First broken bit!

[coryatjohn]

Ok, lots of people have posted a thread like this over the years. I didn't see the same issue I had so I'm asking for advice...

I was at the end of cutting a 1" thick piece of MDF (about 10' total cut length) done in 6 passes. The last pass was about 1/10" and it was on the last leg, about 12" to go and the bit broke. I touched the bit moments after it broke and it was cool to the touch. The chips looked good and it sounded like it was cutting normally before it broke. I'm ashamed to say that I finished the cut with (of all things) a hand saw.

http://www.usnaviguide.com/shopbot/broken-bit-onsrud-52-910.png

I used the default settings for speed and RPM.

I've read that bits are broken in three's so I'm asking for advice before I break the other two. I am unfamiliar with MDF. Did I do something fundamentally wrong? Should I adjust RPM, speed or cutting depth when using MDF?

Thanks.

[steve_g]

Good morning John...
Can you tell me the IPS and the RPM . Did you calculate the chipload when you created the tool path?
SG

[MogulTx]

And tell us if it snapped inside the collet or AT the collet. It is broken very high up. It should be seated farther than that into the collet. If it broke inside the collet, two potential problems are: it may be the wrong size collet or the collet may be bad. ( Is it an ER25 collet on a spindle or a PC Router ?)

MGM

[jhedlund58]

Did that break inside collet? Perhaps not in collet far enuff... and i broke one early on using default settings was told i was going too slow. two things i do... one- i cut half depth of bit not the recommended full ... so .25 bit i go .125 per pass... second... i trust the chip load calculator and my tool db settings which have been tailored to my machine. I have a standard and spindle... many have more powerfull machines etc. my approach may not be correct. Works for me

hope this helps

[coryatjohn]

The bit broke right at the collet.

RPM: 14,000
Feed rate: 6 IPS

Spindle: HSD 4hp.
Collet: ER 25 1/4" (from SB)
Bit: Onsrud 52-901 (from SB)

Chip load calculator? Haven't gotten that far... Maybe I need some edukaton.

Should I just toss this in the trash or can I use it as a short bit? I've done that with the hand router without any trouble.

[steve_g]

Well that's what I thought... much to fast IPS back off to around 2-3.

Chip load calculator is found under tools in SB3

SG

[bleeth]

You should be able to cut through MDF much deeper per pass with that bit. Although Steve is right when asking for more info (feed and rotation speeds), assuming that the break happened below the collet, I would say you didn't have enough of the bit in the collet.
You should always have the bit inserted the full depth of the collet, but never so deep that the bottom of the collet is over any cutting area of the bit.

Assuming your machine is an Alpha feed and speed are not unrealistic, although I would spin somewhat slower. You can also cut deeper. That bit will handle a 3/8 plus pass in MDF. If you have a standard, you are going too fast at 6ips.

[steve_g]

http://www.onsrud.com/files/pdf/2012%20LMT%20Onsrud%20Production%20Cutting%20Tools %20MDF.pdf (http://www.onsrud.com/files/pdf/2012%20LMT%20Onsrud%20Production%20Cutting%20Tools %20MDF.pdf)

Will give you the manufactures poop sheet for your particular bit... the recommendations for that bit in MDF are .006-.008 per tooth at one diameter deep, reduce by 25% for 2XD and 50% for 3XD deep. I routinely cut at 2 X the bit diameter without issues.

SG

[jhedlund58]

I would be affraid to re-use it.

[MogulTx]

Yes. This bit is trash. And some would tell you that you should replace a collet after a bit break. I don't buy into that. If it CRASHED the bit into something, then maybe I would-

but in this case, just replace the bit, modify your speeds and rock on.

[MogulTx]

And what Dave R said: Seat the bit as deep as is feasible without contacting the top ends of the cutting flutes and their run-out area. The grip against the bits will be best that way.

( And some people really torque their bits. I don't. I torque them enough that I know I am sufficiently holding the bit, but not so much that I think I am going to start excessively wearing the threads on my spindle. I need to get a number of years of use out of my spindle- and don't want to pre-maturely wear it out..)

[coryatjohn]

Excellent advice. I will study that chip load calculator.

So in summary, I did a couple of things wrong:

1. Too fast a cut rate.
2. Bit not far enough into the collet.
3. Lack of experience and understanding of the basic process.

