I am designing a Piping system which connects the Turbine exhaust to ACC. Below the turbine exhaust i designed an In-Line Pressure balanced Expansion joint.

The pipe size & Expansion joint size is 1400 diameter and 10 thk. The EJ (Expansion Joint ) weight is 5000 Kg.

I have designed one anchor point at top of EJ approxmately 2000mm ie., turbine exhaust nozzle.

Another anchor point i have fixed 1500mm from the bottom of expansion joint.

I analysed the system, & i got nearly 3000kg weight is acting on the turbine exhaust because of expansion joint weight, it is not allowable as per TG Supplier exhaust nozzle forces and moments.

Then i asked the EJ Supplier - Since the weight of bellow my system is getting failed...for that he said since anchor point is fixed at the bottom of the EJ, the weight of EJ would come only on bottom anchor & it wont come on Turbine nozzle on both operatiing and sustained case. So he asked me to consider 0(Zero) Weight for expansion joint while doing stress analysis, i analysed as like he said & i got values within allowable.

Is it correct what my supplier said..shall i proceed as my supplier point of view.

" 0(Zero) Weight for expansion joint" The Indian culture has produced some amazing things including weightless expansion joints Western societies have no hope in competeing against such amazing technology.

Moorthy,

Anything that is attached to the piping will transfer forces (the weight) to the piping. The weight will be there and the pipe supports and the turbine must accommodate it. Of course pressure balanced expansion joints are particularly heavy (relative to their associated pipe size) because there is so much more hardware involved. Usually, the pressure balanced expansion joint is supported from below (similar to a "base elbow").

The manufacturers of turbines sometimes make useful design manuals available to piping engineers. Many of the design suggestions in the manual generally apply regardless of who the manufacturer of the turbine is. The better of these manuals also provides design suggestions for including various types of expansion joints in the piping design. Also, you should read the NEMA SM-23 Standard to learn more about the design of piping systems connected to turbines.

Good luck my friend.

Regards, John.

Moorthi,

CAESAR II or that way any software related to Applied Mechanics cannot violate the fundamental principles of Mechanics. If the modelling has been made correctly, then wt. of the expansion joint is a must input in the model and if the nozzle takes wt, then that is it.

I feel what you need to do is to see the global element report and see why the nozzle is taking load. This is as per the principle of equilibrium and distribution of loads between elements of different stiffneses ( method of moment distribution).If you yourself can justify why the load is coming to the nozzle ( which I am sure you will, once you carefully look into this report), why you need to ask the vendor or anybody ?

In response to John's reply, I would like to add that Indian culture has never produced expansion joints of zero wt, but yes, many of us don't like to delve into fundamentals of mechanics
and hence puts such questions.

Regards

Well I do actually find many things that Indian culture have given the world of much use to all of us.... for hundreds of years as a matter of fact, my sarcastic humor got the better of me.... (Lets not forget the Western Europeans stumbled into North America trying to get to.... India)

But I have seen many things like this by many people from all over the world, some of them from here in the U.S.A.

Last year about this time a clever person here in the U.S. working for an owner took a pump that was overloaded and with a wave of the magic stress wand made it all better....

He did so by letting the pump slide, and then making the pump weightless, its sliding points frictionless and its inherent resistance to rotation due to its polar moment of inetria non-existent....

So these miracle workers must all go to the same university somewhere in the world (perhaps the North Pole?) Maybe the EJ is weightless????

Thanks to everyone..John C. Luf...the expansion joint supplier is not an Indian Supplier...

The guy here with the pump last year must be working for the EJ supplier!!!!!

Argghhhh ^@#^#@^%@%&#^&*&%^*&(%(

This is not a nationality issue. All of us who have been in this business for a while have our own horror stories, at least a few self-inflicted.

I was talking to my boss this morning, and he told me we have a new stress analyst coming on board soon, He regretted that he had to hire one without a lot of experience.

