Hi Jim,
Please don't take my comments about air bearings as in anyway a criticism of the fantastic arms you have built. The air bearing is one of the best (possibly the best) type of bearing for tonearms, definitely way better than ball-race bearings regardless of how they are implemented.
Unfortunately the effect of the air gap acting as a spring is very real and will have an effect, however small. It is similar, but to a lesser extent, to how the bending modes of of an arm-tube strongly effect sound quality even though the amplitude is in nanometers.
Previously I have just stated that the phenomenon exists without actually providing any analysis or supporting calculations. I was remiss in this.
The reason that the stiffness is quoted by the manufacturer is so the engineer can design so that the resonances do not fall within the operating range of the product being built.
I have gone onto the New Way website and looked at the specification of their 25mm air bushing. I chose this model as I believe it is the one you currently use and because it is the largest, and stiffest, likely to be used for a tonearm. New Way luckily quote the pitch stiffness for this bearing as this is the mode most likely to be excited in our application.
It is 5.3 N-m/mil rad at a supply pressure of 0.41MPa, 60psi.
Rather more difficult to determine is what the moment of inertia of the carriage will be as the carriage is not a simple shape. Working on a simplified model of a carriage with a 75mm effective length and 150g mass, including cartridge, would give a moment of inertia of 280x10^-6kg.m^2.
A longer armtube would significantly increase this as would a heavier carriage.
With this arm the resonant frequency would be 690hz.
Changes to bearing size, supply pressure, carriage mass and effective length will all effect the resonant frequency. Unfortunately the resonant frequency will fall in the audio band.
From an octave or so below this the carriage will be effectively coupled to the rail. From around half an octave above this the carriage will be decoupled from the rail, this is my only issue with air bearings.
The resonant frequency itself should not be to problematic as the use of a silicone damping trough can damp the resonant peek that could otherwise cause a peak in output at this frequency. The air within the gap will also help in damping the peak.
All engineering is a balance of compromise.
My preferred bearing system is to use spikes (pin bearings) as these couple across the entire audio band. They do however have higher friction.
The million dollar question is does the benefit of coupling across the entire audio band outweigh the increase in lateral tracking error due to bearing friction?
As my bearing friction is low anyway resultant tracking errors will be very low, those due to eccentricity etc will be much higher and still present with either bearing type.
YMMV.
Of course there are those who think that an advantage of air bearings is that they DO decouple the carriage from the rail (they're wrong 😁)
Niffy
Please don't take my comments about air bearings as in anyway a criticism of the fantastic arms you have built. The air bearing is one of the best (possibly the best) type of bearing for tonearms, definitely way better than ball-race bearings regardless of how they are implemented.
Unfortunately the effect of the air gap acting as a spring is very real and will have an effect, however small. It is similar, but to a lesser extent, to how the bending modes of of an arm-tube strongly effect sound quality even though the amplitude is in nanometers.
Previously I have just stated that the phenomenon exists without actually providing any analysis or supporting calculations. I was remiss in this.
The reason that the stiffness is quoted by the manufacturer is so the engineer can design so that the resonances do not fall within the operating range of the product being built.
I have gone onto the New Way website and looked at the specification of their 25mm air bushing. I chose this model as I believe it is the one you currently use and because it is the largest, and stiffest, likely to be used for a tonearm. New Way luckily quote the pitch stiffness for this bearing as this is the mode most likely to be excited in our application.
It is 5.3 N-m/mil rad at a supply pressure of 0.41MPa, 60psi.
Rather more difficult to determine is what the moment of inertia of the carriage will be as the carriage is not a simple shape. Working on a simplified model of a carriage with a 75mm effective length and 150g mass, including cartridge, would give a moment of inertia of 280x10^-6kg.m^2.
A longer armtube would significantly increase this as would a heavier carriage.
With this arm the resonant frequency would be 690hz.
Changes to bearing size, supply pressure, carriage mass and effective length will all effect the resonant frequency. Unfortunately the resonant frequency will fall in the audio band.
From an octave or so below this the carriage will be effectively coupled to the rail. From around half an octave above this the carriage will be decoupled from the rail, this is my only issue with air bearings.
The resonant frequency itself should not be to problematic as the use of a silicone damping trough can damp the resonant peek that could otherwise cause a peak in output at this frequency. The air within the gap will also help in damping the peak.
All engineering is a balance of compromise.
