Re-tray project advice - typical plant owner inspection ? advice 3

[mrtangent]

Hi all,

I'm posting this for general advice here. I am leading a major re-tray project on site. We are installing over 150+ new trays on two columns in a few days. The columns have very close tray spacing. Ie inspection and walkthrough once built is very difficult.

We are planning on inspecting 20% Of the trays but I've not been able to find any useful guidance is that normal what would you normally do.

We are using an experienced contactor so we are going for a QA plan on ensureing tray quality but I think we need to do some random checks (ie on the night shift !) to ensure all the trays are installed ok. The balance is every inspection will halt work to complete the checks. So the more we check the longer the project will take.

Any comments most welcome.

 

[sshep]

james1030bruce:

I think a complete run through the tower is a nice idea, maybe even take some pictures of feed and draw-off areas. A quick visual check for debris and loose harware doesn't take much time. If the access panels are left until last, you can usually push (or lower) your way through a tower pretty quickly. Last week I did this on a retrofit tray installation. I took a little stairstepped cutout with me to check that downcomer clearances were within +/-1/8", and that the overflow weirs were also set properly. If these are single pass trays and the tower diameter is not large this type check can be done very quickly (i.e. without climing onto each tray so spacing is not an issue). The main problem with not checking is that should the tower not perform, there are sure to be questions about the internals. If even one of the 80% unchecked trays has a pinched or debris filled downcomer, this could cause the tower to flood.

The items to be inspected for tray towers depends alot on the hardware- i.e. are there adjustable downcomer clearances, etc. Feed entry point and draw-off should be checked against the drawings. Anything that looks funny should also be questioned.

best wishes, sshep

 

[Longinthetooth]

I second Sshep's about pictures of feed and drawoff areas, together with a detailed inspection. There is usualy a manway there so inspection is straight forward. We had a tower that would not perform. We found that the chimney "hats" were installed upside down, restricting the gas flow. The "hats" were, in fact, troughs. - and before anybody laughs- it was not obvious at first sight!!!

 

[lastone]

James1030Bruce, you need to go through the tower and inspect it. This is one of the most important jobs process engineers do. Study all of the drawings very closely so that you understand exactly what you are looking for in each section of the tower. Make a cutout like sshep recommends and check each downcomer clerance and each weir height. Also take with you a small mirror, a flashlight, a level to check each tray's levelness, and a notepad to take any notes. Take pictures of each inlet and drawoff nozzle areas, and pay extra special attention to the bottom trays. Make sure they are installed exactly like the drawings say they should. Take extra care to ensure the bottoms nozzle or nozzles, and the reboiler return are clear. Don't spot check anything. I know it is a pain, and lots of work, but trust me on this, I've learned the hard way.

 

[PAN]

james1030bruce,
I agree with lastone's comment. Random check is not acceptable for me. You should prepare the checklist for the inspector.

 

[mrtangent]

Thanks for the comments, so.

These trays are only about 200-400 mm spacing. I have concerns about going through the column was the risk of not installing the central manways correctly afterwards.
We have well over 100 trays top to bottom.

So to re-itterate you would not expect the tray vendor to sign off on a inspection test record per tray and perform random inspections

 

[epoisses]

DO inspect each and every tray thoroughly. You need only 1 tray with e.g. insufficient D.C. to make your tower flood prematurely.
If you have difficulty inspecting the trays because of the spacing, do you think the vendor would not have difficulty mounting the trays and wouldn't that be yet another reason to inspect the quality of the job done?
Please keep in mind and convince the people around you that an inspection does not halt the work, it's part of the work.
Good luck!

 

[Montemayor]

James:

There isn't much to inspect in a 200 mm (7") tray spacing! Increasing it to 400 mm (15") isn't a great help. I doubt if you can spot defects as you go down a central manway (which is what I presume you have). Epoisses is precisely right - and so are all the rest of the comments. But I don't think going in after the contractor finishes is going to be that effective when you take into consideration that the height of your head is probably more than the tray spacing.

This is a very serious problem. I compliment you on your foresight and initiative in seeking help on this. Making only 20% inspection is a decided risk. I don't know how much you estimate a startup & shutdown are worth, but you may have to go through several lengthy ones if you don't have excellent luck, a good contractor, and a thorough inspection.

