Breaking Router Bit

**cjd** · 05 February 2007, 14:47 Monday

First time you get ANY change in noise/vibration you stop what you're doing and investigate the cause. NEVER keep cutting.

You have to use those listening skills with a router.

Guesses on cause? overheating. Probably caused by feed rate and bit sharpness.

I once had the lower shaft bearing on the router itself come loose (actually the screw-in plate that holds it in).

C

**roccovw** · 05 February 2007, 15:00 Monday

There are a few reasons for breaking a solid carbide bit, the most obvious is striking a foreign object whether in your work or a carelessly thrown tool on the workbench. Carbides weakness is it's hardness, it's not hard to crack or break carbide with a good rap from a chunk of stone or steel. Another cause is a bit that's running too hot then cooled to quickly, the stress of cooling to quickly will crack carbide quite easily. Which brings up the matter of carbide running to hot. Your feed speed versus rpm will determine how hot your bit gets, a too high rpm with to slow of feed speed will quickly heat up your bit and that's were running only sharp bits is essential to maintaining a proper rpm to feed speed relationship. A too dull of a bit and you end up running too slow to maintain a clean cut. The other possibility is that you could have a worn collet or router shaft where the collet sits preventing the router bit to sit properly causing vibrations. Also breaking a router bit while your routering is not all that bad, when they break they just end up getting stuck in the wood. Our cnc would break 1/2 solid double spirals regularly and it was more of a nuisance then a danger.

**speedle** · 05 February 2007, 15:26 Monday

So, since it happened after you applied cooling air, maybe the overheating followed by too-rapid cooling was the cause?

**TomK** · 05 February 2007, 17:30 Monday

I went through an almost identical scenario about a week ago. I attributed the failure to leaving the router bit in the tightened router collet in the garage during a period of sub-freezing weather. The differential in contraction between the steel collet and carbide bit might have been the culprit. That's the last time I'll put the router away without storing the bit.

Whatever the cause, I definately dodged a bullet. I'm seriously looking at some sort of full face protection.

**Gir** · 05 February 2007, 17:45 Monday

One time I was cutting some small pieces of metal with a cutting bit on a dremel. I just started to cut, and then I suddenly realized I forgot my safety goggles. I put them on, then started cutting again. About 4 seconds later the bit broke in half actually lodging itself in the glasses right in front of my eye! Close one!

**JonP** · 05 February 2007, 19:03 Monday

Originally posted by speedle

So, since it happened after you applied cooling air, maybe the overheating followed by too-rapid cooling was the cause?

Yeah, I'm thinking about that, since it was something new I was doing. Also, the heat shock theory, bit getting older, heats up more, collet is a mass of colder metal cooling the end of the shaft... snap!

Or, what I found in that safety paper about wear in the collet causing vibrations, right there inside where it broke.. I'll be closely inspecting the collet and bit shaft when I get home tonight.

cjd, I did break that rule... even as I was thinking I shouldn't, I went a bit longer since I had only half the thru hole to go on the RS28 spot. Really important to stop the moment something changes, even just a little bit!!!

I guess I'm lucky I didn't lean hard on it, and have the bit come loose... I go pretty gently most of the time. A guy on PE chat was talking about having things dug out of him in the ER room, one time having a chunk of carbide go thru his lip, and break a tooth! 8O

It's good getting this feedback, and not just for me. Safety is important, so we all can enjoy fine speakers, with all our body parts attached!

**JonP** · 05 February 2007, 19:13 Monday

Originally posted by TomK

I went through an almost identical scenario about a week ago. I attributed the failure to leaving the router bit in the tightened router collet in the garage during a period of sub-freezing weather. The differential in contraction between the steel collet and carbide bit might have been the culprit. That's the last time I'll put the router away without storing the bit.

Whatever the cause, I definately dodged a bullet. I'm seriously looking at some sort of full face protection.

So, it cracked on you down in the collet as well? Before you started? I could picture the steel shrinking faster than the carbide, or maybe it was the still cold collet and the normally heated bit creating too high of a temp differential.

