solenoid conundrum

Thread Starter


Joined Aug 14, 2005
i might as well give you the story. we have a large milling machine at work, that for some reason kept blowing solenoids in the gear change box. apparently the company that made the machine was asking $1000 each for n.o.s. solenoids, (the machine was made around the 1960's, but when its not blowing solenoids it still runs like a champ!) so i offered to rewind it. i pulled it apart and as soon as i unwound the melted coil the plastic piece it was wound around more or less crumbled to dust. i machined a new one out of aluminium, chucked it in the lathe and wound it full with 20awg magnet wire. once i got it back together and installed in the gear change box i found out it wasnt strong enough to engage a slide that it is attached to. this brings me to my questions.

1. would the use of aluminium for the center piece rather than plastic have any adverse effects?

2. if not, what would be the best way to go about making it stronger:
a. adding more coils of the same size wire
b. removing coils of wire
c. rewinding with larger wire

btw, i found out that the solenoids were getting left engaged inside the gear box due to a faulty switch. they are not continuous duty cycle, so they were over heating pretty quickly. if the machine works properly, they never receive power for more than a fraction of a second, followed by a few minutes without power.

any suggestions appreciated!!


Joined Apr 20, 2004
Soft iron is one of the best materials for a solenoid core, although mild steel (not stainless) will do pretty well. The effect that will enhance the pulling action of the solenoid coils has to do with permeability, which lets a material concentrate the magnetic lines of force.

The old cores might have been ferrite. They would have been formed under pressure (powder metallurgy) and the heat from the coils might have caused them to come apart.

You might try duplicating the solenoids on a cast iron or mild steel core. If the gears don't pull in, adding a layer or so of winding might help the action. Using larger wire could allow more current and so a stronger magnetic field, but that can cause problems with the voltage source and switches.


Joined Apr 27, 2007
It seems that there is some confusion. The piece that crumbled apart was made of plastic or made of ferrite? They might be identical, but ferrite is a paramagnetic substance, and so, is attracted by magnets. It it was indeed plastic, it can be some sort of insulator and it lost its plasticity due to heat (some plastics tend to do that). If it was the actual core of the magnet, yes, you can use soft iron.

Can you post a photo of a complete relay?


Joined May 19, 2004
as mentioned your core material is paramount. realize that ac solenoid coils require different windings (pole shading) then dc coils to avoid chatter.


Joined Jul 17, 2007
I suggest that aluminum is not a good choice for a solenoid bobbin (the spool that the wire is wound on), mostly because of the friction and galling that will occur when the coil is energized. Solenoids are not lubricated, and aluminum is "sticky".

Here's an idea for you - go to this site:
and look at their "solenoid bobbin" section. You might find something that's very close to what you had. If not, you might find a solenoid bobbin that has the same ID/OD for the center section, and longer than the original, so you might could trim the excess or cut it to your custom length. I would give them a call. These folks have been making bobbins for a long time. I'm sure they have their bobbin material optimized, far better than what we could guess at.

Do you have any of the original wire that was wound on the solenoid, even a small scrap to measure it's diameter? Or do you know what the original wire's diameter was?

Increasing the diameter of the wire will reduce the number of turns that will fit on the solenoid. This will reduce the magnetic field created by the coil. Increasing the number of turns will increase the magnetic field (stronger pull). You really wouldn't want to change the AWG size of the wire from what it was originally, as it was engineered for optimal performance.

I would fill the solenoid's bobbin as full as I could get it, winding on the turns as evenly as possible at very light tension (to keep from overstressing/crushing the bobbin), without exceeding the diameter of the spool ends. This should be fairly easy to do on a lathe between centers at low speed. You might have to fab something up to engage a dog plate.
Rig up your magnet wire spool so that it feeds wire with little resistance. For my coil winder, I used a couple of precision bearings I picked up at a place called Skycraft Surplus in Orlando. My spools run horizontally; that's just what works best for me. I don't let the wire between the spool and the bobbin get in a straight line, because if there is any un-evenness in the source spool, the tension in the bobbin being wound will be uneven.


