
I don't use helicoils, but with keenserts or rosans I usually apply the positional tolerance to the tap drill hole to make inspection easier. The downside to doing this is that it does not capture the error between the inner/outer insert thread PDs and between the tapped thread PD and the drilled hole cylinder, but the total of these errors is typically quite small. If you apply the positional tolerance to the installed insert internal thread, the measurement is based on the thread PD according to ASME Y14.5-2009 section 2.9 (see attachment). Inspecting the installed insert thread PD position also requires the use of a special gage.
The positional accuracy of a hole pattern is highly dependent on the manufacturing method used. Assuming the part is fixtured properly, the accuracy of a hole pattern produced using a mill or jig bore is mostly a function of the machine used. A good quality CNC mill can hold .0002" per inch position tolerance on drilled holes under normal conditions, which allows fairly tight true position to be used for most applications. As wktaylor noted, fixed threaded inserts are really only suitable for fastener installations where clearance holes are used in the mating part(s). If the hole patterns are produced using a good quality CNC machine, I would have no hesitation applying a composite positional tolerance of .010" dia MMC within the hole pattern.
Thanks very much for your valuable input Terry. Neat little animation of the gage. QUESTION: Would you hesitate to apply a .005" positional tolerance to a #2-56 threaded insert referenced to the part datums? The greatest distance to the datums is roughly 1.5 inches. The material is a 99% molybdenum alloy.
As long as you are using a good quality CNC machine, the part is fixtured rigidly and accurately, you use machine speeds and feeds that are appropriate, and your machine operator pays close attention to what he/she is doing, then you should be able to hold .005" dia position tolerance between part datum features and the internal thread PD in the base material. But this is not as simple as many people think and it may require running a couple setup parts to get the process right.
I did a bit of research and I found MIL-I-45914 which is a general specification for key-locked inserts (Keenserts). Section 3.8.3.3 specifies maximum runout of .006" FIM between the internal and external thread PDs, which is quite a bit more than I would have expected. You would have to add this insert runout to the positional tolerance of the threaded hole in the part to get the max positional error of the installed insert thread PD. The p/n for a .086-56 UNJC-3B key-locked insert is MS51830-101.
I've attached the relevant sections from MIL-I-45917 for your reference.
