the foil arrived today
the 4.3 micron foil is way too thin for me to remove at any significant length... it should be fine for my purposes (a 1.5" long ribbon) but for a line ribbon no consist width can be removed and the longest piece I have removed is 12"
Dhenry... if you want I can include a few pieces at like 6-10" long of the 4.3 and maybe sandwich them between paper ....let me know
the 4.3 micron foil is way too thin for me to remove at any significant length... it should be fine for my purposes (a 1.5" long ribbon) but for a line ribbon no consist width can be removed and the longest piece I have removed is 12"
Dhenry... if you want I can include a few pieces at like 6-10" long of the 4.3 and maybe sandwich them between paper ....let me know
dhenryp said:
I've been pondering just this a bit, over the past weekend. It's been a long time since I took physics, but it seems to me that there are least three potentially major contributions to the way the ribbon moves...
First, obviously, its intrinsic mass determines how quickly a given force will accelerate it. Second, there is the air mass being pushed/pulled by the ribbon. Finally, as soon as the ribbon attains any significant velocity, an emf must be induced to counter the driving current?
I did a few quick calculations on a napkin but the results seem non-intuitive and I want to revisit them before I make a fool of myself publicly. I'm sure this is all loudspeaker construction 101, but does anybody know the degree to which these different factors matter? Does 1+2 completely overshadow 3, or vice versa? Is acceleration or velocity or even displacement the key attribute?
In other news, I played around with FEMM a bit and am utterly astounded by how much stronger the field is in the gap when everything is surrounded by an iron circuit and crossbars. Factor 10 or more. Wow. We never did anything like that in our ancient prototype...
Audiophilenoob said:
I hastily cut open a capacitor this morning before dashing off to work, and had a similar experience... the foil is just too flimsy to survive almost any kind of manipulation. I'll try a few different approaches, but I am not too optimistic.
In fact, I'm wondering if 7-8 microns is perhaps the correct compromise, at least for my level of proficiency. If McMaster sold that 12 micron stuff in a long roll I'd just get some and not worry about the few dB. Household foil is too thick even for a prototype, though, I want to be inspired by the results.
Zell said:
I hastily cut open a capacitor this morning before dashing off to work, and had a similar experience... the foil is just too flimsy to survive almost any kind of manipulation. I'll try a few different approaches, but I am not too optimistic.
In fact, I'm wondering if 7-8 microns is perhaps the correct compromise, at least for my level of proficiency. If McMaster sold that 12 micron stuff in a long roll I'd just get some and not worry about the few dB. Household foil is too thick even for a prototype, though, I want to be inspired by the results.
no... 5.8 would easily work...
I think I'm sending you some right?
if I am you will see... as long as you carefully move it it's ok
Member
Joined 2003
Noob,
I'm back in the states now and can mail the foil I promised...but it seems you already have more than enough. Do you still want it?
Paul
Ribbon WTW project
I'm back in the states now and can mail the foil I promised...but it seems you already have more than enough. Do you still want it?
Paul
Ribbon WTW project
Hi Zell,
This paper by Hawksford has some information on physical modeling.
http://www.essex.ac.uk/ese/research/audio_lab/malcolms_publication
C55 ADVANCES IN COMPUTER MODELLING OF RIBBON LOUDSPEAKERS, Bank G. and Hawksford M.O.J., 96th Convention of the Audio Engineering Society, Amsterdam, preprint 3837, February 1994
s.html
This paper by Hawksford has some information on physical modeling.
http://www.essex.ac.uk/ese/research/audio_lab/malcolms_publication
C55 ADVANCES IN COMPUTER MODELLING OF RIBBON LOUDSPEAKERS, Bank G. and Hawksford M.O.J., 96th Convention of the Audio Engineering Society, Amsterdam, preprint 3837, February 1994
s.html
I am now testing a different idea:
Since FEMM shows that it is basically impossible to get constant magnetic field across a ribbon, why even bother? The center of a ribbon will always feel less electromagnetic force than borders. If the corrugation makes the ribbon more rigid and it doens´t bend over a vertical axis then you can reduce the weight by removing metal foil from the center.
Naturally then you need to use laminates but hey, aluminum is order 3 times heavier than plastic!
I have 3/4" wide laminate that has 9um aluminum foil on 12um Mylar. Very easy material to handle compared to thin true ribbons. It weight equals 13um true ribbon which is a lot.
Resistance of this material is 0.35 ohm/m when 1cm width. I would like to make 1m long and power amplifier friendly 4ohm ribbon. Then I can calculate that only 2mm width aluminum strips at the edges connected in series will be enough. Everything else can be corroded away by acid.
Now I have 3/4" of 12um Mylar and 4mm of 9um aluminum. It weight equals 6um of true ribbon. But Iit´s far more rigid and all the current goes very close to magnets where field is stronger and furher increase efficiency. Also linearity! And hopefully also ringing of membrane will never be a problem again since Mylar is not so hard that aluminum and has more losses, thus lower Q for resonances.
Now... why anyone should ever use pure and fragile thin aluminum again?
I have no results yet but my laminate works pretty well.