I told him that I doubted that I was any better at using the software than I was at age 32 (longer ago than I care to discuss). He agreeed that that was probably true, but that the difference was judgment. I second that opinion.

If somebody tells you to treat something as weightless and frictionless to make your analysis easier, you should regard this in much the same way as if you were in New York City and a guy came up to you on a streeet corner and offered to sell you a $15,000 Rolex watch for $150. We stress analysis types are not usually the bearers of good news in this industry, and it takes some years before we get over the common idea that we want everybody to like us.

Breen, Luf, Superpiper, Captain Kenny, and others like myself post horror stories from time to time to try to get the point across that you have to understand the whole process involved. Stress analysis is crunching numbers, but that's not the primary goal. The primary goal is to make sure that the numbers you crunch are connected to reality. CAESAR II and other tools make crunching the numbers easy (I have done hand calcs for piping systems!), but this can be dangerous if the input is unreasonable.

Weight can always be removed from a terminal equipment connection in the real world. Sometimes it's complicated, and expensive, but there is always a way. I just spent three days designing about 15 spring supports for piping where a pump vendor was insisting that we bolt 300 pounds of valves directly onto a pump discharge nozzle that is only rated for a 240 pound vertical load. Eventually I found an arrangement of spring hangers that could support some of the weight of the valve station (but not too much) in all of the relevant load cases. It took a lot of trial and error, though.

This sort of problem is not really one where CAESAR II does a good job of selecting spring hangers as yet. I'm sure by the time version 7.0 rolls out of the back room, the folks in Houston will have this figured out, too.

But we always have to remember, CAESAR II is a tool, not an answer machine. When I was in grade school, I read a book called "Danny Dunn and the Homework Machine." It was about a kid who copied all his textbooks into a computer (back before computers were commonly available) and then had it print out his homework in a couple of minutes every night after school. The whole point of the story, though, was that he had to do the homework to get it into the machine in the first place. CAESAR II works the same way - if you want to get good information out of it, you have to understand what the effect of every input item is as you enter it into the software. That's judgment, and it's a precious commodity. Even those who don't like our answers have at least a grudging understanding that, without our efforts, projects are not going to run smoothly.

I delivered a couple of stress reports to one of our projects yesterday, and the project manager sarcastically congratulated me for getting the stress analysis done on time. I just laughed. The reason was that the designer who did the layout had come to me early in the process and we had discussed how the piping should be routed in the vicinity of the equipment to make it unlikely that there would be stress issues. THAT was when I did the work that ensured that my stress report would be done on time.

There are four phases to learning how to be a good stress analyst.

1. You have to learn the Codes.

2. You have to learn how to use the software.

3. You have to learn how people are going to build things that correspond to what you put into the software model.

4. You have to learn how to get people to apply what you have learned in 1 and 3 to their designs so that it makes 2 easier for you.

Support the expansion joint from below without bolting to the turbine nozzle. Adjust the lower support to achieve 0.01 clearance to the turbine nozzle. Insert bolts and tighten the flange properly. There will be minimal weight transfer to the turbine. When the unit heats up, there will be compressive force due to the stiffness of the bellows and thermal expansion. There will be pressure thrust during operation which can be calculated by using the actual cross sectional areas in your specific joints components. The pressure thrust during operation can be adjusted to provide uplift forces to the turbine flange. This will offset the small weight transfer. You must chose the cross-sectional area of the joint to provide the desired thrust. This is not as practical as applying forces and moments to piping systems prior to welding so that the as built condition after construction restraints are released will equal the calculated sustained forces and moments. Which by the way has been used for critical piping systems that have cold spring. Since we are attempting to minimize load transfer we should probably use a manually adjustable vertical support to minimize the transfer. This could work as well as the manually adjustable gap restraints used to withstand expansion and occasional displacements. One would need to know actual operating temperature when the occasional event occurred. Seriously, I suggest that you use springs to support your expansion joint components if they weigh that much. Oh, you could design a lighter weight model of the expansion joint or use some of the support details recently seen in this forum. Obfuscation is humor to the aged mind.