There are three (3) vital area to do a 100% inspection on:

1) The Feed tray section;
2) The Reflux tray section; and
3) The bottom tray section - especially if you have a sealed downcomer.

I presume you have seive trays and not valve trays. If you have valve trays I wouldn't accept anything less than 50% inspection.

I would suggest that you have an owner's engineering rep work side-by-side with the installation contractor during the whole installation job. It's that important and this method ensures that you practically are accomplishing a 100% inspection as you go. It's a crappy job full of hazards and the typical assignment goes to a rookie graduate engineer - but I believe the effort and risk is worth it, not to mention the valuable experience to any young, tranined engineer.

I hope this experience helps and Good Luck!!

 

[mrtangent]

Thanks for the comments and feedback.

Monty is right we cannot inspect after the trays have been installed - again it would be just as bad if the tray manways were not installed correctly.

Our trays are high efficiency trays (will be).

What I am proposing is a:

1. Digital photo of every tray (using IS cameras). - Tray installer.
2. Tray manufurture (not the same company as the installer) supervisors will inspect and sign off for 100% of the trays. They have the experience with access while the works are operating and will have time to inspect all the trays.
3. Procses owner inspects 30% of the trays.

 

[sshep]

James:

One possibility for making a pre-startup functional check of the tower is to conduct a "water run" on either total reflux, or sometimes as a complete system of several towers using hoses. If such a run is practical, it has several advantages like flushing the system of loose debris, operator training on the new system, instrument/control check out, utility side commisioning of heat exchangers, and equipment performance checks (pumps- using an SG corrected head, tower capacity, etc). If chemical cleaning is being done, a water run can often be basically a freebee at the end if water is being used as a final rinse medium.

Advising on the ability of process equipment to handle a water run requires the knowledge of a process engineer with some previous start-up experience to evaluate and develop a procedure, but I am sure your team includes someone qualified to make the needed evaluation if such a functional check seems feasible.

just a thought, sshep

 

[epoisses]

Not a bad though, sshep. Don't forget to empty the tower very thoroughly afterwards, or you might blow your precious trays out when a remaining slug of water vaporizes.

And yes, Montemayor, a very worthwile experience for a rookie! I remember going through my first tower from midnight to 4 a.m. and I did find one D.C. at ZERO inch, heaven knew why. Imagine starting up, flooding, producing tons and tons of off-spec rubbish wondering what the problem might be, maybe doing a gamma-scan, shutting down again, steaming out, going through the tower again.... Several 100k to millions worth of losses.

Alternatively, you could go through the central manways with torch+mirror, as lastone wrote, or you could consider fabricating a camera held by a bar long enough to look into the D.C. from the central manway (what's the diameter by the way?)

Don't trust signatures. Every company who provides this kind of services will tell you they do a quality job and will show you their ISO certificates, but the guys who do the actual work (at night, in the rain, in a smelly and claustrophobic environment) are people who can tight bolts but usually have no more than a rough idea about what a tray should look like and even more vague about what it should do. I don't want to offend anybody, but if you had to do this job yourself without somebody systematically checking your work before closing the tower for several years to come, how much nightly hours would you need before your quality standards start to slip..?

 

[STI666]

Everyone has noted excellent comments about post installation inspections. At the end of it all the responsibility is yours to ensure that all has been installed correctly. If something goes wrong as epoisses mentions, it can cost a lot of money and possibly another shutdown or two to fix.

epoisses mentions performing gamma scans to find problems in towers. I am suggesting that you have a company come in and perform a base line Gamma Scan after startup (provided everything is working well or to find any problem areas). This baseline scan will be invaluable in the future as problems,debottlenecking or expansions take place.

Don Carriere
Scanning Technologies Inc.

http://www.scanningtech.com

 

[mrtangent]

Thanks for your comments.

I think I am going to have to humbley go back to the schedulers and extend the re-tray duration up to inpect every tray. After a lot of searching - and talking with contacts of contacts - I finally managed to speak with someone over the phone that has done a re-tray with the same company recently and in 50 trays they found a few lose bolts and gloves (not many but a few). This is therefore not acceptable for us.

As a survey when you have inspected trays what are the typical findings you have found.
a) 1 piece Random debris every 25 trays
b) 2 piece Random debris every 25 trays
c) 8 pieces Random debris every 25 trays