24" Valve Installation in a 26" Steel Pipe Line

I don't think it's acceptable. 24" valve in a 26" pipe is reasonable, specially if it's a control valve, which are typically at least one size smaller than pipe. But it should be fitted using proper tapers.

Cheers........Codey

In addition to Codey's comments, it sounds like the contractor did not install what was specified.

"although the valve was specified and shown on drawings as 26" installed with 26" slip-on flanges"

The contractor should replace the non-specified valve with a specified valve.

If the contractor felt there was a problem using the specified valve he should have consulted with you prior to installation of a non-specified valve.

In my world, if you (the contractor) change project specifications with out consulting with the engineer, you are responsible when your changes are discovered.

When the contractor went ahead with this work he assumed all responsibility.

I'm curious about the desired refrigerant flow (tonnage and temperature) in this line, along with the C_v of the valve. The attachment method looks goofy, though.

It might be possible to justify this departure from the specs, but I would want to see a really good explanation first.

I have requested detailed information about the possible service conditions for this valve but haven't received the information yet. It is only an open closed valve, not modulating. BTW, I agree with all of the comments about the flow conditions on the inside of this pipe and at a minimum the sharp edge of the WN will have to be ground away or filled, this is another whole issue about whether it is even possible.

If you look closely at the drawing, I know it's hard to read, the line that the valve is installed in is labeled as a 25" rolled seamed connector. They wouldn't call it a pipe because it was a fabricated pipe when the chiller was built. I have since verified that the OD of this pipe is very close to 26", 25 7/8" is the best measurement I have so far, not 25" as they labeled it.

They have also claimed that the pipe is elongated or oval, which it is where it's connected to the compressor, but there is a transition to round pipe where the valve is installed and it is very close to being perfectly round.

They claimed that by forcing the 24" weld neck flange up into the 26" pipe was the best way to connect to this oval, non standard pipe size. I don't agree, I think a 26" slip on flange will go over the OD of the pipe and the OD is actually very close to the standard OD of a 26" pipe.

My question is this, if I was to go along with their argument that this is the only way to install a valve in the line, are there pipe stress calculations that could be done combined with NDE to ensure the weld on the exterior of the weld neck flange has proper penetration and adequate strength to meet the requirements of ASME B31.1 Specifications for the service conditions?

I'm looking at this issue like this, if you build the piping system according to ASME approved procedures using materials that have been manufactured to ASME specs and it passes the NDE tests as specified in B31.1 then the system is rated at the level it was built to. But if deviations are required because of real world conditions, as they might have been in this case, is there a way to go outside the presciptive method outlined in B31.1 and do actual engineering calculations to rate the system. If so what might be required?

I agree with KJK/USA. Unless he was pulling a fast one, why wouldn't the contractor have gotten permission to deviate PRIOR to installation of a non-conforming valve?

DO NOT accept the valve in that configuration!

What will your maintenance guys think of you when it has to be serviced or removed from the line????

Not only the valve, but also the length of pipe involved will need to be replaced. Will you cary a spare?

Also, being on the suction side of the pump only deals with the intended operation condition.

What is the situation when the plant is "off" and presures equalise through the whole system?

Are there other possible system states that could exist where the valve would be under pressure?

What would be the pressure across the valve when it is in the "off" position with vacuum applied?

Totally unacceptable.

The contractor should have gotten a waiver to use a 24" valve. It was probably used because it was available or a lot cheaper than a 26" valve. Once the contractor has deviated from contract requirements and it was accepted, you have bought it. As a consulting engineer, you take the risk of approving an installation that may fail sometime down the line. I would insist the modification be redone according to the original specifications. Show calculations to support your doubts and if they still want to proceed, say goodbye, but charge them for your time spent. No sub-standard installation can ever be tolerated when your reputation is at stake. To approve a possible fix, is not a good idea. It's not a question of it not working properly. It's a question of deviating from a specification.. Deviating is OK, but it takes much engineering to do it without bad results.

I have spent many years involved in government contracts. In my business, specifications are followed to the letter. Deviations cost a lot of money. Most times, an approved deviation ends up costing more than if the work were scrapped and redone to spec.

They actual contend that the 26" Valve is cheaper but would take longer to get, verification required.

I'm looking for a way I could accept this installation, is there any way?

