[voxelman]

I am building enclosures for optical equipment that are based on 7 inch and 12 inch diameter alluminum irrigation pipe. The biggest challenge that I am encountering is how to accurately position and drill various holes on the surface of the pipe.

What techniques do you use to layout the holes radially around the tube? I know the angles of the holes around the surface but I'm finding it difficult to map the positions of the holes on the curved surface of the tube.

The holes are of two general types. Holes that are aligned radially with the center of the pipe and those that are not. The radial holes are easier to drill because the drill enters the material perpendicularly. Holes that encounter the surface away from the perpendicular tend to wander. What techniques do you use to control this problem?

[11bravo]

Let me guess, you're building a telescope and got the pipe from Hastings.
Where in Iowa are you from?
I live up by Mason City and have always thought it would be great fun to build a telescope.
I have not done what you are asking but would be willing to try to help.
Can you give me more on what you have?
Have you tried marking out the holes on kraft paper then wrapping that around the tube.
Have you established lines that run straight along the length of the tube yet?
What size holes, how many?
What kind of tools?
"REAL" tools or the hand held power tools most of us real people have to get by with?

Look forward to hearing from you,

Greg

[Geof]

I am building enclosures for optical equipment that are based on 7 inch and 12 inch diameter alluminum irrigation pipe. The biggest challenge that I am encountering is how to accurately position and drill various holes on the surface of the pipe...

You do not say if you are using CNC equipment or a manual machine. But possibly it does not matter.. in both cases you may find it easier to use a rotary fixture. This allows you to position the holes directly by angular position.

For your holes that do not pass through the pipe surface on a radial line use a center cutting end mill.

And after typing this I realise you may be trying to do this on a drill press...my only advise if this is the case is don't. A milling machine, CNC or manual, is the best equipment with a rotary fixture as I mention.

[voxelman]

Let me guess, you're building a telescope and got the pipe from Hastings.

First, thank you Greg and Geof for your replies.

I did get the pipe from Hastings but its not for a telescope. I'm building prototype weather proof enclosures for very high performance camera/lens systems. It is one of those unexpected opportunities that arise when working for a small company. Our real product is software and I'm the companies Sr Test Engineer. One of our clients had a requirement and my boss knowing of my interest in metal working asked me what I thought I could do. So now I have a sideline of developing these enclosures. OBTW Greg I'm in Fairfield.

Geof I do have access to a milling machine at a machinist we use in Ottumwa Iowa. I'll have to look into a center cutting end mill. What is a rotary fixture? Is it something that needs to be custom made for each diameter of pipe? Also, the 12" pipe would be a tight fit under the head of his small Bridgeport but I could see it under the head of his big one. We actually used the small Bridgeport to drill and counter sink some of the holes in the smaller (7") tube but we just set the tube across the open jaws of the vice and basicly used the mill as a drill press.

I guess I know part of the answer to my question is to do the math to determine the hole positions from the angle off the center line and the diameter of the pipe. What I'd really like to do is figure out how to plot a template for the surface of the pipe using my 3D CAD software (Alibre Design Expert).

[Geof]

Picture shows a rotary fixture. Rotary table, or 4th Axis at right hand end, tailstock bearing at left.

For your application you would probably be able to hold your pipe by using wooden discs slipped onto a shaft between the Rotary table at one end and the tailstock bearing at the other. The 4th axis gives you full program control over the rotary, angular, position of this axis. You would designate one hole as the reference hole and determine the position of all the others with reference to this hole.

[DareBee]

In SolidWorks I would use either the wrap feature or even design as sheet metal and flatten into flat panel.
(For making the template).

Not much math involved manually.
For radial holes - all you need is to figure the arc length from 1 hole to the next.
Off center holes will work the same way, rotate your arc length and then move off-center the correct distance.

It may simplify your work if you were to attach a flat bar (or something) across the center of 1 end. Use that "flat bar" as 0 degrees. Now you can also use a protractor to set up your required angles.

These are fairly crude DIY methods.
Geof or I would put it in our 4 axis CNC's make it dead nuts.

[voxelman]

Geof and DareBee,

Thank you for the additional input.

