Discuz! Board

 找回密码
 立即注册
搜索
热搜: 活动 交友 discuz
查看: 11|回复: 0

Leak testing welded vessels

[复制链接]

9万

主题

9万

帖子

29万

积分

论坛元老

Rank: 8Rank: 8

积分
293221
发表于 2021-8-31 22:56:46 | 显示全部楼层 |阅读模式
We want to validate (leak test) the corner weld integrity of some small light 11 gauge steel pressure vessels (15” x 15” x 28”) we’re having made.  They will hold very high low viscosity mineral oil.  Would it be valid to test them as follows:  Spray the inside corner welds with dye penetrant and immediately close the vessel and pressurize it.  Spray the outside corners (opposite surfaces) with developer and wait about for 30 minutes or so for potential bleed-thrus.
Reply:pressure testing anything with a compressible gas is a risky venture.  your vessel has a volume of 6300in3 and  surface area of 2130in2.  Even a small pressure of say 5psi puts a total force of almost 11000 lbs against the inner surface of the container.  If you don't know what the container can take, you're making a potential bomb.  If you're bound and determined to do the testing this way, buy a 0 - 1 PSI pressure gauge and fill the container to 0.3 to 0.5 psi.  Anything over 2 psi in a tank like the one you describe is risky.  Even with low pressure you may bulge the sides of the container and permanently deform it.Go ahead and spray your penetrant on the inside and the developer on the outside.  Make the seam you're checking the low spot in the container by rolling the container around, if this makes you more confident.  I suspect just doing the normal procedure of spraying penetrant on the outside, wiping down with cleaner, and then spraying developer on the outside will reveal any problem cracks.the other safe thing to do is fill  the contain most of the way up with oil or water and pressurize just a small fraction of the container volume.  Leave it overnight and check for leaks on the floor in the morning or look at the pressure gauge and check for a loss of pressure.Benson's Mobile Welding - Dayton, OH metro area - AWS Certified Welding Inspector
Reply:fill it up with water (takes away the compressible volume) then pressurize with a small amount of air pressure as stated above.  if you don't want to use water, use oil instead.  Hold it for a while.  Check the outside for leaks.  Skip the PT (for the purposes of leak checking anyway)
Reply:I used work fab shop i kind welded every thing ,made custom aluminum boat tank, coast guard guy would come out there code would fill with water with stand pipe above tank connected 15' above it   they stamp there seal on square 4x4 plate tack on tank
Reply:Carl Welcome.I pressure test a lot of stuff I make or repair. Here is just a sample of my jigs.www.tjsperformance.comDynasty 300 DXHTP 240HTP Microcut 380Hyperthem 85JD2 Hyd Bender and HF Hyd Ring Roller all in one =(Frankenbender)Bpt. Mill/DRO4' x 8' CNC Plasma TableInstagram: tjsperformanceYT: TJS Welding and Fabrication
Reply:Just saying, good question and informative answers.
Reply:I would PT from the outside and follow up with filling the tank with the liquid that will be used. Let it the container sit there for at least 10 min to see if it leaks. If this is a 11 gauge tank, it was probably a single pass weld and if it was a short ark GMAW there may be leaks.For the PT make sure you read the instructions or look up how to PT with the specific PT product used. There are many types of PT and they need to be used correctly, including cleanliness, temperatures and timing. For timing on the PT for pre-cleaning, drying, penetrant dwell time, emulsifier if applicable, excessive penetrant removal, post cleaning time and development time is critical and more steps can be critical depending on the PT product used. For acceptance criteria I would not want any blead out or as some call it "PT white".  If you are not up on your PT knowledge and are unaware of issues that may lead to false bleadouts, just fill the tank and see if it leaks.
Reply:All good suggestions! When I worked at Harley, the fuel tanks for the motorcycles were pt'ed to 4 psi under water in a big industrial sink with rust inhibitor mixed into the water and they would look for bubbles. I doubt you would be able to push that big tank under under water but pt low psi with air and spraying soapy water mix on the welds liked we use to check for natural gas or refrigerant leaks (I'm a pipefitter) might work.
Reply:I'd put a liter or so of oil in it, slosh it around to makes sure all the joints are covered and look on the outside for any sign of the oil wicking thru, no pressure needed. It's unreal how thin oil, diesel, light lube oil etc will suck thru a defect, antifreeze is worse.....Mike
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

Archiver|小黑屋|DiscuzX

GMT+8, 2025-12-29 01:15 , Processed in 0.098824 second(s), 18 queries .

Powered by Discuz! X3.4

Copyright © 2001-2021, Tencent Cloud.

快速回复 返回顶部 返回列表