| sreten |
Whist pondering the design of this arm :
http://www.aiko.com/roscoe/airbearingarm.html
and dice45's airpumpless arm :
http://home.t-online.de/home/bernha.../LT-2_index.htm
It struck me that the airflow of the Ladegaard design
could be replaced by a magnetic repulsion bearing.
My estimates indicate that as long as modern magnetic
materials can support say > 500% of their own weight
this should be entirely feasible, at 500% the lateral
mass of the arm would only be increased by 20%.
A small price to pay for no airpump.
Or have I missed something ?
Anyone know what the actual percentage is for modern
magnets ? I would of thought its a lot more than 500%.
:) sreten. |
|
|
| Havoc |
Why not indeed. Only contra-indications I "feel" are that the magnetic circuit of a magentic Ladegaard version would be very "open". So it would generate a large stray field. And while a lot of people atround here dismiss magnetic bearings for platter because of possible cogging, I'm afraid that this is a far larger problem with an arm that is not actively driven. I mean a platter has a large mass and even larger inertia and is driven, so any cogging will be not much of a problem. But the Ladegaard is "driven" by the groove only.
I really need to put mine together. All parts are already finished (more or less) but I never get around finishing the job. Only problem I have is that making the holes 0.3mm drilled through the AL profile, it generates a lot of resistance, so I need a bloody large compressor instead of an aquarium pump.... |
|
|
| sreten |
| quote: | Originally posted by Havoc
Only problem I have is that making the holes 0.3mm drilled
through the AL profile, it generates a lot of resistance, so I
need a bloody large compressor instead of an aquarium pump.... |
I read somewhere you cover the holes (on the inside) with
electrical tape and then puncture this with a hypodermic.
Edit : sorry I misread, you have a pressure problem, not airflow.
:) sreten. |
|
|
| analog_sa |
Hi Sreten
Can you say more about your idea? I assume you'll still need some guiding rails/bearings, right? |
|
|
| sreten |
| quote: | Originally posted by analog_sa
Hi Sreten
Can you say more about your idea? I assume you'll still need some guiding rails/bearings, right? |
you make something very similar to the Ladegaaard arm except
the guide and slide are magnetic material and add magnets.
The slide could be split into two small slides spaced apart,
for stabililty and minimum added mass.
The guide uses bog standard ceramic magnets.
The slide(s) use small high power modern magnets, these
need to be just powerful enough for levitation in the guide.
Should work ?
:) sreten. |
|
|
| Pjotr |
Hi Sreten,
It’s true that magnetic bearings can carry a lot of weight. But they also act as a (non-linear) spring and as such are, uhm well, very springy. Wonder how you gonna deal with this. This is also the case with air-bearings but that “bearing” is very thin and much stiffer.
Cheers ;) |
|
|
| sreten |
| quote: | Originally posted by Pjotr
Hi Sreten,
It’s true that magnetic bearings can carry a lot of weight. But they also act as a (non-linear) spring and as such are, uhm well, very springy. Wonder how you gonna deal with this. This is also the case with air-bearings but that “bearing” is very thin and much stiffer.
Cheers ;) |
Yeah, I was wondering if it would simply refuse to stay in the guide.
But no fear, I have a solution for that - you add another
repulsion system above the arm to keep it in the guide.
This would be a very simple overhead flat linear system.
But I'm beginnning to see the problems :
If you make a stiff triangular repulsion bearing any misalignment
of the surfaces will impart a side force. And any uneveness of
the magnetic field would do the same, this is cogging ?
And there's the vertical resonant frequency to consider :( .
:) sreten. |
|
|
| ingvar ahlberg |
Hi Sreten
What about trying Your idea but with fieldcoil instead, You would get an adjustable system...
If coil is in guide rod there would be several adjust/control options? |
|
|
| dice45 |
Hello all,
not wanting to be a naysayer, but Nd-magnetic stray field? in the presence of 4 air-spaced unshielded phono leads with a 0dB-level of 200-600 µV? :confused: ... and cogging too has been mentioned ... that alone is reason enough not to do it.
While you are musing about Ladegaard-improvements, why not try to feed the air to the slider instead to the rail? Why not figur out a way to lead/guide the (ultra-thin & flexible) air hose in a way that the air hose pull helps the slider to move continuously centerwards?
One hint: with the air fed to the slider, you need only 4 nozzles (with pockets) at the ends of the slider's bearing surfaces and you have 100% nozzle coverage. Hence you can carry much more weight with the same pump as the air pressure does not collapse.
