# voice coil inductance

#### bnilsson

Hi!

I have found a pair of burnt-out vintage Celestion HF1300 tweeters, and I would like to endeavor into the task of re-winding the voice coils to get them working again. The VC "body" seems quite robust, so I have some hope to succeed. However, the copper is so damaged that I have a hard time figuring out the number of turns, or even if it is a single layer or double layer coil, etc.

One hint is the specs, I think I know that this tweeter has a specified coil inductance of 0.2mH, nominal impedance 16 ohms, dc resistance 13 ohms. If I use a coil inductance calculator, and just guess a wire diameter of 0.1mm, coil diameter of 25mm, a double layer of about 4.5 mm length, and a wire length of 6 meters I get 13 ohms and around 208uH in air.

Is this approach reasonable?

I know that an air coil will NOT have the same inductance as when the magnet iron is present, but as the speaker magnet system is not a closed magnetic circuit (only half-way) the magnet might not add too much anyway?

I think I understand that it is not the inductance itself which is the primary goal, but it is rather the motor properties which is important.
But if the resistance and inductance are in the right ballpark for the selected coil geometry, will not the motor be just about right too in the given magnet system?

Comments and suggestions are very welcome.

And YES, I have scouted eBay. Parts for 16 ohms HF1300 seems unusual and are not very likely to be found.

BN

Last edited:

#### JMFahey

Forget inductance which is a "secondary" parameter, you use copper resistivity here which is the "primary" one.

Given VC diameter which gives single turn length and winding length which gives you number of turns you get total wire length, which in due time will give you DC resistance.

Calculate 3 or 4 times , an iterative process, until you find a wire diameter which gives you *around* 13 ohms resistance.
0.1 mm looks very close, but just for safety calculate next higher and lower. (maybe nyou already did)

That said, a jeweler´s loupe will help you confirm whether it´s single or double layer.

And you need access to a lathe or lathe shop to make a tight fitting core to support wire while winding.
Tight enough to hold it round and proper diameterbut of course you must ge able to slide VC off afterwards

Use Epoxy diluted with a little acetone to glue wire, and avoid getting any between former and core or you´ll never be able to separate them.

And (lots of) good luck

Last edited:

#### bnilsson

So you say that if both resistance, inductance and the amount of space it takes is close to the original, it will be about right?

I am using the software "Coil64" which can do a lot of coil configurations, single, double layer etc., so I am now confident that I will not get completely lost.

Thanks for the practical advice, I don't have a workshop but I understand what its needed. I will have to find a way to rig it up. The greatest difficulty will be to get a core of the exact diameter.

#### Cal Weldon

Paid Member

EDIT: I see what you mean about the 16Ω units.

Last edited:

#### DickHertz

Replacement diaphragms with voice coil are available. A lot less hassle than trying to rewind an old one.

#### bnilsson

EDIT: I see what you mean about the 16Ω units.

Yes. The tweeters came in a pair of broken B&W DM2 boxes from the early 70's, that I got for free by a friend. The broken tweeters turned out to be 16Ω when I examined them. As far as I remember ( I was young then) the DM2 sounded excellent, so I really want to get them in shape. If I put some other impedance HF1300 tweeters in there, the crossover won't work.
There were also a pair of Coles 4001K super tweeters (broken as well) but these I could find on the market.

#### bnilsson

Replacement diaphragms with voice coil are available. A lot less hassle than trying to rewind an old one.

You are absolutely right.
If you have found some 16Ω diaphragms somewhere, I would be grateful for a link.

#### bnilsson

Forget inductance which is a "secondary" parameter, you use copper resistivity here which is the "primary" one.

Given VC diameter which gives single turn length and winding length which gives you number of turns you get total wire length, which in due time will give you DC resistance.

Calculate 3 or 4 times , an iterative process, until you find a wire diameter which gives you *around* 13 ohms resistance.
0.1 mm looks very close, but just for safety calculate next higher and lower. (maybe nyou already did)

That said, a jeweler´s loupe will help you confirm whether it´s single or double layer.

And you need access to a lathe or lathe shop to make a tight fitting core to support wire while winding.
Tight enough to hold it round and proper diameterbut of course you must ge able to slide VC off afterwards

Use Epoxy diluted with a little acetone to glue wire, and avoid getting any between former and core or you´ll never be able to separate them.

And (lots of) good luck

To make the beginning and the end of the wire enter/exit at the top of the VC, two layers are necessary. For a single layer, the exit wire would need to pass glued to the outside of the coil, which looks a bit dangerous/unstable to me. What is the common practice for speakers in general? Is it an issue?

#### Sangram

Paid Member
Le is always specified at a specific measurement frequency as it changes quite a lot. This is usually 1KHz at 2.83V (or 1V in some cases), with the coil inside the magnetic gap.

Not sure there is a way to exactly match that, so best to go with DCR and wire guage. While winding it will be apparent how many layers you'll need (I'd guess 2 layers at most). Worst case you can take measurements of the completed speaker and tweak the crossover a bit.

