Self centering small threaded holes

[[Ne...]]

What are the steps to use a larger probe to get location (don't need size) on small threaded holes using the Probe self centering?
Any help would be great. Thanks in advance... Neal

[[Za...]]

I do it so:
1. Create point on plane near hole
2. Create circle in hole
3. Copy coordinate of circle to point (for example X, Y)
4. Set third coordinate bottom of plane (0.5 mm)
5. Set up self centering measurement in strategy setting

[[De...]]

FYI Unless your happy with measuring the location of the lead in of the thread this is not a great method of checking location of the holes.

Not that there are many great methods for small holes, but this is only finding the location of the top of the thread - and that is only assuming the lead in is actually in the same location as the thread itself.

You could scan lines on the minor for a decent result. Of course that is still only checking the minor, however at least that includes orientation in the result.

Best bet for accuracy on small holes is probably to get thread plugs and scan them to actually check the thread itself.

[[To...]]

If you chamfer the hole (minor dia), wouldn't self-centering on the chamfer be almost as good as measuring the hole?

[[De...]]

It depends on the process as much as anything.

I have seen parts processed where chamfer's were put on afterwards on a drill press in a secondary operation and other times milled in place at the same time as the minor and the threads. Obviously in the second case the location of the chamfer would be far more accurate than the first.

So like I said with a known process and enough tolerance, sure maybe its fine - if you don't know the process or if your tolerance is tighter than the allowable locational error of the chamfer to the minor then you can't count on the chamfer and the minor being in the same place.

The main thing I don't like about self centering points for position is that it is only taking a single point, which cannot give the orientation of a hole - if its a true position callout the orientation of the hole is necessary to give a valid answer.

I might worry too much but It could just be that I have worked on tight tolerance aerospace stuff for too long. 🙂

[[To...]]

My bad, I meant to say that both the spot chamfer, drilled hole and tap where done on a CNC during the same set-up. You're right, it depends on how and when the hole is chamfered.

[[An...]]

Probe,chamfer and touching point.
See attached.

Self_Centering_1a.xlsm

[[Me...]]

Andreas.I see the principle but with threaded holes you will still have an erratic feature at the contact dia.
I have nutty engineers who insist on True Position .05mm for threaded hole (WHAAA?). I'm thinking I should use a cylinder probe.
Probably cheaper to shoot the engineer 😉

[[Ri...]]

Neal, if you have a boatload of tolerance for the position of these threaded holes, a self centered point in the chamfer will likely suffice but if the tolerances are snug then plugs or pins will probably be necessary. If there is a MMC or LMC modifier on the position call then measuring the minor diameter with line scans on a cylinder would be best.

Kevin, 50 micron position on a threaded hole is hilarious. You are lucky to have such a fun loving bunch of engineers to work with. It could be that your engineers don't really understand what they are actually requiring here. I like using a technique I learn from a crusty old inspector that I worked with in my long distant youth. Take the position tolerance (50 microns) and divide by three. This will give you a ballpark plus/minus value that is more easily understood by engineers. 50 micron position translates to 16 micron plus/minus from nominal. With that the engineer can determine if that is really necessary.

[[Aa...]]

Please sign in to view this quote.

I see nothing funny here. We've received prints calling for .001" position on threaded holes.

What gets more interesting is when they design assemblies where the components are aligned on threads. The alignment of the assembly is critical to its function, but the threads' positions aren't even toleranced. (E.g., components are held together with a custom shoulder bolt. No concentricity or position called out on the shoulder to the threads of the bolt. Meanwhile, the threaded hole the bolt goes into is the primary datum of the other piece.)

[[Ri...]]

1 micron position for a threaded hole. Hopefully you will have a MMC or LMC modifier to give you a little wiggle room. Do you have to maintain a minimum CpK as well?

[[To...]]

Richard, " = inches...LOL

And, we all know how MMC on the minor diameter impacts the functional location of a threaded hole. Hey, there's no SMH smilie face. Maybe this one is close 🙄

[[Ri...]]

Tom, yep, I missed the ". Let me revise and amend my previous post.

1 thousandth position for a threaded hole. Hopefully you will have a MMC or LMC modifier to give you a little wiggle room. Do you have to maintain a minimum CpK as well?

Better? <13 year old girl rolling her eyes because of something her father just suggested concerning pop culture emoji>

[[Me...]]

I get a lot of this.
I know what they want. We make the part that screws in there. I've suggested to management that we talk to them and modify the design. I got a big "NO. Let them figure it out."

[[Ja...]]

FWIW - “Let them figure it out.” Ouch! I’ve worked for a company with that same philosophy in the past. I understand what they’re saying, but if there’s one thing I’ve learned over the years is that if you help your customers’ engineers look good, you’ll guarantee your company a returning customer – and they’ll probably want to send more work your way on top of it.

At any rate, I agree with most everything that others have said about picking up the location of tightly toleranced threaded holes. One other option you have is to go to the Lavezzi website and see if they have threaded hole locators with the tapered pick-up threads in the size you need. You have a better chance of picking up the actual pitch diameter location with these than with the straight thread versions – At least I have anyway, especially when the threads aren’t exactly perp to the surface.

