Laser cutting masters

[circitfied]

Hi all,

I am a part time PhD student at Manchester University. I have just had my first paper accepted by the Journal of the Audio Engineering Society on how to produce a stereo record using a laser. This process has been protected in the form of a priority filed patent by the University of Manchester Intellectual Property (UMIP).

I have successfully managed to create a stereo recording in a single pass, without digitisation and sampling of the soundwave.Thereby eliminating sampling and translation of the laser across the disc surface in discrete steps. It's still analog!

My next piece of research is adaptive optics to produce audio recordings, with the aim to increase recording speed and stereo separation. So far the mathematics holds up and the bench tests work. This week, a high powered laser will be used. I aim to have a paper on this completed by the end of April.

This research can also be applied to retro-fitting a standard lathe to produce metal masters, or to produce polymer discs. Obviously cutting vinyl discs with a laser is a no go as free radicals and vinyl chlorides are horrible, as well as the vulcanisation of the PVC. However, there are a wide variety of polymers that this technology can be applied to. With polymers and metals, it all a case of matching the material to the laser, whether that is pulse length, or wavelength. Polymers tend to need short pulse length and short wavelength and these tend to be on the expensive side. However, metals (depending on their reflectivity) can be machined with longer wavelengths, which makes it much cheaper to set up.

My University are ready to license this technology, so if anyone is interested hit me up and I'll pass on your details to UMIP. Or if you have any questions, please ask away!

[symatic]

this sounds really cool - do the wavelengths your working with match the same sort of scale that a cutting stylus moves within?

apparently cutting a silent groove means moving the stylus less than 5 nanomillemeters. could these laser cut grooves achieve that level of detail?

would love to see some of this in action if you ever do an open day or something

[misjah]

Isn’t that exactly what “hd vinyl” does?

[circitfied]

Isn’t that exactly what “hd vinyl” does?

HD vinyl digitises the waveform and produce a topology of the groove. According to their latest patent they will use multiple passes to remove the material, leaving the ridges behind. I don't know how repeatable it is having overlapping passes. However, I have produced stepped notches before using multiple passes.

My process will only make grooves, in a single pass, all analog, much like a conventional cutting head.

[circitfied]

"do the wavelengths your working with match the same sort of scale that a cutting stylus moves within?"

Wavelength plays a part on the minimum spot size of the laser. So it doesn't have much relevance to minimum feature size.

[markrob]

Hi,

This sounds very interesting. I was confused by your original post. You indicate that you have made a stereo recording, but later on state that you want to make audio recordings. What did you record? Can you post a clip of the playback or some photo's of the groove(s) you cut? Do you think you will be able to achieve the same or better frequency response and dynamic range as the current state of the art? Is a copy of the patent available for download?

Regards,

Mark

[Greg Reierson]

Isn’t that exactly what “hd vinyl” does?

Let's just say that's what they want to do, some day.

[dmills]

Patent numbers man!
Once we have the numbers we can look up the patent and get the gory details, which is pretty much the point of a patent.

Regards, Dan.

[circitfied]

Hi,
To clarify. This is a priority filed patent. This means it has been filed. However, it is in lockdown for a year, thereby giving me protection on my intellectual property until we have a buyer.

As for audio, or soundwave produced. I have produced a stereo recording. I can play it on a standard turntable. By no means is it pretty. This is because I built the system from a 30 year old laser and salvaged parts from scrapped lasers.

However, I now have access to a high power, UV, short pulse laser. This means I will be able to produce superior cuts from before. The repetition rate of my laser is 400KHz, so pulse overlap will not produce harmonic distortion.

As for producing a record with the same bandwidth. I think I may be able to increase it a little. Today I am testing a new optical beam line with the aim of achieving this.

Of course my published results will be in JAES in a couple of months hopefully. Until then I can't give you my results, otherwise I won't be published.

[circitfied]

Isn’t that exactly what “hd vinyl” does?

Let's just say that's what they want to do, some day.

Precisely.

I already have results and it has been peer reviewed.

[misjah]

looking forward hearing the results!