Actually in some cases the less experienced people have fewer bad habits to unlearn..... ;-)

(yes, some days I include myself in either 'less experienced' or 'lots of bad habits to unlearn' - I'm hoping I've caught most of them)

Shannon


[quote=CraigB](snip)
I was talking to my boss this morning, and he told me we have a new stress analyst coming on board soon, He regretted that he had to hire one without a lot of experience.

One of the grossly misunderstood things is the pressure thrust load on the nozzle due of expansion joint. Many people in the industry, I have come across, feels that in case of untied expansion joints, the full pressure thrust comes to the nozzle ( which is generally wrong as the nozzle should get minimum loading with the untied expansion joints , however the equipment foundation should get maximum loading in that condition ).One of the reasons behind this is CAESAR II, by default uses PXA at both ends of expansion joints whenever effective diameter is given as an input.One should not do it like this. In my opinion the effective diameter should be kept zero and this PXA loading should be applied at the change in directions and only then the program will correctly determine the amount of PXA that the nozzle sees based on relatve stiffnesses of the different members of the system.

Besides the 4 points mentioned by CraigB, I feel that a stress analyst should also be having a good concept of the fundamental principles of mechanics.

Regards

I forgot to tell one thing..

In this case the bellow is supplying by the turbine supplier & i am doing analysis for the turbine supplier..He(Turbine supplier ) Procured the bellow before getting clearance...and i asked him to change the properties of the bellow..that is why i ased the bellow supplier based on what basis he had designed the bellow..for that only he said since the Inline pressure balanced bellow (Pressure thrust will not come for this kind of bellow ) is of tied type the force would not come to turbine flange..but i know when doing analysis with two anchors between the straight length..force & moment will distribute between the two anchors..but the turbine supplier agree with what the bellow supplier is saying (like consider zero weight ) what can i do if my client itself doing like that..that is why i asked the suggestions from you all..thanks alot for your replies...i show this to my EJ supplier..let see what will he say for this...

Dear Moorthy,

Please realize that 'YOU' are designing the turbine exhaust piping system. Will EXJ vendor or your client, turbine supplier- tell what is right ? If it fails, who will be responsible ?

I do not understand why more than one flange weight directly connected with turbine flange will go towards flange nozzle, if you support the EXJ from the bottom ?

If your expansion joint really transmits so high load on turbine exhaust flange, will not the same load pass through the bellow material, too ? Is it designed for that ? Be careful.

http://www.flexicanbellows.com/images/Inline%20Pressure%20Balance%20Bellows.pdf


In case of engineered solutions like in-line-pressure balanced bellow, your conceptual design has to be right.

regards,

sam

yes sam, the type of bellow is In-line Pressure balanced..then is it correct..?

Better support the expansion joint assembly from the bottom by a F type spring hanger of low stiffness. As it is difficult to aportion the division of weight between top and bottom side due to the presence of tie rods, better to take the complete weight of the EXJ assembly out.

regards,

sam

I see plenty problems with bellows in general. Happen to design them. If you have in line pressure balanced unit check if it can have any lateral movement and/or flange rotation. No torsion is normally allowed. Axial spring rate can be high. It is sum of the 3 bellows elements. What ever you do with the unit confirm with the bellows supplier or with person who is expert with bellows.

Inline pressure balanced unit will not have pressure trust that is external. All pressure trust forces are internal provided the unit is designed correctly and the manufacturer had correct tooling. I would think most of the time units are a bit out of balance. Forces to consider are spring force, internal friction and unit mass (Plus imbalance if any).

Bellows elements cannot transmit high forces. Material is thin. Typical thicknesses between 0.6 to 2 mm per layer and 1 to 2 layers. Forces are transmitted through the rods. If you have a case where they are under compression they can buckle if not designed for compression.

"No torsion is normally allowed" This staement is trues for all EJs although I would use the words very little as opposed to no. No = 0 Mom....

