where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Thickness of Rubber Lining in Steel Pipes
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

GE Inspection Technologies GmbH
offers equipment, systems and expertise based on ultrasonics, radiography, eddy curreent and remote visual inspection.

7149 views
00:15 May-02-2007
Pat Devin
Thickness of Rubber Lining in Steel Pipes

My name is Pat Devin and I am currently working for a Mining company in Indonesia.

I am looking for information regarding the use of Ultrasonic technique that will allow me to accurately measure the remaining wall thickness of a rubber lined (hot vulcanised) steel pipe by measuring from the outside of the pipe only. the steel wall thickness is 12mm and the Rubber thickness (natural rubber) is 20mm. I am told that it is possible, but we have tried several probe combinations without success.

Regards

Pat
+628175749513


 
04:49 May-02-2007

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Thickness of Rubber Lining in Steel Pipes There's a small chance that it may be possible, but it will be difficult at best due to interfering multiple echoes from the steel wall, the poor acoustic impedance match between steel and rubber, and high attenuation in the rubber. If there is liquid in the pipe the situation will be even worse due to the dampening effect of liquid at the inside surface of the rubber. Usually the rubber echo will be lost in background noise.

To try it, you will need a flaw detector or a precision thickness gage (not a corrosion gage) with waveform display, and a low frequency single element broadband contact transducer (preferably 2.25 MHz). When you couple to the pipe, you will see a number of multiple echoes from the pipe wall spaced at regular intervals. Because the velocity of rubber is approximately 30% that of steel, the echo from 20 mm of rubber will occur approximately between the 5th and 6th multiples of the steel echo. Unfortunately it will be quite small because of the impedance mismatch and attenuation issues, and when you increase gain you will see, in addition to the multiple backwall echoes from the steel in that region of time, assorted peaks representing mode converted signals and backing and case noise from the transducer.
It is often not possible to reliably distinguish the liner echo from everything that you will see surrounding it at high gain. If you CAN identify it, you can use appropriate gates and blanking to measure the interval from the first metal backwall echo to the liner echo.

Another problem is that if the rubber is eroding, you won't necessarily know where to look amid the multiple steel echoes.

If you have a piece of unlined pipe that you can use for setup, you should record the echo pattern that you see from unlined pipe. Then, any peak that appears in the appropriate time region from the lined pipe that does not appear from the unlined pipe is likely to be the liner echo.

The one situation where this type of test usually works fairly well is where the round trip transit time in the nonmetallic liner is less than that in the metal, so that the liner echo appears between the first and second backwall echoes from the steel. Obviously that involves thin liners only. I'm afraid that in this case, it's likely to be tough to reliably find what you're looking for.


 
20:44 Jul-20-2012
Tim Rice
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Tom Nelligan at 04:49 May-02-2007 .

I was wondering if there have been any "advancements" in technology that would now allow the thickness measuring if rubber in steel pipes using UT as this question was raised in conversation with a colleague.

Regards,
Tim

 
12:57 Jul-23-2012

Wieslaw Bicz

Engineering, -
PBP Optel sp. z o.o.,
Poland,
Joined Feb 2009
237
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Tim Rice at 20:44 Jul-20-2012 .

You should use two focused transducer under some angle in relation to the pipe surface.
This should cause, that you will see only one reflection from the interface between pipe wall and rubber and it is possible to see the reflection from rubber. All parameters (angle, focus, frequency) must be chosen experimentally - depending on the pipe, rubber, etc.

This approach is not easy and may require special software for signal evaluation, but it should bring good results.

 
04:49 Jun-30-2015
Nelson B. Santacruz
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Wieslaw Bicz at 12:57 Jul-23-2012 .

My question is whether it is possible to measure the thickness of polyurethane coating on the inside of a steel pipe from the outside?

 
11:23 Jun-30-2015

Wieslaw Bicz

Engineering, -
PBP Optel sp. z o.o.,
Poland,
Joined Feb 2009
237
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Nelson B. Santacruz at 04:49 Jun-30-2015 .

This should be possible, but requires surely good signal analysis.

 
14:51 Jun-30-2015

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Nelson B. Santacruz at 04:49 Jun-30-2015 .