My preferred bearing system is to use spikes (pin bearings) as these couple across the entire audio band. They do however have higher friction.
The million dollar question is does the benefit of coupling across the entire audio band outweigh the increase in lateral tracking error due to bearing friction?
As my bearing friction is low anyway resultant tracking errors will be very low, those due to eccentricity etc will be much higher and still present with either bearing type.
YMMV.
Of course there are those who think that an advantage of air bearings is that they DO decouple the carriage from the rail (they're wrong 😁)
Niffy
Last edited:
While I'm no fan of air bearing, I must point not all air bearing tonearms are the same just to be fair. Havun's air bearing tonearm is different from Jim's. Judging by his YouTube video, Havun's arm is the type with holes on a shaft floating a round plate carriage that holds the main arm, similar to the Dennesen tonearm, whereas Jim's arm is a bushing captured type that, I believe, is more rigid. There's probably debate which one is better but I just wanted to point out not all air bearing tonearms are created equally. What would happen if Havun uses bushing type air bearing instead and compare to his water floating arm?
Hi Niffy,
It is completely fine. I welcome all criticisms because criticisms help me improve my arms.
I highly doubt the resonant frequency is as high as 690 Hz. However, it is very simple to run a 1 kHz signal. If the resonant frequency is 690 Hz, we can see from the result. Stay tuned. If we can't see the 690 Hz resonant frequency, it means nothing to the design of the air-bearing arm.
Jim
I don't want to be too negative about Havun's creation. I was reluctant to say anything about his air-bearing arm. If you really want my opinion, I would not recommend his design. The half-sleeve is worse than the full-sleeve. I can't recommend the full-sleeve kind. It can't hold the air pressure. In addition to its structure, all parts seem to be made by himself. It makes things even worse. The compressed air can't be distributed evenly.
Please note that whenever I mention air bearing arms, I only talk about high pressure kinds, such as Walker audio proscenium black diamond, moving shaft, or Kuzma airline, moving bearing. I can't recommend all other kinds, so I don't talk about them.
Last edited:
The problem with air is that it is supplied under pressure, which means it makes noise and this is transmitted to the tonearm.
To hear this, it is enough to put the tone arm on a standing vinyl record without moving, turn on the compressor and turn the volume knob of the amplifier and at the same moment turn off the copressor.
This is the main reason why I abandoned the air arm.
Now on the water tonearm the sound is almost like from a laser disc, there is almost no noise in the pauses.
In general, air tonearms differ in that air is supplied to the tube with holes and then part of the air is wasted, otherwise the air is supplied directly to the tonearm bearing, and there are no holes in the tube and then the air is NOT wasted, but in this case the tonearm must be carried by a hose with air and in this case the noise from the air pressure is even greater.
In short, all this is good, but NOT for sound.
To hear this, it is enough to put the tone arm on a standing vinyl record without moving, turn on the compressor and turn the volume knob of the amplifier and at the same moment turn off the copressor.
This is the main reason why I abandoned the air arm.
Now on the water tonearm the sound is almost like from a laser disc, there is almost no noise in the pauses.
In general, air tonearms differ in that air is supplied to the tube with holes and then part of the air is wasted, otherwise the air is supplied directly to the tonearm bearing, and there are no holes in the tube and then the air is NOT wasted, but in this case the tonearm must be carried by a hose with air and in this case the noise from the air pressure is even greater.
In short, all this is good, but NOT for sound.
The diagram is very useful Super.
If yaw is a problem, it would be useful to space the wheels further apart.
This would mean a longer float and trough probably, but it would also accommodate a float with the waterplane area spread further from the arm wand, this will reduce roll (if we continue the boat motion analagy with the wand being the boat.) i believe this is much more likely to be significant in the design and i would expect significant improvements in the measurements and listening tests if this were done.
If yaw is a problem, it would be useful to space the wheels further apart.
This would mean a longer float and trough probably, but it would also accommodate a float with the waterplane area spread further from the arm wand, this will reduce roll (if we continue the boat motion analagy with the wand being the boat.) i believe this is much more likely to be significant in the design and i would expect significant improvements in the measurements and listening tests if this were done.
For your kind of air bearing, you can't use high-pressure compressed air, you may only use 6-7 psi or even lower. I believe anything higher than 10 psi will produce air noise. For my kind of air bearing arms, the bearing and the shaft must be precisely made to avoid air noise. So, for my kind of air bearing arms, you can't make shaft and bearing by yourself at all. In general, you should not make the shaft and bearing for any kinds of air bearing arms unless you have the capacity to make them.
Last edited:
Great to read open minded debate on these subjects with folk acknowledging that all things are a compromise, it's a great way to learn from experienced people and stimulate thought, many thanks to all, M
Compressed air absolutely does not care who made the tonearm and what color it is, it will make noise in all cases.
Sorry, I don't understand what you were saying.
Compressed air absolutely does not care who made the tonearm and what color it is, it will make noise in all cases.
If you were talking about the noise of air compressors, you are right. All air compressors make noise.
If you were talking about the noise from airflow with the arms, the situation may be much more complicated. The reasons to cause airflow noise are mainly structural, design, tolerance of air bearing so on.
The structure and design of all air tonearms is identical, compressed air needs to raise the tonearm above the air rail. As a result, the noise of the air hitting the tonearm part is formed, which propagates through the tonearm tube and is transmitted to the cartridge.That's it.
No. The designs of air-bearing tonearms are not identical. I suggest you need to study the designs of air-bearing arms. Kuzma airline is a good start. It is also the only air-bearing arm I recommend whole heartedly.
Hi Niffy,
I dug out a sweep of a 1 kHz signal I did a while ago for my 1" air-bearing arm. Here is the result. You can see the arm has a 5 Hz resonant frequency. From the sweep, I can't conclude that there is an additional resonant frequency of around 600 Hz. I am not questioning if your calculation is correct or not. I am saying it doesn't exist in the testing result, so, it is not my concern and it has no impact on the performance of my air-bearing arm.
Jim
I dug out a sweep of a 1 kHz signal I did a while ago for my 1" air-bearing arm. Here is the result. You can see the arm has a 5 Hz resonant frequency. From the sweep, I can't conclude that there is an additional resonant frequency of around 600 Hz. I am not questioning if your calculation is correct or not. I am saying it doesn't exist in the testing result, so, it is not my concern and it has no impact on the performance of my air-bearing arm.
Jim
Fans of the roller drive of the disk also say "No" to the rumble caused by this drive and modulated subsequently into a useful signal.
However, this does not cancel the physical process that is undoubtedly taking place.
In your version of the air tonearm there is another very negative point, this is a hose with air pressure that fits directly to the tonearm, this hose is nothing more than a spring and this spring has to be bent by the tonearm along its path.
Well, I don't want to expand this thread into another air-bearing arm thread. Please follow the link. Your criticism was already answered in the thread before. The process of building air-bearing arms was a learning process for me as well.
https://www.diyaudio.com/community/threads/diy-air-bearing-linear-arm.268614/
Hi Jim,
That looks like a pretty clean plot. I assume that it is just a single 1khz tone. Is the 1khz normalised to 0dB? The peek appears to only extend to -30dB. I know this type of software can truncate peeks that don't fall exactly on a sample point.
My concern is not that any resonance of the air gap will produce a peek in the output, the damping of the arm should suppress this, but that the carriage will be decoupled above this frequency. I don't know how well a single tone will highlight something that may be happening a half octave below it.
One of the best commercial air bearing arms is of course the Kuzma Airline. Kuzma also make the 4point conventional pivoted arm.
Both arms use basically the same arm base and VTA tower, the same armtube, the same headshell and similar counterweight systems. They are about as close as you could possibly make an air bearing and pivoted to each other.
On paper the Airline should be better. It has a shorter armtube so should have higher bending mode resonance, a symmetrically mounted headshell, better lateral effective mass and of course much lower lateral tracking errors. Yet the general consensus is that the 4point is the better sounding arm. I believe that the reason for this is that the 4point is much more rigidly coupled to its base via the four spikes from which it derives its name.
Compared to your arm the airline is very different. It only uses a 20mm bushing that is less than half as rigid. As it uses a much longer armtube and is heavier its lateral polar moment of inertia will be much higher. This will result in the resonant frequency of the air gap being much lower, probably in the 100-150hz range, so will only couple at very low frequency. The lower in frequency you go the more energy is transferred from the stylus to the arm so the more important coupling becomes.
As your arm will couple to a much higher frequency than the Airline most of its disadvantage will be mitigated. After all the resonant frequency of the air gap in your arm is still significantly higher than the bending modes of all but the best commercial arms and the bending modes of your arm are probably out of the audio band. I have no doubt that your arm will out perform either of the Kuzma products.
Niffy