Help with Crack off machine

[pugs]

So I have a job cutting 110 to 120 mm x 3mm Boro glass from Schott. I built this Crack off and have it working fairly well.

Could use some tips on how to be better with it, I'm a machinist by trade first time doing anything with glass. I have a new torch tip built with 3 flame holes. Running oxy-propane. Goes well enough could be a little faster maybe with 6 flame holes??

I've been running with the tip of one of the bright blue inner flames touching the glass, is that right way or should I just center the torch in the tube?

I also sometimes get a small axial Crack maybe 1/4" long on one end what causes this and how to avoid. End use is subject to high Temps and pretty sure the axial Crack will propegate longer and longer

Actual cut lengths are about 9" long

http://i839.photobucket.com/albums/zz317/JPMach/Work/20151002_190946_zpsdp2rdorg.jpg
http://i839.photobucket.com/albums/zz317/JPMach/Work/20151003_104342_zpswjp2yfgx.jpg

This was first torch head that only had one hole (several broken drill bits)
http://i839.photobucket.com/albums/zz317/JPMach/Work/20151002_190926_zpsgt8ulp3b.jpg

[Deez]

i believe most crack off machines use hydrogen

[pugs]

I also noticed that if I try to flame polish the cut ends I get lots and lots of axial cracks. Customer is fine with me just touching the corners with a diamond pad though.

[somewhere]

I crack off large tube on the lathe. A couple things I'll comment on but your milage may vary.

A sharp diamond or carbide point. A light but obvious score. I don't go all the way around the tube as I found that causes an uneven break. A single hole tip directly on the score for a count while spinning in the lathe. I don't use anything but a small single tip as I don't want to heat around the score with any blush from the flame. I count and once I find that number I use it as a guid to keep from over heating. I use a wet paper towel on the score if the count goes over and that usually starts the crack.

That is a excellent quality machine you built I wish you lived a little closer. I'd like to see it in person.

I'm sure you will get it dialed in but I was thinking someone operating a crack off machine would chime. Keep in mind on the lathe I'm heating from the outside on the score so your approach with three small burners may be exactly what you want. I do think the quality of the score dictates the quality of the break off.

[pugs]

Yes I have found that if I try and score deep it breaks eradically. Best seems to be a very light mark almost not even visible. I think I'm going to try a couple cuts with the torch centered in the tube and see how that goes.

[somewhere]

If you can find the time I would like to see a pic of the drive system.

[pugs]

I'll get one this afternoon for you. It's a vfd rated 240v motor with a variable frequency drive, chain drive to the shafts. It could stand a rebuild to have a jack shaft put in to gear it down more as it currently only needs about 5hz to run decent speed. And it has to plug into 240vac plug, not a problem for me but others would no doubt prefer 110vac.

[somewhere]

I'll get one this afternoon for you. It's a vfd rated 240v motor with a variable frequency drive, chain drive to the shafts. It could stand a rebuild to have a jack shaft put in to gear it down more as it currently only needs about 5hz to run decent speed. And it has to plug into 240vac plug, not a problem for me but others would no doubt prefer 110vac.

I've been finding treadmill motors with a simple diy vffd work well for some of these projects. We run auto turners for a lot of out work. Started using small stepper motors but had to rebuild for more power. Chain drives with spring tensioner seem to be the way to go but always interested to see how that is configured. Are you having problems running at slow speeds? The gear reduction sounds like a good solution but not without it's own problems considering you don't need much torque I would rather just slow the drive down.

[pugs]

Here's a pic of the drive. The motor is big as it is an inverter duty motor, but only 1/3hp. It mounted to a hinge and gravity tensions the chains.
http://i839.photobucket.com/albums/zz317/JPMach/Work/20151005_150925_zpscoj2yz47.jpg
The whole thing is bolted to one of my saw horses for now, may build a table or cart for it some day, but wanted to be able to tear it down and throw it on a shelf when not needed for awhile

[pugs]

cutting a few test pcs again today with the torch centered in the tube. Same as before I get some that crack nice straight all the way around, and some crack with a 1/8" or so bump or jog in it. Done everything I can to get torch flames all hitting the same spot on the scratch.

Any ideas on what I should be doing to get a flat consistent cut? small bumps are ok and some waviness is fine but sharp jogs that are stress risers wont as the high operating temp will more than likely cause a crack.

