|Poul Ladegaard's Air Bearing Tangential Tonearm
|Page 1 of 1|
|Author:||Roscoe Primrose [ February 28th, 2013, 2:48 pm ]|
|Post subject:||Poul Ladegaard's Air Bearing Tangential Tonearm|
THE THIRD ARM - AND MAYBE THE LAST!
POUL LADEGAARD DESIGNED A WELL FUNCTIONING TANGENTIAL ARM MORE THAN 20 YEARS AGO. IN THE YEARS THAT HAVE PASSED, MANY EXPERIMENTS HAVE BEEN CARRIED OUT. THESE HAVE LED TO AN AIR ARM THAT CAN BE BUILT IN AN AFTERNOON.
By Poul Ladegaard
Translated into English by Thomas Dunker
(pardon occasional clumsy language. I hope it's understandable. -Tom)
After twenty years of having lived, toyed and suffered with home made tangential arms it's about time to give a report on the status.
Partly we are seeing the end of a technology about to be outrun by a newer, perhaps more potent, yet also more alienating technology. Partly, the old analog technology now has reached its mature stage where it is more fun than ever before, where the experience gathered is the greatest and where everyone knows what it's all about.
Furthermore, building your own tone arm is still something most people can do and get away with decent, maybe even superb results if the necessary craftsmanship is present along with good ideas. The emergence of air arms in the past few years has shown new possibilities and given a lot of inspiration. In this respect, this article will give some hints to a DIY arm that not only is extremely affordable, demands no special tools but also represents an almost perfect solution.
Solutions that combine simplicity and practical realization of 'ideal' demands don't come suddenly. They are results of many years of evolutionary development around my own designs combined with close encounters with the countless arms that have passed the critical inspection in the test lab.
A recapitulation of the most important experience and theoretical consider- ations made along the way may be useful.
THE FIRST ARM
The picture shows the final version of the author's (Poul Ladegaard) first prototype. Here he uses a sort of ball bearing to pivot the arm. Later he found that knife blades work equally well. (Stanley blades seem fine)
The arm tube is a carbon fibre tube, actually from a badminton racket. This seems to be a very good choice of material for the arm. The mid graph of 'armgraphs.jpg' shows measurements of vibration done on the arm using a high level sweep. The arm behaves well, as the graph shows.
The arm lift appearing on this picture is made from bent piano wire and an oil damped lift from an old Colton arm.
Translation of descriptive notes:
1 Aluminum angle 3 x 20 x 20 mm with air holes.
2 Araldite (epoxy) glue (silicone can be used successfully)
3 3 mm aluminum angle, cut accordingly. (Or a U-profile of proper size.)
4 6 x 30 mm cross section aluminum block.
5 Support block of black painted wood.
6 Air input tube
7 Tape with 0.3 mm holes.
The armbase is 30 cm long and each end is closed with a thin piece of aluminum sheet (1 mm) filed to the right shape.
As for most of us, it started with the challenge of doing things yourself. The result is guaranteed to be less expensive than the ready made gear. After study of the problem, the challenge often increases into doing things differently as well.
All the others have probably 'made mistakes' somewhere or neglected an 'important' optimization. So the clue is to find a couple of points that seem to have been ignored and which can fuel the enthusiasm of the home designer.
I soon realized that the fundamental problem with a radial arm, namely the mistracking angle, was the root of all evil. First, it is minimized by giving the arm the familiar offset angle. Thereby, the skating force is not eliminated. Rather, it increases, but is usually made easier to correct. However, it never gets ideal. The skating force depends on the friction between the record groove and the stylus tip, and thereby also the cut of the stylus. It also depends on what is recorded on the record and whether or not wet playback is used...the quality of the vinyl...or other lubricants...
No! Optimizing error correction in the third/fourth level, that's not me.
A tangential arm became the only traversable road. Gone were all the afore- mentioned problems and one spinoff was apparent. The arm could be made shorter, lighter, and at the same time less resonant. This way experimentation with the matching of the effective mass to the high compliance of modern cartridges could be carried out quite freely. The arm resonance could, by an 'optimal compromise' be placed between 10 and 15 Hz. Hereby the sensitivity to the subsonic 'junk' on the record is reduced, and a fair reproduction of bass transients can still be had.
