Magnetization and demagnetization

[Dario2002]

I see that many posts regarding demagnetization topic is present in this forum. Let's try to be more practical.

Let's say we have 20" pipe, and we measure 35 gauss on pipe end. Heat treatment is probably not suitable for demagnetization. Next option is demagnetization using welding cables. How many coils we should make, and what current (AC) we should give through those cables. Is there any calculation model for this?

1) Can we help our selfs if start with some stronger current, then reducing gradually? How fast?
2) Can we help if put current through cables and move them out from pipe?
3) Reversing polarity few times?
4) Can it help to connect those pipes somehow before welding (ground bridge)?


And one theoretical question. We have info from pipe welders that magnetization was not uniform on pipe end (circumferential). Is this possible to have areas at circumferential edge with different levels of magnetization? What difference of magnetization in pipe steel can be expected (gradient at one meter or 1 cm)?

 

[BigInch]

You might want to post on the EE forums, if you want to DIY.
Here we use a Zeromag to degause it. No fuss no muss.

http://www.engineerlive.com/content/22345

Independent events are seldomly independent.

 

[hacksaw]

you can use commercial degaussing coils or one from a junked CRT, permanent magnets if you know how, or just clamp a piece of 1/8" to 1/4" thick strip of steel adjacent to the weld (not in it) to siphon the flux away from the cut while you weld

once the two pipe-ends are joined, move or remove the strip

 

[JR97]

Dario,

Two conditions must exist in order to demag: constantly reversing field and gradually decreasing amplitude field. If you are using AC current, you have the constantly reversing part. You still need the gradually diminishing. You can either statically place the cables and lower the current (since AC is running at 60 Hz, 1 to 2 seconds will give enough cycles) to zero. Alternatively you can keept the current constant and move either the coil or the pipe away from each other. Maybe a 15 second time would be sufficient.

to answer your question on number of wraps ... as many as possible. Field is directly proportional to number of turns in the coil. For large diameter pipe it is quite common to have "hot" spots around the pipe.

A couple of potential snags: The field produced by the coil will only be effective for about the radius of the coil out either end. For a long section of pipe you will need to zap it near the end, move the coil and zap it again, etc until the whole length of pipe is inspected. What can happen in this instance is the field can "sneak" behind your progression and you don't end up being successful. In that case try to keep current constant and move the coil along the pipe.

If the field is really stubborn you may need to use DC current. The field produced by AC current will not penetrate into the material as will the field produced from DC. With AC you could be suppressing the field toward the ID only to have it return after the current is removed. DC demag is much more laborious. You need to connect and momentarily hit the current, reverse the leads reduce the current marginally and hit current again. Continue reversing the leads and marginally reducing the current until at zero. Remember the sequence may need to be repeated at multiple locations along the length of the pipe.

As a last resort, try to turn the field from longitudinal to circular. Connect leeches to either end of the pipe and pass current down the length of the pipe. With sufficient current the direction of the field will be changed to around the pipe instead of along it. There will be no readings on the end. Whether this is acceptable or not depends on the end use of the pipe. The pipe is not demagnetized, the field is just in a direction where it cannot be measured and should not interfere with welding pipe to pipe.

Hope this helps.

JR97