What's new
What's new

Calibrate thread wires for 3 wire method?

Piper3T

Plastic
Joined
May 9, 2016
I was wondering what people do for calibration of thread wires?

We have several external thread gages that are too large for our calibrated pitch mics. We normally check these gages with thread wires. To pass a typical audit, would it make sense to calibrate the thread wire set we are using for tractability? The OD mics are traceable, but I'd have nothing to show for the wires.

Wire sets are less than the cost of calibration, but I'm not sure these wires are made to any particular spec, like ZZ or X. I supposed I could set a 3 year calibration cycle on our master set. On the other hand, I could send those thread gages out, but that gets more expensive by far than just calibrating the wire set. Or perhaps I can verify the thread wires with a master calibrated Mic?

I think I'd be most comfortable sending them out for calibration and eating that cost if its reasonable.
Any way, what does anyone else think?

Thank you,
 
Quick look on cheapies and B&S set from amazon claims following: 1" long center section of each wire is ground to nominal size within 0.00015"

Thats 4 um for my metric brain, to measure that with any sort of uncertainty ratio you would want something like high precision MHD series Mitutoyo mike.
(and lets pretend that the wires are perfectly round)

Mitutoyo wire set that I have are specified as +-1um and that probably explains part of the price.
 
And regarding the price of new set vs. cost of calibration:
What is the cheapest set that comes with any sort of half-credible looking calibration certificate?
 
The wires deform slightly under measurement pressure, so you measure them between a flat anvil and a cylinder (3/4" diameter for most thread wires). You also want to control the measuring force. Here's a good document on measuring thread wires:

https://www.nist.gov/sites/default/files/documents/calibrations/on_the_measurement_of_thread_measuring_wires.pdf

Keep in mind also that the wires are used to calibrate gages, so they need to be more accurate than the gage. NIST says that working thread wire usually have an uncertainty of +- 10 microinch. The type of equipment that can achieve that uncertainty isn't cheap. You can calibrate a lot of 3-wire sets at $10 apiece before you save enough money to buy measuring equipment that's good enough.

How big are these thread gages and are there a bunch of different thread pitches? Even with known diameter wires, is your measuring equipment/environment accurate enough to measure the gage using them?
 
Keep in mind also that the wires are used to calibrate gages, so they need to be more accurate than the gage. NIST says that working thread wire usually have an uncertainty of +- 10 microinch. The type of equipment that can achieve that uncertainty isn't cheap. You can calibrate a lot of 3-wire sets at $10 apiece before you save enough money to buy measuring equipment that's good enough.
I don't think you get +-10 microinch thread wires 10$ per set.. :scratchchin:
More like 20x that price.
 
I was wondering what people do for calibration of thread wires?

We have several external thread gages that are too large for our calibrated pitch mics. We normally check these gages with thread wires. To pass a typical audit, would it make sense to calibrate the thread wire set we are using for tractability? The OD mics are traceable, but I'd have nothing to show for the wires.

Wire sets are less than the cost of calibration, but I'm not sure these wires are made to any particular spec, like ZZ or X. I supposed I could set a 3 year calibration cycle on our master set. On the other hand, I could send those thread gages out, but that gets more expensive by far than just calibrating the wire set. Or perhaps I can verify the thread wires with a master calibrated Mic?

I think I'd be most comfortable sending them out for calibration and eating that cost if its reasonable.
Any way, what does anyone else think?

Thank you,

I agree with Conrad Hoffman (post #2) but how accurately do you think you measure thread pitch diameter with thread wires?
Thread flank angles have a tolerance too and many tend to either forget that or ignore it.

http://f-m-s.dk/Flank angle tolerances.pdf

I've no problem with you using wires to check your plug gauges but if you are in doubt send out for calibration.
I'm against a general fast calibration time interval. What is important is:
1. How often used?
2. How fast do they wear?

Keep a record and that proves you are monitoring.
 
I don't think you get +-10 microinch thread wires 10$ per set.. :scratchchin:
More like 20x that price.

Good question. I don't know the uncertainty my calibration shop does on thread wires. I got the price off one document and the uncertainty off the other. Now you've got me wondering; I'll shoot them an e-mail and ask.
 
Good question. I don't know the uncertainty my calibration shop does on thread wires. I got the price off one document and the uncertainty off the other. Now you've got me wondering; I'll shoot them an e-mail and ask.

I tought that you were talking about wire set prices, not calibration prices. (OP was talking about how the wires cost less than calibration)
Any case 10usd for calibration is very cheap if the calibration is done even half as complicated as in the NIST document.
Something non-contact like laser mike would allow making the measurements pretty fast.

These apparently fullfill the +-10 microinch spec:
Thread Measuring Wires (+/- .2") - .4"-.5" TPI
 
So we do a very wide variety of diameters and pitches. I think I'm somewhere around 200 different plug gages. The sizes we can't do with pitch mic start at 3 inches and go up to about 10 inches and a few larger. I haven't gotten anywhere with design with any sort of standardization, and every year we get 4 or 5 more new threads to inspect.

Anyway, here is the set I was thinking of purchasing to be our new "master."
Amazon.com: Brown & Sharpe 599-4816 48-Piece Thread Measuring Wire Set: Industrial & Scientific

I'm not too concerned with the gages wearing out. For one, each gage doesn't get used very often because we have such a wide product line.
Also, using wires in the past has very clearly identified the gages we need to replace.