[Riley]

Cutting these tubes for propane radiant heaters by chance? Did you switch to hydrogen? All the ones I've seen in pics and read about are hydrogen. Only other variation in torch placement I've seen have the torch running down the inside of the tube with a sparker to ignite the flame. Probably a bit more sketch to have gases running inside the tube but there must be something to it if people have gone that route.

Without actual experience on these that's as much as I can offer. If I ever chuck a tube on the lathe and score it then try a sharp propane flame I feel like a way hotter more abrupt flame of a hydrogen torch would actually provide the necessary thermal shock. Without a crack off and no desire to check full tubes on the lathe I tend to just wet saw a deeper groove 1/4 the way around a tube and using a hot rod you can thermal shock a decent cut.

[aREa541]

A light score, flame as hot but precise as you can manage with no splash around the score. Too large of a heat signature will give you issues. Let it turn with the flame on the score for a little bit (amount of time will vary with size of tubing and flame) then use something to apply a drop of water or two to the score. It should crack off perfectly when the water is applied if the heat and score are correct. I found that water can be the difference maker in getting good results.

Really what you want is a rapid severe shock in as thin an area as possible that is directed around the tube by the score. If you heat up too large of an area or score too deep or long it can spread the stress more than you want making chipped ends. The water helps with the severity of the shock without overheating or overscoring. Like mentioned, hydrogen is the preferred gas and is what is on the machine in the shop.

The pic of your flames make them look rather large. I would go for more ports smaller in size which will make them more precise. As small as you can get the ports while still maintaining the candle length needed at a good intensity will most likely be ideal for you.

Great build on that machine though! I am sure you will get it dialed in with some trial and error.

[pugs]

Thanks for the tips, will try with some water.

I'm going to try and build a new torch tip that uses mig welding tips for the orifices as I can get those in 0.023", 0.025", 0.030", and 0.035" sizes and I think I have plety of room to use them with the large diameter tubing I'm working with and that might help to get the smaller flames and still have them out at about right diameter to work. Saves from having to drill the tiny little holes, maybe it will work, maybe it wont but should be worth a try. Will do that with 6 nozzles as well.

Someone asked me about having to anneal these after cutting, do you guys recommend it? I didn't think with the quick thermal shock break that I put enough heat in to the glass to where it would require annealing. I do have a heat treat furnace for my metal stuff that I could do it in, just not sure if its needed to help prevent small cracks from forming or not. And if it is required what temp and how long would it require? My furnace is capable of 2,300 F and does have a ramp up/ down controller on it.

[Deez]

Hydrogen

[pugs]

I am looking into that too, was hoping I could get the propane to work atleast for awhile before spending more money on another bottle or 2 and regulators, torch body etc.

[Groundjoints.com]

Hydrogen, and make sure that your scorer is super sharp and you aren't overscoring. A flame slightly bigger then the diameter of the tube. And take your time. Is the glass being supported as it cracks off? Sometimes the weight of the cutting process will cause a tooth.

[vetropod]

Interesting thread!!!

Depending on the application these rings are being used for, annealing may be necessary - residual stress can definitely cause cracks to travel.

You can make a super-cheap polariscope with a pair of polarized sunglasses and an iPad. Turn the screen of the iPad white (such as a blank web page), hold your clear glass over the iPad and look at it through the sunglasses. As you turn/rotate the glass, stressed areas of the glass will appear colorful, link an oil sheen on a puddle of water.

There's lots of info on this forum about annealing, but if you need to do so the annealing temperature for borosilicate is 1050F, holding time varies by size/thickness of the glass, and more critical is ramp-down time to the stress point (950F.) Here's a little info about annealing: https://mikeaurelius.wordpress.com/2008/01/30/garaging-and-annealing-two-different-aspects-of-the-same-beast/

[pugs]

Thanks that helps a lot. Haven't got much time on it for a bit.

The short little test rings I'm cutting didn't have support the longer production lengths do. I'll watch that a bit as well

[nicko0]

i read some where to use a spray bottle with a needle in the tip to apply the water directly and only to the score.

the unevenness may b arising from the inconsistencies in the tube. be it uneven stress and or wall thickness.,, both from the manufacturing process.

im just speculating..... ive only played with my crack off a bit.