True. "Very little" can be even defined. EJMA (www.ejma.org) has a formular for it. Normally only the manufacturer can do the calculation because the calculation reports and drawings they give out do not have sufficient information. Even if I could do the calc I would ask manufacturer to say yes or no on any critical item.

Absolutely correct 110%!!!!! I concur whole heartedly....

Once again my supplier confirmed over E-mail what he had said.

EXJ supplier Reply: With fixed support at bottom and the expansion joint(Inline Pressure balanced ) being tied type, the dead weight will not act on turbine flange as a pulling load.

Even for a tied bellow, portion of weight would have to be shared by turbine end, if one does not separately support the bellow assembly with a spring.

For such a symmetric type of inline bellow configuration, 50% of total 5000 kg should go to the turbine end unless you support the exj assembly separetely with variable spring supports.

Anyway, due to the tie-rods the weights will go directly to turbine end and the bottom end as tensile and compressive forces respectively, thereby not overloading the bellow elements.

Best way to verify is to test a small such bellow in vertical orientation. Ask the EXJ vendor to demonostrate what he/she claims.

regards,

sam

These pressure balanced units are not very common. They are complex and every time when I design them I have to "reinvent them". I try again :-)

Sam's post above has an image. In this design one set of nuts is away from the plates. They should be at the plate. The idea is that when the unit is compressed or expanded volume stays the same e.g. 2 small bellows units compress and the big unit expands. Small units have lower spring rate. If the nuts are loose and there is no pressure then small elements compress and nothing happens to the centre bellows. See http://www.jat.co.za/worksamples.htm for similar design. (Internal forces on rods and plate can be huge.)

If you position the EJ vertical, no pressure inside and top flange is loose then it will get shorter due to its own mass. Amount can be calculated using spring rates. As a result if the top flange is bolted to something it will pull that part downwards. For me by putting support under vertical EJ I would use 1/2 mass on top flange as downwards force.

Imbalance under vacuum or internal pressure will add to end forces.

Forget internal friction forces in vertical installation.

You need to add spring force e.g. spring rate * axial movement. Manufacturer has to give spring rate. They can calculate EJMA spring rate, which is not working rate. Working rate is less. It cannot be calculated, only measured. Assume 30% tolerance on given spring rates.

Yes. What Jouko has opined is right.His illustrations are very good indeed.

In my posting,no set of nuts should be away from the plates. I noted while posting it from a manufacturer's dwg; but, it is so obvious, I thought that everyone must have read EJMA handbook & know the difference between control rods & tie rods!

regards,

sam

Ahhhhhhhhhhhh Ha!! The picture become more clear.

The weight is to be completely supported from below.

Unfortunately the supplier has a misunderstanding. If you like you can refer them to me. I can discuss the issue with them. You can also send unit calcs plus DRWG to my E mail. If the loading is so critical are you sure that the unit is in balance? Will check the design (Results not given here). No charge :-)

Some comments about supporting from bottom (as I see it):

Unit is vertical and it has 4 plates. Two of these are connection flanges. I call plates 1, 2, 3 and 4 starting from top. 1 and 3 are rigidly joined. Same with 2 and 4. Assume number 4 doesn't move but 1 is coming down. When looking the setup assume bellows convolutions are not there. They are paper compared to all other items.

If you put fixed support on 4 then 1 and 3 are not supported but 2 is. You cannot put fixed support on 1 or 3 as they move. These 2 are supported by spring force (elements) and bolting at 1. If you put spring support on 3, rods will see additional compression force (buckling) plus you increase the overal spring force. If you put on 1 then only spring force goes up.

I do not like the idea of putting springs:
- space is limited
- they have to be absolutely symmetrical
- if springs are not in balance EJ sees uneven loading
- all 4 plates are under heavy loading. Any additional spring load would have to be checked by the supplier.

I overlooked one possibility. Bill's posting earlier has also the idea. EJs are supplied normally locked into neutral position and they are installed normally with locking on. This way you will have the mass issue.

