Speaker magnet glue

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have an 18inch subwoofer that the magnet has slipped and I need to glue it back into position. I have re-aligned it into position and I have a new re-cone kit for it ready to be installed.

What’s the glue that factories use?

I have a tube of loctite 324 and it’s used to glue magnets into electric motors and metal etc…. I was wondering if that would bond a speaker backplate to the magnet? The magnet has a lot of wait to it being an 18inch

Any epoxy will do, I'd assume. However, if you took the magnet assembly apart, the magnet might now be partially demagnetized, as they are usually charged after assembly, and the closed magnetic circuit keeps them at a level of magnetization that they would not hold without (at least this is what conventional wisdom here on diyaudio holds).
I’ll give this loctite 324 a try and let youse know how it goes and if it works maybe I’ll post some pictures for other people that are in the same boat to work off.

well??? how did it go?? I have a new set of MB Quart car speakers and one has a loose magnet assembly.

The ceramic magnet and steel pole piece are still attached as one assembly. it's the joint between the whole magnet assembly and the stamped steel basket. the factory applies some sort of adhesive to the assembly and then rivits or swedges the stamped steel basket to the magnet. on one speaker I have, the magnet assembly is slightly loose from the steel basket and it rattles.

I need to use some sort of permanent, slightly flexible adhesive around the magnet/basket area to permanently bond the two together. and I'm curious what to use.

Epoxy will be too brittle/stiff/hard I think. Contact cement might do it but I'm not sure if the bond would be strong enough. I need something that is sort of in between the two.
Last edited:
If you use the Loctite 324, try to get a hold of and use Loctite 7075 Primer.
I work at a place that glues magnets into linear motors and we have found that the primer, as well as surface preparation (cleaning with both polar and non-polar solvent, and some mechanical cleaning like abrade with scotchbrite) makes the 324 work way better.
Linear motors? Hmmm. A loudspeaker motor with moving magnets instead of moving coils, and many kW's of continuous power rating.
Too bad they are basically a single frequency design. Ohhh... sonic weapons:rolleyes:

I am very interested in this subject and I am one step away from just calling up Eminence and asking what they use! which honestly may not be a bad idea.

I have struggled with this for many years now and have had many instances where I needed to re-glue a speaker magnet and have never really found a satisfactory glue. Currently I have 2 projects that i need to find a solution for. one is a pair of 5" drivers where the ceramic magnet and the top pole piece are separated. the glue the factory used dried up and when the speakers where dropped in shipping, the ceramic magnet slipped from the mounting of the top pole piece pinching the voice coil. I will get back to these later...

But my first project is as described above. So to test various types of glue. I have a case of these little crappy 3" sony speakers that have now become test mules! I cut some pieces of scrap sheet metal to simulate the speaker basket and planned to glue them to the flat backs of the speaker magnets. this should be a pretty close way to simulate the actual job at hand. I stopped by my local hardware store and picked up 3 types of glue that all said they worked on metal and ceramic that looked like they might be promising.

My past experiences with epoxy have proven to be way too brittle. so I ruled those out to start with.

I first tried Gorilla glue. I have never used Gorilla glue but have heard many people say it's great stuff. However i was unaware that it is a foaming type glue! and it is cured by moisture. which makes this a very UN-ideal glue. But out of the 3 glues i tried. it had the strongest bond.

2nd I tried Elmers Probond. consistency of watered down elmers glue. almost no bond and in fact was still wet inside after 24 hours of cure time. don't waste your time with this stuff.

3rd was Loctites GO2 glue. this is interesting stuff. thick, dried clear has a dried consistency somewhere between hot glue and silicone. almost no bond, but I would use this stuff for gluing down crossover parts, connectors etc. I love this stuff for that application but not for gluing metal/ceramic speaker parts.

So it is clear that I am going to need some industrial type product! local hardware store bought stuff just isn't going to cut it.

