Scribing Parts

[mabeasch]

I have never seen this done with the CMM and I am looking for some help in this matter. We have a spoiler that the DRE wants to have blue and scribed parts for a build event to match up the body lines from mating parts. I do not even know where to start.

Thanks

[craiger_ny]

I used to scribe parts with the CMM with a hard probe scribe. As long as everything is done in the same tip rotation you can do it without qualifying. You need the "Take hit" short cut key for doing your alignments. I can't remember what that key was or if it is still available. If it is not then you will have to do all read points for alignment, just be sure to offset the scribe diameter/2. Best to scribe points than lines, then use a template to connect them. This is all "invented" by me as I never had anyone show me how (presumably the right way) but it works. Perhaps someone here has a conventional method that makes more sense.

I have only done "2D" scribing, in other words 2 axis I am not sure what to tell you if you need all 3 axis. You can travel in all 3 as long as you are scribing the whole time in the same 2.

Craig

[Matthew D. Hoedeman]

Using a CMM for scribing a part is a BAD idea. THe machine was design and made to be moved by the controller, NOT by hand and to scribe, you have to move it by hand. This is the easiest way to throw the machine out of calibration. How are you going to move it by hand? By the quill? That could throw the XZ out of sqaure REAL quick. The old-style controllers (real old) had a button you could press to 'bypass' the probe tip requirement for the machine to move, but not the new ones (Sharpe32) and I think that is a good idea since you really should NOT be moving it by hand, and I do NOT mean the joy-box, that is STILL the controller moving the machine. You should NOT grab the machine and move it by hand.

[Andrew Gardner]

Closest I've come to that, was to calibrate a small probe, and bring it down close to the part, and scribe a mark by hand - eyeballing the center of the probe. It usually ended up a within .005 of were it needed to be. I didn't have the option of a hard probe, and didn't want to clamp anything onto the PH10 or Z axis.

[craiger_ny]

Come on man it is the same as manually measuring, there is nothing wrong with manual. What do you think people did for the 30 or more years before DCC became the mainstream? If your machine allows manual measuring then it is perfectly fine. Just do like I said and only scripe a small point to connect later.

Matt what is your beef with manual measuring anyway? You and I are 180º on that for some reason.

I have faith in manual measurments, I came up on a manual. We have machines however that do not allow it so no I can not condone grabbing the Z rail and trying to move it on one of those. But if the machine design allows it you are fine.

Craig

[Matthew D. Hoedeman]

Come on man it is the same as manually measuring, there is nothing wrong with manual. What do you think people did for the 30 or more years before DCC became the mainstream? If your machine allows manual measuring then it is perfectly fine. Just do like I said and only scripe a small point to connect later.

Matt what is your beef with manual measuring anyway? You and I are 180º on that for some reason.

I have faith in manual measurments, I came up on a manual. We have machines however that do not allow it so no I can not condone grabbing the Z rail and trying to move it on one of those. But if the machine design allows it you are fine.

Craig

AH! THAT is the difference!

A MANUAL machine is MADE and BEEFED UP (designed, if you will) for being hauled around manually, almost all machines that are controlled by a JOYSTICK were made to be moved around BY joystick, especially any within the last 10 or 15 years. Not grabbed and hauled around. I know for a fact that my machine is held square in the XZ plane by 4 screws. YES, 4 screws and only 4 screws and this is a BIG, heavy machine. No dowels, no keys, nothing but 4 screws going through clearance holes in the z-rail assembly into the bracket on the top of the X rail assembly. NO WAY would I trust that to hold square if someone were grabbing it and pulling/pushing it side to side, against the screws. No way! I have no problem with manually measuring things, but only with a manual machine. Most of the machines today are so light weight and wimpy, I could easily see them twisting or warping out of sqaure from someone grabbing the quill and pushing/pulling them around.

(EDIT)
BTW, we have an old portage machine we use for scribing parts, when we can't talk them out of it.

