| Railway Preservation News https://www.rypn.org/forums/ |
|
| UT technical questions https://www.rypn.org/forums/viewtopic.php?f=1&t=7948 |
Page 1 of 1 |
| Author: | Ray Dewley [ Tue Dec 30, 2003 4:20 pm ] |
| Post subject: | UT technical questions |
What effect, if any, does boiler scale and or mud have on the readings obtained from a UT? Are the instruments calibrated only for the density of the steel? |
|
| Author: | Roger Stabler [ Tue Dec 30, 2003 5:13 pm ] |
| Post subject: | Re: UT technical questions |
> What effect, if any, does boiler scale and > or mud have on the readings obtained from a > UT? Are the instruments calibrated only for > the density of the steel? If your UT tester is of the type that produces a graft you should be able to determine a false reading of thickness by observing the graft, the person using the ut needs to be very familiar with the tester practice with it to get used to how echoes from scale will show on the graft. The digital only ut testers work fine on clean metal but have problems with scale and rust exspecially around staybolts and rivits which I find is the area that I'm usually the most concerned with large areas of the barrel can be looked at visually and backed up with UT, If there is pitting or wasting in the barrel you should be able to see it clean the area and get a good reading clean as much scale as possible before ut testing. |
|
| Author: | BobK [ Tue Dec 30, 2003 6:08 pm ] |
| Post subject: | Re: UT technical questions |
We did the required boiler survey and so have a little eperience with a UT machine. We found that scale would cause no reading rather than a false reading. False readings were caused mostly by bad contact with the probe. Basically the more you pay for a UT tester the less problems and better accuracy you will have. |
|
| Author: | ge13031 [ Tue Dec 30, 2003 6:38 pm ] |
| Post subject: | Re: UT technical questions |
> We did the required boiler survey and so > have a little eperience with a UT machine. > We found that scale would cause no reading > rather than a false reading. False readings > were caused mostly by bad contact with the > probe. Basically the more you pay for a UT > tester the less problems and better accuracy > you will have. Bob hit a very important point, namely clean contact for the probe. Start off with a standard that is clean and thicker then the material to be tested so you set the scale on the scope where your back reflection will be. You set your attenuation to give a strong back reflection from the standard. When you put the probe (using the proper couplant) on the clean surface of the piece to be tested your scope will show a strong back reflection from the true metal surface and possibly a second weaker reading from any crap or corruption on the back face. You then scale your true metal thickness from the scope. The scope will also show you any interior metal defects as back reflections dependant on their size, be they non-metallic inclusions or voids. You can also get some idea of the damage to the inner face if you start to lose back reflection because of heavy pitting or corrosion ...this can be overcome by using a narrower probe. Can you tell I don't buy a digital readout UT system ??? lamontdc@adelphia.net |
|
| Author: | jc boykin [ Tue Dec 30, 2003 8:40 pm ] |
| Post subject: | Re: UT limitations |
What Dennis means is that the small, digital readout ultrasonic thickness testers are almost useless with corroded or pitted boiler plate. An oscilloscope readout instrument costs not a few thousand dollars, but will allow you to measure real metal. I have seen a lot of "measurements" chalked on boilers which are greater than the original thickness of the sheets. The oscilloscope allows some utility in crack detection in other than flat plates as well. The best instrument would be a shear-wave ultrasonic tester, but we're talking real money there. In addition, with suspicious parts or or complicated areas of a boiler, ultrasonic testing alone is *not* very reliable. Dye penetrant will show some surface cracks. Magnaflux will show deeper flaws and, I think, should always be used to confirm ultrasonic. In areas where cracking is suspected, x-ray examination is often indicated as well. This is another expensive procedure. It must be done by professionals (who usually must have the geometry of the part carefully explained to them). Perhaps Dennis could recommend an oscilloscope-readout ultrasonic machine within reach of most boiler operators. johncb@u.washington.edu |
|
| Author: | buddy bob [ Tue Dec 30, 2003 9:32 pm ] |
| Post subject: | Re: UT technical questions |
> What effect, if any, does boiler scale and > or mud have on the readings obtained from a > UT? Are the instruments calibrated only for > the density of the steel? I was wondering what are trying to ut. Is kent fra? regards bb |
|
| Author: | ge13031 [ Tue Dec 30, 2003 11:31 pm ] |
| Post subject: | Re: UT limitations |
