Get a fitting chart showing Tees with all standard branch sizes.
Chose one of the standards, if it exists.
Otherwise use a weld-o-let.
**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)
I have never used a pipe-to-pipe connection on a project in Oil & Gas. I know they are allowed up to full diameter to full diameter, but that is the weld that busts more welding tests than any other so I'm a bit afraid of getting it done right. The profile of the end of the stub seems to be a difficult concept for a lot of welders. In the old days people would shove a square-end into the run an weld it on. The intrusion in the flow can really tear up a pig running by, I haven't seen that technique called out in a decade.
The company I worked for used tees where they were available (you can usually get a reducing tee with the branch 1/2 or greater than the run) and weld-o-lets where tees are not available. I've seen similar rules at other companies.
If they don't make it, you'll have to use a w-o-l.
I use Tees when the outlet diameter is > 1/2 straight run diam.
**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)
As an aside (since BigInch's link included them), I love reducing elbows. I once did a very precise measurement of the pressure across a reducing elbow vs. a concentric reducer before or after the elbow. The instruments were calibrated to +/-0.001 psi. The test was trying to change a policy that excluded reducing elbows so I had latitude to spend a bit of money, but it wasn't unlimited.
I ran the test at 4 different inlet Reynolds Numbers (all in gas, all turbulent) in each direction. For the reducing tee in reducing direction (6 inch down to 4-inch), pressure increased at each Reynolds Number. For the reducer-and-elbow in the reducing direction, pressure decreased at every Reynolds Number (my hypotheses was that we got boundary layer separation in the reducer that added to the dP, but I never did further tests to asses the hypotheses). Going the other way, the reducing elbow gave me a lower dP than the reducers and elbow every time. In short, the reducing elbow out performed the company-approved methods in both directions at every Reynolds Number tested with 6-inch and 4-inch pipe.
Granted, I needed every bit of my calibrated precision to show the differences (they tended to be on the order of 0.01 psi difference), but I made my case and got the stupid company standard changed. My real point was that it saved a weld, but better hydraulic performance was valuable icing on the cake.
Before anyone asks, this report was company internal at a time that the company did not allow publishing any technical data. It isn't published anywhere and I couldn't give you a copy if I still had one.
Were your test points at the start and stop points, or did you measure pressure T several points going through the elbow, or the reducer and elbow? (Just being curious.)
I built spools for each configuration and took my data points 4 pipe diameters upstream and downstream of the minimum end of the radius (assuming the minumum end of the radius was based on 1.5D of the 6-inch pipe). No intermediate data points at all because I was concerned about the Heisenberg Uncertainty Principle (i.e., putting in thread-o-lets could easily cause pertubations that would be bigger than the changes I was looking for).
What do the Pipe Spec's call for? If you don't have Pipe Spec's., you'll need to hire Piping Engineers to write them for you. There is no general rule of thumb on when to use weldolets, sockolets, thredolets or pipe stubins or even tee's. Those decisions have been hashed out and decided by the Engineering/Purchasing Groups in the Pipe spec's.
zdas04, Now try and find reducing ells on the shelf at suppliers or the manufacturers. To special order they will tell you that there is a nine to twelve month lead time. Can the project wait that long?
That is true in some sizes, but I've installed several hundred 3X2, 4X2, 6X6, and 8X6 reducing el's using fittings that my local parts house kept on the shelf. I installed one 36X30 in a drip that took 6 weeks to get, but the 30 inch pipe cap took two months.
Same with reducing tees, but that is another story.
API RP 14E has a table showing recommendations for when to use O-Lets vs reducing tees. This is not a mandatory set of requirements, just recommended practice.
