What's new
What's new

Machinable magnet bar stock

cross hair

Aluminum
Joined
Jan 16, 2007
Location
Ohio
We need to make some custom magnets to fit in an assembly. The magnet will be on a moving part and be used to activate a proximity sensor.
I have not been able to find anything such as machinable magnet bar stock. I am thinking of ordering a large magnet and trying to machine it down. What I end up needing are wafers 1.870 Dia. x .050 to .075 thick. I would like to have a hole in the middle as well to make installation easier.
I would appreciate any input from those who know of a source for magnet bar stock or have any experience machining magnets. My guess is we won't have much success but unless someone tells me not to even bother I will order a magnet and see what happens.

Thanks,
Don
 
I was under the impression that magnets can't be machined. Think, where do the chips go?. If you really have to machine them, then first you must demagnetise them and remagnetise them after machining.
You don't say exactly the field orientation, is one face N, the other S, or is one edge N the opposite one S.
Magnetic fields can be conducted to the working area by bits of soft iron or some exotic magnetic material, so if a magnet can be "hidden" somewhere, its field could be conducted to a more useful location.
Frank
 
All depends on what type of magnet we're talking about, there's so many.

Most of them will grind ok but slowly and can/will fracture very easily.
If grinding, keep wet as the dust will burn with some of them.

Wire EDM is how many of the shaped magnets are made.

Easiest and often best approach is to have a magnet manufacturer in China take care of it all.
 
You don't say how strong the magnet needs to be, I'll guess that to activate a proximity switch it's not too strong. I think our friend Peter has offered you the best solution, just punch them out of magnetic sheet.

As to machining magnets, magnet being a sort or 'relative' term here, only thing I can add is that we drill a .125" dia. hole .625 deep in N35 neodymium magnets for a product. They are a pretty powerful magnet to say the least. We made up PVC tube for the lathe chuck, a blast of compressed air deals with the chips. Done on an old cone head engine lathe that is appropriate for such use. Kind of hard on HSS drill bits but for the magnets we drill it works fine.

Gus
 
IBM frequently timed mechanical operations using magnets with mercury-wetted reed switches and Hall effect transistors, usually with an adjustable mechanical moving shunt made of steel that would shunt the field, on their old I/O typewriters, printers, and tape machines. These were permanent magnets and were usually about .125" X .125"X .250" long in size. They looked like they were sliced out of a larger magnet. Maybe with a slitting saw. I like the idea of using air to remove the chips. Probably like machining cast iron without the carbon. May try it. Joe
 
I am not sure how strong the magnet needs to be to activate the prox switch. Polarity doesn't seem to matter. There will be .360 thick wall of aluminum between the two but our test have shown that not to be an issue. The part the magnet will be mounted on is basicly a free floating piston (no rotational lock), we want to limit the vertical movement but still allow it to float rotationally.

I like the magnetic rubber sheet suggestion, I will pull that web site up and see whats there.

Having magnets custom made may be an option if we could keep the cost down, the problem is we only anticipate needing 300 a year.
Tough Tool offered the link to KJ Magnetics, I will ask them to quote the size we need, maybe we'll be surprised and the price won't be so bad.

Another suggestion was to magnetize tool steel to use. Wouldn't the steel lose it's magnetism over time?

Thanks to all,
Don
 
use the right kind of sensor and you can use ordinary steel/any metal, why make life difficult?
would not like to try machining samarium cobalt under any circumstaces
mark
 
Mark, no one here has much experience with sensors so it's possible we are making things more difficult then necessary.
Our thinking was we have an aluminum piston in an aluminum sleeve so we needed something non-metallic to trip the sensor, so a magnetic sensor made since. Maybe we're wrong, let me know.

Thanks,
Don
 
Can you machine a pocket into a piece of aluminum or even plastic then glue a magnet in that ? If that's an option it seems like it would be easier than altering the magnet.

Edit to add, is this an electronic valve ?

Clint
 
Clint,
Is we locked the piston into position so it wouldn't rotate we could do as you suggest, we have discussed that option if we can't come up with a magnet to fit. That approach creates other problems in machining and assembly.
The assembly ends up being a safety device on a robotic welder. If the welding neck gets knocked out of position the robot shuts down.
 
I have worked with proximity switches a good bit, not on the switch, but setting them, replacing them when they get damaged etc. I have not researched them, so we may have called them a generic term, and may have been using something else. This is a black plastic case about the size of square piece of bubble gum, the switch did not use a magnetic trip, only a metal trip as the sensor metal. They even have a red LED light on the back so when you set the correct distance between the sensor and the metal (not magnetic) the light would come on and all was well. If there is more than one kind of proximity switch maybe you could try one that only requires a piece of metal to trip the switch rather than a magnet. Just trying to help out with the main problem, I have no help at all with machining the magnet.

We use these to measure throughout a 360* rotation, to activate (switch) timed events, and also for a confirmation at certain points in a 360* rotation. You can use the switch to turn on or off, or in other words to activate or deactivate. I hope I have not added more confusion to your project.

Clint
 
capacitive or inductive

Previous posts have me thinking about this and a quick look into a Grainger catalog confirmed my suspicions that they are right. You are looking at inductive prox probes. If you use a capacitive prox probe it will sense ferrous metal and eliminate your need of a magnet. I am no expert but have seen and serviced them in the past. Take a look around and I bet you find that you can use one of these in the mount that you have designed and do your machining on steel instead of a magnet. Someone smarter than myself will probably provide you with better insight!
 
One way to approach this problem would be to have a disc of soft iron on your piston. Make pole pieces with a magnet in series with it. Basicaly like a pair of metal bars say .125 thick and say 1" wide, at the cylinder end cut out a round groove in each one so they are a good fit on the out side of the cylinder.The unmachined ends have a powerful magnet + iron spacer between them. The theory being that when the disc (+ ali + air space) is in line with the pole piece the magnetic field will increase compared to when the disc is missing. With any luck your proximity detector can be set to detect this change of magnetic flux.
If the polepieces are two close together the field will "leak" directly between them and the change of field will be smaller.
Frank
 
Don,
Thanks for sharing your application details. If the piston is not inside of an internal combustion cylinder but part of a sealed unit, I think one of the thin round magnets slightly smaller in diameter than the piston, glued on top of the piston with a Hall effect transistor mounted outside the cylinder may work. You may need to play around with field orientation but it looks like the field orientation can be any direction. see here at: http://www.kjmagnetics.com/magdir.asp
Joe
 
Last edited:
the magnetic type of sensor uses a reed switch [tiny glass tube with BeCu spring and a bit of Fe spot welded on, in magnetic feild it closes, if you use a proximity sensor you can acheive the same useing induction, the sensor has an oscillator thingy in it and a coil, like a baby metal detector, does require a voltage google proximity switches for all the techy stuff
regards
mark
 








 
Back
Top