Comment by numlocked | Hacker News Reader

Comment by numlocked | Hacker News Reader

At work, we give Cliff’s Klein bottles as the quarterly award for engineering excellence. When I ordered the first batch, I called the number on the Acme Klein bottles website to ask whether they could be engraved (the site states the CLIFF won’t engrave the bottles but I wanted clarification if ANYONE would engrave them). It rang and went to voicemail. I started leaving a message with my question when suddenly there was a lot of noise as the phone was picked up (apparently it was a real answering machine) by someone shouting “don’t hang up! Don’t hang up!”. It was Cliff. I was briefly star-struck, told him what an influence The Cuckoo’s Egg had on me when I was younger, and then received a 15-minute oral history of engraving methods and why none of them would work on his Klein bottles. An absolute highlight of my year, and I think they’ve been well received as engineering awards...even if we have to use a label-maker to add the recipient’s name :)

Did he say why laser engraving wouldn't work? I can see the curves causing some distortion but you should be able to compensate for that.

Edit: He answers the question in his FAQ

> The borosilicate glass laughs at laser beams

..which makes sense. Seems like sand blasting would be the best way.

In short - engraving borosilicate is difficult, because it does not expand very much when heated. So when hit by an infrared laser beam, the glass doesn't locally expand and spall off a tiny speck of glass .... instead, the glass locally melts. When the beam moves on to the next pixel, the first site either solidifies or sometimes dribbles away. Laser engraving of borosilicate just doesn't work well. And, because of the compound curves of the Klein bottle, you need a big depth of field as well as a rotary bottle holder. All laser engravers that I've played with have a very short (10mm or so) depth of field. Diamond engraving doesn't work well because of the curves. Sandblasting with a rubber mask works, but making the mask is a pain. So, after lots of attempts, I've given up on engraving these guys. Also: mistakes are expensive -- this is hand-blown glass.

(Everyone should make sure to note the name on the above comment)

Thanks for answering, and congratulations on your book's anniversary.

My job gives out copies of your book as a somewhat prestigious award to those who find and report security problems, and I'm very happy to have both received and read a copy.

This is basically exactly what you told me on the phone :) We also had a fun digression about sand-blasting tombstones. And of course, we covered the hand-engraving/dremel option.

So...label maker.

Can't acid be used to etch glass?

Also, why not use glass that's easier to etch?

With quartz glass, it's going to be hydrofluoric acid and I invite you to read up about its toxicity and lethality (user stories like: splashed a few ml on a leg, did all the emergency procedures exactly right, got quick emergency care, leg amputated, died anyways).

S04E20 of ER had a patient die due to contact exposure to hydrofluroic acid, I had to look it up after reading this comment to make sure I got it right - don’t know why but it’s one of the most memorable to me.

Seriously, don’t mess around with this shit - even IF you notice exposure (hard, because you may not even notice until it’s too late) there’s a good chance it’s already a done deal.

I'm glad that I don't have to work with the stuff after reading up on emergency medical procedures. I once found a very long list of what to do by an expert MD and it's really not pretty.

I'm not an expert in either chemistry or medicine but what I gathered was:

The medical team needs to try and offset what the fluor is doing to the balance of ions in your body, to keep your cells alive. HF will diffuse within pretty much all of your tissues (including diffusing into bone like it was a sponge iirc) and a medical team will check your blood ion levels every five minutes or so while sticking you with CaCl (iirc) injections along the path the HF may diffuse through your body (if you're lucky enough to have hit an extremety like a hand). While the HF is diffusing, you will not feel anything. They cannot just pump you full of the proper ions to counteract the effects of the fluor and maybe bind it because that would kill you as well, they have to hit the right elevated concentration locally to offset what the fluor is doing.

Edit: I've read up on it again and some of the things I wrote aren't right and I misremembered, like HF exposure being painless. At first it may be so but once the calcium levels drop, exposure is connected to extreme, "deep" pain. Treatment guides stress to not admister aneasthetics because the pain is the best indicator for successful treatment. Also, Calcium Gluconate is recommended, not Calcium Chloride.

This [0] is a detailed source I found (PDF).

[0] (PDF) https://ehs.unc.edu/files/2015/09/hfaexposure.pdf

Hydrofluoric acid is also what they use to dispose of bodies in Breaking Bad, although they did exagerate the effectiveness.

Yeah. A digester tank would probably be more effective... essentially a "hot tank" like used to be used to clean engine blocks for rebuilding, except for dissolving organics.

