MuMetal/Metal to Reflect D.C. Magnetic Fields

Only shield I can use is in the one direction will be a flat plate, can not make an enclosure. Sounds like I need to stick with a flat metal ferrous plate which may be my best bet

There is no way to block a magnetic field that I know of....

In addition to experimenting, you might find it useful to download some free magnetic modeling software.

A sketch or drawing would be very helpful.

The purpose of the shielding is not completely clear, but if I understand correctly, you are trying to shield something from the field of all of the magnetic solenoid valves. ...or are you trying to shield the valves from each other?

It's also not clear whether the values you provided are due to the permanent magnets alone, to the solenoids alone, or to the combined solenoid and permanent magnet, presumably with the solenoid energized in the direction to augment the permanent magnet's field.

7cm is a very large distance for any common magnet to have a significant push or pull!

Trying to protect equipment below solenoids 13cm - 30cm away (relays and magnetic core memory). hope this image is readable:

Agree with all your comments; however this is an already built system ready for installation and currently my only option is to attempt to block the fields. If what I gather together so far is correct, I think the ferrous metal plate will at least disrupt the magnetic field which hopefully lowers the peak level in that one direction. A Gauss meter will verify this for me.

Yes, this is what I think will happen once I place some sort of ferrous metal in the direction of concern as you have shown here, my hope is that the peak measured readings will be significantly reduced.

The're not reduced at all....just the path altered....The strength of the field reduces at the square of the distance....just a little distance makes all the difference...

Yes. The drawing helps, as does your clarification! I was visualizing the 6th solenoid in the same plane as the axes of the group of 5...

There should be no problem with relays 13cm or more away from the solenoid valves.

Magnetic core memory! Why? I thought that technology had disappeared from new development at least 30 years ago!

This is being installed in an existing system with old technology. Once I get to make measurements to the entire system, I will have a better understanding.

Found this....

https://www.tindie.com/products/kilpelaj/core-memory-shield-for-arduino/

That is an interesting board and article, but I read essentially all of it, as well as another similar article linked within it, and saw absolutely nothing about magnetic shielding. Although I do have one, I've done very little with Arduinos. I get the impression that they use the term "shield" to refer to any auxiliary board that plugs in to the Arduino.

Similarly, I probably still have a small core memory unit stashed somewhere. I know I had at least one some time in the distant past, and tend not to throw such small items away, but I never did anything with it except show it to my students, 25 or 30 years ago.

Yeah I don't think there is any actual 'magnetic shield' that exists....at least in the form that the OP is envisioning....

...Deflecting the field with encapsulation is about as good as it gets...

TheWilddotter, you are spot on to my problem!

We have not glibly discounted A.C. magnetic fields, I just know that moving magnetic fields is a whole other issue which is not my problem. I only want to address D.C. magnetic fields here since A.C. would open up a whole different discussion..

BTW-I forgot to answer a previous question which was that all my concerns are in the powered off state, these latching relays are only powered for ~100msec to change their positions....

Q1. No fortunately two totally independent systems and not sure yet of the shielding the other system has but will investigate this once I measure the peak of the fields I am dealing with.....I will then do an analysis on the potential impact on the relays/magnetic core memory.

We measured the A.C. Magnetic fields and fortunately since the controller has a built in 100msec delay between energizing the solenoids, we will not have to deal with those, especially collapsing fields, summing since the measurements for. A.C. magnetic fields showed this is definitely not a concern... although I like your troubleshooting technique for this and will keep it in my back pocket if I run across a future issue on a different system; but I'm not sure it would help with my D.C. magnetic field concern.

OOOOHHHHMMMMM, OOOOOHHHHHHMMM, OOOOHHHHMMMM

OK. If you can discount AC magnetic fields, I am going to assume that you can also discount DC magnetic fields from external sources other than your solenoid assembly. I note, as has been well discussed that a complete multi-layer wrapper of mu-metal around the "victim" of stray DC fields is optimal shielding. I agree with this but there may be a rather Zen alternative of wrapping your shielding around the aggressor of interest(your solenoid set). Realize that if someone comes in and puts a crusher magnet opposite your solenoids with your memory in between, that the Zen alternative will be a total disaster. If you say "not my concern" then you can achieve major shielding success by placing a multi-layer wrapper of mu-metal around your solenoids. If that, too, is beyond practicality then a plate shield is probably not the best you can do. As is apparent in Solar Eagle's excellent drawings (and I think he said it as well) magnetic fields travel in circles. The shortest magnetic loop available is the preferred one. High permeability material paths such as mu-metal make the effective magnetic path length shorter where that path is through the high permeability material.