[aaron]

Wasn't Souri doing something with a laser cutter a few years ago...?

[circitfied]

Wasn't Souri doing something with a laser cutter a few years ago...?

Souri was indeed doing so. I think groove geometry was a stumbling block. Well it was for me.

After lots of pondering and then being introduced to new laser technology I've solved that too. I'm in the process of planning a paper on groove geometry and how it affects playback.

[Scott_HUll]

Groove Geometry does seem like it would be Very tricky. You have to remove more material in the base of the groove and progressively less towards the top. You said this is a continuous analog process - but is it real-time? As with any new tech - it's hard to imagine it overcomes previous obstacles... but that the whole point right?

[circitfied]

Hi Scott,

You are correct, a Gaussian beam does indeed produce a "U" shaped beam rather than a "V" because of the power intensity distribution. My research this summer is to fix this problem. I have some solid ideas and if I have the time will be working on it.

Yes the process is analog and is in real time. I'm not currently recording in real time because of the 25 year old laser I'm using. However, my new experimental set up will solve this problem. I'm hoping it will at the very least produce a recording in real time. Hopefully, I can get this set up running this week. I have aligned the beam and bought material. Fingers crossed it is as good as I have mathematically calculated!

[misjah]

At this point in time I think it would be most interesting if it can be used on material other than lacquers.

[dmills]

I could maybe see some tricks with changing the frequency of a Q switched or mode locked laser that might work, but you would be needing a very high PRF, more in the mode locker then Q switch territory, and mode locking usually means a gas tube of some sort or a long cavity YAG, difficult to do either with a internal cavity diode and ECDL is getting expensive or fiddly.

Pulsed sources are good as they ablate rather then melt, so the chip problem goes away in a literal puff of smoke.

Could an alternative groove geometry work? What I am thinking is that a spherical stylus will track (on the surface) given a U shaped groove smaller then the stylus radius, obviously a nightmare for surface noise, but what would happen if you cut a groove wider then the relevant contact points with a narrower one inside it that defined the stylus contact points?

By stacking two U shaped grooves of different sizes you would get the M component in the usual way, and the S component by varying the depth of the larger cut (Power or focal point).

Code: Select all

------|         |------ Disk surface
      --|    |-- Depth of this step defines the L-R component.
         --- Stylus runs on top corners of this inner groove (Depth does not much matter here as long as it it deeper then the maximum L-R depth of cut).

A Blue ray DVD writer head seems like it has most of the required drive coils, photocells and optics, might need an extra polarisation beam splitter for the second beam (used to cut the outer groove to give a better contact surface), but the doings are cheap.

I would suggest that hitting a nitrocellulose coated disk with a Q switched laser is possibly a poor plan, pick another substrate.

[CutR]

Some guy who messes about with lasers had a try at embossing but with very poor results. Granted he's not very versed in the cutting game but his attempts were entertaining and an eye opener for a noob like me. Looking forward to seeing your work sir.

[circitfied]

DMills

Q switching a laser is a messy business. It works fairly well if you require a set frequency. However, if you start changing the frequency then the pulse energy changes. Making it challenging to regulate power accurately.

Another option most modern lasers use is PWM control. This equally upsetting as you end up releasing bursts of pulses. Obviously when you are trying to machine a groove will result in irregular groove geometry. Especially when the angular speed is 500mm per second or above.

Luckily I am about to have access to a laser with a rep rate of several hundred KHz.

As for groove geometry. I like your idea of two beams co-axially aligned. It would be an absolute bugger to align! I think I have a more efficient option.

Material wise I do have some Apollo lacquers they were kind enough to give me. I'm loathe to use them as they will probably catch fire, much like the factory. I'm looking to process stainless steel in the coming months. I'm also going to have a go at making a stamper. Not sure if anyone would be keen to try a stainless steel stamper though.

[circitfied]

Oh and good news my JAES paper has been fully accepted. Finally. I didn't realise it would take so long. They now need some suitable author photos and some quick bios.

Tomorrow I will carry on with my literature review for my second paper. Hopefully the turn around will be much quicker this time. I have too many ideas!