> In addition, with suspicious parts or or > complicated areas of a boiler, ultrasonic > testing alone is *not* very reliable. Dye > penetrant will show some surface cracks. > Magnaflux will show deeper flaws and, I > think, should always be used to confirm > ultrasonic. In areas where cracking is > suspected, x-ray examination is often > indicated as well. This is another expensive > procedure. It must be done by professionals > (who usually must have the geometry of the > part carefully explained to them). > Perhaps Dennis could recommend an > oscilloscope-readout ultrasonic machine > within reach of most boiler operators. Unfortunately I can't, my equipment knowledge is several years out of date ... a local testing lab would surely provide some help or hints. Ultrasonics work rather well when looking for defects IN sheet, plate, pipe and small bars because the grain structure has been refined in the rolling process (unless the product was rolled too hot and contains "blown grains") Once you get into heavier sections your internal grain structure gets larger towards the center and you can get some strange results. At Homestead Wheel and Axle they had a rather large ultrasonic unit that could send a pulse from one end of a railroad axle to the other ...rejection was based on the percent of signal that got through. For this same reason it is difficult to drive a signal through a casting with its coarse grains and in the case of grey iron, large graphite flakes. Probably the most demanding product we ever ultrasonic teated were the seamless pipe we produced for submarine ballast tanks ... 26"OD, 0.75" wall, 43 ft long, 100% sonic test with wall thickness, defect dia. and location plotted for USN approval. This is definitely an application where you need a zero failure rate. There was research going on with submerged ultrasonics, where a large awkward part was submerged in water ...but I digress. We never had much luck with magnaflux showing up subsurface defects ... we would get a "fuzzy" indication, test grind it, and maybe once in awhile find something ??? The advantage of the magnaflux process is that you can calibrate it to only show those surface discontinuities you consider deep enough to cause problems whereas dye penetrant picks up everything. If it is absolutely imperative that you "prove up" a large casting or weldment then x-ray provides the most definitive answer. I do believe this was done with cylinder castings for several locomotives in Australia. I watched them use x-rays to check welds on nuclear waste containment cylinders in England ... the welds were multiple pass submerged arc machine welded approx. 5" wide at the top and 12" deep. Back to real life ... Calibrate your equipment against a good clean standard ... Clean both surfaces for sonic testing ...visual inspection of both surfaces to give you confidence in the UT process and you should have no surprises. I don't see UT inspection replacing a good pull the tubes, clean, and visual inpect all surfaces inspection because it can't get into all the nooks and crannies, poke around rivets etc. that visual inspection can. lamontdc@adelphia.net |
|
| Author: | Earl Pitts [ Wed Dec 31, 2003 12:18 am ] |
| Post subject: | Re: UT limitations |
> We never had much luck with magnaflux > showing up subsurface defects ... we would > get a "fuzzy" indication, test > grind it, and maybe once in awhile find > something ??? The advantage of the magnaflux > process is that you can calibrate it to only > show those surface discontinuities you > consider deep enough to cause problems > whereas dye penetrant picks up everything. > If it is absolutely imperative that you > "prove up" a large casting or > weldment then x-ray provides the most > definitive answer. I do believe this was > done with cylinder castings for several > locomotives in Australia. I watched them use > x-rays to check welds on nuclear waste > containment cylinders in England ... the > welds were multiple pass submerged arc > machine welded approx. 5" wide at the > top and 12" deep. > Back to real life ... Calibrate your > equipment against a good clean standard ... > Clean both surfaces for sonic testing > ...visual inspection of both surfaces to > give you confidence in the UT process and > you should have no surprises. I don't see UT > inspection replacing a good pull the tubes, > clean, and visual inpect all surfaces > inspection because it can't get into all the > nooks and crannies, poke around rivets etc. > that visual inspection can. It sounds like we're talking about two very different things here. "UT testing" as part 230 refers to it is strictly to be used for determining the thickness of the metal, not looking for cracks or other defects. There are, as has been pointed out, other NDE methods to look for cracks, pits, laps, voids, etc. As has been pointed out already, it is very important that the tester be completely qualified on the machine and the methods, that he use a machine with a graphic readout (such as a Krautkramer), that he know how to read and interpret the results, that he calibrate the machine with test blocks, cleans the surface, and uses the proper couplant. The little digital-only machines are not much use in pressure vessel testing. They are fine for testing things like the thickness of the tender sheets, but they have a hard time telling the difference between scale, dirt, paint, etc. and steel. The thickness of the boiler and firebox plates is essential for calculating the Form 4. It's also important to test in the locations where the steel is likely to be the thinnest, such as around the staybolt holes, where most corrosion is the worst. And yes, there are people out there blithly testing these things that have no earthly idea what they are doing. |
