Cheap galvo project idea

[MechEng3]

I had some thoughts today of the mechanical workings of a hard drive reading arm assembly. I have played around with the bits I have here and wondered if it would be far fetched to make a galvo of sorts from a pair of these guys. Seem to have all the driving mechanism and electronics in the unit. Seems it would be simple enough to figure out a way to signal these to move in a controlled manner. Maybe just a bad idea....but then again???? I have some machining chores to complete for Spec and after that I would be able to put some time and parts towards this. Any ideas or coments? The biggie here is cheap and readily available parts.

[allthatwhichis]

I've seen this used once. But it was crude and it was more of an osilliscope idea. They put a mirror on the reader arm for an x movement and a speaker for the y movement. I Can't seem to find the link though...

[buffo]

I agree with Aaron - I've seen this used before. If I remember correctly, they actually worked fairly well. Something like 10 kpps speeds, though that was a long time ago. Perhaps the stepper motors for hard drive heads are faster these days. (Google should help you find it...) Still, it's an interesting project idea...

Edit:
A quick google search didn't turn up as much as I thought it would. I found these two sites, but neither one is the one I had seen previously:

http://spt06.chez-alice.fr/00/scan1.htm
http://sean1983.awardspace.co.uk/?laserharddrive

I may have confused the hard drive project with this guy's website where he shows you how he built his own galvos from scratch. He did get pretty decent speeds out of them, so maybe that's what I was thinking of, and not the hard drive galvos...

[steve-o]

I took a small triangular piece of neodymium magnet, epoxied it to a piece of antenna tube (very light-weight) and put it close to a low Z coil (probably 15 turns or so) put the coil in water to keep it cool and applied 12V @ 60 Hz from a hefty transformer and the thing about jumped out of my hand! (right before the transformer started smoking)
It was oscillating at 60 Hz pretty good--anyone know what that would translate to in points? (Probably 5K or less?? )

[Zoof]

I took a small triangular piece of neodymium magnet, epoxied it to a piece of antenna tube (very light-weight) and put it close to a low Z coil (probably 15 turns or so) put the coil in water to keep it cool and applied 12V @ 60 Hz from a hefty transformer and the thing about jumped out of my hand! (right before the transformer started smoking)
It was oscillating at 60 Hz pretty good--anyone know what that would translate to in points? (Probably 5K or less?? )

<technical input>

Any spring mass damper system will oscillate at frequence of the input (if that is single frequency like 60Hz). The parameter you want to know is the phase shift, thus how many degress the sinoid waveform of the output has been shifted from the input. For scanning: the lower the better and should be near 0 degrees. At the resonance frequency the phase shift will be 90 degrees and beyond near 180 degrees. Scanning can only be done well below the resonance frequency of the system. To increas it: lower the inertia of moving parts and increase spring stiffness.
At the resonance frequency the amplitude of the output is maximum to the same input.

</ technical input>

[steve-o]

Zoof,
Thanks for the input. As we speak I am dissecting a speaker to try a "voicecoil galvo" I bet the resonant freq of this mother (without the mass of the cone) will be about 10 khz
----------------------------
Am I crazy? Yes.
Am I Sane? Well I guess that depends on your definition of..
(not you Zoof I'm just trying out a new signature )

[The_Doctor]

For scanning: the lower the better and should be near 0 degrees. At the resonance frequency the phase shift will be 90 degrees and beyond near 180 degrees. Scanning can only be done well below the resonance frequency of the system. To increas it: lower the inertia of moving parts and increase spring stiffness.

Good info. One thing I've wondered about is adding extra torque and power, at the expense of taking up space. Is it feasible to use those small 'bookshelf' loudspeaker bass drivers? I once considered using that and a carbon fibre drive shaft epoxied to the coil former, in a closed loop controlled by a variable capacitor made from ultra-thin fibreglass PCB fitted to the mirror axle. I gave it up at design stage because I couldn't think of a way to stop it overshooting and being generally disgracefully clumsy. I still think that the idea of ducting in extreme torque and power to the mirrors on a small axle from a larger motor is a good one though.

Nature might first suggest not, as it's the small creatures like ants and crickets that have fast controlled responses and extreme power to weight ratios, but on the other hand, whales can create well controlled high frequencies with powers that can cross oceans, so the problem seems to be one of efficient and precise coupling.


Edit: I see Steve-o got there a fraction before me. Speakers do seem like a good cheap source of very good torque though, and if well coupled, can surely make 30kpps? Newer horn-loaded tweeter drivers might be especially interesting. If a tad huge.

[Zoof]

Usually the resonance frequency of loudspeakers is around 100Hz (only). This is deliberately below the audio frequency range and the output will have the 180 degree phase shift which is not a problem for audio. When using speakers for positioning as in scanning, the phase shift becomes a problem.
With closed loop feedback the scanning freq can be pushed beyond the speakers natural freq - I haven't seen closed loop speaker scanners anywhere yet so that will be a novelty Going that direction the current in the coils will also be more than usual so anyone trying that will want some heavy duty speakers. (thinking of car stereo systems...)

[The_Doctor]

Speakers would have a higher frequency if they had the air-coupled cone removed, and horn drive units already might have a very high resonant frequency.

I'm not really interested in using speakers though, they have moving coils, I'm interested in moving magnets, that way the coils are more easily cooled.

My main point is I'm wondering if driver size is always a limiting factor on speed. I still think that it should be possible to make a larger motor that couples to a small axle and mirror, at the expense of extra noise and heat, possibly, but still be faster. As far as I can tell, CT scanners already do something like it, the CT6810 seems to do it by extending the motor while keeping the diameter small, but maybe there are other geometric arrangements that can get faster speeds from larger volume motors.

[steve-o]

Usually the resonance frequency of loudspeakers is around 100Hz (only). This is deliberately below the audio frequency range and the output will have the 180 degree phase shift which is not a problem for audio. When using speakers for positioning as in scanning, the phase shift becomes a problem.

The resonant frequency of audio speakers varies with size (as well as other factors. A 15" woofer will have a lower resonance frequency than, say, a 5" midrange. When removing the paper part (the cone) leaving only the spider and the coil the resonant frequency will also change.

Phase shift (current lag) is present in all inductors (including speaker voice-coils) I'm not sure you mean about the 180 degree shift. In an inductor the voltage will lead the current by 90 degrees if I remember my college days correctly (its been 27 years )

--and yes. It'll be a trade-off: bigger size, more bulk for cheaper price and higher speeds.