Since FEMM shows that it is basically impossible to get constant magnetic field across a ribbon, why even bother? The center of a ribbon will always feel less electromagnetic force than borders. If the corrugation makes the ribbon more rigid and it doens´t bend over a vertical axis then you can reduce the weight by removing metal foil from the center.
Naturally then you need to use laminates but hey, aluminum is order 3 times heavier than plastic!
I have 3/4" wide laminate that has 9um aluminum foil on 12um Mylar. Very easy material to handle compared to thin true ribbons. It weight equals 13um true ribbon which is a lot.
Resistance of this material is 0.35 ohm/m when 1cm width. I would like to make 1m long and power amplifier friendly 4ohm ribbon. Then I can calculate that only 2mm width aluminum strips at the edges connected in series will be enough. Everything else can be corroded away by acid.
Now I have 3/4" of 12um Mylar and 4mm of 9um aluminum. It weight equals 6um of true ribbon. But Iit´s far more rigid and all the current goes very close to magnets where field is stronger and furher increase efficiency. Also linearity! And hopefully also ringing of membrane will never be a problem again since Mylar is not so hard that aluminum and has more losses, thus lower Q for resonances.
Now... why anyone should ever use pure and fragile thin aluminum again?
I have no results yet but my laminate works pretty well.
One thing I didn´t mention.
Even if my laminate compares with 6um pure aluminum foil in weigth, it´s resistance is not the same. So how the efficiency compares?
Let´s assume that needed power relates to resistance and current in a following way:
P = RI^2
<->
I = sqrt(P/R)
Lets assume that power is 1 watt. Then currents that go through the magnetic fields and cause electromagnetic forces for equal masses are...
I(6um aluminum) = sqrt(1/0.525) = 1.38 A
I(laminate) = 2x sqrt(1/3.5) = 2x 0.53 A = 1.07 A
2x is because current is looped once.
Now even if the laminate seems to have a bit less current and force that is not true. Current deviation is different and with the laminate, current goes through higher magnetic field causing more force.
Therefore I can well say that my laminate equals 6um pure aluminum foil or thinner and has other superior features.
Have anyone else tested similar things?
Even if my laminate compares with 6um pure aluminum foil in weigth, it´s resistance is not the same. So how the efficiency compares?
Let´s assume that needed power relates to resistance and current in a following way:
P = RI^2
<->
I = sqrt(P/R)
Lets assume that power is 1 watt. Then currents that go through the magnetic fields and cause electromagnetic forces for equal masses are...
I(6um aluminum) = sqrt(1/0.525) = 1.38 A
I(laminate) = 2x sqrt(1/3.5) = 2x 0.53 A = 1.07 A
2x is because current is looped once.
Now even if the laminate seems to have a bit less current and force that is not true. Current deviation is different and with the laminate, current goes through higher magnetic field causing more force.
Therefore I can well say that my laminate equals 6um pure aluminum foil or thinner and has other superior features.
Have anyone else tested similar things?
APi said:
Your calculations seem a trifle off to me... sure, running your current through 2mm rather than 1cm will concentrate the charge density, and running it twice will double it. From a different point of view, though, the scarce resource here is not current but flux. With 4mm total aluminum in a 1-cm gap, 60% of the flux is completely wasted. That just seems painful to me.
That said, I'm very interested in hearing how this turns out!
Zell
APi said:One thing I didn´t mention.
Even if my laminate compares with 6um pure aluminum foil in weigth, it´s resistance is not the same. So how the efficiency compares?
Let´s assume that needed power relates to resistance and current in a following way:
P = RI^2
<->
I = sqrt(P/R)
Lets assume that power is 1 watt. Then currents that go through the magnetic fields and cause electromagnetic forces for equal masses are...
I(6um aluminum) = sqrt(1/0.525) = 1.38 A
I(laminate) = 2x sqrt(1/3.5) = 2x 0.53 A = 1.07 A
2x is because current is looped once.
Now even if the laminate seems to have a bit less current and force that is not true. Current deviation is different and with the laminate, current goes through higher magnetic field causing more force.
Therefore I can well say that my laminate equals 6um pure aluminum foil or thinner and has other superior features.
Have anyone else tested similar things?
I've looked for a long time trying to find a 2um mylar laminated with 1um of aluminum
no one makes it that thin... but theoretically it would be around 1/5th the weight of 6um aluminum
Zell said:
First I have to point out that this design is aimed for very long ribbons from, let´s say, 1m up, where you really need to think, how to make your ribbon last. One good thing is that with long ribbons you can get enough resistance to be driven by a amplifier directly.
My opinion is that 60% of the flux is not wasted.
Let´s think about a traditional ribbon, where a conductor goes also in the middle of the gap. Flux there is smaller than next to magnets because the flux short circuits around magnets´corners. The more distance you have from the magnets the less flux you have. Even if the gap seems to be relatively short and both sides have magnets.
Also the flux in the middle is less linear than close to magnets. The same reason: flux short circuits and curves. So a true ribbon membrane can be moved very little if no distortion is a goal.
Of course that short circuit is no problem if you use huge magnets compared to gap width but you don´t have to waste so much money to have a linear ribbon.