Ignorance is expensive! I toasted a $25 brand new bit.

[steve_g]

John

http://www.talkshopbot.com/forum/showthread.php?t=15176 (http://www.talkshopbot.com/forum/showthread.php?t=15176)

If you haven't found this thread it might help with basic bit info...

SG

[coryatjohn]

Steve,

Excellent document. It should be included with a new bot!

[steve_g]

Thanks John (blush, blush) that document was truly a group effort. I had planed to do something similar regarding hold down methods after my hip replacement... but I never slowed down long enough to do it!

SG

[coryatjohn]

Hold down seems more intuitive. This chip load stuff is science with a "bit" of alchemy.

[curtiss]

I like to change out the whole collet rather than just change the bit. I suppose that is a somewhat expensive with a spindle collet ...

I also use some mid-strength lock-tight on the bit before inserting it, It dries and "I believe" fills up any small gaps and limits some wobble...

[gerryv]

Curtis, out of interest, do you use your collets for more than one bit? I'm thinking of the locktite setting up unevenly in some but not all of the grooves in the collet leading to imbalance. Just wondering.

[coryatjohn]

Ok, I got the chip load thing figured out. Certainly makes a lot of sense and is easy to work with, at least on paper. I'm sure the reality of it requires a handful of broken bits at least.

[curtiss]

Curtis, out of interest, do you use your collets for more than one bit? I'm thinking of the locktite setting up unevenly in some but not all of the grooves in the collet leading to imbalance. Just wondering.

I have about 4 1/2 PC collets and 4 1/4 PC collets, seems they were about $6 each. I use the lock-tight between the 1/8 shank bits and the insert bushings I get here. I usually have one bit per collet in the ones I use often.

http://www.mlcswoodworking.com/orderstatus/html/smarthtml/pages/bushing_ball_bearings.html

Probably not the high dollar set up, but they work pretty well. The film of lock-tight sometime prevents the electrical contact and you have to make sure the clip is on the bit before the ZZ.

I would guess the film is much less than .001 thick once it sets up.

[supertigre]

Over tightening the collet will also cause it to break where it did.

[coryatjohn]

Guy,

I just put enough oomph into the collet as it needs to stay where it needs to be. TJ at the ShopBot class in Durham was pretty clear on what tension is needed. I think that class was quite useful.

[curtiss]

Have not seen Brady help out on this thread.

I guess he has never broken one... :confused:

What is the most expensive bit anyone has broken / or even purchased ?

[Brady Watson]

Have not seen Brady help out on this thread.

I guess he has never broken one... :confused:

What is the most expensive bit anyone has broken / or even purchased ?

Nope. Never. Not even once.

I'm Scotch. It is against the rules :D

Y'all are doing just fine without my input...carry on ;)

-B

[dana_swift]

@John: Unlike Brady, I have broken many bits. I suspect I will break more in the future. I guess he can deduce the optimum feeds, speeds, and bit mounting for any application and material. I on the other hand, have to break bits to learn!

About the only thing I can do with guaranteed accuracy is predict when the collet will collide with the stock, meaning the bit needs to protrude a little more. After that, I have found "safe zones" for feeds and speeds, but nothing is guaranteed. Any defect in the carbide, dirt between the bit and the collet start the process of crack propagation in the bit shank.

Heat is a huge culprit in eating bits. Keep the feed rates in the sweet spot where the chip load calculations say it should be and that is the best place to start. Then the unknown is what should the pass depth be? No equations to help us there. Just experience. Shallow cuts at the correct feed rate are easiest on the bit. Deep cuts get the job done sooner "sometimes". Sometimes we find ourselves mid-cut having to hit the space bar and change to a new bit.

The depth of cut is determined by the horse power available to push the cutting edge of the bit through the material, as well as the flex resistance of the bit. If the bit is thin, that becomes the limiting factor. If its a half inch bit with lots of meat, the horsepower is the limiting factor on the cut depth.

Its always safe to go shallow, but its not safe to go slow. A year or two from now you will find there is a good sized sweet spot you can go to with little fear of replacing bits too soon. If the chip load calculation shows a feed rate of "X", you will probably be just fine with "X-1" or "X+1". (Did you know you can change the feed rate while the bot is cutting?) Experiment.