Mr. Santacruz, please refer to my May 2007 posting above that describes the situation in detail. Unfortunately, physics hasn't changed since then! Polyurethane, like rubber, is highly attenuating and a poor impedance match to steel, so the same principles apply. The measurement will be very difficult at best and likely not practical unless the coating is so thin than the transit time through the coating is less than the transit time through the steel wall.

 
15:24 Dec-06-2015

Christer

Sweden,
Joined Dec 2015
7
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Tom Nelligan at 14:51 Jun-30-2015 .



Hi everyone,

I'm working in a project where I need to measure the thickness of rubber through steel, i.e. steel between the transducer and the rubber. The steel is ~2 cm, and the rubber is around ~10 cm or ~20 cm. Initial set-up in lab will be in contact mode, but later on, squirter is required in echo or pitch-catch shear wave mode. The interface contact pressure of the steel-rubber is very high.

I have learned, partly from the forum, factors to consider:

- Low frequency
- High voltage
- Square wave
- Pitch catch shear wave (possible in the ~10 cm set-up)
- Signal analysis
- Tone burst (perhaps no advantage over 220 Vpp with standard piezo-ceramic elements)
- Hit the steel's (first) harmonic

Currently, I need to find equipment that have a high potential to work. Would it work with an Olympus 5058PR (900 Volts) together with an transducer of 0.1 MHz (Olympus V1011) and 29 mm steel (1st harmonic at 5800 m/s)?

RITEC (suggested by INSIDIX )has some high power UT, e.g. RPR-4000 at 8 kW tone burst, giving 500 Volts at 0,1 MHz, but up to 2000 Volts at higher frequencies. I also saw that the RPR-4000 has 200 us pulse width while 5058PR is at 4 us. However, I don't know what those figures mean, e.g. 8 kW tone burst. Also found Innerspec PowerBox 1200 Volts.

Further thoughts are:

- Does Olympus 5058PR have square wave pulse mode?
- It's noted here in the forum that USM 23LF is at 1200 Volts. Is there any replacement for USM 23LF? Or is there any pulser-receiver more suitable than Olympus 5058PR 900 Volts?
- If required, would it be any problem to make wedges, e.g. for V1011?
- Should I add an preamplifier? e.g. 5660B to 5058PR
- Should I go for a lower or higher frequency?
- Any concern regarding resolution? I would like to measure the thickness with +/- a few millimetres.
- optel.eu has sopme 1300 Volts pulsers bot up to 170 Hz. Is that for some long range UT or could I use something similar?

The rubber has a v_L of 1566 m/s. With 1 MHz spike at 300 Volts, I can measure up to 36 cm (6 reflections from 60 mm sample).

 
00:07 Dec-07-2015

Wieslaw Bicz

Engineering, -
PBP Optel sp. z o.o.,
Poland,
Joined Feb 2009
237
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 15:24 Dec-06-2015 .

It is a difficult problem, since you will have many strong reflections in steel and a very weak reflection from rubber (the amount of energy transmitted from steel to rubber will be quite small and attenuation in rubber high). I would assume, that it will be difficult to find a signal from the opposite wall in rubber in the noise, generated by the sender in steel. From my point of view the use of two transducers (separated sender and receiver) placed under same angle in relation to axis perpendicular to the surface of steel can give more chances to select the proper signal, especially, if you chose correct geometry, causing, that direct reflection in steel will not reach the receiver.
The solution of the problem has not much to do with equipment chosen, but much more with suitable geometry of the setup (wave propagation) and signal analysis.

 
01:38 Dec-07-2015

John Norman

Consultant, owner of business
NTS Ultrasonics Pty Ltd,
Australia,
Joined Oct 2012
107
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 15:24 Dec-06-2015 .

Hi Christer.

Wieslaw is correct that a two transducer configuration will give a better chance of detecting the backwall echo from the rubber/air interface. If you can get the reflection from 36 cm using 1 MHz and the equipment you mentioned, there is probably no need to use anything more "exotic". I have done a lot of work inspecting thick sections of rubber (think earth mover tyres and similar off road tyres), and a high sensitivity 1 MHz transducer always gave me the best results (although, to some extent this depends on the properties of the rubber). A problem with trying to use lower frequency transducers for rubber is that you may need to start winding and fitting inductors and modifying the matching of the transducer (plus cable) to the flaw detection equipment.