Joined Aug 8, 2005
If you didn't place any tape on the aluminum form, there could be shorts.

Using an aluminum coil form instead of plastic will definitely change the characteristics. If you can't make one from plastic, you could try wood, or a piece of plastic pipe that is close to the same diameter as the original.

Also, did you wind the wire in nice carefully made sequences, or is it wound haphazardly in a bunch?

Thread Starter


Joined Aug 14, 2005
thank you for all of your suggestions, i really appreciate it, as im sure my co-workers also do. (none of us enjoy changing the gears by hand with a prybar)

well, i guess its my turn to answer your questions!

im 99.9% sure the original bobbin (sorry, i couldnt think of the word for it in my first post) was made of plastic. i made the new bobbin to the same dimensions, based on a newer, but same model solenoid.

the bobbin has no parts actually sliding against it, there are a pair of brass 'gibb' strips in the middle, that center piece slides on.

my first attempt at making one was actually with wood, but it proved to be impractical with the tooling we have in the shop. (and the lack of anything other than 2 and 4 by 4's)

i covered the aluminium with loads of electrical tape, specifically to prevent shorts. after winding i checked the resistance against the number i got when it was all on the spool, and the continuity between the bobbin and the wire, all was ok.

i did wind the coil somewhat haphazardly, it was kind of difficult to guide the wire due to the shape of the bobbin (square), and the way it was chucked in the lathe. if winding it neatly will help out, i have no problem trying again.


you mentioned ac solenoids are wound differently. i noticed the newer solenoids look like they have two different coils of wire on the same bobbin, with a separator in the middle. they also look like they are two different sizes of wire.

could the effect of the aluminium bobbin be that detrimental or should i shoot more for rewinding it similar to the newer ones? or both?! lol

thanks again!


Joined Dec 27, 2007
I think I may speak for many when I say that, "You must have ferrous material in order to provide an efficient flux path". Wood, aluminum, etc. are not ferrous material, so they will not allow flux flow, thus, no strong electromagnetism.

I do work with magnetic bearings. I can do full analytical modeling and have done FEM with them, I am quite sure you must use some sort of ferrous material..

Also, beware of shorting the magnet wire. It happens incredibly easy! Typically, the core is sprayed in an insulating varnish, then wrapped in fiberglass tape. Even after these precautions, they usually dip or VPI the device.



Joined May 19, 2004
are you exciting the coil with ac or dc? if ac, then i beleive you must incorporate a shaded pole - which retains some amount of magnetism during the zero crossover - or you'll experience chatter during phase reversals. or you can rectify and use a single coil of windings.

the ferrous metal core will provide additional flux path for a stronger mechanical transfer. it may or may not be required for your application.


Joined Mar 11, 2008
The magnetic flux a coil can generate is determined by it's amp turns (A*T). If you increase current or turns you get more flux, squared if you do both. I think you will need to wind it with the same size wire if you don't know how many turn there was originally.
On question 2, need to work it between a and c also AC or DC voltage will ditermine the amount of turn.


Joined May 16, 2005
I concur that the same size wire as the original, or smaller wire than the original, should be used. If the bobbin is "scramble wound," by all means go with smaller wire. If you've got change gears on that lathe (you're in a machine shop, yes?) you might get away with using the same size wire and winding a precision coil. You can do this by setting up some kind of tensioner on your tool post.

tom foxe

Joined Apr 4, 2008
If the old bobbin was plastic the new one shouldn't be metallic, even if it's non-magnetic it will have electric current induced in it, reducing the power of the solenoid. The whole point of the solenoid is to attract some iron/steel part so it is either drawn into the centre of the bobbin, or so that it stays there until the current is switched off. The overheating indicated that the current is big and the bobbin not a good conductor of heat or electricity If you have solved the overheating problem then a bobbin can be made from plastic, wood or other non-metallic solids. I strongly agree with the person who suggested you contact the experts, bobbin-makers who may even be able to supply a bobbin identical to the original.