Also one last note- I keep several spares of my most used bit sizes. Sometimes I change a broken bit and "fix the problem" to discover it was a different problem breaking the bits.

Also- NEVER drop a carbide bit on a hard surface. That starts cracks in it and it is just a short time till it breaks under load. Shattering is a possibility. You do wear safety glasses don't you?

I'm sure Brady has never had a bit shatter either! But since he is one smart cookie, I bet he wears safety glasses (just in case of a manufacturing defect I suppose)..

D

[Brady Watson]

Don't be fooled...I got a whole bin full of broken bits ;) ...I'm not sure exactly why I haven't just thrown them out. Maybe I am thinking they could be used for something else. I have taken a few of them and sharpened them to a point and used it as a precise indicator for setups etc.

If you run a job shop and get into some of the more exotic materials, you are going to break some bits trying to find that chipload sweet spot. Generally speaking, I care little about chipload. I am focused on cut quality first, then if I can eek out a few more parts by dialing down RPM/rubs then great. The problem with using chipload calculators is that there is zero consideration for how the part are held down - that is to say, that it assumes that the geometry of the tool and your hold down solution are not fighting each other to the point that it will cause the parts to lift or move. Sometimes you have NO other choice than to grossly lighten up chipload in order to keep the part from breaking free from the hold down...so you kill a tool or two...You get the job done, and you get paid.

So...I don't want to invalidate the science of chipload, because it IS 100% valid from an engineering point of view. I'd love for tools to last a lot longer because I was able to run at ideal settings. BUT, the hard truth is, with our machines, running ideal chipload doesn't really happen as often as you'd think - and there are even more variables that you have to consider, such as the shape of each part & whether you are going to defy the laws of physics with acceleration and deceleration on parts that have convoluted curves. Large parts, like those for cabinets etc, are much easier to dial in than say, a bunch of smaller 'amoeba' shaped parts. So keep that in mind.

If you get in the 'ballpark' with chipload settings, then you're golden...and just to dispel some of the mystique behind chipload, if you let the router rooster-tail some chips into your cupped hand, and you mic out the thickness of the chip, that's you're chipload. Nothing more than that.

Truth be told - No. I don't always wear glasses. :eek: If I am just cutting sheet goods and lumber, no. If I am hammering @ 12" per second with chips flying everywhere - with so many matchsticks you can't even use dust collection - then yes. If I am cutting any kind of HDU, metal or aramid/graphite - yes. Static electricity is a strange thing...move your part with the machine off & static charges can propel chips in your face. Ask me how I know!

-B

[gerryv]

According to the Clan Watson Society of Canada, the Lord Lyon has confirmed that the clan does have a chief, but the title has not been claimed since it was held by James Watson of Saughton in 1818.

Hey maybe we could start a movement to have Brady named chief and he could invite us to his Scottish castle.

[FSICM]

pull your collar off and blow it out make sure there is no dirt or dust in it
I broke 2 back to back because the collar didn't tighten properly.

[chiloquinruss]

I get asked often how I learned to run a cnc machine, and I reply 'experience', then I go over to my toolbox and get the bag of 'experience' to show them the broken bits. Most of them are becuase of cockpit error and not a faulty part or setting, just DUMB stuff! :D Russ

[beacon14]

Save up your broken and dull bits and send them in for ca$h:

http://ibuyscrapcarbide.com/

I haven't tried it yet but I found this guy at the IWF and plan to give it a go. I probably have a couple hundred bucks worth of used carbide in the jar - I knew there was a reason to not throw them out.

[coryatjohn]

Save up your broken and dull bits and send them in for ca$h:

http://ibuyscrapcarbide.com/

I haven't tried it yet but I found this guy at the IWF and plan to give it a go. I probably have a couple hundred bucks worth of used carbide in the jar - I knew there was a reason to not throw them out.

Since we all have a bunch of this stuff lying around and individually, it probably doesn't amount to a hill of beans, perhaps someone has a charity that would be a good recipient for such donations. It might be a fun drive to contribute to.

[Brady Watson]

Ha! Thanks David!

See? I'm not a pack rat...I just knew it would only be a matter of time before I could cash these in for big bucks! :D:D:D

-B

[beacon14]

It may seem like big bucks until you add up what you probably paid for all those worn out and broken bits. I've been buying my compression cutters 20 at a time - somewhat painful to do the math on what is in that jar of scrap!