Regards
John Norman

 
15:17 Dec-07-2015

Tom Nelligan

Engineering,
retired,
USA,
Joined Nov 1998
390
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 15:24 Dec-06-2015 .

Mr. Christer, you had a few questions about one of my company's products. Here are my answers.

"Currently, I need to find equipment that have a high potential to work. Would it work with an Olympus 5058PR (900 Volts) together with an transducer of 0.1 MHz (Olympus V1011) and 29 mm steel (1st harmonic at 5800 m/s)?"

The V1011 transducer is a high power, very low frequency contact transducer that is designed for measurement of highly attenuating materials or across very, very long sound paths. It would typically work well in direct contact on very thick rubber, but as you know the steel causes major problems. You will see a large number over very high amplitude multiple echoes from the 29 mm steel. In this case, because the rubber is so thick (100-200 mm), there is a small chance that the steel echoes will decay before the backwall echo from the rubber arrives, however my guess is that the steel echoes with mask any reflection from the rubber. Attenuation in steel at 0.1 MHz is so low that there will be LOT of reverberating energy in the steel.

"Does Olympus 5058PR have square wave pulse mode?"

No, is 5058PR is a high voltage (900v) spike pulser. Our square wave pulser/receiver is the Model 5077PR.

"Or is there any pulser-receiver more suitable than Olympus 5058PR 900 Volts?"

I don't think the challenge is in driving the transducer harder, rather it is in identifying the backwall echo from the rubber amid all the reverberations generated in the steel.

" If required, would it be any problem to make wedges, e.g. for V1011?

I'm afraid that is not something that we offer, but of course you could build one yourself.

"Should I add an preamplifier? e.g. 5660B to 5058PR

Your 5058PR already contains an integral 30dB preamp that can be inserted into the sound path. This is in addition to the primary 60 dB receiver amplifier. Check your operating manual. However, again, the issue here is not absolute sensitivity but rather signal-to-noise. Adding yet another amplifier to the system will do nothing to improve s/n.

 
21:40 Dec-07-2015

Christer

Sweden,
Joined Dec 2015
7
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Tom Nelligan at 15:17 Dec-07-2015 .

Thank you very much for the replies. I will make some sketches of pitch-catch models and order another transducer. I will get back to the thread when I have some sketch of a set-up.

 
10:27 Dec-09-2015

Anmol Birring

Consultant,
Birring NDE Center, Inc.,
USA,
Joined Aug 2011
588
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 15:24 Dec-06-2015 .

This is a long shot. You are dealing with high attenuation in rubber, large thickness, and high impedance mismatch between steel and rubber. I doubt if it can be done with ultrasonics.

 
11:16 Dec-09-2015

Wieslaw Bicz

Engineering, -
PBP Optel sp. z o.o.,
Poland,
Joined Feb 2009
237
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Anmol Birring at 10:27 Dec-09-2015 .

It can be done, but is difficult.

 
15:15 Feb-04-2016

Christer

Sweden,
Joined Dec 2015
7
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 21:40 Dec-07-2015 .


Hi, I said that I will return when I have made some sketch of a setup. However, I had a successful test on 9 mm aluminium vulcanized against 96 mm rubber (v_l equals 6300 and 1566 m/s). I used a 1 MHz transducer at 300 volts 500 ns square pulse excitation. The transducer was handheld against the aluminium in an echo-setup. No signal processing, averaging or such. Interestingly, I had to tilt the transducer 1-2 degrees to get a good signal. The ringing reduced. Likely the sound had time to reflect away laterally in the aluminium.

 
13:56 Dec-18-2016

Christer

Sweden,
Joined Dec 2015
7
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Christer at 15:15 Feb-04-2016 .


The project report for those interested is available through the link below. Comments and experience in the topic are appreciated.

[url=http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-23514]Ultrasonics for Monitoring of Mining Mill Linings[/url]

 
10:10 Dec-19-2016

Simon Amallraja

Consultant, Advanced NDT Methods
Anu Consultants,
India,
Joined Aug 2013
38
Re: Thickness of Rubber Lining in Steel Pipes In Reply to Pat Devin at 00:15 May-02-2007 (Opening).

You can use Tone Burst PR system - may be RAM 5000 Mark V or VI from RITEC will help for this application. We have supplied one system in India for similar application.

Kindly check the specifications of RAM 5000 and it will help you to understand your application in a better way.

 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|