Has anyone DIY a Field coil?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Has anyone machined and wound their own field coils for those drivers with replaceable magnets? (Lowther, AER etc)

I fail to see how they justify such high prices for what is essentially an electro-magnet with a soft iron cover.

Can anyone please direct me to a thread or webpage that shows in detail how these are put together?
 
Yeah, your right, sucks don't it. You must bear in mind the drivers you are referring to aren't thrown together by robots.

The was an article in EA May 1990, page 114, detailing a little on the Rice and Kellogg 6" EM driver.

It's not that hard to remove the backplate / pole piece assy and magnet from modern drivers. Dropping one can break the glue.

Older drivers had screws to keep the works together. They would be easier to convert, though I don't know if it would be worth the effort.

Many modern drivers have the front polepiece riveted to the basket.

This is where it all started:
http://mixonline.com/TECnology-Hall-of-Fame/1925-dynamic-loudspeaker/

If I can find a source of soft iron, I'll let you know. Until I do, I'm stuck with steel. It might be possible to melt down old gal waterpipe. That's not got much, if any carbon in it.

Voice coils can be sourced from speakerbits in Melbourne.

Geoff.

PS. here's another.

http://history.sandiego.edu/gen/recording/rice-kellogg.html
 
Hi,
cast iron has lots of carbon in it and in big crystals that make cast iron brittle.
Soft iron has most of all the impurities removed. It's the removal of all the impurities than makes it soft iron and makes it good for passing flux.
Adding zinc to the melt, from galvanising, will probably make an alloy that is worse than soft iron, but I'm guessing on that one.

There are a couple of threads in this Forum on manufacturing a coil magnet for a big speaker and another on test/ listening results for a commercial coiled electromagnet driver.
 
"cast iron has lots of carbon in it"

That would be cast steel? It might be possible to separate the zinc.

The gal waterpipe I am referring to is malleable, not the cast pipes used for drainage. I have thought of using iron from old drivers. That means getting rid of cadmium. I'd prefer to mess with zinc.

Cast steel is better than cast iron, though not as good as pure iron or armco, on a b-h curve, according to my reference.

Edit: Andrew, you could be right about the zinc, there is no mention of it in the various mixes. And I don't recall seeing it when I worked for a magnet producer, and they were mixing high grades of alnico.
 
Hi,
back to a suitable iron for passing flux around the magnetic circuit.

Very low content of any alloys would be essential. But particularly the alloys that interfere with the magnetic properties.
Some types of steel may be second best after soft iron.

I wonder how close wrought iron gets to fulfilling the need? Problem may be getting blanks big enough to turns poles and face plates from it.
 
The thin plates from transformer iron would be easy to melt down.
What about pig iron briquettes? Old radar magnetrons have easily accessible thick iron plates. You'd need quite a few. Wrought iron has extra lead added to make it more malleable so I don't think it's suitable.

I mentioned those drivers because they usually have their magnets retained with screws.

Given a billet of the right material I think I could turn a pair up on a lathe in a day easily. It might be possible to use a whole roll of magnet wire, spool and all - no winding. I see no reason why a lump of metal and a roll of wire should cost a couple of thousand euro.
 
OzMikeH said:
....... It might be possible to use a whole roll of magnet wire, spool and all - no winding.
I don't think so.
You want the maximum volume of copper in there to minimise the heat build up for the necessary AT (ampere turns) to generate our field strength.
If we can't get above 1T then we are wasting our time.
I believe some of the cheaper grades of steel allow around 1.9T to 2T to be generated.
2.1T to 2.2T should be possible with soft iron. I wonder if we can get higher?
Do we need to get higher?
Or even as high as 1.9T?

What are the T/S benefits from high flux?
What are the sound quality characteristics accruing from high flux.
Gilbert Briggs did some work on this?
Does anyone have links to ANY papers on this flux strength advantage?
 
Hi,
it looks like that site has not been updated for at least a couple of years.

I had a look at copper and found that he is predicting about 2.4A for 2T. He has 1000turns and DCR of 3r1.
That is about 17W to be dissipated. This coil is going to get pretty warm. A big heatsink bolted to the back end will help keep the iron cooler, but the coils are pretty big and internal temperatures are going to be a lot higher than surface temperatures. =max volume of copper.

It is also obvious that the critical part of the iron circuit is right next to the gap. Elsewhere he appears to be using at least twice the area of iron in the magnetic circuit to keep iron flux below 1T.
It's a bit like a resistive circuit passing current. keep the resistance in the cables low and allow the main resistance where the power is to be used. here the flux is replaced by current, the area of iron is area of copper cables and flux density is current density. The gap is equivalent to the resistor dissipating the power. So concentrate the flux into the gap and make the rest of the circuit low flux density (= low resistance).

If we take that a little further, it may be that the central pole piece and the top plate use the best flux carrying iron and that the backplate and outer annular ring can be any lower quality steel that can easily pass the flux density in those areas. You can see that these two components can be made any thickness without impinging on the volume where the copper coil is. I think that copper volume is the critical design parameter that determines the achievable flux density in the gap.