And another hint .... keep the nozzle pockets as small as even possible as they store pressurized air ... => could cause oscillations at higher frequencies. |
|
|
| sreten |
Was just a thought, wondering if magnetic is possible. I don't
pretend to know much about air bearing arms, other than what
appears to be common sense, and I'm sure their are people
who've given air bearing arms a great deal of thought.
| quote: | | not wanting to be a naysayer, but Nd-magnetic stray field? in the presence of 4 air-spaced unshielded phono leads with a 0dB-level of 200-600 µV? ... and cogging too has been mentioned ... that alone is reason enough not to do it. |
DC magnetic fields are not a problem. Don't think cogging is either.
| quote: | | Why not figur out a way to lead/guide the (ultra-thin & flexible) air hose in a way that the air hose pull helps the slider to move continuously centerwards? |
Just point out that the arm generally travels much further back
and forth at the eccentric frequency than it does at the very
slow average speed towards the centre. IMO helping the slow
average speed towards the centre is not a technical issue.
regarding the basic idea :
With the comments so far I can see as usual the problem is far
more complicated than it first appears. And modification of
design using a different system is far from a "good" solution.
Pjotr :
You say magnetic bearing can carry a lot of weight - interesting.
If this is true then a very small system is required.
I need to give this some further thought.
P.S. all :
I'm just as interested in why it can't work as if it can.
:) sreten. |
|
|
| Pjotr |
| quote: | Originally posted by sreten
...........................
Pjotr :
You say magnetic bearing can carry a lot of weight - interesting.
If this is true then a very small system is required.
I need to give this some further thought.
P.S. all :
I'm just as interested in why it can't work as if it can.
:) sreten. |
Hi Sreten,
The force a magnetic field applies onto two poles with an air gap in between can be expressed as:
F = B^2 x A / (2 x µ_0)
Where F in Newton, A in m2 and B in T (or Wb / m2). Taking into account a max saturation flux density of the necessary iron of 1.5 T, the max force you can get is then app. 100N/cm2. This is equivalent to a pneumatic force of 10 bar.
Cheers ;) |
|
|
| sreten |
| quote: | Originally posted by ingvar ahlberg
Hi Sreten
What about trying Your idea but with fieldcoil instead, You would get an adjustable system...
If coil is in guide rod there would be several adjust/control options? |
Hi IA,
Can't see anything wrong with the approach, especially for
an experimental set up, I'm now thinking along the line of
something that could use a coil or permanent magnets.
:) sreten. |
|
|
| sreten |
| quote: | Originally posted by Pjotr
Hi Sreten,
The force a magnetic field applies onto two poles with an air gap in between can be expressed as:
F = B^2 x A / (2 x µ_0)
Where F in Newton, A in m2 and B in T (or Wb / m2). Taking into account a max saturation flux density of the necessary iron of 1.5 T, the max force you can get is then app. 100N/cm2. This is equivalent to a pneumatic force of 10 bar.
Cheers ;) |
Thanks Pjotr,
I can see this sort of force in attraction.
But I don't think we would be dealing with saturated poles.
One of the problems seems to be the "springyness" of the
repulsion, a geometry is needed that rapidly varies the air
gap as you approach the surface to "stiffen" the bearing.
I've got an idea that I'll draw up and post later.
:) sreten. |
|
|
| sreten |
Hi,
I've pondering what could work well as magnetic bearing but a
given is that the "stiffness" of an air bearing is not possible.
Also thought it would be nice to include a magnetic vertical
bearing in the basic idea as well, though knife edges could
still be used if a v block is used top and bottom.
The basic principle is shown in the diagram, I'm not suggesting
this is the best way to do it, just trying to show the idea.
The slide consists of two hollow steel tubes energised
by two small high power magnets placed each end.
(For weight considerations these could be steel tube sections
at each end mounted on aluminium supporting hollow tubes)
The guide is fairly self evident and a DC field coil
could easily replace the ceramic magnet(s) shown.
Now as I see it the top is inherently unstable and the slide
would simply fly out sideways, to stabilise this the arm is
underhung from the slide, the length of the underhang being
enough to provide the classic slight increase of tracking of
force as the arm lifts from horizontal, for improved tracking.
Seems to me it would only work if the bottom slide piece sits
at equilibrium very close to the V-groove wall, I'm thinking
the \o/ interface would cause this to happen as long as the
pole piece field is not too strong.
Do you think the \o/ interface would be quite stiff ?
The upper |o interface I envisage having quite a gap.
(Some mileage in adding a concave section
to the pole piece , i.e. an "(o" interface ?)
Obviously the whole thing would need to be made overlength
so the ends do not affect field strength, theoretically
bucking magnets at each would help even the magnetic field.