The result can be characterized as a geometrically optimized, light arm with a traditional cardanic suspension. But the mechanical problems associated with such a DIY design were substantial. Motor and gears for movement were not silent and naturally caused a stepwise movement of the arm. Countless experiments with gliding rails for the lateral bearing, contacts for starting and stopping the motor, lift design etc., was needed. Still, the operation was only for the expert and the numerous service interruptions were only accepted by the designer, who was put to the test. He knew how difficult they were to avoid. But times of musical enjoyment of the highest quality are remembered, often of several minutes' duration, and occasionally, an entire record side! And such moments are not easily forgotten by a do-it-yourselfer.
THE SECOND ARM
The foundation for this design was divided in two: A whole new idea for a movement entirely without motors and complex controller was to be tested. At the same time, a theory about a more optimal and well considered placement of the arm resonances was to be tested in real life. The latter in the attempt to split resonances below 20 Hz in the vertical and horizontal planes with independent responses, thereafter optimizing them separately.
The basis was/is that there is only groove modulation in the horizontal plane at lower frequencies, below about 200 Hz, and that unwanted disturbance from warps on the record exist in the vertical plane. Due to the desire for the best possible transient response, the frequency response in the horizontal plane should be as smooth as possible, with a soft 6 dB/octave rolloff down towards 1-2 Hz, so that the only disturbance from eventual eccentricity can be eliminated.
In the vertical plane the goal is to achieve the best possible rolloff of the warp frequency below 10 Hz. This means that the arm resonance should be between 12 and 16 Hz, and it can very well be undamped, so as to make the rolloff as steep as possible. In real life this means that the effective mass of the arm in the horizontal plane has to be 25-50 times greater than in the vertical plane, where in a light arm it's typically about 10 grams. The total effective mass of the moving system in the horizontal plane will then be in excess of 250 grams, and depending on its distribution, the total mass can get very high. One experiment was made.
The necessary inertia was achieved by using two large lead weights coupled to the vertical axis of a cardanic bearing, at a distance of about 5 cm at a right angle to the arm tube. The arm was now ready to be tested - at first as a radial arm. Listening tests as well as measurements confirmed the idea.
Now the challenge was to establish the tangential movement of the arm and arm base, now weighing more than 500 grams. It turned out to be a problem for the expert! Any sort of stepwise movement would instantaneously provoke any modest asymmetry in the arm and cause torque momentum, sending the diamond on an involuntary skating trip. Here, the idea of a new principle of movement had to be tried. The mass of a lead weight was to be transformed into the needed horizontal pull force using a pulley.
The problem now is to control the movement of the arm so as to ensure tangential tracking at all times. A heating wire slowly melting through stearin wax had the necessary properties and could by means of varying the electrical currrent be smoothly adjusted from a full stop to the necessary maximum speed. The wax was placed in a small tray in connection with the pulley wire. A wire in the wax along the edge of the tray was fastened to the arm base and thus it could steer/slow down the turning of the wax tray and thereby the movement of the arm.
The mistracking angle of the arm was measured with a photo cell, whose signal was converted into an evenly pulsating current in the heating wire.
It worked and yielded completely smooth and silent movement. But in order to get the whole thing working the heavy arm base had to be furnished with a great number of precision ball bearings as wheels and for steering.
Although it sonically fulfilled all expectations, it was to critical to use in practice. The very large inertia weights were very demanding to adjust to the required balance, and changing tracking force or replacing the cartridge would take hours. Furthermore, it turned out that the wax changed properties after having been melted through a few times, and had to be replaced. Dust deposits on the roller rail of the heavy arm caused unwanted friction and service interruptions.
Soon the arm was covered with dust altogether, and the 'spare' Moerch DP-6 took over the daily duty with ease of operation and reliability that contrasted 'the second arm'.
The wax control is theoretically ideal, it even works in reality, but requires a carefylly chosen type of wax in order to become reliable.
