Member
Joined 2003
Member
Joined 2003
WOW! How does it sounds??
And one question for all of you. I have been listening recently a pair of Apogee Centaur Minor speakers which have ribbon midrange/tweeter and classic bass speaker. Ribbon is made of mylar with aluminium stripes on it. The sound is amazingly opened even at low levels and you can hear depth of the room or concert hall exactly as if you are there. That is something that i never heard so good on convencional speakers (and i have been listening hi end stuff). So do you have that feeling on those pure aluminium ribbons and is that characteristic of all ribbon speakers?
I have made small ribbon speaker with magnets from hard-drive just for fun, but i have been pushing 200w of power just to hear very low earphones volume. And those magnets are pretty strong with fingers. Anyway, the sound was so good, that my Sennheiser headphones were jealous. If those magnets don't do any harm to my brain, i should build ribbon headphones for sure.
And one question for all of you. I have been listening recently a pair of Apogee Centaur Minor speakers which have ribbon midrange/tweeter and classic bass speaker. Ribbon is made of mylar with aluminium stripes on it. The sound is amazingly opened even at low levels and you can hear depth of the room or concert hall exactly as if you are there. That is something that i never heard so good on convencional speakers (and i have been listening hi end stuff). So do you have that feeling on those pure aluminium ribbons and is that characteristic of all ribbon speakers?
I have made small ribbon speaker with magnets from hard-drive just for fun, but i have been pushing 200w of power just to hear very low earphones volume. And those magnets are pretty strong with fingers. Anyway, the sound was so good, that my Sennheiser headphones were jealous. If those magnets don't do any harm to my brain, i should build ribbon headphones for sure.
Member
Joined 2003
HI guys!
You can can look att the swedish diy forum artikel!
In swedish, but you can look at some butiful pictures.
http://www.hififorum.nu/artiklar/Dahlbergsband/del1/Dahlbergsband.htm
My ribbon is 112cm.
It play down to 400hz.
If i play fullrange with just the ribbon and no crossover i can hear bass but wery low level.
Crossover frekvens is 400hz.
You can can look att the swedish diy forum artikel!
In swedish, but you can look at some butiful pictures.
http://www.hififorum.nu/artiklar/Dahlbergsband/del1/Dahlbergsband.htm
My ribbon is 112cm.
It play down to 400hz.
If i play fullrange with just the ribbon and no crossover i can hear bass but wery low level.
Crossover frekvens is 400hz.
Mx said:
Thanx... a very well documented project.
(Some of the links to big images on Page 2 are broken -- most seem to go off site (ie http://www2.hififorum.nu/artiklar/dahlbergsband/del2/bilder/laminat0013.jpg))
dave
Yes, very well written and very well made speakers.
I have just downloaded Femm, and first thing i must ask you guys with more experience, is can we simulate this 3d problem in Femm? I am very interested in claims that this magnet arrangement is better (gives more uniform and stronger magnetic field) so i would like to measure that.
And how can i measure magnetic field in reality, on those Apogee speakers?
Thanks
I have just downloaded Femm, and first thing i must ask you guys with more experience, is can we simulate this 3d problem in Femm? I am very interested in claims that this magnet arrangement is better (gives more uniform and stronger magnetic field) so i would like to measure that.
And how can i measure magnetic field in reality, on those Apogee speakers?
Thanks
Femm will only do 2d simulations. What I have done is to model my ribbon with femm once from the face-on direction and then from the top down direction. This works well to refine your design but I'm not sure either way will give you an exact simulation of a single point in 3d space.
I don't know if there is any way to measure the large field strengths with anything approaching consumer cost equipment. If there is, I've not found it.
I don't know if there is any way to measure the large field strengths with anything approaching consumer cost equipment. If there is, I've not found it.
Thanks Dhenryp. Then it will be matter of trial and error.
Someone wrote before that long ribbons with stripes can have resistance high enough to be driven directly. I didn't buy aluminium foil yet, so i must ask you, if you can confirme that.
I am very interested in building mylar/aluminium ~1m long ribbons, and i should get strong magnets in near future.
Apogee ribbon have flat mylar/aluminium stripe, and very big magnets with classic (left/right of the ribbon) arrangement.
I suppose that i should get all the materials, make frame, and then experiment with different riibon types.
Keep on diy!
Someone wrote before that long ribbons with stripes can have resistance high enough to be driven directly. I didn't buy aluminium foil yet, so i must ask you, if you can confirme that.
I am very interested in building mylar/aluminium ~1m long ribbons, and i should get strong magnets in near future.
Apogee ribbon have flat mylar/aluminium stripe, and very big magnets with classic (left/right of the ribbon) arrangement.
I suppose that i should get all the materials, make frame, and then experiment with different riibon types.
