What's new
What's new

Guaging intersection

Ox

Diamond
Joined
Aug 27, 2002
Location
Northwest Ohio
Sorry, kan't think of a good title... :o


Part that I am running right now has an angled cross hole going out the bottom of a hydraulic port on a 20* angle.

The spec shows the C/L of the cross hole tool to be .098 from C/L of the port it'self - at the surface.
How would you guage this?

I put it on the shadowgraph as recommended, but the pic is fuzzy to say the least on round shiny guage pins.

Doo we need a better shadowgraph? A better opperator? Or something different?

I'd rather not post pics of the parts, sorry.
Hopefully my 1000 werds is as good as a pic?


--------------------

Think Snow Eh!
Ox
 
Yeah.

Pin in thread bore, and then a pin in the angled hole.

I get a fuzzy edge line - just like my chum told me I would on the rounds. Neither one of us use a shadow often.
I was using his unit. It's not a fancy unit.

The surface edge was sharp tho.


---------------------

Think Snow Eh!
Ox
 
Does the shadowgraph have a stop on the light source?
A stop is like an adjustable iris between the bulb and the lamphouse lens.
You close it down and the image gets a lot dimmer but the edge on round parts becomes much cleaner.
If not, sometimes you can clean it up by moving the bulb location or filament orient.
Some compartors have a colored filter you can flip in to help and you can do sort of the same with a green 35mm camera filter taped to the front of the lens.

You get the fuzzy picture because the light source is not telecentric and it "bleeds" around the curvature.
Also acid etching and cold blackening the pin helps prevent ghosting. Maybe try a black drill bit shank and see it that helps.
Bob
 
So forget the optical complication. Pins have a measurable length. You have depths of the bores. From the values you compute via trigonometry. That’s what I’d do.
 
it might be a tad overcomplicating things, but to get a real accurate measurement, I would set it up on a sine plate and measure off to a tooling ball.
 
Flat some sacrificial pins to get a crisp edge? I've done this exact thing before on common optical comparator setups.
 
Flat some sacrificial pins to get a crisp edge? I've done this exact thing before on common optical comparator setups.


Originally I thought about grinding flats on the pins, but then I figgered that the little bit in the middle that I couldn't grind away - is likely the exact part that is causing the trouble. If I ground flats on it the other way might work I guess. Even if it wasn't clocked quite right - it should be the same on both top and bottom I guess....

Or maybe not ..... I guess the light would reflect off of the slightly mis-indexed flat from the light source end and give an unreliable line again? The other side would prolly be crisp tho... ??? This would likely give a wrong result too, as I found that my DRO showed that my pins were .010 smaller than they are, so if the reflection was the same, I would get a -.005 result on one side, and a crisp (?) line on the other side? Resulting in a .0025 C/L error?




But then overnight I got to thinking that I wouldn't need BOTH sides flat - just one! (I think?) So if I ground the pins in half and pointed the flat towards the light - that should give me what I am looking for - right? Or even slightly past half to make sure that I'm not slightly miss-indexed and getting that light hitting the rad even just a little? The pin would seem undersized a little maybe, but C/L should still be intact - which is all that I would be looking for anyhow... :scratchchin:

And I guess you came up with the same thing.
Have you done this before?


Personally - I am not following how this tooling ball, or trig'ing off the lengths of the pins works? There is no bottom to the one hole, and the other hole is better known I guess, but that sounds highly complicated and unreliable to me.

???


What I'm thinking is that I should have bought that Keyance machine that was here.
That should have been able to doo it eh? ;)


---------------------------

Think Snow Eh!
Ox
 
I've done it both ways. Flatting one side and then turning the flat perpendicular to the light, make sure both edges are crisp and it really doesn't matter if your flat is on center or past, etc. Or, flat both sides or even square it with 4 flats, and then you can get a more reliable centerline more quickly by making sure the upper and lower edges are crisp. Pointed pins never made me feel comfortable with the focus. With multiple flats, you obviously should be able to get the DRO to read you the flat perfectly and make you feel good about the numbers.
 
That really looks like a custom gauge to me.
A go/nogo plug that indexes off the face and the drilled bore of the port maybe?
You can't use a ball and pin because both the angle and the offset have tolerances.
On edit...that looks like an ORB, if so then use that for the concentricity.

Confocal microscope looking down the diagonal bore?

Flatten the pins...the fuzz is diffraction, make the pin as thin as possible.
As mentioned, stop down the source.

Nice problem.
Ask how the customer will QC it, use their method.
 
Let's try my second attempt at Flikr.
This has taken me WAY too long!





---------------------

Think Snow Eh!
Ox

So, recreating this, sort of, mount your part onto a sine plate with the port at 0 deg. Mount a tooling ball somewhere convenient. It doesn't matter where it is, just so long as it stays in the same spot.

Capture.JPG

Now, you can calculate the numbers by hand, but it's a lot easier, if you have a cad system, to just draw the balls location as shown above. Now, rotate the view 20 degrees and find the distance from the angled hole to the tooling ball as shown below;

Capture1.JPG

Adjust your sine plate to 20 degrees, and this is the dimension you should have.
 
Just a guess, but if I am having trouble seeing a cylinder, it would seem that locating a sphere wouldn't be any better.

???

It's not like I can't measure it regularly with pins, I just couldn't focus on them. A flat on one side should solve that issue.

Not saying that your method wouldn't work - other than you may want to mount your ball outside the part - but it only seems to complicate the process to me?


--------------------------

Think Snow Eh!
Ox
 
Just a guess, but if I am having trouble seeing a cylinder, it would seem that locating a sphere wouldn't be any better.

???

It's not like I can't measure it regularly with pins, I just couldn't focus on them. A flat on one side should solve that issue.

Not saying that your method wouldn't work - other than you may want to mount your ball outside the part - but it only seems to complicate the process to me?


--------------------------

Think Snow Eh!
Ox

my method isn't using a comparator. Just a height gage and a surface plate.
 
Sorry, kan't think of a good title... :o


Part that I am running right now has an angled cross hole going out the bottom of a hydraulic port on a 20* angle.

The spec shows the C/L of the cross hole tool to be .098 from C/L of the port it'self - at the surface.
How would you guage this?

I put it on the shadowgraph as recommended, but the pic is fuzzy to say the least on round shiny guage pins.

Doo we need a better shadowgraph? A better opperator? Or something different?

I'd rather not post pics of the parts, sorry.
Hopefully my 1000 werds is as good as a pic?


--------------------

Think Snow Eh!
Ox
Correct spelling might help.
 








 
Back
Top