Increasing inertia in the horizontal plane with side weights fastened to the vertical bearing is too critical to adjust practically and is very sensitive to disturbance.
Separating the arm resonances in the horizontal and vertical planes and the corresponding optimization however, is beyond any doubt desirable.
THE THIRD ARM
Translation of descriptive notes:
1 Carbon fibre arm tube from racket, with an outer diameter of 7.5 mm.
2 Bracket for fastening of arm tube, made from 3 mm brass with 3 mm thread.
3 Counter weight (made from lead), tightly fitted and then carefully squeeze fastened to the arm tube after the right tracking force has been set.
4 Lift, made from bent 2 mm thick wire and supported at two holes in the supporting blocks.
5 Knife blade
6 Rest for knife blade if a harder material than aluminum is desired. Alternatively it can be used as a simple height adjustment for the arm.
7 3 mm nut.
8 Flange, made from a piece of 2 x 15 x 15 mm angle aluminum profile.
9 'Floater' made of 2 x 20 x 20 or 3 x 20 x 20 mm angle aluminum. Length is 18 cm.
10 Arm base with air holes of diameter 2.5 mm drilled 2 cm apart. Angle aluminum as in 9), but 30 cm long.
11 A narrow strip of double sided adhesive tape, optionally with a little contact glue to get the joint between the tape pieces perfectly air tight.
12 Aluminum angle as in 9), but 30 cm long and with a hole for a 5 mm piece of pipe to connect to air hose.
13 Wires to cartridge. Choose only the best and emphasize flexibility highly. Teflon insulated wires are too stiff.
14 Tape with 0.3 mm holes, made with, for instance, a syringe needle.
15 Air film of thickness 20-60 micrometers.
16 5 mm pipe for air supply.
17 Air hose.
18 Support block. Can be acryl, or as here, made from black painted wood. Can easily be made from two pieces with a 45 degree cut glued together. The support block can finally be glued to the turntable for continuous use. Double sided tape works fine in the experimental phase, and perhaps even permanently.
NB! The drawing is not exact to scale.
Photo of the final arm built according to sketches above. Note that you can add mass to the 'floater' and trim the lateral inertia. This is difficult with most conventional arms (using a sliding tube arrangement).
If you don't have enough space on the plinth for arranging the floater symmetrically like this, you can place the arm at the side of the floater and add weight on the other side to balance the floater. It's important that it has the mass center on the middle, or else more air will be needed.
Many years went by. The advantages of the tangential principle were still not forgotten. Neither were the many problems associated with the previous solutions. If a third arm was to be built, it decidedly had to be simple. Furthermore, it had also become a family demand that it should be as humane to deal with as the friendly Moerch arm.
Maybe an air arm could be the right way? But, even though the first ones I encountered in measurements for test reports were fascinating, their design was certainly not suited for kitchen-table cheating. Also, a number of details seemed less than well considered. It had to be possible to do it in a simpler way and to do it 'right'.
Independent of the design it is a welcome property of air bearing arms that they almost automatically have the desirable big difference in vertical and horizontal inertia. The whole mass of the arm contributes to horizontal inertia due to the translatory movement. It should be possible to use this in developing a good solution.
Johnny Soerensen's description of his 'Kolibri' arm in 'high fidelity' nr. 7/8 and 9 1990 was sufficient provokement. And if it hadn't been for the expensive 'laser hole' tube there would probably already be a copy on which to base new experiments.
One single idea, however, set of a minor avalanche and (almost) all the pieces of the puzzle fell together:
"An air arm based on angle profiles, where only the bearing for horizontal movement is air bearing, where the vertical movement is controlled with a traditional bearing based on balls, pins or knives."
This immediately gives the following advantages:
The air bearing does not necessitate precision turned or matched tubing. Readily available angle aluminum can be used.
The air holes can be drilled on a 'kitchen table' level from th back of the angle profile with bits of comfortable oversize. The size of the holes is reduced by applying adhesive tape which can be punctured with a 0.3 mm syringe needle.
The vertical bearing, whether with knives or ball bearings is automatically centered and requires no further fastening.
