Remagnetising a tiny rod magnet

[Chris Hogg]

I have a very small rod magnet, 4mm dia x 8mm long, part of the tipping-bucket mechanism in an automatic rain gauge. The magnet has lost its magnetism to the extent that it will no longer close the reed switch that registers when the bucket tips. I have tried replacing it by a similar sized neodymium magnet, but this is too strong; the movement of the magnet during its travel away from the reed is not sufficient to allow the reed to open.

I have tried putting a couple of 'keepers' either side of the Nd magnet, which do reduce its strength sufficiently for the reed to open and close properly, but the additional weight of the 'keepers' is such that the bucket tips too early and completely upsets the calibration of the gauge.

I have looked for a traditional Alnico rod magnet of the right size, but cannot find one. Plenty much bigger, and plenty of Nd magnets of the correct size, but nothing in Alnico.

It seems my only solution is to remagnetise the old rod magnet. I have tried heating it to dull red heat and then picking it up with a little Eclipse horseshoe magnet while it was still hot, expecting it to retain some magnetism as it cooled, but it didn't work. I don't understand why not; I thought that was a standard method for magnetising steel.

I'm open to suggestions as to other methods for remagnetising the magnet, bearing in mind it is fairly small.

[Bob Minchin]

Assuming the poles are wanted at the ends (?) Wind a few turns of copper wire around the bar and flash the ends across a car battery a few times - should work.

[Rob Morley]

Why not use a smaller Nd magnet? You can get disc magnets 2mm wide by

1mm thick - one or two of them should do the job.

[Dave Liquorice]

Are you sure that the magnet had lost it's magnetisium not that the reed switch had got tired? The neodymium magnet being so strong would make even a tired reed work. The TBRG and anemometer here both use reed switches and I've replaced both several times. The magnets in both are orginals, from 1999.

The one in the anenmometer gives up just from the sheer number of operations, one per revolution of the cups. 20 rpm = 10 million operations/year...

The rain guage one gives up either due to corrosion or the wire/glass seal failing. After the repairs these days I give they circuit board assemblies a good all over squirt with a conformal coating to keep the damp out.

Now you definitely have a non-magnetic magnet I'm not sure how you'd get it back. I think you need to magnetically "shake" the crystals and apply a strong magnetic field, then reduce the the "shaking" with the strong field still applied. But the field strength required might be very high, there is a property (whose name I forget) were the field has to be above a certain level before the crystals stick in that position.

Wonder if wiping your neodynium magnet repeatedly in the same direction would work? Trouble is we don't know if the poles where end to end or face to face. I guess google will indicate the orientation of the field to operate a reed. I have a feeling it's a pole, rather than the field bewteen poles.

[The Natural Philosopher]

the way we did it at school was with a big solenoid. a few turns of quite thick wire connected via a fuse and a switch to the mains.Or it might have been a car battery You threw the switch, the fuse blew and anything ferrous inside got magnetised.

It was advised to remove your watch before throwing the switch.

[Tim Streater]

Won't work with AC, now, will it. Or are you hoping the fuse blows in less than half a cycle?

[Bob Eager]

How about those things that magnetise screwdrivers?

[Andy Wade]

Try it and be surprised...

If the prospective s/c current is large compared with the fuse rating (say 30x or more) and it's a sand filled fuse then that will generally be the case. 'Current limiting' is the term used - the fuse opens and the arc is quenched within a fraction of a cycle, such that the full prospective current is never reached.

[The Natural Philosopher]

generally blows at full cycle.

it works

[harryagain]

Get your neodymium magnet and repeatedly "stroke" your old magnet in the same direction.

[Brian Gaff]

Yes you need a very big field to remagnatise anything. If it seems easy to do then it won't last very long hot or not!

Hence screwdrivers having to be done over and over. Brian

[Chris Hogg]

Many thanks for all the suggestions. I like the idea of giving it a belt with the mains; fairly easy to set up. But I have a couple of queries. Will an ordinary 13A domestic plug cartridge fuse in a standard plug be OK, or does it require something special? Bearing in mind the length of the bar (8mm), any suggestions as to what wire I should use? Insulated, obviously, to avoid shorting between turns, but I only have bits and pieces of 1mm^2 and 2.4mm^2 cable handy that I can strip out a single wire from (although I could probably find some similar sized flex). The thickness of the insulation will limit the number of turns I can get over 8mm, but if I go for the thinner wire (i.e. the 1mm^2 stuff), it will not have such a low resistance and may reduce the peak current. Any suggestions as to how many turns are necessary?

If that doesn't work, RM's suggestion of a much smaller Nd magnet is a good idea, although I will have to compensate for the lower weight in order to keep the correct calibration of the rain gauge, but that shouldn't be difficult.

If all else fails, I'll try Harry's suggestion of stroking it with a Nd magnet, although the small size and the technique will make it fiddly. Still, if it works.....

[The Natural Philosopher]

I think at that size I'd be inclined to use teh coil and car battery approach. essentially get a small cardboard tube and wind a couple of hundred turns of simple pvc covered wire over it, and hold it all in place with masking tape and just 'flash' that across a car battery.with the magnet thing inside

[Chris Hogg]

Thanks Dave. When the gauge failed (at the end of last month's dry spell, which is how I noticed there was a problem), I had a good look at it and found that even a small horseshoe magnet wouldn't operate the reed, so I assumed that the reed was the problem. This opinion was reinforced when the one of the sealed ends of the reed cracked open as I removed it, even though not using any degree of force; I assumed a crack had developed in the seal, possibly due to corrosion of the wire, and damp had got in. But having replaced the reed, the existing magnet still didn't have enough power to close it, even though the little horseshoe magnet did. I was a bit surprised at this, as I had always thought that magnets lasted almost for ever, at least in that sort of application. Perhaps I got the wrong sort of reed; I just bought one of the right size. Maybe they come with different compliances, if that's the right term, with some requiring a stronger field to close them than others.

For sure the original magnet is no longer magnetic, but I think it was just a simple N-S bar, judging by its effect on a compass needle, and it lies parallel to the reed.

[Dave Plowman (News)]

Buy a magnetic door switch for an alarm? They are cheap - one part contains the reed switch, the other the magnet. If you'd like the dimensions of the actual magnet, I'll measure one for you.

[Dave Liquorice]

Reeds are very fragile, particularly the wire to glass seal. When you form the leads to fit holes in PCB or WHY *always* support the wire between the bend and reed with pliers (and not too close to the reed!) so as to not apply any stress to the wire/glass bit.

DAMHIKT...

I'd agree with that, I've never come across a magnet that is no longer a magnet unless it has suffered a lot of abuse, like being heated to cherry red. B-)

That's basically what I did... I think they do come in different sensitivies and/or field line orientation but I wouldn't expect it to be critical for this application.

Which AWS is the rain guage from?

[Chris Hogg]

If that wouldn't be any trouble, then yes please, although I'd be surprised if it was anything like the same size. If by chance it is, would you let me know details of the door switch please?

[Chris Hogg]

A Davis. They do a repair service, but at a price. I've had it getting on for ten years now and the whole system is beginning to show its age and really needs replacing completely, so I don't want to spend much money on it now. I'd rather get a new system, but probably not a Davis. I've had to replace a number of major components since I've had it. It doesn't seem to have coped very well with the high humidities, winter gales and salty atmosphere we get down here on the coast.

[polygonum]

Have you got your eye on any particular alternative system?

I have been wondering about jumping from my very basic, non-connected Oregon kit and keep shying away from any decision.

[Chris Hogg]

Not yet! If you want advice you could ask on uk.sci.weather .