Stiffness in all planes would be pretty good, my only concern
would be the vertical resonant frequency, not sure about this.
Don't bash me too hard if there is something fundamentally
wrong with this, I'd like to know what that problem is.
Just seems to me getting rid of the airpump is a great convenience,
at the cost I admit of rigidity, but I assume there would be be
some noise advantages over an air bearing arm - there must be ?
:) sreten. |
|
|
| Havoc |
Don't think the stray DC fields are a problem on the leads. They might upset the cartridge, but that can be far-far away.
As for damping, making the carrier out of copper should give nice magnetic damping without adding much weight.
While supplying the air to the carrier in a conventional Ladegaard may have advantages, the stiffness of the feeding hose will be a problem. I was amazed when I found out how low friction this arm bearing was. Even my phono leads (two 0.1mm Cu twisted pairs) have influence. |
|
|
| sreten |
| quote: | Originally posted by Havoc
As for damping, making the carrier out of copper should give nice magnetic damping without adding much weight.
|
Isn't using aluminium just as good ?
Though I'm having a hard time seeing the induced
current flow that would provide some damping.
I can see that an aluminium oval placed as shown would provide
lateral damping and could be incorporated into the construction.
Nice point though ! I wasn't considering it.
:) sreten. |
|
|
| sreten |
Just realised the aluminium oval will not provide damping in
that position, the induced currents would cancel each other.
:) sreten. |
|
|
| Havoc |
| In order to provide damping, the fieldlines have to cross the material at right angles. A eddy current or foucault brake for a rotating axis works by putting a copper disk on the axis, and then have a C-type coil former with the disk running through the gap. Don't know if I'm clear, but the fieldlines would go through the thin disk section, the C-section going over the rim like a disc brake caliper. |
|
|
| sreten |
| quote: | Originally posted by Havoc
Don't know if I'm clear, but the fieldlines would go through the thin disk section, the C-section going over the rim like a disc brake caliper. |
Seems to me eddy current effects would be
minimal, but I stand to be corrected on this.
I was initially thinking of induced shorted turn
damping, but not possible as far as I can see.
:) sreten. |
|
|
| dice45 |
sreten,
i would not be happy concerning the hanging architecture, i prefer to have my vertical pivot axis as close to the records surface as possible; warps and pressing bubbles have the fewest effect on effective stylus speed then.
Nevertheless i think you should try out if your idea works :) Ready, steady, go! :up: As you have drawn it i don't see any danger of cogging, just mighty stray fields ... :yinyang: If you think this doesn't affect sonics.
Maybe it's me only who is incompatible to the sonic effects caused by a small magnet or a piece of iron located in close proximity to a signal-conducting wire (or worse: around that wire, add to that list: faintest trace of magnetic attractability in any passive or active component except Xformers). If you have read my last post in the Nagaoka MP-50 thread, you may have wondered why i waste my energy in designing RCA connectors and such. That's why.
BTW, i also use a magnetic bearing inside my tonearm's lateral force compensation mechanism but the magnetic circuit is tinytiny and closed (not much stray field) and far apart from the phono leads. |
|
|
| sreten |
lets be clear on this :
Although I find parallel tracking arms interesting I have
no interest whatsoever in replacing my Rega tonearm.
Its mounted on a suspended subchassis turntable.
However as a why not ? if not ? and thats just wrong ?
person I find the the basic principles of alternative
approaches interesting, especially if they are dodgy.
Here I'm firstly interested if its possible, and secondarily
if possible what are the compromises involved. I feel if
its possible unless a serious disadvantage is revealed
that it would be viable approach for arm builders who
accept the performance trade-offs.
:) sreten. |
|
|
| Pjotr |
Hi Sreten,
The system you propose in the drawing is not stable. It only works if the moving magnetic system is perfect vertical. But if it turns a little there will be a horizontal force on the upper pole of the moving part applying a force on the tone arm, either up or down, depending on the direction of the turning.
Regarding the force possible, a flux density of 0.1 T shouldn’t be a problem with standard ceramic magnets, i.e. those from magnetic door locks.
Regarding eddy-current braking, you can put a strip of copper at the inside of the upper stationary pole. But anyway the speeds involved are too less to let an eddy-current brake be effective IMHO.
Cheers ;) |
|
|
| sreten |
| quote: | Originally posted by Pjotr
The system you propose in the drawing is not stable. It only works if the moving magnetic system is perfect vertical. But if it turns a little there will be a horizontal force on the upper pole of the moving part applying a force on the tone arm, either up or down, depending on the direction of the turning.
