Tube Alignment

[dixie11]

I am fairly new to this stuff, and there seems to be a lot of people on this forum with helpful advice.
We have a tube with two brackets, and I am trying to develop a useable alignment to perform a full layout. There at least 4 bends in this tube. The 1st cylinder center line is Datum A, and then there is a collar on the cyliner which is Datum B. The one bracket face is on a 45 deg angle from the centerline of the cylinder, and the other bracket is 90 deg.
I have had difficulty running a iterative alignment. Any suggestions?? Someone did mention to me about using a best fit alignment, but to be honest I have no real idea how to use that type of an alignment.

Any help would be really, really appreciated!!

Thanks in advance
Jeff

[Wes Cisco]

Jeff, I am not in the Automotive field. I do everything with the good old 3-2-1 method. It never fails me.

3 = 3 points for a plane
2 = 2 points for a line
1 = 1 point for origin

level to the plane, (or cylinder), rotate to the line, (can be constructed, between two circles for example), assign origin points. (for true 3-2-1 you must construct the point that is zero for all axis, with pc-dmis you can have a different feature be zero for each axis, saves some time on unnecessary constructions).

HTH (Hope this helps)

[Ray Wilson]

Tubes are a very funny animal and is one of the worse things a new user could be trying to get started on.
What I have done in the past is run a much smaller alignment at the very end of the tube. Once I have this and it works then move out to take in the other area of that cycl. Do NOT try to get to agressive and think you are going to do one alignment for the whole tube. Once you have that small cycl measured then let the machine do a search with a wide prehit to find the next cycl. Once it has found it then do a whole new alignment and measure what you need in the area of that cycl. Continue to do this all the way down the tube for each cycl. When you have gotten to the end and everything is measured then build one more alignment and this will be the one that everything is dimensioned to. Normally I like to dimension as I go, but this is one time I will purposely wait until the end of the program before I ever try to dimension. In fact I don't even do all the construction of the intersections until that finial alignment is done. Just remember these tubes are very easy to bend and because of that you have to write a program that can handle all of that, so putting in large prehits and searches is a must to find that cycl you are working with at the time. Oh and one more thing, when you have that finial alignment in place be mindful of your work plane when doing the dimensioning, it makes all the difference in the world.
Good Luck!

[dixie11]

thanks

hey guys thanks for the replies. I do have an iges file for the part and I have tried to use an iterative alignment, but with no success. I guess what is really confusing me know is how do you do a search as Ray indicated. As I said earlier I am fairly new at this and have doen well with most sheet metal parts, fixtures, etc., but it is a unique challenge to program this tube.

[Jeffrey Lovely]

Keep in mind that if using a best fit - use it as a rough alignment
lets say point A = 0,0,0, and point B = 100,0,0
part measures 110 A= -5 B= 105
when actualy the profile of the end is + 10 not 5

I have measured plenty of tubes- they are tricky
I was doing some very Long exaust systems
I never got to use this software when doing so
But Iterative should work very well for you
you are still leveling to 3 or more points
rotating to 2
origin to 1

[Don Ruggieri]

Tubes are your friend

I've worked with tubes in the past, with ni IGES file. It's difficult to drive (with the jogbox) at the compound angle to take good hits on the cylinder section(s). Here is a trick I used.

1) Measure a section as a cylinder with 3 hits each on two cross sections. This should solve as a cylinder.

2) Do Alignment/Level to that cylinder, set the origin to the cylinder also in X/Y/Z. You can ignore the rotation for now.

3) Keep in mind that Z is along the cylinder centerline. Program a new Auto cylinder with the same diameter as the old. Add about 1/2 the length of the cylinder to Z to get the "top" of the cylinder. Define it to measure as many level as you need, but only using the top 180 degrees that are accessible. Let it execute this feature slowly to see if it is what you want. All hits should be perfectly normal, spread evenly, levels evenly spaced, etc.

4) Once you have what you want, delete the original cylinder and the "temporary" alignment to it. When prompted to update commands below, say YES. This will put the newly created cylinder back in the original coordinate system.

This works best if you are adding each new cylinder to the end of the program, so as not to get confused when you are adding and deleting alignments as I described.

Also, it has been a while since I've done this, but I think I've described it in sufficient detail. Any questions, just ask. This made it real easy for me to get good looking (measurement wise) programs with evenly spaced sections and good normal hits on each.