With my arrangement the current goes as close to magnets as possible where the flux is the strongest and efficiency can be made better. Also the flux remains pretty constant there if one doesn´t drive the membrane too close to magnets´ corners. That improves linearity and reduces distortion.
The negative thing in my design is not "wasting flux" but increasing membrane´s weight to conductor area -ratio by using laminates. But my idea is to compensate that ratio loss by using flux more efficiently, close to magnets.
And it seems that efficiency of originally 21um thick laminate can be as good as 6um thin true ribbon, when part of the aluminum is corroded away and remaining strips are looped.
Compared to a true ribbon this have some pros:
- Mecanically strong membrane. Even re-corrugatable!
- No need for a current transformer or a resistor in series.
- Less distortion.
- Efficiency good enough.
Cons:
- None?
APi said:
First I have to point out that this design is aimed for very long ribbons from, let´s say, 1m up, where you really need to think, how to make your ribbon last. One good thing is that with long ribbons you can get enough resistance to be driven by a amplifier directly.
My opinion is that 60% of the flux is not wasted.
Let´s think about a traditional ribbon, where a conductor goes also in the middle of the gap. Flux there is smaller than next to magnets because the flux short circuits around magnets´corners. The more distance you have from the magnets the less flux you have. Even if the gap seems to be relatively short and both sides have magnets.
Also the flux in the middle is less linear than close to magnets. The same reason: flux short circuits and curves. So a true ribbon membrane can be moved very little if no distortion is a goal.
Of course that short circuit is no problem if you use huge magnets compared to gap width but you don´t have to waste so much money to have a linear ribbon.
With my arrangement the current goes as close to magnets as possible where the flux is the strongest and efficiency can be made better. Also the flux remains pretty constant there if one doesn´t drive the membrane too close to magnets´ corners. That improves linearity and reduces distortion.
The negative thing in my design is not "wasting flux" but increasing membrane´s weight to conductor area -ratio by using laminates. But my idea is to compensate that ratio loss by using flux more efficiently, close to magnets.
And it seems that efficiency of originally 21um thick laminate can be as good as 6um thin true ribbon, when part of the aluminum is corroded away and remaining strips are looped.
Compared to a true ribbon this have some pros:
- Mecanically strong membrane. Even re-corrugatable!
- No need for a current transformer or a resistor in series.
- Less distortion.
- Efficiency good enough.
Cons:
- None?
efficency will be the same as with a 6um as long as the weight is the same
also naturally going obscenely thinner... say a 6um laminate with 1um of aluminum could gain you up to 6db in added efficency... a massive number
also makes for a more robust ribbon and mylar is much stronger than 6um alum...
too bad that stuff is impossible to find
and given the choice between the same efficeny composite ribbon and a true aluminum I'll take the true everyday for sonics
Audiophilenoob said:
You just haven´t heard a good laminate yet.
BTW: Does anybody have measured results why true ribbon could sound better that light laminate?
My feeling is that a true ribbon cannot be better than laminate. Not at least for wide range use. As a short and pure tweeter I got your point. Anyway... It is extremely challenging to designs true ribbon so that...
- ...it hasn´t any resonances and distortion when played loud due transformer nonelinearities. Laminates can be made to meet power amplifier resistance requirements with no reactive components.
- ... it has exactly linear magnetic flux in the gap. Also in the middle. With laminates you can optimize current density versus flux.
- Pure aluminum tend to ring easily if not corrugated exactly right and even after that it is more sensitive to any wrinkles caused by misuse.
With laminates you can make less bad designs and assumptions than with true ribbons. Of course people like different sounds and some love certain distortion figures that f.ex. tube amplifiers and true ribbons can generate
APi said:
You just haven´t heard a good laminate yet.
BTW: Does anybody have measured results why true ribbon could sound better that light laminate?
My feeling is that a true ribbon cannot be better than laminate. Not at least for wide range use. As a short and pure tweeter I got your point. Anyway... It is extremely challenging to designs true ribbon so that...
- ...it hasn´t any resonances and distortion when played loud due transformer nonelinearities. Laminates can be made to meet power amplifier resistance requirements with no reactive components.
- ... it has exactly linear magnetic flux in the gap. Also in the middle. With laminates you can optimize current density versus flux.
- Pure aluminum tend to ring easily if not corrugated exactly right and even after that it is more sensitive to any wrinkles caused by misuse.
With laminates you can make less bad designs and assumptions than with true ribbons. Of course people like different sounds and some love certain distortion figures that f.ex. tube amplifiers and true ribbons can generate
I really haven't heard very many laminates at all actually.... I just can't DIY laminates
Audiophilenoob said:I really haven't heard very many laminates at all actually.... I just can't DIY laminates
You don´t have to. There is a lot of companies that produce laminated foils for business use. One I found is Neptco and their Neptape 1001 which I am using. I ordered 3/4" wide foild and it´s exactly that all the way.
http://www.neptco.com/website/neptco.nsf
At the moment I don´t have enough measurements to say if this foil has any drawbacks, except weight, but it sounds pretty good anyway.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.