The arm tube with counterweight, bearing and cartridge can be optimized as a readily detachable separate unit.
With added mass on the arm carriage such as lead weights or shot, the effective mass of the arm in the horizontal plane can be adjusted to the desired value without influencing the vertical inertia.
The arm does not have to be placed on the middle of the carriage as long as the mass center of the carriage with arm and lead weights is kept at the middle of the carriage. This yields possibility for a more flexible placement on the turntable.
The distance from the horizontal bearing to the bottom of the the arm base can be be made as little as 10 mm, depending on the thickness of the profile. Thus, the arm base can be fitted so that it is suspended over the record if a very short arm is desired. This without serious consequences for the optimal placement of the horizontal bearing - in level with the record surface.
HINTS AND TIPS REGARDING THE PRACTICAL DESIGN.
The choice of profiles and preparation in the two models that have been built are solely based on the materials that were already available in the workshop junk box. As such, it is not the result of many experiments or considerations. The air bearings described here can take a useful load of up to 500 grams with a good pump. If the supporting surface is halved, it may be able to support 250 grams, and that may be sufficient for a light weight design. The two prototypes described here should have plenty of margin.
THE FIRST PROTOTYPE is based on a piece of standard 3x20x20 mm angle aluminum that was cut into a length of 30 cm for the arm base and 18 cm for the carriage. This gives the arm a net travel of 12 cm. This is sufficient to cover the engraved part of the record of about 10 cm and allows room for 'parking'.
After grating off corners and edges, the profiles were lightly honed together with a piece of fine sandpaper, coarseness 240 or finer (grade figures for sandpaper may vary from metric to imperial. Use your heads! -Tom) Coarser paper is not recommended. Loose grains may settle in the soft metal, causing disasterous consequences later. There is very little room when the thickness of the air film is less than 50 microns! There shouldn't be too great a need for polishing unless the profiles are very rough. Fine steel wool is recommended for finishing, and in many cases it will handle the whole job.
A helpful tool consisting of two strips of wood, for example 22x22 mm fastened to a wooden board so that the distance corresponds with the thickness of the profiles is almost indispensible. Partly, it protects the kitchen table, and partly it makes it a breeze to measure and drill the air holes from the outside of the profiles. Good results have been achieved when the holes are drilled along the middle axis of the profile (on the outside) and with a diameter of 2.5 mm, the holes placed 2 cm apart. Hereafter the holes should be grated off around the edges and you can polish again if necessary - or wait with the polishing until all the holes have been drilled. (RECOMMENDED!!!!!! -Tom) The holes can now be covered with tape, such as 3M Scotch or similar with long term durable adhesive. The final holes can now be made with a 0.3 mm wire or syringe. (I found that normal Scotch tape was a little bit brittle and the holes wouldn't get perfectly circular. I wound up using two layers of paper masking tape cut narrow and lined these with PVC insulating tape to ensure airtightness. -Tom)
All you need to do now is to close the angle profile at the back and to fit an inlet for the air hose. The easiest way to close it is to use another piece of aluminum, fastened with a strip of double sided tape. See sketches.
THE SECOND PROTOTYPE is built with anodized profiles measuring 2x20x20 mm. Here, no polishing is possible, so the surfaces had to be smooth from the beginning.
If the profiles have slightly different inner and outer angles, they can be made equal. Cover the long piece on the inside with aluminum foil and cover it with a thin layer of epoxy glue. Now the carriage piece can be pushed into the cement. After the cement has hardened, the carriage will fit perfectly with the bottom piece.
As a horizontal bearing, one with 3 mm ball bearings as well as one with 'hobby knife' blades have been tried. Details can be seen on photos and sketches. Both gave outstanding results.
The arm tube is a carbon fiber tube from a discarded Badminton racket with a lead block as counterweight. The arm is fastened to the bearing and cartridge with small clamps.
There is nothing particularly critical or expensive to anything in this design. Try first with the most easily available materials and methods.