DareBee - I haven't put any time into your suggestions for the layout template but I will.

If this project goes anywhere and I have to do volume production it looks like the rotary fixture is the way to go. Also, since we are turning aluminum end caps for them we can easily make a pair just for use in the fixture with indexing designed in. The end caps are attached using 4 allthread rods running the length of the tube inside with the front end threaded into the end cap and the rear end bolted through using SS barrel nuts from McMaster-Carr.

I have the prototypes finished and powder coated. Boy does powder coating cover up a multitude of sins. The Hastings pipes have a lot of surface texture from the forming and what I'm pretty sure is a weld line but the powder coating hide almost all of it.

[Geof]

Geof and DareBee,

Thank you for the additional input.

DareBee - I haven't put any time into your suggestions for the layout template but I will.

If this project goes anywhere and I have to do volume production it looks like the rotary fixture is the way to go. {b}Also, since we are turning aluminum end caps for them we can easily make a pair just for use in the fixture with indexing designed in.[/B] The end caps are attached using 4 allthread rods running the length of the tube inside with the front end threaded into the end cap and the rear end bolted through using SS barrel nuts from McMaster-Carr.

I have the prototypes finished and powder coated. Boy does powder coating cover up a multitude of sins. The Hastings pipes have a lot of surface texture from the forming and what I'm pretty sure is a weld line but the powder coating hide almost all of it.

Careful now you are starting to think like a machinist; actually considering how to make production more efficient! Horrors!!!!... engineers aren't supposed to think like that, they are supposed to make things more complicated and less efficient

And yes powdercoating is a good coverup, just make sure they prep the aluminum correctly so it does not chip off.

[voxelman]

Careful now you are starting to think like a machinist; actually considering how to make production more efficient! Horrors!!!!... engineers aren't supposed to think like that, they are supposed to make things more complicated and less efficient

And yes powdercoating is a good coverup, just make sure they prep the aluminum correctly so it does not chip off.

Well I majored in civil engineering, you know dirt, steel and concrete. I've always been very hands on, worked in construction and have a carpentry shop.

I've only become aware of how cool metal working is in the past year. My first exposure was when I designed a filter mechanism for a camera. I did the design in Alibre and sent it off to a machinist in Colorado. When he saw the design he said it was kind of limber. This was a totally unfamiliar usage of the word limber for me. I asked him to explain what he meant. Thus began my education in metal working. I have found the whole experience very fulfilling. Every machinist I've worked with so far has responded positively to my questions about what I can do to make the design better and their job easier. I see them as my hands and I try to apply what I've learned to improve the quality and reduce the cost of the equipment I design.

The powder coating shop I work with performs a pre-etch of the parts prior to coating. So far I've been very pleased with the quality. I live in rural Iowa and word travels fast when a shop does shoddy work.

[Geof]

It is difficult to find a good link between Civil Engineering and this:
"building prototype weather proof enclosures for very high performance camera/lens systems. It is one of those unexpected opportunities that arise when working for a small company. Our real product is software and I'm the companies Sr Test Engineer"

[Geof]

Because it is Friday evening and I have nothing better to do I went to find out what a voxel is...and found this:

"A voxel represents the sub-volume box with constant scalar/vector value inside which is equal to scalar/vector value of the corresponding grid/pixel of the original discrete representation of the volumetric data. The boundaries of a voxel are exactly in the middle between neighboring grids. Voxel data sets have a limited resolution, as precise data is only available at the center of each cell. Under the assumption that the voxel data is sampling a suitably band-limited signal, accurate reconstructions of data points in between the sampled voxels can be attained by low-pass filtering the data set. Visually acceptable approximations to this low pass filter can be attained by polynomial interpolation such as tri-linear or tri-cubic interpolation."

Now if you can explain this is words of one syllable. and using fewer than a thousand or so, I promise I will genuflect in the direction of Iowa at least once or twice in the next day or so.

I'll bet they don't use this kind of jargon in Civil Engineering.