They're used at a local University for disposing of cattle remains from the veterinary school... takes several hours, but then the waste liquid is diluted and sent down the drain.

Why not bury or cremate?

There is a "chemicals I won't work with" on HF

Technically "...won't touch" because he did/does work with HF.

https://blogs.sciencemag.org/pipeline/archives/2004/03/03/th...

> the only times I even use the solution forms of the reagent are on a very small scale and in weakened form

HF is also used to wash cars by people wearing no protection, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5779585/

E.g. "As a case example, one worker (case 1) splashed his left leg while transferring a cleaning solution of HF and sulfuric acid between containers. He did not irrigate the area and continued to work for approximately 1.5 hours with soaked pants and shoe until he developed an uncomfortable burning sensation" ...

HF diluted to <1% is used to wash cars, and only idiots don't wear skin protection while doing that.

To reliably etch glass, you need concentrations of 10% or preferably more, at which point you really want a PVC apron, a face shield and a heavy-duty fume hood to work with it, to avoid the sort of exposure in [1].

[1] https://www.uvm.edu/cosmolab/safety_manuals/Fatality%20due%2...

"...Upon evaluation, the patient was reported to have a quarter-sized brown necrotic area on the anterior left ankle and burn to the anterior left lower leg. Emergency medical technicians irrigated the area with calcium gluconate and transported him to a burn unit, where he received a calcium gluconate injection. He sustained a small area of full-thickness skin loss requiring excision and debridement with a skin graft. The worker received outpatient burn therapy and returned to part-time work 6 weeks after the injury. A foot paresthesia developed, and the worker received a permanent partial disability payment."

I'm going with "things I won't work with", Alex.

HF etching cream for glass is sold in Home Depot (really H2SO4 + NaF powder)

The only tricky part of working with it is getting clean edges with a stencil.

Could you combine laser engraving with a jet of air, to blow away the molten glass?

AFAIK basically all laser engraving/cutting is air assisted already. Turns out smoke is really good at getting in the way of lasers, and air is a really good way to get rid of it. That said I don't know if it'd be a firm enough blast of air to move any significant amount of molten glass.

I've used my laser cutter to cut a mask for acid etching when the laser itself couldn't etch the material needed.

How thick is the rubber you use for sandblasting? Could it be laser cut to reduce the pain of making the mask? It sounds like you've exhausted all workable options but my curiosity is piqued.

I currently work in the laser engraving industry, and I can confirm everything you've said to be true. However, I wanted to let you know about some extremely cool technology that I had the opportunity to experiment with a year ago or so. Some companies have developed optics with adaptive focus that use liquid lenses to mark beyond a 10mm depth-of-field - see https://www.optotune.com/technology/focus-tunable-lenses for an example, and https://youtu.be/ss-tkXcClMk for a video of the system in action.

I still don't think it would work well for your purposes - borosilicate just doesn't mark well, and you'd need to start with an accurate 3D model of the object - but I hope you enjoy the video!

(And thanks for writing the book!)

Sand blasting would work, although less well than on other glass types. Acid etching would work, although the chemicals needed are dangerous and it's apparently quite a finicky process to get a good result - over etching results in a less visible etched pattern than optimal exposure to the acid.

To laser etch borosilicate you'd have to use an expensive laser, either an abnormally high powered IR laser or something higher frequency like a UV laser. It wouldn't be enough to just melt or heat stress bits off the surface, you'd have to micro structure (microscopic 3d carve) it to form something that would reflect light well, otherwise you wouldn't be able to see the etched pattern except in certain light.

On top of that the depth of field issues for a laser apply as well as the incident angle for the beam - you'd want something close to a surface normal angle to get a consistent, controllable result, so you'd have to use a 5 axis machining setup to make sure that happens. A 5 axis precision UV laser would probably cost more to run than the bottles are worth, despite their quality.

What I'd probably do if I had to label bottles like these with names is make a small soda glass plaque or tag with a mold and possibly some color tint for contrast, then etch it using an appropriate method (or maybe even mold the name in) and affix it to the bottle either by heating and slumping it into place or molding it to a matching curve and then using optical adhesive to affix it.

Or, just make a nice display box for it and put the name on a brass plaque on the front :)

A lot of people are trying to solve for etching and engraving here, but if I wanted to personalize a glass piece that already (kind of, in some world) looks like it belongs in a lab, I think I’d try an enamel paint process instead.

May have a nicer result than a label maker, depending on taste.

Reply on news.ycombinator.com