[craiger_ny]

When moving manually your input should be nearest the bearing surface. If you move in the X you should be moving with your hand on the X nearest the X rail. If you move in the Y your hand should be pushing (on bridge style) the bridge all the way down by the Y rail. The biggest mistake people make is inputting X and Y moves at the Z rail. Only input at the Z rail for Z moves and in a pinch balance and stability but deliberate X Y moves should be made at or as close to their respective rails. Sometimes when I need to vector in X and Y I input at the X rail for both axis.

If the machine supports it go for it. I agree that machine quality is down lately and probably not many disigns (especially larger ones) are considering manual measurements like they used to. I can not say bad things about the practice though if you do it smartly.

It does take a knack.

Craig

[John Kingston]

I have to agree with Matt. I wouldn't scribe with a CMM. I'm guessing Matt and I are considered "oldtimers" (40+ ain't old!) by many. Why can't you blue and scribe on a plate with "traditional" measuring tools? At times I find that to be theraputic, something we all need here in QC.

[craiger_ny]

How do you scribe out a part that is asymetrical and made up entirely of curves on the surface plate? Faster than with a CMM mind you.

I'm not afraid to be a dissenter here. I'm not old but some of the CMMs I've run are (Bendix, Cordax machines built in the 60's and 70's). I will not say a CMM should replace the fundimentals I am a proponent of being able to inspect on the bench everything you inspect on the CMM. But when you have done your time and you understand the fundamentals (yes I am a good "bench inspector") you use the technology in concert with that fundemental knowledge.

Blue that sucker up, put it on the CMM and scribe away. If you don't know what you are doing on the bench you won't know what you are doing on the CMM and it will be screwed up either way.

Craig

[Jan d.]

I have heard from someone that makes something like a "stamp". You can probe change it is and go to a location and "press" it on the part. It leaves some ink. based on the location where the ink spot is, you can then sort the parts. This device was very cleanly springloaded, so you do NOT hurt your CMM.

I did some quick searches on Google, but for the life of me can NOT come up anymore with who made this thingy. Maybe it was homegrown... Shouldn't be difficult. Take one of them Renishaw stylus changer modules. Bolt a spring/stamp assembly to it. You can even locate an ink supply on your machine. So the routine for that stylus would pick it up, go get some ink and then deposit a stamp in a certain location.

Disadvantage: can be rubbed off quite easily.

I am old too.... I'd be hesitant to scribe with my CMM. Craiger knows what he is talking about though. Especially when it is about beer. Was this about beer? I forgot. Early Korsakow syndrome I guess....


Jan.

[Matthew D. Hoedeman]

How do you scribe out a part that is asymetrical and made up entirely of curves on the surface plate? Faster than with a CMM mind you.


Craig

Well, for about 99% of the jobs we have had in here since I started, we have had dedicated holding (checking) fixtures. Simply place the fixture on the protage table, square it up, put in your scribes and scribe away. 95% of the jobs we do don't require blue and scribed parts anymore. The only jobs we have had that didn't ahve holding fixtures didn't require blue and scribe. One of the benefits of automotive, I guess, holding fixtures.

Now, to get into technical aspects:

You can scribe with a DCC driven CMM if you have some technical and mechanical skills WITHOUT ever touching the machine yourself. I have designed (but not built since the portage table still works) an attachment to put on the quill of the CMM that will hold the scribes and the scribes will pivot like they do on the portage machine, thus allowing you to DCC move the machine and have the scribe scribe on the part. Not exactly sure if it will work as I have planned, but just about anyone COULD come up with their own design to fit their own machine. AND, by doing it this way, you could:

1) Align the fixture with the regular tip
2) Remove the head from the quill
3) Attach the attachment
4) Use a 4-way pin (or simular item on the fixture) to set the scribe and offset the alignment
5) Program in some basic moves and have it scribe
6) repeat for each of the other 2 axis

Sounds like a scribers dream, but it is really still in the dream stage for me at this point in time.