Anyone have any feeling about overtightening being a cause? I have read, and assumed, that you pretty much want to tighten it as much as humanly possible. And a steel shaft bit and steel collet would be pretty similar. But it seems like TiC expands faster than steel as opposed to slower, so that maybe you could overstress the shaft and crack it when it heated up?

**bob barkto** · 05 February 2007, 22:20 Monday

It's getting hot!!

I'm sticking to my worn collet reasoning for this as well!

If the collet was worn then overtightening might cause a fracture as you could get some bending stress going on. It would take quite a bit of wear tho and your bit would probably see the extreme angle of the bit sticking sideways out the collet.
If the collet was within tolerance then, no, the force on a cylindrical surface from another will be pretty equal and you should not see anythig but even, compressive force. Maybe the bit will get a little bit longer when sqeezed but not enough to exceed the strength of carbide. Think about the load on the bit when it's cutting and now imagine the supposed stress from sqeezing the bit in the collet and youll see where I'm coming from.

Just an addendum to the heat issues.:Z:M

Most of the fears about heat fractures pertain to microscopic damage to the extreme cutting edge of the tool, Not bit breakage! It's damage that is barely visible! 8O

Sometimes you'll see some chipping that will be visible to the naked eye. Mostly what you will see with your naked eye is a dull bit.

Router bits have more than enough mass to absorb and dissipate the heat generated from cutting wood to be relatively immune to heat induced fracture of the cutting edge, let alone the whole bit. If that weren't the case none of us would be using solid carbide bits. They'd all be dull and broken on first use.

Now think about where your bit failed. At the shaft deep inside the collet. That is the least heat sensitive spot in the whole equation. The bit is thickest and is held in what amounts to a giant insulating heat sink. No way is the temperature diferential going to be great enough there to cause a fracture. There will be practicaly no temp dif there, even with forced air blowing on it.

Bit breakage is caused by exceeding the yield strength of the material the bit is made from or from mechanical shock. Not heat.

If you want to see the effect of heat on your bit try this experiment.
Put on your safety glasses, heat the cutting edge remnants of the bit with a torch until it is hotter than when you were cutting, drop the hot bit in cold water.
What happened?

Originally posted by JonP

Anyone have any feeling about overtightening being a cause? I have read, and assumed, that you pretty much want to tighten it as much as humanly possible. And a steel shaft bit and steel collet would be pretty similar. But it seems like TiC expands faster than steel as opposed to slower, so that maybe you could overstress the shaft and crack it when it heated up?

**TomK** · 05 February 2007, 22:57 Monday

Bob,

Thanks for your contributions on this topic.

After the bit fractured, it scored the outer lip of the collet. I went ahead and replaced the collet when I replaced the carbide bit. $10.95 is areasonable charge for peace of mind.

**wkhanna** · 05 February 2007, 23:47 Monday

Just a passing thought, since Carbide is V brittle, the bit may have first suffered a minor chip, causing the initial vibration, which in turn caused the catastrophic failure?

Having carbide cutting edges ‘glued’ onto a steel carrier helps the tool endure some shock loads since the steel is more elastic and acts as a sort of shock absorber.

Replacing your collet was a wise decision, regardless, IMHO.

**Paul H** · 05 February 2007, 23:47 Monday

This may not have been a factor here, but it's still good advice - do not buy cheap router bits - find somewhere else to save money, on something that won't be spinning at 25,000 rpm inches from your hands and face.

**bob barkto** · 06 February 2007, 00:51 Tuesday

Possibly. Vibration isn't good!
I think it would need to be a fairly good chip tho. I haven't seen small chips cause any noticeable effect.

Yes, tipped bits are much more durable than solid carbide. But it's really difficult to create a spiral tipped bit, particularly in smaller diameters. You see some in planers and moluders etc. 3-4" diameter.