I think everyone agrees that I have every right to reject it.

while you are thinking of a way that you can accept this absurd deviation from the plan, perhaps you should also be thinking of a way to keep your name from being listed in the defendant/liable party section of any legal action that is sure to follow... Call those clowns and tell them either make it right or go away and forfeit payment plain and simple due to non performance. If they took the time to screw around and try to pass off the 24"valve they surely could have spent whatever extra time to (a) do it the right way, and (b) communicate the prob getting the correct part before using the " better to ask forgiveness than permission" theory

No code can accept that very bad design and construction. The WN flange is designed to be welded from its neck (grove/butt weld not fillet weld). Another malfunction is the throttling of the pipeline 26 NPS, where it must be checked from process point of view related with flowrate, pressure drop, .. etc.

There is a chance for occuring a cracking for that fillet welds especially with that large size of pipeline 26 NPS. Two months ago, there was a problem for 24 NPS of the same process due to lite settlement of that heavy line which causes a cracking of the body of a gate valve 24 NPS.

So, the installer must supply the butterfly valve with full bore 26 NPS and its flanges 26 NPS where it shall be assembled.

You have not stated the operating temperature of the refrigerant, BUT at -40F, the saturated pressure is still 0.6 PSIG so it is likely NOT in a vacuum, just lower pressure.

A weld-neck flange will create massive turbulence on BOTH sides of the connection due to the weld-neck protrusion. Slip-on flanges MIGHT have been acceptable EXCEPT that 26" slip-on flanges and a 26" valve were specified. Therefore, they are the ONLY satisfactory solution.

I cannot believe that a 26" valve would be less expensive than a 24" valve- just the "odd" size would trigger a significant price increase.

Additionally- be sure that the valve actually installed is rated for refrigerant service. I have never seen a butterfly valve used in ANY refrigerant system, except occasionally in ammonia. I cannot see how the valve will seal effectively.

"I cannot believe that a 26" valve would be less expensive than a 24" valve- just the "odd" size would trigger a significant price increase."

I fully agree with the above comment from energygod. In fact I doubt whether the contractor may get this size non-standard valve so easily.

Additionally- be sure that the valve actually installed is rated for refrigerant service. I have never seen a butterfly valve used in ANY refrigerant system, except occasionally in ammonia. I cannot see how the valve will seal effectively.

I do not agree with this. I have seen a saisfactory performance of "butterfly valves with SS disc & nitryle seal" for refrigerant R-12 service for over 20 years.

In my opinion the best way to sort out the problem could be to introduce 26" to 24" reducers on both sides and use 24" butterfly valve with slip-on flange.

"You have not stated the operating temperature of the refrigerant, BUT at -40F,. . . "

. . . Steel has a significant loss of strength around -30 to -40°F! During early WWII, several "Liberty" supply ships were lost near Alaska due to hulls cracking because of this factor.

What is the operating temperature and pressure of the 26" pipe? What is the pipeline velocity (which affects pressure drop)? Have the Contractor's licensed engineer provide the calculated line stresses from pressure, pipe supports, and based on the submitted calculations, have the Contractor submit written certification that the un-authorized revision is safe and suitable for intended service and the temeratures, pressures, and flow velocities expected.

Since the Contractor has created this problem, the Contractor should either re-do the work or provide sufficient proof that the unauthorized change is not a problem.

Good morning Sampson45w.

Assuming the contractor received prior approval from the designers/owners, and that the piping arrangement used was vastly cheaper than that specified, and that internal pressure is negligible, and that no extreme high temperature is to be present, and that there is minimal vibration, and that certified welders were/are used, and that the slight turbulance caused by the 24" weld neck flange would be no greater than that caused by the valve's disk, I would approve the fabrication on one condition: if the 24" wn flange could be welded on the inside to the 26" pipe/connector, thus adding the strength of the second weld, and filling the void, thus decreasing the turbulance.

But the specification is for an installation that meets ASME B31.1 and we don't want to relieve the contractor from meeting that spec. We are considering whether there is a way to meet that spec with documentation, testing and modifications to the current installation

Since the installation of the WN is not done according to ASME specifications or procedures, what should be done to document the strength and acceptability of the welds to meet that spec. I'm thinking pipe stress calculations and X-Ray the welds but what would be the standard to judge the welds against.

Is there an acceptable method of documenting an installation that did not follow ASME standard installation procedures but have it meet the ASME B 31.1 specification? I don't think just having a PE stamp a drawing is adequate without some backup calculations and NDE.

Since the contractor is obviously oblivious to the specification's requirements- like totally redesigning the programmed installation using absolutely NO logic with regard to the impact of the installation- JUST SAY NO.

