API653 - Low Nozzle Problems.

[KiwiTanks]

I have a tank which has a very low nozzle which I have been asked to inspect to API653. It has never been inspected before, and the drawings have some references to API650.

This is a 6" nozzle, which has been installed touching the floor to shell joint. The strake is 17mm thick, and the nozzle has been installed with a 25mm doubler plate.

I have told the client that the nozzle doen't comply (tank built in 1980) and probably never did. I have told them that the nozzle needs to be cut out (if not required) or replaced with an insert plate, and a nozzle to the API650 height.

My 1st question is - have I told them the correct thing? Should this really be cut out? (Also bear in mind that New Zealand is a seismic zone)

Then I had to inspect another tank (unknown code too) which has a manhole almost identical - touching the floor-to-shell weld. They have asked if the 6" nozzle doesn't comply and has to be replaced - what happens with the manway on the other tank??

Can anyone offer me some advice??

 

[gr2vessels]

Kiwi,
Just like in Oz, the Kiwis can use the flush nozzles, which are allowed bu our big brother, as per the API 650. Obviously, you have to be the judge of work done properly, but by the sound of it, it seems OK.
cheers mate,
gr2vessels

 

[IFRs]

More information is required. What is the shape of the reinforcing plate? How much distance is there from the tank bottom to the nozzle neck? If this is truly a flush nozzle, does it meet the design details shown in API 650 for flush nozzles?

 

[khardy]

If it is a regular nozzle (i.e. pipe and flange as opposed to flush type cleanout) this it is certianly not in compliance with API 650. The manway doesn't meet code either. That said, with customer approval, under certian conditions it may be ok to leave them. For example, in shop tanks, i've put in nozzles like your 6" because the customer required a flush style connection but didn't want to spend the money on the flush type detailed in the code. The tank was small, stresses low, and the client requested and apporved the design knowing it didn't comply with API.

In your case, the 6" might be ok left as it is but a proper evaluation would take into consideration diameter, thickness, corrosion, risk of failure, etc. I would be far less inclinded to leave a manway in like that.

 

[KiwiTanks]

The details are not in compliance with a flush-clean-out type fitting. It is a simple 6" nozzle stabbed in at the very bottom of the shell, (ie touching the floor plates) but has been reinforced by the 25mm doubler (Shell strake thickness is only 17mm). The tank is 19m diameter, 17m tall and the stresses are relatively high.

In the tank with the low manway, the tank is only 8.5m diameter, x 6m high, with low stresses - so I'm less worried about this one.

 

[jrjones]

I agree with khardy about the "proper evaluation". I have run into this exact senario a few times for a few clients. Rather than looking for loop holes, etc, I have just built quick FEA models of the nozzles to show that even during hydro conditions the nozzle stresses are acceptable.
You may want to do the same thing, the geometry is usually pretty simple.

 

[khardy]

Based on a 19-meter tank i'd be inclined to leave it if:

1. It has been in service for a long time in this configuration. (i.e. if it works, don't fix it) and 2. there was little or no corrosion.

But if there are other repairs to be done on the tank, I'd go ahead and fix it since the crew will already be on site.

 

[KiwiTanks]

Thanks for all your advice.
We are removing the low nozzle altogether. We have a door sheet and 2 strake plates also to be replaced..
Cheers

 

[JoeTank]

I'd like to add to jrjones' comment about using FEA results to judge suitability of the nozzle. My greatest concern about such details is not the stress level, but rather the poor quality of the weld details near the 6 o'clock position. Access is poor, so weld quality usually suffers.

Joe Tank