TESTING AND MEASUREMENTS:
The objective was to get the horizontal plane arm resonance down to 2-3 Hz. The arm and carriage weigh 160 grams altogether, which was incteased up to a total of 300 grams with two triangular lead weights. These were molded in the bottom of a leftover angle profile. This yielded the desired results. See measurements:
Graph one (upper):
Arm resonance measurement, executed with the regular test record. Note the very smooth response in the lateral ("vandret") plane all the way down to 3-4 Hz. The arm resonance in the lateral plane at about 2.8 Hz could be damped to prevent susceptibility to vibration. The sound will hardly benefit from this, since the flat response from 4-5 Hz on upwards already guarantees a perfect reproduction of bass transients.
Graph two (middle):
The high level sweep is unable to provoke bad behavior or flexing in the carbon fiber arm.
Graph three (bottom):
With a special vibrator developed for the purpose, the arm's behavior at low frequencies can be studied more precisely than with a test record and with better flexibility. By connecting the air pump to a variac it is possible, simply and repeatably, to study how the damping of arm resonances depend on the thickness of the air film. Curve A is with normal air flow (220 volts on the pump). At 160 volts curve B is obtained and at 150 volts, curve C. Apparently, a curve between B and C would seem ideal, but the situation is unstable. The suspicion is that beginning friction between slide and arm base is what 'helps' here. A damping concept must be based on something else than the thickness of the air film.
Now one might fear that poorly centered records would provoke the stylus too much when it has to move so much 'baggage'. In the old days - in the age of radial arms - there was somethign called a 'skateometer'. It came from Dual and could be fitted instead of the cartridge so that one could see if it had been properly adjusted. With this fitted on a tangential arm, the effect of excentric records could be evaluated. On the arms desribed here, the load was typically around +/- 50 milligrams. This is insignificant compared to a normal tracking force of 1.8 grams. More critical is the adjustment of the arm base to perfect level. But here is yet another advantage of this design. Simply lift the arm tube off the carriage and see if the carriage remains still.
The optimal thickness of the air film can give rise to many considerations. Immediately one would assume that it should be as thin as possible. Thereby you get the leas vibration and a thinner film might give a desired damping in the horizontal plane. It's easy to experiment by varying the air supply. Based on preliminary results, an air film with a thickness of 50 microns will be sufficient. This can be achieved in practice and can be controlled with known strips of paper or foil by slipping them into the air gap. Normal 80 gram copier paper is about 80-90 microns thick. If that can pass, there is too much air. Normal LP audio tape is about 35 microns and standard cassette tape is down around 23 microns. As shown in one of the measurement graphs, a damping effect is noticed as the air film is reduced. But here it is possible that beginning friction begins to affect the response. The temporary conclusion is that it's a bad idea to base damping on the air bearing.
In order to get a precise picture of the performance of the arm down at a couple of Hz, we have developed a little vibrator to use in addition to test record measurements. It's a modified midrange speaker that can simulate a test record from 1-50 Hz. The results of measurements using this device completely confirm the test record measurements.
The ideas and suggestions presented in this article should make it possible for most of you to join in and also to get a good result. The presented solutions deliver, in my opinion, something near the closest it's possible to get to the ideal regarding the dynamic and geometric properties of a tone arm.
Of problems still awaiting solution is to establish some sort of damping of the resonant peak at 2.5 Hz. If this can be done, the arm will be more immune against external 'shaking'.
When a couple of ideas for solutions have been tried out they will be presented in this magazine and will be easily added to the existing arm. (No such improvements have surfaced yet. -Tom) But, under any circumstance it's almost a requirement for good results that it be mounted on a stable wall to remain in level. Thus, problems with picked up shaking will also be reduced.
|Author:||Roscoe Primrose [ February 28th, 2013, 2:49 pm ]|
|Post subject:||Some comments by Thomas Dunker|
Hello all DIY friends!
Now I have finished adapting the article on Poul Ladegaard's excellent DIY air arm into English. I have not converted metric into inches. You who still use inches better get used to metric, the sooner the better!
The main file is a complete translation of the original article as it appeared in the June issue of the Danish magazine 'high fidelity'. Lest you think this is a violation of copyright, I can assure you that the policy of the source magazine is that all material can be used without permission if it isn't to be used for commercial purposes.