You could support the unit from bottom, install the unit vertical without bolting to the turbine flange and remove the locking devices from EJ. EJ will shorten half mass/spring rate. Adjust the closing gap as Bill recommends. No mass on turbine.

To do above;
- check EJ installation instructions if this is what they recommend/allow
- if installation is with locking on (or no instructions) the supplier has to check the design. EJ is not installed into neutral position. There is a presetting, which has to be considered.

I did a check on the unit shown on my site. It is 54", 10 bar, mass 5000 kg. Installing it vertical without locking would compress the unit about 4 mm. Design movement -12 mm. This specific unit cannot handle the extra 4 mm without a risk. Each design is different and has to be checked.

Thanks Jouko sir...if not bolting the turbine flange..how can vaccum pressure will maintain..

Forgot to mention that after setting the gap helium filled titanium bolts should be used to close the joint.

Dear Jouko,

Why do you require helium & titanium bolts?

regards

sam

Sam,

Jouko I believe is referring to the start of the thread where the EJ was declared "Weightless" The helium would cancel out the weight of the bolts nuts gaskets etc. I am guessing.

A joke Sam.

But if the helium cavities in the bolts are too big (we must after all maintain the bolts total metal area) then we will have to go to larger diameter bolts. Then we have the problem of having to redesign the flanges as these will not be B16.5 flanges. Also consider substituting unobtanium for the titanium for reduced weight (this is Cleveland, Ohio technology). With that material we could grossly reduce the required bolt area and make room for the helium cavities. Perhaps Mr. Luf can suggest a supplier.

Just a suggestion.

Regards, John.

When I retire I intend to work on unobtainium and non-expandium I have some metal matrix ideas in mind.... I may resort some ideas found in text written by the legendary Merlin... magician and alchemist of legend

A current material that is similiar to non-expanidium is abresist lined piping.... the stuff does not expand hardly at all... it can be lined with a variety of materials the one I looked at was volcanic basalt!

Sirs,

Now, when in the days of climate change everyone is switching to LEDs, one can think of Tungsten too, which is having even smaller thermal expansion coefficient. But, as its cost is high, pipe will be of nano scale only - just to be believed to existing, not to be seen!

regards,

sam

John C. Luf,

You better not post a reply to anybody's post.

It will be good for the "FORUM and its Users".

Thanks

Bajwa... Who appointed you high imperial ruler of an open forum???

If you don't like my posts don't read them! As for you I have looked through some of your posts your contributions to this forum indicates a lot of questions some of dubious nature... so take your own advice....

luf with a huff!:D

seriously, without breen/luf/diehl/ay
this forum would be useless.

Either of them could call me a fool and i would respect it.
Every time someone post's a question here you are leaving yourself open to critizm, but that is a cost i am willing to pay to get it right.

My questions on relief systems could well be answered with "hey buddy go figure" but thats a risk i take.

please dont upset the few people on here who genuinlly know what they are doing.

Tim

Cool...Cool..Cool...thanks for everyone for their contribution in this topic.

The commissioning date of the plant is announced...Commisioning date is on 15th March...so dont want any pilot plant study for this Zero Expansion joint...let see what will happen..today i replied to my turbine supplier (Client) if anything will happen on that day..i m not responsible for that..

If anything will happen, surely he will comeback..at that time i will also come back to this topic.

So..we will wait for some 3 months..

Hi ,

I have a question about modeling the cast basalt liner(A53 Grade B with basalt liner) and stress analysis of that also i have some ceramic liner?
Please advice me.

"One of the grossly misunderstood things is the pressure thrust load on the nozzle due of expansion joint. Many people in the industry, I have come across, feels that in case of untied expansion joints, the full pressure thrust comes to the nozzle ( which is generally wrong as the nozzle should get minimum loading with the untied expansion joints , however the equipment foundation should get maximum loading in that condition )."


Does it matter? The pump it will take it all.
And if the pipe is not perfectly aligned with the nozzle, what it will happen?