Magnetic personality
I recently sorted through a stack of old IDE hard drives that have accumulated here at Casa de Loco over the years. The Connors were all toast, so today I took them apart to see what was what. One had a very obvious sticktion problem, that's where the media surface gets a tacky coating from left over cleaning and lube liquids recondensing on the platter. Over time that gets bad enough the heads will stick and hold the platter, keeping it from starting to turn when turned on. In this case it wasn't that bad yet, but there was a definite defect right where the heads come to rest when the drive powered down, ruining track 0 and maybe track 1. The other two I couldn't see anything wrong with, but still not working.
So I dug the magnets out. Why? I dunno, I've always had a thing for magnets, ever since I dug a pair out of a toy's DC motor back in the 50's. Magnets were magic to me as a kid. I loved how I could make one move on top of the coffee table by moving the second one underneath it. I was seven. Lots of things were magic to me back then.
The ones I got out of these drives were scary. I've not played much with neos (Neodymium) before. And these things are brutal. I use a hex drive handle with a magnetic bit holder with a large variety of bits for those special 'secure' screw heads that are so popular today. And while unscrewing the magnet mounts, the bit got sucked into the magnet pole. I didn't think anything of it until it was time to put my tools away and couldn't get the darn bit out of the drive head. That one contact was enough to super magnetize the bit, enough that I had to use pliers to pull it out. Now I wonder if I need to make a demagnetizer for the bits.
Totally unconnected magnet story -
Back around 1996 I was designing a motor drive and control system for a tiny permanent magnet motor that used neos. They were glued to the armature and the coils were on the stator, the reverse of the common ferrite magnet motors. But it was brushless and that was the goal. I was using an 8 pin DIP Microchip MCU for the control. Two outputs controlled a pair of MOSFET H-bridges and two pins were for sensing the armature position, the remaining two pins were for speed up and down push buttons. And +5vdc and ground pins. Simple system. Turn on the MCU, push and hold the faster button until the desired speed was reached, then push the slower button to slow down.
The motor maker wanted a max speed of 1000RPM, so that was easy to set in software. But I had to play with it a bit and set a max of 2000RPM just to make sure the whole thing would still work. Bad idea. I found out why the motor was to be limited to 1000RPM, the glue and the coating on the neos wasn't strong enough to hold at 2000. I slung one magnet off the armature. The coating had pulled away.
I did some research to find out why there was a coating, and it's because neos corrode in air so they have to be coated, usually in something epoxy based. And that coating limits their performance in motors due to the imposed additional air gap. Plus it weakens the mounting and makes them susceptible to scratches. The coating forces the designer to use a larger neo than he should to compensate for the air gap.
Ever since, I've thought neos were an expensive and inefficient thing to use, but I suppose there's a cachet to advertising your product uses the most expensive and powerful Neodymium magnets. Enough to make up for their extra cost. I think Cobalt-Samarium magnets make better sense and plain old Al-Ni-Co will suffice in just about every use. And ferrite is still king for cost effective designs.
So I dug the magnets out. Why? I dunno, I've always had a thing for magnets, ever since I dug a pair out of a toy's DC motor back in the 50's. Magnets were magic to me as a kid. I loved how I could make one move on top of the coffee table by moving the second one underneath it. I was seven. Lots of things were magic to me back then.
The ones I got out of these drives were scary. I've not played much with neos (Neodymium) before. And these things are brutal. I use a hex drive handle with a magnetic bit holder with a large variety of bits for those special 'secure' screw heads that are so popular today. And while unscrewing the magnet mounts, the bit got sucked into the magnet pole. I didn't think anything of it until it was time to put my tools away and couldn't get the darn bit out of the drive head. That one contact was enough to super magnetize the bit, enough that I had to use pliers to pull it out. Now I wonder if I need to make a demagnetizer for the bits.
Totally unconnected magnet story -
Back around 1996 I was designing a motor drive and control system for a tiny permanent magnet motor that used neos. They were glued to the armature and the coils were on the stator, the reverse of the common ferrite magnet motors. But it was brushless and that was the goal. I was using an 8 pin DIP Microchip MCU for the control. Two outputs controlled a pair of MOSFET H-bridges and two pins were for sensing the armature position, the remaining two pins were for speed up and down push buttons. And +5vdc and ground pins. Simple system. Turn on the MCU, push and hold the faster button until the desired speed was reached, then push the slower button to slow down.
The motor maker wanted a max speed of 1000RPM, so that was easy to set in software. But I had to play with it a bit and set a max of 2000RPM just to make sure the whole thing would still work. Bad idea. I found out why the motor was to be limited to 1000RPM, the glue and the coating on the neos wasn't strong enough to hold at 2000. I slung one magnet off the armature. The coating had pulled away.
I did some research to find out why there was a coating, and it's because neos corrode in air so they have to be coated, usually in something epoxy based. And that coating limits their performance in motors due to the imposed additional air gap. Plus it weakens the mounting and makes them susceptible to scratches. The coating forces the designer to use a larger neo than he should to compensate for the air gap.
Ever since, I've thought neos were an expensive and inefficient thing to use, but I suppose there's a cachet to advertising your product uses the most expensive and powerful Neodymium magnets. Enough to make up for their extra cost. I think Cobalt-Samarium magnets make better sense and plain old Al-Ni-Co will suffice in just about every use. And ferrite is still king for cost effective designs.