The picture files are grayscale jpegs scanned from the article and are very valuable if you are going to build this arm. Each picture file has an explanatory description.
It is my profound hope that you will find the needed answers to all questions concerning the project in the translated files. I have commented to some paragraphs where I found it useful. I have built this arm myself, so I am aware of the problems. Please read the files thoroughly before asking me any questions. I had only the article to lean on and it was enough for me.
I don't have a lot to add, but I do have a few points:
Pumps intended for use with fish tanks are great for an air arm. Bear in mind that some pumps are noisier than others, so think about where you can place the pump. A good idea is to keep it sufficiently far away from the listening room to avoid the noise they make. You should add a 10-15 liter (abt.3 gallons) tank to smooth the air flow. And it's crucial to have some way to adjust the air flow from the pump. Most pumps use solenoids that are powered directly from the wall voltage. These can be made quieter by inserting a step down transformer. Here you Americans and others with 120 V wall voltage have an advantage.
If the temperature difference between the room where the pump is and the room where the record player is is sufficiently great, condensation may be a problem. In that case you need a gas drier. This I know nothing about.
I use a RENA 301 pump which works great and is cheap. But it's noisy. Check out pumps like Whisper 2000 if you need a quieter pump. Quiet pumps are generally more expensive.
ALL the metal parts of the arm assembly should be grounded to the preamp ground. Experiment with different arrangements for best results. If the carriage and arm base are not connected electrically (with a flexible wire) a static buildup tends to develop due to the friction of the air flow onto the carriage, finally discharging to cause loud pops in your speakers. This I found out myself.
Make sure you are able to adjust the level of the turntable with great accuracy. I added a separate level adjustment to the arm assembly, because it's pretty hysterical about perfect level. This means that the arm in the present form can not be used if the plinth of your turntable is mounted on springs or other elastic principles. The large moving mass would upset the mass balance enough to make the arm mistrack. Keep this in mind before you start building.
The vertical Tracking angle is clumsy to adjust on this arm. Ingenious souls may wish to add some sort of adjustment feature for this. I'll scan a picture of a different DIY arm with a simple VTA adjustment that should be easy to copy. This will be made available here some time this month (may '94).
The very idea of tangential arms is to have the needle tracking the record with a perfect tangent at all times. So you need to carefully align the arm base, arm and cartridge to get all the angles right. An easy way to do this is to find an awful old record that you hate so much you never want to listen to it again. Now draw a narrow line across the exact middle of the record with a tiny, sharp object. Put it on the platter and make sure the stylus tip falls into the tiny groove you cut at both the beginning and the end of the record side. If it does, you have perfect tangential tracking, provided that your cartridge and arm are at perfectly right angles. How to do that you'll have to figure out yourselves.
THE ARM ITSELF:
Can be made from any familiar material you have handy. Don't despair if you can't find a broken Badminton racket or other source for carbon tubing. Arms have been made from all sorts of materials in the past, so feel free to experiment.
Is measured with a suitable device intended for the purpose. If you can find that cheap Ortofon plastic thing it'll be enough to do the job. (Comes with most Ortofon cartridges when you buy them. Ask a dealer.)
Now I can't think of anything else to add, except that you owe it to yourself to build this arm. Read the article and find out why this may be the superior alternative to any arm, and at a ridiculously low cost. DIY projects of such brilliance aren't seen very often. Why Poul Ladegaard didn't patent it I don't know.
|Author:||Roscoe Primrose [ February 28th, 2013, 2:52 pm ]|
|Post subject:||Pics of other implementations|
Thomas Dunker's table/arm:
JP Garigon's table/arm:
Some detail of JP's arm:
|Author:||Roscoe Primrose [ February 28th, 2013, 2:53 pm ]|
|Post subject:||More information|
Jeremy Epstein's Ladegaard arm FAQ http://home.earthlink.net/~jeremyepstein/ladegaardfaq.html
|Page 1 of 1||All times are UTC - 5 hours [ DST ]|
|Powered by phpBB® Forum Software © phpBB Group