Really, I'm just wondering about tractability and how to show it to an auditor in a reasonable way.
The way our quality manual is written allows us to calibrate in-house most items. I normally send pin gages out and thread wires would fall into that category in my mind. But at $1 per pin, I pay $35 for the set and $48 to calibrate? And I'm not sure it would pass calibration to what, a ZZ or X spec? The description of the B&S set says ground to within .00015 in the center so I guess that's good enough for ZZ.

I'd like this new procedure to be both reasonable and worthwhile. Using a calibrated mic to verify the pins seems reasonable for this application, as long as I write the procedure that way.
 
I'd check first if that .00015 tolerance is good for any sort of thread gauge you use?
Methinks its intented for shop floor use when someone wants to measure the thread he just turned.
 
So we do a very wide variety of diameters and pitches. I think I'm somewhere around 200 different plug gages. The sizes we can't do with pitch mic start at 3 inches and go up to about 10 inches and a few larger. I haven't gotten anywhere with design with any sort of standardization, and every year we get 4 or 5 more new threads to inspect.

Sent you a PM but not sure if it went through or not. Do you only do external threads? If internal then what?
 
Good question. I don't know the uncertainty my calibration shop does on thread wires. I got the price off one document and the uncertainty off the other. Now you've got me wondering; I'll shoot them an e-mail and ask.

And the answer is... 12.1 millionths. So slightly worse than NIST/Mitutoyo factory uncertainty, but probably good enough for standard gages, especially on larger sizes where the tolerance band on gage pitch diameter is a little larger.
 
Last edited:
I'm not too concerned with the gages wearing out. For one, each gage doesn't get used very often because we have such a wide product line.

A few ways to decrease your calibration frequency/cost if the gages aren't wearing out:

1) Just use a longer calibration interval. You could make it 2 years, but you have to be able to justify it to the auditor (that's the tricky part depending on the auditor).

2) If every gage doesn't get used every cycle, seal them after calibration with a wax dip. When the next cycle rolls around, you only have to check the gages that were used. This also protects the gages.

3) Slowly increase the calibration interval on gages over time. There are very complicated formulas for doing this using statistical quality control (which you can read about in NCSLI RP-1 if you have a copy, or NASA has a guide that's free, NASA-RP-1342). Alternatively, if you're confident that a 1 year interval is too short but don't want to do a bunch of math, Savannah River National Labs came up with a simplified version, which basically boils down to increasing the interval by 10% each time unless you have an out-of-tolerance, in which case you shorten the interval. It's in a document called "Simplified Calibration Interval Analysis".

In cases like this where you need to keep auditors, be they ISO or customer, happy, I find it very useful to have a standard to point to. It's much easier to tell your auditor "we're using a longer calibration interval, calculated per this formula from Savannah River National Labs" than it is to tell them "we're using a longer calibration interval because it feels right." In the former case, the onus is on them to articulate why the standard you've chosen to follow is wrong.
 
I'd check first if that .00015 tolerance is good for any sort of thread gauge you use?
Methinks its intented for shop floor use when someone wants to measure the thread he just turned.

Yes, that's probably true. So, let me re-state my original question with regards to calibrating thread wires. An inspector uses thread wires and a calibrated micrometer to inspect a part. The micrometer is traceable, but do the thread wires need to be? Maybe inspection should be using much better thread wires? In our shop we're always used a basic set, but it has no tractability.

And we are tracking tool use and through digital check-out and check-in. We're using those guidelines to extend less-used tools. Eventually it will be integrated into our ERP system. For now its in an excel chart that lists every gage and how many times it was used in a given interval. I'm confident this can justify extending some tools that are used rarely. We've also used this data to calibrate and deactivate the tool until next use. The wax is a good idea though.
 
Sorry, we got a bit off the original question. Yes, you should ideally have traceability on the thread wires, whether that's done by an outside shop or in your inspection department. My cal shop charges $10 per 3 wire set, so a full set would be $480.

After they're calibrated, make sure you use the exact size (from the cal certs) not the nominal size when you do the pitch diameter calculations. Any error between the actual diameter and the diameter you use in your calculations will be multiplied 3-fold. The actual diameter of the wire theoretically doesn't matter if the thread flanks are perfect as long as you use the correct value in the formula. What using a best size wire does is decrease the effect of flank angle errors. You'd have to do some math to figure out the effect on pitch diameter of a +-.00015" diameter variation given the maximum allowable error in flank angle.

All that being said, you need to make sure that the micrometer you're using is accurate enough. For comparison, the pitch diameter tolerance on a 4" plug gage is +-.00025
 
Yes, that's probably true. So, let me re-state my original question with regards to calibrating thread wires. An inspector uses thread wires and a calibrated micrometer to inspect a part. The micrometer is traceable, but do the thread wires need to be? Maybe inspection should be using much better thread wires? In our shop we're always used a basic set, but it has no tractability.

I don't see any way around it, you need to calibrate the thread wires one way or another.
And to check thread gauges you really need something better than +- .00015" thread wire set.

I'm 100% out of my competence zone here but X tolerance seem to be most common on commercial thread gauges and if you want to calibrate X tolerance gauges you probably need that .000005" accurate wire set. The cheap thread wires are 30 times worse than that!

If you think your life is bad just take a look on Euramet guide on thread gauge measurements:
https://www.google.fi/url?sa=t&rct=...ller%5D=Base&usg=AOvVaw3JCkdka_IfcqZ9iBurHZWc
:D:willy_nilly:
 








 
Back
Top