[craiger_ny]

Never mind. I don't know why I bother. You can not scribe with a CMM. It is not only impossible but it is morally unacceptable and Illegal everywhere except Malaysia and somewhere inside the Arctic circle I forget exactly where. If you were able to do it, it would ruin you machine, your reputation and your life, also your kids will not get into college.

Might I suggest taking up a life of heroin addiction and unsafe *** with African prostitutes instead, it would be much wiser than scribing with a CMM.

Sheez, I tell the guy it is legit to do something I have done numerous times and it is like I am telling him it is OK to test a nuclear devise on the Korean Peninsula or something.

I take back everything I said. It can't be done. It was all a lie to increase my postcount.

Craig

[Matthew D. Hoedeman]

Never mind. I don't know why I bother. You can not scribe with a CMM. It is not only impossible but it is morally unacceptable and Illegal everywhere except Malaysia and somewhere inside the Arctic circle I forget exactly where. If you were able to do it, it would ruin you machine, your reputation and your life, also your kids will not get into college.

Might I suggest taking up a life of heroin addiction and unsafe *** with African prostitutes instead, it would be much wiser than scribing with a CMM.

Sheez, I tell the guy it is legit to do something I have done numerous times and it is like I am telling him it is OK to test a nuclear devise on the Korean Peninsula or something.

I take back everything I said. It can't be done. It was all a lie to increase my postcount.

Craig

Dang, is it time for your ritalin man?
You did just go over a 1000 again, didn't ya? Should have made you happy, at least for a little bit!

Hey, I have no problem with YOU using your CMM to scribe a part, but to say that it is absolutely 100% possible with a cmm? No, I will have to disagree with you on that. On a manual machine, GO FOR IT, all the time, no problems. On a machine designed for strictly DCC control (either joystick or computer), bad idea. Maybe not for you, but would I trust even 1% of the people out there to do it WITHOUT damaging the machine? NOPE. Sorry, I wouldn't trust them. Hilton, Monkeyboy (kbotta, you thought I forgot about that avitar, didn't ya!), Kingston and a few others, yeah, I probably would trust them not to hurt the machines, but everybody on this board even? No, I would not be able to say that all of them could do it without hurting the calibration of the machine. HEY, nothing against ANY of you, but those are just the facts. These machines are highly critical to every aspect of OUR livelyhoods. Too many ********* already do not want to trust our numbers when we say something is out, so we really want to add the (possible) effects of man-handling our machines in ways they really were not designed or built to be used? What happens when it is time for the annual calibration and the tech says, "Hey, you were 0.040" out of square" or .100" or 0.200"? Does that give us all a nice warm feeling about all those numbers that we told the machinists (or others) were bad? Then, what happens when they come in with a part YOU rejected the day before the machine is calibrated and have you check it again and now it is good? Will THAT help you defend that the numbers from the machine as being right? There is just too much at stake! Yeah, this is an extreme case, but you ahve to try to think of ANY and ALL consequences of your actions. Sure, you can scribe a line with ANY CMM, but, what will the CMM be capable of after that?

This is one I would really like to hear Hioton's opinion of, he is the GOM (grand old man) of inspection, I am sure he has SOME opinion of this.

[Andrew Gardner]

I have heard from someone that makes something like a "stamp". You can probe change it is and go to a location and "press" it on the part. It leaves some ink. based on the location where the ink spot is, you can then sort the parts. This device was very cleanly springloaded, so you do NOT hurt your CMM.


Disadvantage: can be rubbed off quite easily.

Jan.

A couple of times, I got the probe close, put a drop of Dykem on the probe, and took a Z hit. It usually worked out alright (I was working on aircraft parts, nothing square, no fixtures). Take a few hits like that, and connect the dots later using conventional methods.

Just be sure to clean the probe when you're finished. Can't have a dirty probe

[John Kingston]

This has opened up some awesome ideas. (Nice of everyone to share. )
I've thought about it different times, but due to the type of parts, it was easier to use the plate.