Designing a custom pipe run with potential measurement inaccuracies

[Obscurity]

Hi all,

I'm a relatively young engineer, and I'm looking for some advice. I've been tasked with designing a steam pipe run that leads from a newly installed steam drum to an existing pipe connection across the room. The piping consists of multiple 6" dia carbon steel spool sections with flanged connections for various stop valves and flow meters. My plan is to model each pipe spool as accurately as possible, have them welded off site, and then have them shipped back for install. Modeling the piping itself is fairly easy, but because the start and end flanges (steam drum and existing pipe connection) are so far apart and in slightly hard to reach locations, its hard to get a precise measurement of each flange's location in regard to each other. My tech's have provided dimensions as accurately as possible with a tape measure, but I'm worried that the tolerances involved in that process will cause the flange connections to be misaligned slightly and cause issues during install. So I ask the question: how would you approach this issue?

Option 1: Figure out a way to get more accurate flange measurements. Any recommendations for this that doesn't involves really expensive surveying tools? Also, how much tolerance do you think I would need for a pipe assembly of this size to be properly aligned? Is there any wiggle room, or does it need to be pretty tight?

Option 2: Get a pipe fitter on site to cut and weld the last few pieces as needed. This one won't be very feasible for this job, but I am still eager to learn about the process for future projects if this is what you would do.

Option 3: Add in a few expansion joints/adjustable fittings in both the horizontal and vertical sections. This would be a great option because it would allow my techs to adjust things on the fly without the need for welding. I simply just don't have any experience on the matter and can't find any products that are used exclusively for this purpose. I am aware of expansion joints for thermal expansion and things like that, but could they also be effective for this application? If so, any recs on specific products/types of fittings?

Main goals are minimizing the amount of back and forth between us and our welder, as well as making my technicians lives as easy as possible during install

Any advice will help. Thanks!!

 

[Compositepro]

Sounds like my first project out of school in a chemical plant. Replace several thousand feet of corroded 3" vent piping that ran the whole plant with new Kynar lined pipe. Just used a tape measure and isometric graph paper. I had to crawl though many crowded pipe racks. You, or your technicians, have to be very, very meticulous. Also have a supply of spacer spools ready at installation. It definitely helped that I was a gymnast in high school. That was in 1982. Installation went very smoothly using the plant's staff of very experienced pipe fitters. I'm not really aware of an "easy button" for this type of work except pay someone else to do it.

 

[LittleInch]

When you have a complex bit of pipe work with fixed end points, you basically need to design the isometric with some "green lengths" in X, Y and Z and ditto any angles if you have then as well. You need to make these big enough to accommodate the final size, so minimum length about 500mm. Often shown as something like 1250* as the dimension. don't try and go fitting to fitting without some spool piece able to add or subtract a certain amount

Then build it out from your fixed points and then measure the final tie in spool super accurately and then tack it in place and offer it up.

All pipe spools have tolerances and things like gaskets are allocated 3mm, but might be 4 or 2 once they've been compressed. So trying to get every dimension super accurate (+/- 2mm) will just leave you cursing and swearing trying to fit the last bit in. Don't even try.

Expansion joints are not designed to accommodate this sort of misalignment and movement and cause other problems as they then need to be restrained as otherwise they start to move all over the place when you add pressure or temperature to the system.

If you get a 3D laser scan you might get pretty close, but even then, there are just too many tolerances with a multiple spool system.