Cheers ;) |
If you read the text with the original diagram this is discussed,
some optimisation is possible, whilst on the subject :
| quote: | Originally posted by dice45
Iwould not be happy concerning the hanging architecture, i prefer to have my vertical pivot axis as close to the records surface as possible; warps and pressing bubbles have the fewest effect on effective stylus speed then. |
I agree there would be a motion vector along the groove.
However using double v blocks and a knife edge vertical
bearing as previously mentioned would address this if it is
a concern. both approaches have merits / disadvantages.
:) sreten. |
|
|
| sreten |
| quote: | Originally posted by dice45
Maybe it's me only who is incompatible to the sonic effects caused by a small magnet or a piece of iron located in close proximity to a signal-conducting wire (or worse: around that wire, add to that list: faintest trace of magnetic attractability in any passive or active component except Xformers). |
Certainly magnetic effects cannot be ignored.
And a DC coupled phono pre amplifier would not be a good idea.
Asumming the phono preamp has a sensible subsonic
filter I can't see any major problems with the magnetics.
Though I would presume any vibrations in the tonearm wires
would produce a signal, so some care is needed here, twisted
pairs at least I would of thought.
:) sreten. |
|
|
| Pjotr |
| quote: | Originally posted by sreten
If you read the text with the original diagram this is discussed,
|
Oops missed that point. Maybe something along the lines of the attached pic may get rid of it. But I don’t think it will do completely and such a thing is not easy to make. You need a skilled instrument maker for it with the appropriate tools and machines.
Cheers ;) |
|
|
| sreten |
nice diagram !
But it misses some of the other points, namely the deliberate
\o/ bottom interface to "increase stiffness ?", the top section
is suggested as a possibilty in the text, the text also suggests
the top gap is much larger than the bottom.
The hanging design makes the structure easy to make, the idea
is the instabilility is deliberately exploited to reduce the increase
of tracking force over warps due to the hanging design.
:) sreten. |
|
|
| Pjotr |
Hi Sreten,
I tried to make the air gap as small as possible, just to minimise play and as such increase stiffness. I didn’t intend to suggest any further construction details. The hanging construction seems to me an obvious way. You need to put the counterweight somewhere and anyway you can’t put is through the magnets :D
Cheers ;) |
|
|
| hifiZen |
Has anyone considered active servo control of the magnetic suspension to improve the stiffness? An actively driven coil setup with optical (laser) feedback system could concievably maintain extremely high precision positioning.
Specifically, I'm thinking about the laser heads and servo systems used for CD/DVD players, as I have some experience with these. Off the top of my head, the focus accuracy needs to be something like 5 microns. Focus and tracking of a DVD laser head has a frequency response of roughly 2kHz bandwidth, due to the low mass of the lens and positioning coils. Obviously this kind of response would not be possible with the mass of a tonearm carrier, but if a very tiny gap could be maintained between the carrier and it's rails, then one could rely on the air film for additional stiffness and damping, just as with an air bearing system...
Comments? |
|
|
| sreten |
Hi hifiZen,
Good to have another viewpoint.
But IMO a non starter for the following reasons :
a) AC signals through the field coil would be very difficult to suppress.
b) As far as I understand an "airfilm" 5um gap requires
very high precision and finish in the slide and guide.
c) once you introduce a sevo system is begs the very obvious
question of servoing the horizontal positioning, and this has
been comphrehensively done before.
Including servo absolute centering of the record on the platter.
I'm really interested in if a passive magnetic approach is possible.
As far as I can tell so far the basic idea with an adjustable
field coil could be a practical simple workable solution.
:) sreten. |
|
|
| hifiZen |
Hm. I think you're right.
After sleeping on it, I've decided that the possibility of having vibrations introduced directly to the tonarm carrier by the servo coils may well outweigh any benefits. As I recall, the easy way to tell when a CD/DVD head had gained focus lock is the 'hissing' noise it makes, by virtue of the fact that the objective lens and it's coil act just like a tiny tweeter... once the loop is closed and the servo feedback takes over, any noise in the system manifests itself as that hissing sound. As you can imagine, SNR in these little feedback loops isn't wonderful. Considering the system complexity, and add to that the reasons you give above... it doesn't look too good. But, it was fun to think about! :) |
|
|
| ktigerb |
Hi Sreten , Guys
If idea stupid , please stow the flame throwers , but why not a magnetic rod small diameter and a close fitting magnetic sleeve to hang the arm from ?
regards
keith |
|
|
| sreten |
In principle it could work, with a cross magnetised rod.
What I'm not sure about is the amount of torque due
to its intrinsic instability that would be generated.
:) sreten. |
|
|
|