IF you want to try to move forward as quickly as possible with a COMPLYING installation, then using a eccentric (flat on bottom) reducer from 26" to 24" and then using WN flanges for the 24" valve SHOULD comply with the code and allow for a functional installation. The minor pressure drop created by the small constriction should have no impact on chiller operation.

By the way- the eccentric reducers were to avoid any possibility of trapping oil upstream of the valve- ANOTHER screw-up that I had overlooked before when discussing problems with the current installation.

No!!!! That is not to be done and is unacceptable! First, the reductions in the line size will cause you problems, second this will not pass inspection or any code ASME, API, any fire protection codes and the company that insures the owner of the plant will demand it's replacement. Third thing is the alignment could be a problem. I would suggest that you do not accept this work and I would have that Company that is the contractor and the Foreman of the job and the welder drug tested and then escorted off the site for shoddy workmanship and lying to you. Then I would bring in a competent contractor to complete the job correctly. My final question to you is if you caught the contractor doing this shoddy work, what other type of shoddy work has he done on your job site that would jeopardize the integrity of your facility???

A few issues-

First- WHY would a properly installed welded assembly not pass ASME or any other inspection?

Second- The MINOR reduction in bore, with appropriate transitions, will have NO adverse impact on the operation of the system. The sketch provided shows this pipe to be a branch of a larger main.

Third- Going back to the OP again, what engineering firm designed this installation in the first place and specified B-31.1 assembly rather than B-31.5? THAT is the beginning of the problems, the shoddy workmanship and disregard to the specifications came AFTER the original engineer screwed up the design.

I agree that the installing contractor be escorted off-site and prohibited from bidding any further activity on-site (PLUS billed for the needed corrections as well as not being paid for the non-complying work).

First- Weld neck Flanges are not designed to be inserted into a pipeline and the pipe welded on the outside shoulder on the WN. I've see slip-on flanges in the ASME Code but no where have I seen a WN Flange used as Slip-In Flange. I don't think this installation meets ASME B31.1 or B31.5

Second- I have asked for process information so we can evaluate the effect of the bore reduction, I also suspect the effect to be small. The problem is made worse by the raw edge of the WN inside the pipe but that could be minimized by filling or grinding. The 26" pipe comes from a shell and Tube HX'r, a "Cooler", the 16" branch is going to a smaller compressor for part load operation.

Third- The spec and design was provided by the vendor who hired and supervised the installation contractor and deviated from their own approved submittal without notifying the engineer or construction manager.

The installation contractor is on notice and the final resolution might be just what you suggest, especially when they insist that this is the best way to install this valve. The deadline is approaching.

The deadline is approaching. Sounds like the lawyers may be waiting for their cue to come in also!

At this point, the discussion becomes contractual, not engineering related. Who bears the cost of fixing the screw-up, and who bears the cost of the missed deadline. Who is respnsible for damages if the cobbled up flange installation fails or causes operating problems. Not sure I have good advise to prevent the fecal material from hitting the air-moving device.

For the record- I totally agree that NO flange should be used as a SLIP-IN flange. My suggestion was to install reducers to take the 26" primary pipe down to 24" ONLY at the "existing" valve- purely as a means of getting the installation in a manner that COULD comply with the B-31.1 code quickly and using materials already on-site. It was NOT intended as a "perfect" solution- that would be an installation with (2) 26" slip-ON flanges and a 26" butterfly valve designed for refrigerant service.

Unfortunately- since the 24" weld neck SLIP-IN flange has been installed, the "ideal" solution will also include cutting out the offending flange (by the way, was only ONE flange installed this way or was one on each side of the intended 24" valve installed?) or flanges and installing some length of "new" 26" pipe to attach to one of the 26" slip-ON flanges. Since the existing 26" pipe is not apparently a standard pipe- this could be somewhat problematic.

Energygod, I looked at a fuzzy blow up of the photo attached to the original post. There are 24" weldneck "slip-in" flanges on both sides of the valve. Also, space for pipe removal and replacement is limited to approximately 18" between a tee and the suction side of the compressor.

As you understated, "this could be somewhat problematic." I believe the correct solution is to cut out the 18" +/- of non-standard pipe and replace with standard 26" pipe with the 26" valve that was originally designed. Or have the contractor provide adequate engineering calcs and evidence that the current configuration is safe and not detrimental to the refrigeration system. And offer a credit for the difference in price between the 26" specified valve and the 24" un-approve substitute.

The bore reduction if done properly would not be as much of a problem as the uneven bore of the interior of the pipe would tend to collect particles and have additional wear at that point causing a failure of that joint. I would suggest that this portion of the piping be replaced and done correctly. Any job big or small do it right or not at all.