Collets take a lot of abuse. I know mine do! Always changing bits for this and that. And the major point of wear is the opening, right where you least want it.

Originally posted by wkhanna

Just a passing thought, since Carbide is V brittle, the bit may have first suffered a minor chip, causing the initial vibration, which in turn caused the catastrophic failure?

Having carbide cutting edges ‘glued’ onto a steel carrier helps the tool endure some shock loads since the steel is more elastic and acts as a sort of shock absorber.

Replacing your collet was a wise decision, regardless, IMHO.

**bob barkto** · 06 February 2007, 01:05 Tuesday

I hope my comments were helpful.

Yeah, $10.95 is really cheap insurance.
I would recommend anyone change a collet that shows any visible signs of wear or damage.

Originally posted by TomK

Bob,

Thanks for your contributions on this topic.

After the bit fractured, it scored the outer lip of the collet. I went ahead and replaced the collet when I replaced the carbide bit. $10.95 is areasonable charge for peace of mind.

**JonP** · 06 February 2007, 03:45 Tuesday

After talking with a machinist kind of guy at work, he was suggesting ways to check the collet for various kinds of wear... mostly involving a 1/4" rod or pin gauge, just barely tightening, then running it up and down to feel for an uneven amount of grab at different depths... and moving it from side to side to check for cone shaped wear that would allow such vibration.

Doing this tonight with another router bit shaft DID feel uneven and there seems to be some side play... I also think I can see just a touch of wear marks on the carbide, but it's so damn hard I'd think the steel wouldn't leave much. Looks like we've found the culprit! I'll take router, collet and all in tomorrow, and let him look it over. He says this kind of thing happens often... collets wear, milling bits start to chatter and sometimes break.

The real pi$$er for me tonight... Yeah I can go to Lowe's and get a Bosch spiral bit for $20 to replace my probable Rockler or whoever I paid $25 for... BUT... can I buy a new 1/4" collet for my Dewalt that they even now have for sale on the floor?? NO!!! I even had to spend time with the guy wearing the "lead tool department administrator" badge and explain what a collet is... :M

I'm wondering how much of my Jasper Jig I'll mow off, if I try to use the 1/2" straight cut bit to finish the tweeter thru hole?

**JonP** · 06 February 2007, 04:11 Tuesday

Originally posted by bob barkto

I'm sticking to my worn collet reasoning for this as well!

If the collet was worn then overtightening might cause a fracture as you could get some bending stress going on. It would take quite a bit of wear tho and your bit would probably see the extreme angle of the bit sticking sideways out the collet.
If the collet was within tolerance then, no, the force on a cylindrical surface from another will be pretty equal and you should not see anythig but even, compressive force. Maybe the bit will get a little bit longer when sqeezed but not enough to exceed the strength of carbide. Think about the load on the bit when it's cutting and now imagine the supposed stress from sqeezing the bit in the collet and youll see where I'm coming from.

I think we're looking at a worn collet, as per my prior post. That would make the force uneven, and focused at a smaller point. I don't know what kind of expansion difference might occur with dissimilar mateials, but the bit apparently did run and cut for a while with only 1/4" or less in the collet and not fly out. (thankfully!!) And it then fell out on to the table 15 seconds after the router turned off. But, I think the vibrations from the unsupported shaft were the primary action, as per the safety paper I linked, and my machinist friend's comments. The compression and heat differential probably just added to the stress.

Now think about where your bit failed. At the shaft deep inside the collet. That is the least heat sensitive spot in the whole equation. The bit is thickest and is held in what amounts to a giant insulating heat sink. No way is the temperature diferential going to be great enough there to cause a fracture. There will be practicaly no temp dif there, even with forced air blowing on it.

Bit breakage is caused by exceeding the yield strength of the material the bit is made from or from mechanical shock. Not heat.

If you want to see the effect of heat on your bit try this experiment.
Put on your safety glasses, heat the cutting edge remnants of the bit with a torch until it is hotter than when you were cutting, drop the hot bit in cold water.
What happened?

