We installed steam turbine inlet piping with spring locked near stg nozzle and showed rotating equipment people flange alignment with nozzle flange opened.

Now, when the spring is unlocked, the nozzle flange gap changes from 6 mm to 50 mm and flange alignment with springs unlocked is not achieved.

Will I change spring with higher size spring to keep both cold and hot load within allowable range and cold load of the spring determined at site with spring unlocked and 6 mm gap achieved.

Other approach proposed is to add (50-^0=44 mm spool by cutting 300 mm length and adding 344 mm in vertical leg to achieve 6 mm gap with the existing spring.

Is the second approach is cold long (matl 19) to have CS load case to add in stress analysis and revalidate the analysis ?

For the sake of argument, can we eliminate the spring altogether and add 100 mm sag in unlocked condition with 94 mm spool ? After all, a spring saved can be used elsewhere!

Is the second alternative a right on neo-right proposition!

reg,
sam

There exists a third approach proposed by one of my young colleagues.

He proposed to get the alignment achieved with a higher size spring brought on loan from nearby spring stocks in the establishment. Then, find out the difference between the load reading of the spring on loan with the original cold load mark to be keptwith spring unlocked.

If the difference is low enough to accept for STG vendors nozzle aloowable Fr = 3xNEMA allowable /3, in our case, we can allow the piping with spring locked and first spring brought to cold mark by adjusting the spring load flange height in unlocked condition.

This looks to be acceptable; isn't it ?
reg,
sam

I noted that nobody has commented on this issue so far; let me explain this issue a bit more clearly.

The problem is simple - one steam turbine inlet piping was installed with all the spring locked and API 686 flange alignment was achieved at nozzle.

Now, when springs in this piping were unlocked, piping flange was sagging by appreciable length of 60 mm below steam turbine inlet flange.

Now, a piping spool has been cut in inlet vertical run of 300 mm and a new spool of 360 mm is put back inits place.

API 686 flange alignment is restored at nozzle; but is it a cold long in inlet piping ? If yes, will we recalculate the piping stress with this effect ?

regards,
sam

If you add or remove pipe length at installation to take up pipe sag to button up the system, you are adding cold spring to your system. If your system is linear or if this extra or missing pipe does not change the activation of nonlinear restraints, there will be no consequence in terms of expansion stress range - your cold spring will only change the installed and operating loads on your connections.
I would include the cold spring in the analysis for both the operating and installed conditions. The expansion stress range will be the difference between these.
I guess the question I would have is how do you know those springs are all set to the proper load.

No, springs are not set to their preset cold load. Rather, to use the bought springs kept within operating range say -10 to 80 mm for dv70 witch variable spring, we added 60 mm to fill sag with springs unlocked to meet all 686 alignment requirements.
No doubt, cold spring will have impact on operating condition nozzle load, not expansion stress, as Dave Sir, has rightly pointed out.
Reg,
Sam

Sam,

If I get you correctly, you in all probability are dealing with turbine nozzle alignment with springs blocked and unblocked and are having issues.

I have encountered such issues many a times and there are two possible scenarios for your particular case:

- The misalignment values in your turbine piping calc with springs locked and unlocked are within API 686 limits and the alignment issue is only at site.
- The misalignment values in your turbine piping calc with springs locked and unlocked matches the site readings.

For the first case, check with site about the lengths of piping, construction sequence & most importantly whether springs were locked when the construction was in progress. Since you've not mentioned the size, also ensure that the flange weights (for 24" above) are as per B16.47A orB as there is significant difference between the two.Another check to be made for valve weights used in calc and actual. If everything is OK, then its definitely a constructional error and this can be rectified by accommodating the misalignment in FIT-UP spool. If there is no horizontal misalignment, task is relatively easier.And if that is also present, this horizontal misalignment can be accommodated in fit-up spool and taper welded.

For Second case, change spring selection and bring alignment for both locked and unlocked cases within API limits. Also, ensure NEMA compliance!

Hope it helps!!