Thanks to el`Ol for the link.
 
I could get shot down for this. I think the more flux the better. Recent experiments I have done indicate the magnetic circuit of many speakers is not saturated, ie it will carry more flux.

Adding an additional magnet on some full rangers provided an extra 3db in sensitivity across its range. I can show curves to back that up.
Transient response is also improved. according to my ears.

Transformer laminations and cores in old 3000 type relays were silicon iron. From memory, the core in those relays is only about 5/8" dia. OK for high freq drivers, not much use for bass use.

Other sources could be starter solenoids from cars, and the rotor from an AC synchronous motor.

To get maximum flux, we need to fill the gap with the right no of turns of the right gauge. More turns of smaller gauge will increase resistance needing a higher voltage, and increasing losses.

Using less than optimum material for the poles/circuit could be offset using thicker plates. The centre pole piece will be the limiting factor.

Melting down existing materials could be a challenge. Back in the factory, the alnico mix was brewed in a high frequency furnace. No flame to add carbon. Also the varnish on the laminations would need to be disposed of.

Modifying existing drivers could be interesting. I don't think I could carve up a Goodmans or Wharfedale, but maybe a Richard Allen.

I might hide for a while. LOL

Edit: "It is also obvious that the critical part of the iron circuit is right next to the gap."

Yep, that's where it's needed.
 
The power dissipation is probably not a big issue. In the old console radios I dragged home as a kid, the large openback cabinet also had a chassis with 5 or 6 octal tubes. The field coil doubled up as the B+ filter choke. The magnets ran warm, but not hot.

Did I have fun at show and tell in 6th grade!
 
Here is what prompted the original question.

http://www.fullrange-speakers.com/eng/treiber-e.htm

According to that page: Variable field = variable Qts

I don't understand the comment "More turns of smaller gauge will increase resistance needing a higher voltage, and increasing losses"

Does it matter?
Assuming the magnetic circuit is sufficient the amount of power (Watts) applied should be approximately proportional to feild strength whether through many turns high voltage or few turns high current.

I'd love to see a cross sectional drawing of that feild coil, is it almost completely full of copper wound on the pole piece?
Is there 2 windings inside, one on the pole piece and the other wound the opposite way on the inside of the outside cylinder?

This thing is huge, What is the limiting factor for the size, heat dissipation or the amount of iron needed for the magnetic circuit around the outside?

Sorry for the novice questions, I just can't see why an electro magnet has to be 5 times the size and 10 times the price of a permanent magnet.
 
OzMikeH said:
I don't understand the comment "More turns of smaller gauge will increase resistance needing a higher voltage, and increasing losses"
more turns requires less current but has more resistance so the heat stays roughly the same. If you do the full calculation I think you'll find that the Watts stays the same if the volume of copper stays the same.
OzMikeH said:

Does it matter?
Assuming the magnetic circuit is sufficient the amount of power (Watts) applied should be approximately proportional to field strength whether through many turns high voltage or few turns high current.
But push in more current and you get more AT and thus more flux in return for more heat.

OzMikeH said:

I'd love to see a cross sectional drawing of that feild coil, is it almost completely full of copper wound on the pole piece?
Is there 2 windings inside, one on the pole piece and the other wound the opposite way on the inside of the outside cylinder?
It will probably be 10layers with 100turns per layer or something similar that gives 1000turns in total. Basically the space is crammed full of neatly wound copper to minimise the power dissipation and maximise the AT.
OzMikeH said:

This thing is huge, What is the limiting factor for the size, heat dissipation or the amount of iron needed for the magnetic circuit around the outside?
The limiting factor are the losses around the magnetic circuit. This depends on how close each material in the circuit gets to it's inherent flux density limit. There comes a point for each material that the flux will not increase any further. Even reaching that point where the flux density vs AT curve has flattened out the copper will have probably reached the insulation failure temperature. That's the second limiting factor.
OzMikeH said:
I just can't see why an electro magnet has to be 5 times the size and 10 times the price of a permanent magnet.
the cost of labour in accurately machining all those parts. The labour in assembling it. The capital investment in turning and winding machines. Buying in the specialised/expensive materials. The expertise in designing and developing the product for tiny sales volume. Compare that to moulding a ferrite block and zapping it with a high flux magnetic field. No contest.

But then ceramic cannot approach 2T
 
more turns requires less current but has more resistance so the heat stays roughly the same. If you do the full calculation I think you'll find that the Watts stays the same if the volume of copper stays the same

Are you sure?

The dissipation in the coil P = I^2 x R. For the same AT product, if you double the amount of turns this more or less doubles the resistance and cuts the current in half. Thus the dissipation will be reduced by a factor of four.

I also came across Finn Hammer's site and dug up the original drawing from Voigt's patent:
An externally hosted image should be here but it was not working when we last tested it.


This construction is supposed to minimize flux leakage and maximize the saturation at the airgap

The elements of this construction can be found in the
Lowther PM4A

If you take a look at the PM4A you'll see that they use permendure (2.4T saturation) only for the top plate and the pole piece.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.