[[Je...]]

The most repeatable method that displayed an acceptable level of accuracy for locating a thread that I have found is to use 4-6 line scans along the axis of the threaded hole. Keep scan speed to under .3"/s with point density of .0002-.0004". Scan in the outward direction (start probing at bottom of thread) so the stylus drags along the minor as not to bounce the stylus by pushing it into a 30° thread pitch. Change evaluation to "maximum inscribed" to simulate a pin in the minor. Filtering will affect the diameter reporting if you have it enabled by default. Error does present itself at times but this is the best method I've found. With some coarse threads I have found will require changing the evaluation of the probing vector to be locked.

To try and keep and inventory of every thread locater needed would cost tens of thousands of dollars for our company and increase iinspection time. For that reason we don't bother unless the print call-out is ridiculously tight (less than tp.003") and the customer refuses an engineering change to correct the over zealous imagineers idea requirement.

The LMC and MMC modifier is moot for a threaded hole as measured by a CMM as the modifier references the pitch diameter rather than the minor. Minor diameter does not factor into LMC/MMC and who really has time to measure actual pitch diameter of an ID thread to calculate MMC "bonus" rather than verifying pitch diameter with a go/no-go plug.

[[To...]]

Please sign in to view this quote.

[[An...]]

Please sign in to view this quote.

[[Gr...]]

I am confused.
What is the point to self centering a circle?
How is it different from just measuring it normally?
The manual (as always) does a terrible job of explaining.

I actually have a problem on one of my parts where the hole moves around more than the diameter of the hole. My 1mm stylus hits the plane without going into the hole. So for me (at least at the moment) isn't during a scan, it is getting to the scan in the first place. Is there a method to use to find the hole's rough location (A), then use A as a reference to use in a strategy for a properly accurate location (B)?

[[Me...]]

From the studies I've read, measuring as a helix at the pitch is the most accurate way to measure threaded holes on a CMM. The probe rides on the same portion of the thread, regardless of where it starts. There was a great Lunch and Learn however many years ago where they tested all the methods: circle, cylinder, linear scans, helix, etc. The conclusion was the helical sweep was the most consistent and accurate.

Ditto on the MMC on the pitch diameter. Forget about it. Not worth the agony.

[[Ri...]]

Robert, a helical scan on thread pitch does an okay job but it's problem is that you cannot depend on the size finding. The stylus can be riding on the thread crest on one part, then off on the next. Line scans for a cylinder is the routine I've found has the best repeatability for location, orientation, and minor diameter size when used with the Maximum Inscribed Cylinder evaluation. That being said, my experiments where on average sized threads, 1/4-20's, 5/16-18's with about 1/2 inch to 3/4 inch depth. The need for lots of data is crucial for this. 4 lines is simply not enough. I like 6 lines for a starting point in my programs and will increase that as needed. Also, lots of points per line. My starting point for that is 500 points per line. The one problem with this routine is that it takes the longest amount of time. This is where I've used helical scans but, I would run the helix in the opposite direction of the thread. As with Lines, a large amount of data is needed with the Maximum Inscribed Cylinder evaluation. Double check the diameter finding with a gage pin.

[[Pe...]]

Please sign in to view this quote.

Start with a different probe at least a little larger than the hole. Scan the plane with lots of points in a tight grid strategy focused around the hole so that the probe will dip down into the top of the hole as it goes over it. Add a Minimum Coordinate construction based on the plane. This will find the spot where the larger probe dipped into the hole. Now use the coordinates of the min point in formulas in your circle. This will guide your 1mm probe into the hole.

It's not a bad idea to go into Resources >> Feature Settings Editor... and select Travel >> Speed from the drop-down and turn the approach speed down for the circle to help protect the small probe as it approaches the circle as well.

[[Me...]]

Hi Richard,
I understand what you're saying. I check threads first with a gage to confirm the pitch, then use the helical scan and have good results on location. I use a ball size as large as possible for the thread. The minor diameter isn't used for material condition so if I need that I'll use a pin, mold, etc. I've tried it both ways... linear scans and helix. The ball size seems to have the biggest impact on the results. They're both good ways to check location. I thought the study I read was really informative and that's probably why I lean towards using a helical scan most of the time.

[[Ri...]]

Robert, if MMC or LMC aren't involved then a helical scan on the thread pitch is perfectly fine. If you are referring to the same study I saw (I was working for Zeiss at the time), there were some minor tweaks to the strategies so that measuring the minor diameter become a more solid method. As with everything Calypso, the answer to virtually any questions begins with, "well that depends,,,"

[[Me...]]

Correct me if I'm wrong, but MMC or LMC doesn't apply to the minor diameter. It applies to the pitch diameter, which isn't realistically measured by a CMM unless the threads are pretty big and accessible. Material condition on a minor has little to no bearing on the position of the fastener, as threads are self centering on the helix of the thread once you tighten it down. A loose fastener might give you more initial ability to assemble, but when you tighten it down it's going to center itself. If not able to do so, it will put a lateral load on the fastener, which isn't usually desirable.