Dear DJ,
There is one problem. Moot question is whether adding spool after locked spring nozzle alignment for unlocked spring case sag at nozzle flange is cold spring!
Reg,
Sam

SJ, can you please indicate the paragraph and page from API 686 where is specified the alignment with unlocked spring? Really I can not find...
Thanks

@Danb, refer clause 4.7 Note 1 (P6-28)of API 686 of Chapter 6. Moreover, all renowned vendors insist on alignment check with both locked and unlocked springs for m/c guarantee clause by vendor.

@Sam,

It depends on which scenario I stated above are you dealing with. If everything is as per design and construction is also in line with design, then this shouldn't have occurred at first place. One more point: is the first support near to the nozzle (I presume its table top m/c) is from steel structure/portal.IF yes, confirm with Civil that this is as rigid as possible.
Coming to your question, if this is what is going to be done at site, then , yes.It is cold spring (or rather cold pull). This needs to be well documented and issued to client. Also, records of relaxation need to be maintained after every shut down and if required, the necessary cold spring required to be also tabulated and provided to maintenance team.

Read it. No mention on removed stops while flange is disconnected.

Regarding cold spring, it may be for the pipe but not for the flange. With cold spring you will have reactions in the flange while with this adjustment no.

Seems you're right. 686 doesn't state to make check during unblocked spring case. But vendors do insist for unlock case check.
Thanks for making me to read API-686 again. In the subsequent paragraphs, it is clearly stated that preset locks need to reinstalled before making alignment check.Also, Figure D-1 of Appendix D clearly states to install Preset lock during alignment check.

well.... one thing that is to be mentioned is that 686 is a Recommended practice.
There are many requests and vendors insist because they can do this.

It would be nice if the Project Manager review the contract and see if is somewhere specified the request.

But how do you explain that 686 do not in fact ask this?
Are they not aware?
Or is something else.
If you take the drawing it have an allowable load table that have some loads. It is not specified that you can not have loads in cold case.
Vendor have to prove that with these loads will not have misalignment.

I know that it is a sensitive subject, but my opinion is that like everything in this life, better is the enemy of good.

Trying to remove the strain in every condition may lead to situation like this. Now the line is unbalanced, interventions were done, nobody knows how will behave and what loads will be in the machine in operating state. Just because will be combined with the other ones from the other nozzles.

So?

Project management is more sensitive topic than this!!

:-0)

But on a serious note, its not wrong to ensure alignment with both spring locked and unlocked case.

What's your take on this? After all, huge costs are involved for such m/c.

".... its not wrong ...." It is always good to have a perfect situation, but in reality things are not ideal.

Please read pages 168-169 clause6.4.4 of Pipe Stress Engineering by Peng to understand that for steam turbine and centrifugal compressor piping, many equipment vendors expect flange alignment both for locked springs and unlocked spring condition.

Project owners in their design basis incorporate such requirements to be followed in detail engineering.

Sustained load calculation is not accurate unless carried out with as-built pipe spool, valve and specialties weight alongwith integral attachments of supports connected with piping. If piping analysis is not carried out with springs chosen by cold balance method - avaiable in Caesar-II hanger toggle (tick on), such requirement of flange alignment both for locked springs and unlocked spring condition can't be achieved.
reg,
sam

If is contractual you have to follow, but not because "many vendor expect" or similar. This is what I'm trying to tell, generally speaking.
Is only between you and your client.
If is not contractual and if you want to add this condition as an extra measure, fine, is perfect.
And if you can comply with all other requirements, is perfect.

But mandatory is to stay within vendor allowable and if lock or unlock is the only thing that is in doubt you will sit with the client and will find a way.

Sometimes it may be the last thing to be concerned.

Regards,

I read again the Mr. Peng pages you mentioned and I see that while he agree with the method, he also recognize that this subject lead to disagreement between Machinery engineer and stress engineer. Also he state that the adjustment is done by changing the spring load after is unlocked. This is in fact an interference with the decision of the stress engineer and it is seen as a source of friction. He also recognize that stress engineers prefer hot balance. In addition he mention that the method is valid for low or moderate temperature and for gas or vapor. To be mentioned that there is gas with high density even sound odd. He doesn't mention any document in favor of this method and also he doesn't state that this method is mandatory.

So as I said, it is a sensitive subject.

Regards,

Yes. You are right! It is owners requirement.
But, in the end, adopt the approach you can justify! Any short cut can prove to be costly, if rotating machinery fails!
Reg,
Sam