For the core you can try using a knob (the ones used for volume controls etc). They are usually available in close diameters, and you can make up some more diameter by using masking tape over the knob. The shaft hole will help you rig it up to a spindle for easy turning.

Last edited:

#### bnilsson

Le is always specified at a specific measurement frequency as it changes quite a lot. This is usually 1KHz at 2.83V (or 1V in some cases), with the coil inside the magnetic gap.

Not sure there is a way to exactly match that, so best to go with DCR and wire guage. While winding it will be apparent how many layers you'll need (I'd guess 2 layers at most). Worst case you can take measurements of the completed speaker and tweak the crossover a bit.

For the core you can try using a knob (the ones used for volume controls etc). They are usually available in close diameters, and you can make up some more diameter by using masking tape over the knob. The shaft hole will help you rig it up to a spindle for easy turning.

Thanks for the very practical advice about the knob, I would never have thought of that myself.

#### JMFahey

To make the beginning and the end of the wire enter/exit at the top of the VC, two layers are necessary. For a single layer, the exit wire would need to pass glued to the outside of the coil, which looks a bit dangerous/unstable to me. What is the common practice for speakers in general? Is it an issue?
Standard 2 layer is easy, both exit same side (towards diaphragm or cone); in single layer tweeters they start from diaphragm side, and return inside the "slot" created because former material edges go around but not meet each other.
Since former material is typically 0.1 mm thick, no big deal.

On speakers which use much thicker wire, it returns on the outside but a slot is cut on top plate to give it space (Bose 901 1 ohm woofers).

As of:
The greatest difficulty will be to get a core of the exact diameter.
Common practice is to use regular cold rolled steel cylindrical bars in standard diameters as polepiece material, to save on \$\$\$ and machining, so your tweeter polepiece must basically be a machined piece of 1" bar, and to allow movement VC former must be slightly larger diameter, typically 0.2 or 0.25mm (tops) more.
Which means if you get a piece of 1" iron/brass/aluminum/nylon roundbar or even pipe for free or peanuts and you wrap it in a single layer of 0.1 or 0.125 mm Mylar sheet (old X Ray sheet cough cough) you will be about perfect.

#### bnilsson

Standard 2 layer is easy, both exit same side (towards diaphragm or cone); in single layer tweeters they start from diaphragm side, and return inside the "slot" created because former material edges go around but not meet each other.
Since former material is typically 0.1 mm thick, no big deal.

On speakers which use much thicker wire, it returns on the outside but a slot is cut on top plate to give it space (Bose 901 1 ohm woofers).

As of:
Common practice is to use regular cold rolled steel cylindrical bars in standard diameters as polepiece material, to save on \$\$\$ and machining, so your tweeter polepiece must basically be a machined piece of 1" bar, and to allow movement VC former must be slightly larger diameter, typically 0.2 or 0.25mm (tops) more.
Which means if you get a piece of 1" iron/brass/aluminum/nylon roundbar or even pipe for free or peanuts and you wrap it in a single layer of 0.1 or 0.125 mm Mylar sheet (old X Ray sheet cough cough) you will be about perfect.

#### Cal Weldon

Paid Member
I can't locate any 16Ω diaphragms but in another thread here someone suggested contacting Coles Electroacoustics. Worth a phone call at least.

#### Attachments

• Screen Shot 2019-10-02 at 11.29.29 AM.png
74.8 KB · Views: 126

#### bnilsson

I can't locate any 16Ω diaphragms but in another thread here someone suggested contacting Coles Electroacoustics. Worth a phone call at least.

I actually did that, but they did not have any such parts any more.
They once had their own replacement for the HF1300 called "3000", but this is no longer produced. I bought the super tweeters 4001K from them, though.

#### Cal Weldon

Paid Member
ok, well good luck with the project.

#### GM

Cal,

FWIW, your web page is 'unreachable': Error 523 Ray ID: 51f9626e6a6bc87f • 2019-10-02 20:21:48 UTC
Origin is unreachable

GM

#### JMFahey

You´re welcome

One extra detail: earlier I suggested diluted Epoxy as adhesive, and that´s the "high power" standard because it´s the highest temperature/power standing one, but it´s unforgiving, once set it´s set, for better or for worse, and does not admit retries.

But that´s an old style tweeter, and former material seems to be some kind of brown Kraft paper (Celestion used it for decades) son instead I suggest you glue wire with Nitrocellulose type adhesive.

Model builders call it DUCO and is available in small quantities from airplane builder shops, maybe some craft shops too.

Or visit your friendly car painter and ask for a little clear nitrocellulose paint base or its modern replacement "Universal" type which is fit for Nitro (old style) or Acrylic (new style) car paint finishes.

Personally I use a lot of it, so I buy a can of the transparent base and leave it uncovered some time so solvent evaporates and it thickens from "paint" to "glue" state.

Advantages are that it´s very light and worst case , errors can be wiped out with some acetone or thinner and you do it again.

Last edited:

#### Cal Weldon

Paid Member
Greg,
Thanks for the heads up. If you mean the one in my sig line, it's works here just fine.

#### Attachments

• Screen Shot 2019-10-04 at 9.51.02 AM.jpg
259.4 KB · Views: 90