[voxelman]

Well the courses that I really nailed were all Industrial Engineering computer programming related. When micros first made their appearance in 1975 I was captivated and threw myself into learning the related electronics and software topics. To keep a long story short, construction and Civil engineering were left far astern. I worked in a variety of early microcomputer related ventures and then in 1990 I went to work for Vital Images, Inc. as an applications engineer. The company's focus was using high performance workstations (Silicon Graphics) to visualize volumetric (3D) data. As an applications engineer it was my job to explain not only what a voxel was but how a volume composed of voxels could be displayed in a way that allowed it to reveal its secrets.

Numerous techniques are used to acquire volumetric data. One of the most common is the density measurement procedure know as a CAT scan. A CAT scan is typically performed using an X-ray source and a linear array of X-ray detectors that are rotated around the object to be visualized. Measurements of X-ray intensity are taken at different angles around the object and mathematical techniques are used to backtrack to the density of the material at points between the x-ray source and detectors. The X-ray source and the detectors have finite size and number and therefore the measurements that are resolved from them also have a finite size. In current use this results in an X by Y image resolution across the space being analyzed of about 1000 X 1000. The term “finite” is also important because the mathematical processes use a large amount of computer power. The end result of the process is an array of about 1 million numbers each representing the average density of a small volume of space.

The conventional view of an X-ray image is as a 2D picture composed of 2D picture elements or pixels. Each pixel is assigned an intensity or gray scale value in the image that identifies the average density of the region represent by the pixel. However, as can be seen from the above description each picture element also has a finite depth and a resulting volume. From this is derived the term voxel or volumetric (picture) element.

Characterizing the qualities of a voxel and the relationships between voxels is an area of intensive study. In this example accurate density is required to correlate the various image regions with blood, bone, muscle, other tissue and regions of mixed composition. Regions of mixed composition are particularly problematic because they produce nonspecific densities. The primary way to address this problem is to increase resolution as the resulting regions of mixed density become less significant.

Continual improvements in computing resources have allowed the combination of many individual volume slices into extended spatial volumes that with proper encoding can provide surprisingly realistic anatomical representations. These aid the radiologist in his role of communicating the nature and extent of injury or pathology to primary physicians and surgeons. Furthermore, CAT scans are used in engineering as a nondestructive testing tool for high value/risk systems such as the solid fuel rocket motors used in man rated systems and fan blades used in jet engines. See www.vitalimages.com for examples of current 3D medical visualizations.

When Vital Images moved from Fairfield, IA to Minneapolis, MN in 1998 we parted ways. I was unwilling to accept a 45 minute commute to the workplace and the greatly increased costs of living in a large metropolitan area. I provided consulting in the field for a couple of years and worked as a manager for a local competing telephone company for three years after that. Then I received the proverbial offer that I couldn’t refuse. The owner of a small business that had spun off of Vital Images early on asked if I would like to work for him on a wide range of projects with a primary focus of software testing. I've been with Vaytec for 2 plus years and needless to say my job description has expanded. You can see some of what we do at www.vaytek.com.

So now you know the rest of the story… Hope this long ramble has proven interesting.

Murray aka voxelman

[Geof]

Oh dear now I have to find what direction Iowa is from here . Although you used more than one syllable in places so you so only get a brief nod not a full genuflect.

I know a little bit about CAT and MRI scans. I am a Machinist, the old fashioned 5 apprentice type, but I got bored many years ago. In '75 I was part way into a doctorate in Biological Chemistry. One of the profs I knew at that time was doing NMR on polysaccharides and glycoproteins. Some of his stuff was subsequently used as background information when people started developing Nuclear Magnetic Resonance Imaging. I kept track of things for a few years because it sort of amused me; this guy was in some ways derided by his fellow profs because he was just characterizing proton resonances in various environments. Nobody could see any application for the information then but once the imaging application came on the scene it did serve a purpose. I finished by degree and left academia, I had got bored again , not really, I didn't want to leave BC and I had irritated so many people at the university I could never get a position there so I started my own business.

[voxelman]

Just nod towards the East

From Vancouver: Azimuth roughly 80 degrees East magnetic, range roughly 1600 miles.

Points out how important basic research is. You just never know where the next breakthrough is going to occur.