Need your help folks

[kbotta]

Ok, here is the scenario:
My boss was approached by a few 'higher ups' about moving my gage crib from the pressroom to the warehouse out back. Realestate is at premium at the current moment.
The warehouse is neither heated or cooled - and is at a much different temp most of the times then the actual pressroom is.

Here is what he wants done:

Grab a fixture (a relatively small and tight toleranced one) and
A) Measure it as it is.

B) set it out in the sun (or a hot place) and measure it after a day of 'outside temps'

C) stick it in the fridge, on a controlled temp and measure it after a day.

He can then have ambient, hot and cold temps (from the data) to show the 'others'.

We are worried about other issues also - condensation, security, etc..but this is his main focus right now.

sooo....
What do you think about his plan?
What types of differences do you think I will see?
And lastly, would change any of his plan?

Thanks for your help - and looking forward to your responses and ideas,

Kevin

[John Kingston]

I'd try it. Keep the data between you two for now. Maybe do a more controlled study of multiple measurements. This looks like one for Hilton.

[kbotta]

Yep, he approached me with a "covert" operation - so it wil be between just us for the time being...

[ag162]

Hey Kevin,
Not many of the ppl here have an idea of the Georgia sun. However, this is the best time to do this kind of test. July and August sunshine could be too much for it. As for the varying environmental changes, I ( in my opinion ) believe that the measurements will be based on the material of the fixture. At my previous job, the fixtures stayed on the production floor and that was the best way to check a part coming off of a prog die at 80 deg F. Which is ok for a wide toleranced part. But if you are in a tight tolereance situation,,,,,,I wouldn't do it. The fixture may not be the same after the wide variables and will need some adjustment to be certifiable again.

A.Gore

[kbotta]

Alton,
I never even thought of that. Very good point - and thanks.
I'll use a obsolete gage.

Also, it is an aliminum base stamping gage.

[Paul Sarrach]

Other than humidity, causeing rust, and most gages are not made of steel any longer. At what temp the storage is should not come into play if you give them enough time to soak in the lab to temp.

I do not see you getting any different numbers once it get back to room temp.

But I have been wrong before. It times like this when its only an opion, that I get nervous. Because before our CMM God speaks, he usally has some sort of data to back him and I am usally wrong

[Hilton Roberts]

If the goal here is to justify a controlled environment for your cmm, then what you need to do is take advantage of the variation that you know exists and use it to justify your position.

In your controlled environment, qualify and measure at the same speed and make sure the part you have is stable.

When you simulate the part either being extremely hot or cold, calibrate fast and run at a different speed than you calibrate at and change the prehit if you can get away with it.

If you can select the part to be measured, get a part made of aluminum and as large as you can get on the machine. The COE of aluminum is about 13 microinches per degree times the length of the part.

Because aluminum reacts quickly to temperature change, make your program to check the distant features first as the temperature effect will be more pronounced there. In other words align the part and measure the features on the extreme ends of the part and then work toward the ones closest to the center of mass. The change is proportional so you will see the most variation there. In a controlled environment, the change is still proportional but the part will not be changing nearly as much as one either cooling off from being outside or warming up from being taken out of the cooler.

The key is to use the variation that you know exists to help you make your case. Give me a call at 480-891-7263 if I can help.

I really like this thread......

H

[kbotta]

Ok, great advice!
The part to be measured is going to be an actual check fixture (gage) with hopefully some cnc cut profile areas, stabs, go-nogo's etc...
So - by changing the prehit - do you mean by making it bigger, or smaller? What effect will this have - except an amount of variability? I don't want to introduce something that is not 'fair' to it, but to show em it's not a good idea. A true study if you will.
What do you think?

Great advice about grabbing the large length's first...
(the gage has to be small enough to put into the fridge)...
Wish I had a walkin cooler

Kev

[Paul Sarrach]

Kev. do they want to move just the gage crib, of the CMM too?

[kbotta]

Paul
what I am afraid of is, when they set up a die (that whole 1 minute change over thing) they don't get the gage till it's time fo rthe 1st piece inspection. They could potentialy grab an ice cold gage (warehouse doors are never closed, it gets down to the teens in the overnight here at times) and try to measure it with attribute techniques - and some bright "smarter then me" die repair man is going to whip out the old die grinder and go to town -because Billy jo said so.
Get my drift? I need to prove this.
They are not prepared enough, or given enough of a warning to 'prestage' the c.fixtures. This would require additonal storage - which they tell me no-no on...
Thanks for the rust issue, that is one of our 'pocket aces'.
Keep em coming guys, I'm counting on ya...

Kev

[kbotta]

I don't think the CMM. I wouldn't work out on the shop floor. Not my cup of tea. I need the AC (high blood pressure) and I'm not from here (georgia heat).
I'd wither away and die....

They would have to build a new building for that (controlled) and I don't see them doing that. - great -now I won't be able sleep - thanks Paul....J/K

[tlnzz]

Sounds like someone is going to great lengths to get you out of the bulding and as far away from manufacturing as possible. LOL

[Hilton Roberts]

You don't have to have a walk in cooler. You can use a co2 fire extinguisher or get dry ice from the store and pack around the fixture or part you select to cool it. Put a couple of thermometers inside and insulate them with silicone caulk on the top of the bulb and cool the part until the core temperature read on the bulb gets down to the test temperature you need. For aluminum, it will not take a lot to introduce a large change. The idea of changing measure speed and calibration speed and prehit distance is just to introduce more variation into the process.

Then there is the old tried and true trick of the halogen lights to better "see"....
They introduce an enormous amount of thermal energy and if they happen to shine more on the bridge.......on the free end......goodness sakes alive...that sucker moves up and bows.......who woulda' thought.....

Those lights can simulate sunlight in an open shop that radiate on the machine at different spots through out the day with different intensity.

You should have no trouble making your case.

H

[Lee Johnson]

Aluminum expands at .00001244 inches per degree per inch depending on alloy.
If you have a part that is 10 inches long and it is made of aluminum a 20 degree change in temperature will result in a .00248" change in size. I work in the range of .0005" or less all the time so temp is very critical to me. If you have a +/- .025" tolerance you will be ok. Most steels fall at about half that somewheres between .0000067" and .000008" .
All ferrous gages will rust when exposed to extreme temperature variations just from condensation nevermind the uncontrolled humidity. You will need to coat everthing with a heavy coat of some sort of grease to protect them. or you might as well just throw them away. Many of my tools got ruined becauase it was raining on the day we moved into a new building and that was just overnight.
Hope this helps
Lee

[John Kingston]

The idea of changing measure speed and calibration speed and prehit distance is just to introduce more variation into the process.

Then there is the old tried and true trick of the halogen lights to better "see"....
They introduce an enormous amount of thermal energy and if they happen to shine more on the bridge.......on the free end......goodness sakes alive...that sucker moves up and bows.......who woulda' thought.....

Those lights can simulate sunlight in an open shop that radiate on the machine at different spots through out the day with different intensity.

You should have no trouble making your case.

H

Hilton,
Are you suggesting slanting his study in his favor!!
A true Statistician! Use the numbers to you advantage.
Great ideas.
John