Never Mind RE marty007's point 2: How is it even possible to "back-calculate exactly what load the flange can take"?
Expression in CC2901 is sum (sic) of moment and axial force terms. You can have of all of one, none of the other. Each term may be a resultant of other terms, for example, L, C, R directions.
I agree there needs to be provision for replacement equipment, successful service.
Regards,
Mike
The problem with sloppy work is that the supply FAR EXCEEDS the demand
Thanks for dig1 & martty007 valuable comments.
I have had the same experience as martty007, it did not pass by vessel engineer but passed by piping engineer in different ways to calculate. Before both use a same way, to meet vessel engineer requirement it has to lowdown the external loads…..but I have not seen any cases to increase the flange rating decided by piping due to the nozzles external loads. For this issue, piping engineer play a major role and PV engineer just check, that is just because they use different ways to calculate. It should not need PV engineer to check if the both use a same way for CC2901.
To settle this issue in roots is using a same ways to consider the nozzle external loads on the flange by ASME VIII (pressure vessel)and B31 (piping).
Marty007 hit it on the nail with 1 and 2. Also with replacement coil sections. As an engr for heat exchanger appendix 13 plate box headers,this is exactly how 90% of our customer purchase our pressure vessels. It will take extreme education to get all the competitor bids to be apples to apples and I assume we would take a hit while others ignore 2901. With the understanding of the intent(as a fabricator I definitely agree and understand your 2901) its not uncommon for my customers to want 2-3 X API Nozzle loads or some arbitrary loads as they do not want to calculate the actual loads till much later. Then 12 weeks later they design the piping around it or do not even tell us the action piping loads. How do I avoid charging the customer for a stronger flange design then he is going to use? The simple answer is he must tell me the actual loads, but this would be a culture change.(although welcomed) We currently are still trying to figure out how to approach this issue. The customers that want 2901 calculations are also abusing the understanding of working together and want to request 2901 review only AFTER pressuring us into agreeing to 2-3X API nozzle loads based on PRG's 661PRO(FEPIPE) or FEA analysis. What I am noticing, similar to the above, smaller flanges <4" are failing. Example: 3" 150# @300F SA105 Pd=150psi, nozzle loads will have a API Nozzle load Fyaxial= 760ft-lbf and then M1z and m2x of 600ft-lbs each. Fm=1.2 gives us P2901= 506 psi. But the Pm = 16m/(pi*G^3)= 593 psi. Without even solving for Pf, I'm already failing at 2 X API Nozzle loads and forced to provide 1X only or redesign to 300# flange ratings.(check me please) Since API Nozzle loads are arbitrary at best, it almost seems that when requesting 2901, they should supply us the Actual Pipe loads so I can use 2901's Fe/M1/M2 values correctly. right? Otherwise, I only have two variables to change...decrease Fe/M or increase flange rating per 2901? I can also see heat exchanger industry pushing back that the coolers not be used as pipe hangers...which I would be fine with on floating headers.
im interested in others thoughts....especially when it comes to back calculating all active jobs and replacement jobs in the field.
LOL, I had semi assumed this but I also thought maybe I could just use the mean inner and outer diameter of the gasket dimensions.... What else have I not considered
I'd just prefer to use a fictitious gasket mean diameter to go with the fictitious nozzle loads to tell the customer his specified flange rating is insufficient
Edit: and he's getting the adder. Double
Regards,
Mike
The problem with sloppy work is that the supply FAR EXCEEDS the demand
As a fabricator for a customer that is not going to provide gasket/piping forces/etc till the units are shipped(some even by packages that have no end user yet and are for rental fleets), how does API's nozzle load tables play into affect? Is the idea that we should no longer quote arbitrary API Nozzle Load table multipliers? If I do, then I could see automatically jumping a flange rating even when its not required at the end of the day. OR can we interpret API Nozzle Load tables as referring to the nozzle pipe to pressure vessel attachment, and the flange to nozzle pipe attachment would follow Code Case 2901? The issue is in the Code 2901, if i use API Nozzle load table as the axial and moment force,I'm showing failures on anything past times 1 of the table.
Whenever calculation according to CC 2901 is not satisfied, do you think it can be considered acceptable if I perform an alternative check according to Div.1 Appendix 2 considering an equiv pressure calculated on the basis of the additional loads FA & ME OR according to Div.2 Para.4.16, using y=0 MPa?
I mean, I'd try to calculate my std flange according to Appendix 2 or Para. 4.16, taking into account the effect of the external loads, but excluding the seating condition (since no external loads are expected to be applied during the seating phase).
For sure it would be only an attempt to solve the problem in case the seating condition is the governing one.
Interesting feature (bug?) with ASME Section VIII, Division 1 is that is actually no role for the buyer/owner/user. Just the manufacturer and the AI - from the strict Code perspective. Any owner/user/buyer involvement is strictly commercial/contractual.
ASME Section VIII, Division 2 doesn't have this feature (bug?). There are defined roles for the User (or the User's designated agent), which does give them Code-enforced requirement to accept/reject things. This (among many other reasons) is why I recommend Users to get vessels designed and fabricated to ASME Section VIII, Division 2, Class 1 as a default, rather than Section VIII, Division 1.
(Perhaps we take this specific discussion to a separate thread...)?
TGS4, you must remember that Sec VIII, Div 1 is also used for mass produced items where no contract is entered into with a buyer. Nevertheless the equipment must be designed and fabricated to some recognized standard. This Code (or something like it) must accommodate such equipment.
The problem with sloppy work is that the supply FAR EXCEEDS the demand
TGS4, I agree, only between the AI and Manufacturer as they are the jurisdiction.
but i have question on your "TGS4 (Mechanical)25 Apr 19 23:01
SnTMan - only with respect to contractual requirements. That is not a Code requirement. Unfortunately."
TGS4- when you mention having customers stop using arbitrary nozzle load tables, which i agree, but once this gets implemented into 2019, how are you expecting OEMs to handle API Nozzle Load tables. Take full exception or just beef up the coolers? Its not uncommon for us to sell 20-40 lease compressor package pressure vessels that have no actual end user projects assigned to them yet. These vendors want the vessels rated to 150# or 300# so they can use them in the next 2 years as jobs come up.