Keep on diy!
FEMM
Femm will only do 2d simulations. What I have done is to model my ribbon with femm once from the face-on direction and then from the top down direction. This works well to refine your design but I'm not sure either way will give you an exact simulation of a single point in 3d space.
FEMM is, I believe, very accurate if the missing dimension is "repeated". In other words, say we're modelling a line array. If the FEMM slice we use is perpendicular to the long axis, and also there are no variations along the missing dimension, it should be fine. There will only be a small difference at either end of the line. The FEMM examples show a simulation of a conventional speaker: this works because the missing dimension is simply rotated around the center.
In J. Verhagen's excellent book he measured a Raven (or was it a Cantus driver?) using a hall-effect sensor and found that it agreed very closely with the FEMM model.
Unfortunately, dhenryp's ribbon design does not seem to meet these simple criteria, which makes it harder (because he has cross pieces offset along the vertical). So, in this case, I guess, FEMM is producing inaccurate results, perhaps.
Femm will only do 2d simulations. What I have done is to model my ribbon with femm once from the face-on direction and then from the top down direction. This works well to refine your design but I'm not sure either way will give you an exact simulation of a single point in 3d space.
FEMM is, I believe, very accurate if the missing dimension is "repeated". In other words, say we're modelling a line array. If the FEMM slice we use is perpendicular to the long axis, and also there are no variations along the missing dimension, it should be fine. There will only be a small difference at either end of the line. The FEMM examples show a simulation of a conventional speaker: this works because the missing dimension is simply rotated around the center.
In J. Verhagen's excellent book he measured a Raven (or was it a Cantus driver?) using a hall-effect sensor and found that it agreed very closely with the FEMM model.
Unfortunately, dhenryp's ribbon design does not seem to meet these simple criteria, which makes it harder (because he has cross pieces offset along the vertical). So, in this case, I guess, FEMM is producing inaccurate results, perhaps.
fortitudine - I didn't think measurement would be hard because of my design, I just assumed that it would be hard for consumer devices to measure the high field strengths involved. I've not read the book you referenced, but it seams that I was wrong in my assumption. My crosspieces are nothing very unusual, they just provide for a path to complete the magnetic circuit.
Mx - if you make you aluminum conductor thin enough (thin in both the thickness and width) and long enough you can drive ribbons or planar devices directly with amplifiers. With planars (e.g. foil stripes on mylar) you can get very long effective lengths by connecting all the physically parallel strip in series electrically. You could also use very thin ( in thickness) aluminum because the mylar can provide the physical strengths.
I agree with you, the best way to build something is to start building something
Mx - if you make you aluminum conductor thin enough (thin in both the thickness and width) and long enough you can drive ribbons or planar devices directly with amplifiers. With planars (e.g. foil stripes on mylar) you can get very long effective lengths by connecting all the physically parallel strip in series electrically. You could also use very thin ( in thickness) aluminum because the mylar can provide the physical strengths.
I agree with you, the best way to build something is to start building something
Because i will build classic magnet arrangement ribbon speaker (without cross pieces), it really makes it easier to simulate in femm, like Fortitudine said. Only problem are places between magnets (small gaps), i can't simulate that in femm (in 2d), but i assume that it doesn't affect much magnetic field (and distortion of speaker) because ribbon is much longer in places between magnets.
With some aluminium foil that i have on 2mm wide strips i got 8Ohm on ~1m of lenght, which is bad. I would like 1m long ribbon, so i must have about 5 to 6 stripes in parallel, but i will find some other foil. Good thing is that i found mylar in different thicknesses (from 5 to 20 um), so I will experiment with all of them.
And one more question. When I am buying magnets, where can i see in specifications direction of magnetization?
Mx
With some aluminium foil that i have on 2mm wide strips i got 8Ohm on ~1m of lenght, which is bad. I would like 1m long ribbon, so i must have about 5 to 6 stripes in parallel, but i will find some other foil. Good thing is that i found mylar in different thicknesses (from 5 to 20 um), so I will experiment with all of them.
And one more question. When I am buying magnets, where can i see in specifications direction of magnetization?
Mx
Your picture is neat - what software did you use to build it?
It would be best to eliminate the spaces between adjacent magnets. Aside from the cost of the extra magnets there is no technical reason why you can't do this. I would also suggest reconsider using crossbars. Completing the magnetic circuit with cross bars will increase the field strength in the gap significantly (I can't remember the number now but I thing it's something like a 40% increase).
It would be best to eliminate the spaces between adjacent magnets. Aside from the cost of the extra magnets there is no technical reason why you can't do this. I would also suggest reconsider using crossbars. Completing the magnetic circuit with cross bars will increase the field strength in the gap significantly (I can't remember the number now but I thing it's something like a 40% increase).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.