I might try that experiment! I see what you're saying about the heat transfer. I'd bet that the carbide might not break all by itself, like glass, when heavily thermally stressed. But, at a squeezed point, with the carbide expanding faster than the surrounding steel, that would multiply the force on the shaft. And, the max differential point would go a bit deeper into the "heat sink" with a lot of heat traveling in for a while, so I could buy a peak thermal stress point as much as 1/4" in.. Peak physical stress by bad collet support, thermal stress adding more physical stress, and vibration all focusing at the same point...

Ah, conjecture is nice..... might even be so. :W

At any rate, we can see collet wear, I did hear this occasional strange vibration, that got more and more common, and didn't stop to check it out.
I'll report in with what I see tomorrow...

**Martyn** · 06 February 2007, 17:14 Tuesday

I'm not a metalurgist, but I'd guess that this is a classic brittle fracture. The contributing factors would be:

* a surface defect such as a crack (probably invisible to the naked eye), or a chip or other localised damage (possibly doe to scoring caused by a badly fitting collet or whatever);

* high loading on a 1/4" shank.

There could also have been a fatigue component due to the high number of stress cycles, or a flaw in the material itself or its heat treatment. Heat caused by routing is unlikely to have been a factor. A "macro" photograph of the fractured surface would be very interesting.

Stick to quality brands and avoid "Made in China" no-names.

**bob barkto** · 06 February 2007, 20:16 Tuesday

Your machinist friend is very wise! :B

Sucks that Lowes doesn't have a collet. Not really surprised tho, many people could be persuaded to buy a new router...

Check out contractor supply houses and tool repair shops. Most stock those types of parts (collets). DeWalt is pretty popular so there's a good chance you can find one localy. Also check for the nearest major brand service center. Most stock parts for more than several brands of tools and will do 'will call' for something like that.

Originally posted by JonP

At any rate, we can see collet wear, I did hear this occasional strange vibration, that got more and more common, and didn't stop to check it out.
I'll report in with what I see tomorrow...

**JonP** · 07 February 2007, 04:22 Wednesday

Here's the update...

I looked at the collet again, cleaned it well, and checked it over with the microscope at work. Nothing drastic to see, but there did seem to be uneveness you could just make out naked eye. Under the scope, it was more visible. When just barely tightening it by hand, and sliding another 1/4" shaft in there, it felt like it grabbed farther down, rather than evenly. It also had a noticeable amount of side to side play.

My friend did an inspection as well, coming to the same conclusions. He spent a fair amount of time looking at the break edges... hard to say the results. Looks like a small dirty break, then a clean one thru the majority of the shaft. Flaw, just the start and end of the break... ?? It is just about at a 90 deg angle to the length. I might try shooting a pic thru the camera...

On the side to side play, I wish I had a new collet to compare with. The cone shaped wear mentioned in that router safety PDF would look and act exactly like that. But, it could be that the collet will look like that when not fully tightened, and I won't know till I get a new one. My friend thought that it could be a likely failure mode, the worn collet doesn't support the bit all the way up, you get a vibration mode going, and CRACK!

And... the less used 1/2" collet did not seem to be as loose, or grab faster down low when doing the barely tightened feel test.

I'll call Dewalt's support number tomorrow and find my best way to get a new one. (or two, or three...) And make a regular schedule of feeling the fit and checking for that wear!

Upside on this... I only carve away a few thousanths of my small Jasper, with a 1/2" straight cutter, so I'm not paralyzed. Was fearing my future ability to countersink those little PE ND20 tweeters and other small objects.... but no problem!

**Martyn** · 07 February 2007, 17:03 Wednesday

Originally posted by JonP

Here's the update...

....Looks like a small dirty break, then a clean one thru the majority of the shaft...

Like this: http://www.metaltech.co.uk/images/FractureA_PIC04.jpg?

I'd guess that the difference between the sizes of the two areas is greater in your case...

Breaking Router Bit

Breaking Router Bit

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment