Project Ryu - DIY Field Coil Loudspeaker
***First i want to say I'm not sure if i should post this under Fullrange or Multiway. Since at this point, there is a high posibility of the need for a supertweeter i chose Multiway forums. I won't mind tho if moderators think it's more suited for Fullrange forums.
About 6 months ago, i started thinking about a DIY fieldcoil speaker. I know opinions vary quite a lot on this matter and so i will do my best to offer as much data as possible for things to be judged objectively. Since that time I have been talking with some fellow countrymen and audio enthusiasts about this project and these conversations helped me in making some of the design decisions. The work is still in progress and i hope your opinions and comments can help tune it to higher performance.
To begin, here are the basic aspects about this project:
- field coil underhung motor structure
- 12" diameter
- Wide range (up to 10kHz), hopefully fullrange (up to 20kHz)
- high efficiency
- to be used in BLH
My personal taste in sound is that i like it when it comes out of large radiating surfaces. Large SD, large horn mouths etc. I think it's a necesity when one tries to reach real, live music levels and at the same time keep a good response to instruments' attack, sustain and decay. This is the reason i chose 12 inch diameter. I would've liked 15" better but it would've been too difficult to make it wide range. Perphaps a future project.
Field Coil Motor Structure:
I have spent a lot of time looking for other field coil projects, for any information on FC design or about available products and i have to say there isn't
much out there. From the pictures of opened FC units i could see a basic shape of the pole pieces. I have to say this is amongst the first design issues i encountered, there isn't much freedom to design these pole pieces, not like the case with permanent magnets anyway.
The next thing one can notice is that the new FCs are low impedance type and this is easily understandable. I have decided that my field coil will also work with 12-15V supply.
Underhung design means thick top plate. Of course this will cause a drop in flux density in the gap. The compromise here i found to be between flux density value [B] and voice coil excursion. Since this speaker will mostly be used horn loaded i sacrificied Xmax to make a thinner top plate. I use standard 1010 steel with a maximum of 0.13% carbon, the best i could get my hands on.
I ran simmulations in FEMM for different values of top plate thickness, FC's number of turns, dimension, wire diameter etc. Here are the results i stopped at:
- magnetic gap height: 12mm
- magnetic gap width: 1-1.5mm
- 5.3 ohm FC 1400 turns of 1.25mm copper wire
Below you can find a rather crude sketch of motor structure:
Here are some results from FEMM analysis:
I ran simmulations with 1010 steel as well as 416 steel (has slightly more carbon typical 0.15%) to see the differences.
1010 Steel [B] Curve over 20mm and 12mm:
416 Steel [B] Curve over 20mm and 12mm: (actually the curves are smoother had to lower resolution a bit because of my computer)
AS it is, at 2.2 Amps we have 25W for the FC. I don't expect to have too much trouble with heat at this power. Test will tell tho, depending on how well the metal does the job of sinking the heat. The [B] value predicted by FEMM lies around 1.6 Tesla. In reality i believe this value will be smaller 1.4-1.5 T. In tests i will drive the FC with 3 amps and see how much it heats up.
In this design the gap width is 1.5mm. Somewhat large, I'm thinking to add a 0.5mm thick copper ring right in the gap (yellow part from picture below). Now
it's true i have never seen a picture with a FC unit having a faraday ring and I'm not sure if it's necesary. I guess the FC itself has a role in that too.
I'm also making a 1mm gap variant (without copper ring) in hopes to get more flux density. The FEMM simmulation shows:
Here are some parts of the magnetic structure after machining:
Voice coil is 2" diameter. Haven't decided on the former material yet, and there is still some designing work to be done. I will use 0.13mm copper-clad aluminum wire (hoping to receive it Monday).
Data for the ECCA wire is as follows:
Copper Vol: 15-17%
Density [g/ccm]: 3.7
Resistivity [ohm*sqmm/m]: 0.0255
I will come back to the voice coil issue soon.
Cone will be made of japanese paper called washi (fullrange crowd might now why hehe). I got a few rolls of washi of different types and will test a few variants.
Next week work will start on cones, most probably will be made from 12 petals. I will test both straigth and curved cone. An exponential curved profile like in the picture below has a calculated flare rate of 0.387/cm which should correspond to a freq close to 1050Hz. Have to look into whizer cone as well.
Thats it for now, let me know what you think so far. I will post more info as i go with this project. I hope it will awake more interest into fieldcoils and
shed some light on the matter.
Wow, Hentai, this is great.. I and a number of other members have been waiting forever for someone with the requisite skill set to bring a FC speaker project to the forum.
I think Pano will be equally interested in your project.
Probably a bit beyond my current machining skill set, but I will be following avidly. :D
Hurray! Some real info on field coil drivers. Thanks Hentai. :up:
1.4 tesla is 14000 gauss so this seems like a very reasonable flux density in the gap, not too many PM speakers do better than this. I'm assuming (naively perhaps) that this design will have a pretty high BL?
Hello, glad you find the topic interesting.
Yes, a high BL is mandatory for high efficiency. I am also looking to tilt the response upwards in frequency, as in higher sensitivity while going up in band. This will be very helpful when using horn loading.
Of course electromagnetic overdamping will be an option on how hot you run the FC.
But as you identified, i aim to get 1.4T and not overheat the FC.
I will post later about Voice coil and its part in the BL figure. A high length of wire makes a high BL but also increases mass and i want to keep mass as low as possible for efficiency and wide bandwidth. However i made some calculations on how much i gain either by increasing BL and mass alltogether or reducing BL and mass as well. I will post them later after some checking. BL on its own doesn't say the whole story there is also the issue of DC resistance of wire that i will touch on later as well.
Congrat, Hentai. Your FC already looks great.
I read your posts with great interest, respect and admiration.
I will need to dig out my old text books to refresh my understanding of the math and phyisics.
Thank you for starting this thread.
Wondering about the baskets, are you going with stamped steel or will you use cast aluminum in the actual driver build? I would suspect that the cast aluminum are likely to be more rigid and less likely to exhibit any sort of resonances..
Hello and thanks for replies.
The baskets will definitely be aluminum. The motor is very heavy, already it weights
8.8kg and that is without all the copper and metal.
Im not sure if steel baskets can hold the weight.
|All times are GMT. The time now is 10:00 AM.|
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio