Why so much cutter head power?

[Viewmaster]

Hello, I'm a new kid on this block.
I am building a cutter head to cut Edison wax cylinders and progressing reasonably well using a 4 inch loudspeaker as the drive for the stylus.

Have studied many interesting threads on this forum and also many web sites about cutter heads, I wonder if someone can answer, please, a big puzzle that I cannot find the answer to......

Professional cutter heads take big wattage inputs to the coils, up to 500 watts I have read !
Now the stylus is linked directly to the drive and not through any mechnical leverage. So why does the stylus still only move that tiny amount with hundreds of watts going into the driving coils? Is the stylus suspension very, very stiff and so requires high power to move it?

My loudspeaker drives my stylus way too much with just a tiny fraction of this power, so why do these pro cutters require all this power even though some degree of negative feedback is used?
Thanks for any replies to explain this puzzle.
Albert.

[cuttercollector]

You wrote (in part)
"why do these pro cutters require all this power even though some degree of negative feedback is used? "

I am not an expert but I would say that the pro cutters require all this power
precisely BECAUSE negative feedback is used.

You are correct, to put some sound in a groove requires only fractions of a watt. After all, the cylinders you are experimenting with were recorded acoustically using just the concentration of sound from a horn to a small diaphragm connected to the cutting stylus to cut the groove.

Most non-feedback cutters from the 40s-50s will probably take only about 40 to 50 watts peak program before burning out. The long term average is much lower.

The issue is trying to produce linear flat frequency response at high cutting levels. Early cutters, even pro ones before the feedback era, used various
mechanical damping schemes to prevent resonances and make the response somewhat linear, removing big resonant peaks etc.
This is not to be confused with record/playback equalization such as the RIAA curve, though the two do inter-relate.

So, negative feedback was developed, as it had been in amplifiers, as an electrical way to damp unwanted resonances and peaks in the frequency response, and lower distortion.
They did it as a servo-feedback system wherein the feedback coil generates a voltage from the movement of the main driven coil and stylus assembly. This voltage is applied to the beginning of the amplification stages as an out of phase "correction signal" that lowers system gain as the distortion produced by the moving mass of the cutter coils, and thus is in the waveform generated by the feedback coil, is applied to the whole system and thus forces it to more accurately reproduce the exact audio analog waveform and cut that into the groove.

As the demand to cut at louder and louder levels on the disc developed not to mention the need for 2 of everything to cut stereo, the coils and magnets got bigger and more heavy duty. Despite attempts to lower mass and make things more electro-mechanically efficient, they became heavier

Now, if you want more mass to instantly "be here now" so as to record a low-distortion waveform at something near 20Khz. at a high level, well, you have to pump a lot of energy into the system to make it move precisely as you want at the right time to replicate the wave form. The more mass moving at a greater level (louder) at a higher frequency, with lower distortion (better, more Hi-Fi sound) starts to require even hundreds if not thousands of watts to control and move the mass of the coils to make the stylus move exactly precisely in a low-distortion replication of the desired analog audio waveform. Most of it is dissipated as waste heat controlling the movement of the mass rather than as actual cutting work which still requires comparatively little power.

It is pretty much a "brute force" approach.

[Viewmaster]

Thanks very much, cuttercollector, for your reply and taking the time to write all that explanation.
Amazing to me that 500 watts (or more!) only moves a stylus such a small amount just because of neg feedback, but as you say, moving mass fast requires power.
I am trying too to get more power into my loudspeaker voice coil driver by using 20/1 mechanical leverage, but this does introduce more material that will resonate. Obviously neg feedback is the way to go.
Anyway, thanks again.
Albert.

[cuttercollector]

I personally think that given modern materials science and computer aided design, it might NOT be the way to go. (Negative feedback) It is more of a correction for a problem than an elegant solution. I have long wondered if it might be possible to produce a much more efficient design that is less power hungry. Some of the older lower quality home cutters used crystal (ceramic) cutter elements and were able to cut to decent volume levels at 78 RPM up to 7-8Khz. This with only 1 or 2 watts of available undistorted amplifier power.
I wonder if you could make a modern stereo cutter using piezo elements and produce a uint with low mass thus moving the resonance up out of the desired pass band or at least pushing it high enough that it would aid a falling response curve at the high end. I think this would be somewhat self-equalizing to the RIAA curve due to the impedance curve of these devices and the fact that piezo elements are more displacement sensitive than velocity sensitive. I just don't know if you could still get wide and flat enough response. I don't think level would be an issue. But the moving mass would be incredibly low. In fact you would probably need to add mass to the whole head assembly for something to react against as it cut. And it would only take a few watts as all the energy would not need to be absorbed in elaborate damping and/or feedback schemes.

[cuttercollector]

Flo, (or others) did you ever attempt to make a cutter with piezo elements experimentally?

[vinnymeyer]

Hi Folks,

The high power isn't required because of the negative feedback. High GAIN is required for this, but not high POWER. The high power is necessary because you have to accelerate the mass of the coil, coupler, torque tube, and stylus through the resistance of the cutting medium, and at high frequencies this takes a LOT of energy. Add to that the high-end rise of the RIAA curve, and that output power is logarithmic not linear, and the numbers get big fast.

Bear in mind that while the amplifiers have to be able to sink 500 watts or so RMS, they RARELY have to do that. Mostly this is there to handle peaks cleanly.

Hope this helps,


Vincent Meyer

[Viewmaster]

The high power is necessary because you have to accelerate the mass of the coil, coupler, torque tube, and stylus through the resistance of the cutting medium, and at high frequencies this takes a LOT of energy.
Vincent Meyer

I confess that I have never actually seen any good cutter heads but that torque tube you mention is shown on a photo I have of a Haeco head. I believe Westrex etc may have them too.

These torque tubes, which hold the stylus, look to be quite heavy brass and I wonder why they are needed as this must add greatly to the power required in order to move them at high frequencies.
Albert.

[vinnymeyer]

No, the torque tubes, while they look heavy, really don't have that much mass. They translate the linear motion of the transducer for a channel to the rotary motion on the shank of the cutting stylus, to translate THAT to the modulation of the corresponding groove wall.

The difference between the power required to move a stylus in free air verses the power required to move it through lacquer is similar to the difference between walking and walking through water in a swimming pool - there's more resistance, so it takes more power.

This can be seen in early cutting systems before heated styluses where the EQ changed with the radius of where you were cutting, to compensate for the energy required to move through the lacquer at different linear speeds.

Vincent Meyer