Can I use a SMT inductor as a small electromagnet?

[zigunov]

Hello, fellow nerds! I am working on some rather "unusual" project and I was wondering if I can use an SMT inductor as an electromagnet.  (https://www.digikey.com/en/products/detail/w%C3%BCrth-elektronik/74404064102/5353075 is an example, but it could be any part of the kind)

I need very small electromagnetic forces (like order of 1 to 2 grams), and the Neodymium magnet I'm working with is going to be just one or two millimeters from the inductor. I don't have one of these inductors sitting around, otherwise I would just test it. But I was wondering if anyone has tried this before.

I'm not very sure about the construction of these inductors. Do they close the magnetic loop (i.e., the ferromagnetic material loops around the coil)? Or is it common to have inductors with open magnetic circuits?

Sorry if I wasn't very clear. I'm still figuring this out!!

[amyk]

Yes, the magnetic circuit is closed in basically all the inductors I've ever seen, from tiny SMT ones to huge dozen-henry "chokes" for power supply smoothing.

[pigrew]

I've never actually tested it, but in principle, this is definitely a plausible design. I forget how to do the force calculations...

The part you selected is listed as semi-shielded (encased in ferrite epoxy). For your plan to work, you will want an unshielded component. In addition, you will want to have no metal fill on your PCB below and around the inductor. A "grid dot" pattern of many unconnected small islands of copper might be helpful.

I'd first try something like Murata 22R153MC to check if it works in practice.  Be sure to stay within the current limit of the inductor. There will be some sort of tradeoff between inductance, current, and DC-resistance.

[zigunov]

Thanks guys I really appreciate the help! So as far as I understood the magnet would stick to the ferrite core, so that wouldn't work for a motion application. It might work with steel, though, which would be fine to me.
I think I'll try to harvest one from an old power supply and see if it works.

The reason I'm doing this is because I've built some PCB electromagnets but they're always too weak and spread out. I need something more compact even if it's a little taller. I couldn't find proper smt electromagnets, otherwise I would go with that. Perhaps I just don't know what to search for?!

[pigrew]

Perhaps I just don't know what to search for?!

I think "unshielded" wire-wound is what you're looking for. Unfortunately, most of the inductors you will find on scavenged boards will be shielded or semi-shielded.

I just did an experiment with a 1 mH through-hole ferrite-core unshielded inductor. It was easily able to hold up a 5g slug of steel when driven with 500 mA (about 1/2 watt dissipation). I think the surface-mount parts will be similar.

[zigunov]

Hi, pigrew, I'm sorry I didn't see your replies. I'm really grateful for your input. Because I was completely oblivious to your post , I ended up scavenging two wirewound inductors from an old graphics card. They didn't produce any appreciable amount of force even after I drove them with 6A, to my dismay!!

But now with your input this makes more sense. I want the unshielded ones, for sure! Thanks, again, for reporting your experimental findings!! This gives me extra confidence to get some varied parts to play around with and find what works best to me.

By the way, this is part of a project I'm working on, where I want to drive an array of about 1000 pneumatic solenoids in an array. I want it to be scalable and somewhat cheap, this is why I turned to something that is already mass manufactured.

Thanks for the help, pigrew!!!

I'll post updates once I get more stuff =)

[fourfathom]

Yes, the magnetic circuit is closed in basically all the inductors I've ever seen, from tiny SMT ones to huge dozen-henry "chokes" for power supply smoothing.

I use SMT inductors in the 0.1 - 10.0 uH range and these have all had ceramic forms with a single-layer solenoid coil -- definitely not closed magnetic loop.  You're not going to be developing much of a magnetic field with those small uH-range coils though, although if you place them next to each other you will get enough coupling to affect the tuned-circuit response and inter-stage coupling.

[Berni]

You need a very large number of turns on your coil to make it a good electromagnet.  The reason these ferrite core indutors have a small number of turns is because they have to avoid saturating the core at large currents. The sort of ferrite that is optimized for low AC losses can't handle as large of a DC field.

For your application you are probably better off getting something like a steel M3 standoff(There are even SMD solderable ones) and winding a lot of hair thin wire around it. Then use iron as a target to attract.

[zigunov]

You need a very large number of turns on your coil to make it a good electromagnet.  The reason these ferrite core indutors have a small number of turns is because they have to avoid saturating the core at large currents. The sort of ferrite that is optimized for low AC losses can't handle as large of a DC field.

For your application you are probably better off getting something like a steel M3 standoff(There are even SMD solderable ones) and winding a lot of hair thin wire around it. Then use iron as a target to attract.

I see. Yeah if the number of turns is small it might not be nearly as effective, which is a bummer. I will still try them, because the alternative (your suggestion) is a grim path... I plan to do about 2000 of these inductors distributed across three boards. I don't think I can wind that many inductors, consistently, without going crazy in the process. This is really why I'm turning to those mass-produced ones...

[mikeselectricstuff]

I've used inductors like that for very short range inductive data coupling, but I doubt they would be useful for anything mechanical - the core would attract the magnet and saturate, so any change in field by energising the coil would be minimal.
To attract a magnet you need a coreless coil. Maybe look at something like this

[zigunov]

Hi, Mike, I know Carl's work. The issue is that I have about 6mm diameter to print the solenoid. I went through some iterations in the past with PCB coils, but I simply can't fit enough turns in the coil to generate the forces I need even with 4 layers. I know what I'm trying to do is quite hard in on itself.

I wish that there was some miniature electromagnet as a product that I could buy but it doesn't seem to exist. I'll give the ferrite ones a try, but if they don't work I will have to make my own miniature electromagnet factory 

[S. Petrukhin]

I think you can break a small-sized electromagnetic relay and get a coil out of it.

[atot]

Hello, did you have a solution to your problem? I am facing a similar one right now.