|
| Author: | dave [ Wed Dec 31, 2003 8:28 am ] |
| Post subject: | Re: UT limitations |
> The little digital-only machines are not > much use in pressure vessel testing. They > are fine for testing things like the > thickness of the tender sheets, but they > have a hard time telling the difference > between scale, dirt, paint, etc. and steel. > And yes, there are people out there blithly > testing these things that have no earthly > idea what they are doing. Digital instruments are actually very good indicators of thickness of material provided they are used properly on correctly prepared surfaces. Cleaning and visual inspection should be done prior to UT, and then there will be no paint or scale present to confuse the untrained inspector. BTW, especially in flanged or formed sheets, it is possible to find small areas where the thickness is more than original spec due to upsetting in the forming process. More potential confusion. The benefit of the new Form 4 process it seems to me isn't so much in the exact technicalities as in the overall requirement to actually make a real and complete assessment of some boilers which hadn't been legitimately surveyed in decades. Even if your measurements are a couple thousandths off, any patently unsafe conditions will make themselves known to even semiskilled inspectors. dave irondave@bellsouth.net |
|
| Author: | Ray Dewley [ Wed Dec 31, 2003 8:35 am ] |
| Post subject: | Re: UT technical questions |
> I was wondering what are trying to ut. Is > kent fra? regards bb bb, Nothing to do with Kent, #5's boiler is fine - CAMA's RR is not under FRA - doesn't cross public roads and as you know, insular. I am just curious about UT and how much stake thier lives on it being absolutely accurate. I have heard a story which causes me to wonder (I'm following up on now), is it the technology, specific device, or operator failure..... BTW, bb, what are you up to these days????? Ray |
|
| Author: | ge13031 [ Wed Dec 31, 2003 8:51 am ] |
| Post subject: | Re: UT limitations |
> It sounds like we're talking about two very > different things here. "UT > testing" as part 230 refers to it is > strictly to be used for determining the > thickness of the metal, not looking for > cracks or other defects. There are, as has > been pointed out, other NDE methods to look > for cracks, pits, laps, voids, etc. You are absolutely right ...the advantage that a scope system has is that it will give you visible record of a goofy result. If, as Dave mentions, the plate is thicker due to fabrication the scope will show it as a clean back reflection. If you get a thin spot you can check to see if this is due to actual dimensions or an internal defect. The digital reading is a result of mathematics replacing the trained technicians eye for go/nogo result. All this jazz may be more then you want to know but it sure would give you more confidence in the results. We used digital UT in checking wall thickness on finished pipe ...good clean surfaces ...for a spec check and a remaining wall thickness check after repairs. Worked fine but any funny readings were verified with a scope unit. lamontdc@adelphia.net |
|
| Author: | ge13031 [ Wed Dec 31, 2003 9:17 am ] |
| Post subject: | Re: UT technical questions |
> I am just curious about UT and how much > stake thier lives on it being absolutely > accurate. I have heard a story which causes > me to wonder (I'm following up on now), is > it the technology, specific device, or > operator failure..... The situations I have been personally connected with; 1. Every US nuclear submarine has one source for its ballast tanks namely seamless tube made at USS Lorain, OH and finished in Christy Park, PA. Despite senator Proxmire this is the only operation that can do it. Needless to say this product is very carefully inspected and finish UT is critical ...we have never had a failure in use! 2. Along the same lines the gas cylinders for the space shuttle launching platform are manufactured under the same specs. and have never failed. 3. Axle materials for railroads, trucks, automobiles, combines etc. undergo automated UT on a routine basis since they are safety critical parts. Like any other technology it has its faults and limitations, including equipment malfunctions, operator error, falsification of records and just plain failure to perform. If you are going to hang your hat on Ultrasonic Testing you had better make sure you are doing it right. lamontdc@adelphia.net |
|
| Page 1 of 1 | All times are UTC - 5 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|