I want to try that loctite 324 but whoa is it expensive!
Last edited:
Here are some pics


  • IMG_20130523_175550_593.jpg
    76.3 KB · Views: 698
  • IMG_20130523_175602_002.jpg
    65.5 KB · Views: 648
  • IMG_20130523_175615_660.jpg
    69.4 KB · Views: 664
one of the issue i have with my first project is that I cannot fully separate the magnet assembly from the basket. I just have a small gap to work with. I need to be able to inject a small amount of glue into the gap and then apply a bead all the way around the outside and hope it holds. so i need some strong, resilient stuff.
Where I work we glue magnets with an aerospace adhesive that is similar to JB Weld.
Really, the JB Weld is very strong.
Most important factor in gluing, anything, is surface preparation.
It has to be clean, use both polar and non-polar solvent.
Dawn dish soap and then Toluene or Hexane will work fine too.
Roughen it up with an abrasive like scotchbrite or use a Dremil with a sanding drum or roll..
Roughen it after cleaning or else the abrasive will push dirt and grease into the substrate and epoxy will never stick.
Then clean both parts again.
As to maintaining a glue gap, I just put a sprinkle of glass beads meant for a blasting cabinet into the epoxy while mixing it.
Typical glass beads are 0.004" and less which provides the perfect gap for epoxies when clamped.
If you want to spend more than JB Weld at hardware store, order some Hysol EA9394 2 part epoxy system.
It holds helicopter blades together and bonds magnets very well.


And PS, the thin glue layer, although having a low permeability, is thin so it makes little difference.
Last edited:
WE HAVE A WINNER!!!! at least partly!....

I talked to a guy that works in a large motor shop and hey said they use Loctite 620 for holding magnets in place. which is really best used for holding cylindrical objects inside a tube...but I tried it. and this is an aerobic type so it needs an air tight seal. the liquid inside dried but any on the outside stayed liquid even after 24 hours of cure time. as expected. But it pulled apart easily using just my hands.

I emailed Eminence and they said they use a structural acrylic adhesive. So I am looking for some of that. But in the meantime I tried some rear view mirror adhesive and WALLA! wow is this stuff STRONG!

I cannot even pry this apart!!! the one corner wasn't exactly flat when i glued it and that's the only reason i was able to get under it and pry and bend it like that! that is not coming off there!!! SO...it looks like this stuff would work well for gluing magnets. next I have a speaker that the magnet has fallen off of it so I will try the bond between ceramic and metal and see how it goes.

And while the rear view adhesive will work for that application. I dont think it will work well for the steel to steel application I need it for.

I am going to contact the manufacturer and talk to them about it a bit and see if I can buy this stuff in larger quantities.



  • IMG_20130602_161157_804.jpg
    203.9 KB · Views: 372
  • IMG_20130602_161231_856.jpg
    99.2 KB · Views: 376
  • IMG_20130602_161207_393.jpg
    222.8 KB · Views: 354
Back to my question: does the glue joint represent a break in the magnetic circuit? If so, the factory (and you) want the smallest possible spacing and have to choose a glue-type accordingly.

It is quite a different matter to glue a magnet to, say, the housing of a starter motor. Or gluing a magnet assembly to the speaker basket. Neither of those glue joints are in the circuit or could affect the qualities of the magnetic field.

Back to my question: does the glue joint represent a break in the magnetic circuit?

The glue joint does represent a break in the magnetic circuit and will add to the overall actual air gap, which I will guess is something like 5 to 8 mm.
(It adds like series resistors..5mm + 0.1mm = 5.1 mm gap)
So the 0.1 mm gap at the magnet changes the overall magnetic circuit very little.
Actually the permeability of the magnet material is way lower than steel so the thickness of the magnet also adds quite a bit to the overall air gap if it is in series with the magnetic path.
Loctite 326 is essentially the same chemical as the typical rear view mirror adhesive and is available from Mc Master Carr for ~ $22 for a big bottle.
It requires the use of an activator that cost a few dollars more that the actual adhesive.
I used this for magnets, both SmCo5 and NdFeB types but it has a temperature rating that was too low for our needs and kept failing.
I do not know how hot a speaker motor can run but the Loctite Speedbonder lets loose at around 120 C.
In bonding magnets in 5 figure ($$) machines we went to the high temperature aerospace adhesive and never looked back.

Hope this helps,
I'll have to take your word for how to quantify gaps although my intuition is more about power-laws, not series addition. Yes, very helpful analysis.

But isn't the gap (the one filled with the VC wire... sometimes even edge-wound in the fine old days of Stephens drivers) very tiny, like closer to 2 mm than to 8 mm?

Most glues would result in insignificant added space in the gap. But some, like Gorilla family, not as small.

Last edited:
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.