If you frame a rectangular box along its edges and place a plate toward your target, it becomes very difficult for the field to trace short paths out of your solenoids which will reach your target. All solenoid generated DC magnetic paths but paths directly perpendicular to cube faces will therefore remain within your mu-metal plate and mu-metal cube frame and not bother your memory. This means vanishingly small DC magnetic fields from your solenoid array will reach the core memory. Bob's your uncle.

thewildotter

P.S. I have ignored the effects of saturation. Use enough high permeability material to make this irrelevant by adding an appropriate amount of mild steel(does not saturate easily but is somewhat less permeable) inside(or even replacing) your costlier mu-metal. I have also used "cube" loosely when I really meant rectangular box(or parallelepiped).

Yes, all other AC/DC magnetic field contributors will be 2-3 meters away. No other equipment will be present accept the occasional laptop but that will still be at least one and one-half meter away and probably has no magnetic field contribution.

"you can achieve major shielding success by placing a multi-layer wrapper of mu-metal around your solenoids" Ok - this may be what I am looking for, even though I am concerned in only one direction I think you are telling me to go ahead and treat my whole assembly and this will take care of it in all directions.

THANKS, it is much clearer now on what I can or need to do going forward, you've been a big help!

Here is some good information.

http://www.magnetic-shield.com/faqs-all-about-shielding.html

Can you make a mu-metal shield around the core memory? It may make more sense in that the mu-metal is less likely to saturate.

"Should the source of interference or the sensitive device be shielded?

The answer to this question depends on several factors. Shielding the source may involve stronger fields, and therefore thicker materials. One must be sure that all interference sources are shielded, or the sensitive device will still be affected. The usual approach is to shield the sensitive device. This prevents interference from both present and future sources."

http://www.magnetic-shield.com/pdf/should_the_source_of_interference_or_the_sensitive_device_be_shielded.pdf

While completely surrounding the core memory in a mu-metal box would be optimum, this is probably not feasible and the best you can do is to cover the core memory as much as possible with a continuous covering of mu-metal to provide a low magnetic flux resistance path around it.

This equipment with magnetic core memory has probably worked for a very long time. Since you need to "fix" the magnetic field issue there must have been some sort of a change to the design. Can the change go back into design review with the additional specification that the change needs to work in the equipment that it is installed in?

Design change or review is not needed at this time..... I'm just preparing for the new installation of these solenoids (plus 10+ others which are not located near any potential sensitive equipment). Also today found out that the older equipment may just have a magnetic disc drive (as apposed to magnetic core memory) so it will still be sensitive to D.C. magnetic fields (remember in the old days when someone could use a speaker magnet to wipe out a magnetic or floppy disc drive?). Also will soon have the actual distances between the installation of the solenoids and the magnetic disc drive where now I am thinking that they will be far enough apart that the fields will not be strong enough to effect the disc drive. So looking less likely to be an issue. Also suspect with the spacing of the magnetic solenoids being at 2.1 inches apart, the summation of fields will start greatly rolling off for more than 3 solenoids (based on previous tests to other D.C. magnetic field producers in close proximity) Today we did model this with some software which is also looking promising. Will hold off on any request for a new design review since I will know for sure in two weeks once I get to measure the actual fields from the delivered complete 6 solenoid assembly.

A magnetic disc drive is a WHOLE LOT LESS sensitive to external magnetic fields than core memory.

Erase a floppy with a speaker magnet, sure. But erase a hard drive with a speaker magnet, I'm not so sure. It has been a few years since I last tried to do that, but at that time, it did NOT destroy the disc nor the information on it.

Long before they started switching from Hard Drives to Solid State Drives, Apple computers have used rather powerful magnets to hold screens and other parts (like the MagSafe™ power connector) in place, and as latches to hold the cover closed.

Back when I had a laptop with a hard drive, I used it while resting on 75kVA and larger transformers, sitting on machines that did include solenoids of various kinds, and once even on a large granite table that had an array of hundreds of magnets to hold parts in place. On that occasion, I had quite a scare, as the screen went black, but then I realized that the magnets were there, and as soon as I moved away from that table, the computer returned to normal, with nothing lost. Due to the magnetic field, the computer "thought" that I had closed the cover, so went to sleep. The standard sleep sensor in a Mac laptop notices the presence of the latching magnet in the top of the display, to put the computer to sleep whenever the screen is closed, to save energy.

It is definitely a good thing to look ahead to avoid possible problems, but I think in this case, you're worrying too much.

Agree but just want to be prepared for my upcoming test. I'll only have one shot to correct any issues if they exist, which does seem less likely. I will know once I measure the magnetic fields. It will not be in a lab environment so will have to bring what I need and the information gathered here will help determine what that should be.