Band Saw blade fatigue questions

[David Ragan]

I a MM-16 with 1 1/4" Resaw King blade from Laguna. I love it.

Also, I have a Starrett bandsaw blade tension gage (anyone want to buy it?) I routinely tension to 30-35K. (Now, I know the finger push or flutter method is perfectly fine.)

My WW is on wkd only-and light duty then.

The blade was/is detensioned @ end of every workday.

About 3 years ago, I left the blade tensioned all night. Came back down the next day, and it broke. Awful racket. Blade was 2-3 yrs old.

Sent it back, couldn't be repaired, so I bought another one. I figured that it was my fault for leaving the tension on all night.

Some time ago, I read that there is no need to detension the blade----I have been a little put out with Laguna ever since I read that. They never asked me how it occurred, just said it couldn't be welded and that was it.

Besides the debated tension issue, did I do too much back and forth de/tensioning and cause metal fatigue?

Do I have a beef with Laguna?

Whats the deal?

[John TenEyck]

As long as you are below the fatigue limit adding/removing tension should have no impact on when the blade failed. 30 to 35K seems awfully high, however, but I don't know what Laguna recommends. Still, I'd be suspicious that tension level is the reason it failed.

John

[Barry Richardson]

Haven't used the Resaw king that much, it got dull and I put it aside and have never got around to resharpening it. However, I use a Laguna 16 and find that hard back blades like lennox dimaster will often fatigue and break before they are dull. I have read somewhere that 16" (and even more-so 14") wheels are too tight of wrap for big blades and the can fatigue prematurely, I think the theory has merit.

[David Kumm]

Tensioning and detensioning wasn't the cause. I'd look for hairline cracks at the gullet. Sometimes the sharpening can cause stress at the gullet that eventually fails. Spinning the back bearing fast can heat the band and lead to premature failure too but at 30-35K tension that probably didn't happen. That tension sounds OK although I run the trimaster which has a thicker band. I run 1/4: bimetal at that tension and it has lasted until dull. Dave

[Erik Loza]

My entirely non-scientific opinion: De-tension at the end of the day. No idea why the OP's blade failed but I do recommend de-tensioning after use.

The amount of tension your average 14" saw is capable of exerting on a blade nothing like the pressure a big, steel-framed bandsaw can. Or at least one of our steel-framed bandsaws.

Erik Loza
Minimax USA

[ian maybury]

From an engineering perspective the factors that determine the fatigue life are the stress level (typically at the most highly stressed point), and the number of reversals. That's stress on/off cycles.

15 - 20,000psi working stress/tension is often recommended for carbon steel blades, and 25 - 30,000 psi or more for stronger alloy/bimetal steels. Tension gauges infer the stress from the strain. (ins stretch/in length) By a quirk of the maths they don't need to know the width or thickness of the blade to do so - but they do presume all steels have the same Young's modulus value which may be a slight approximation. The notching for the tooth gullets probably also introduce some variability.

30,000 psi is a pretty high level of stress. Heat treated higher carbon content and alloy steels can be multiples stronger, but mild steels can often be close to starting to deform/take a permanent 'set' at this stress level - so it's not insignificant. Meaning that blades tensioned to this level are subject to fatigue effects/working in a zone beyond the fatigue limit. i.e. will eventually break if the load is cycled/the stress reversed enough times.

Graphs or fatigue curves record test data which says how many cycles a steel can handle for a given stress level. Fatigue doesn't arise below a certain minimum stress level (up to the fatigue limit), but as the stress level is increased beyond that failures can occur. The number of cycles needed to cause failure is initially very high, but reduces initially slowly and then progressively more quickly as the stress level increases - until eventually the piece will fail in a few cycles or just by simple overloading.

e.g. too small band wheels (below about 400mm) can for example cause premature failure in some (thicker) blades as a result of the tighter than recommended bend required to get around them shoving the reversing bending stress level (which adds to the stress already caused by the tension applied to the blade) high enough so that fatigue failure quickly follows. Correctly set up blades will if run for long enough likely eventually fail through fatigue, but should at least be set up so that they wear out before they have been through enough stress reversal cycles for this to happen.

It's tough to predict fatigue life in reality, but if a blade is failing too quickly when run on a suitable saw at the recommended tension/stress level then several possibilities come to mind (among the very many that could be in play):

1. The blade is overtensioned for some reason. (gauge problem?)
2. Some sort of defect like a nick at the bottom of a roughly cut gullet, a weld or material defect in the steel is resulting in a stress concentration/notch effect. These can easily multiply the local stress by a factor of 2, 3 or more - greatly reducing the number of stress reverals required to cause failure.
3. The metal has at some point been made brittle (tolerates fewer cycles) or much weaker (suffers fatigue effects at lower than intended stress levels) as a result of e.g. previous bending and straightening/work hardening, overheating, or disturbance of the heat treatment or composition during brazing or welding.
4. Something about the saw set up is adding stress to the blade, or increasing the number of stress reversals. Maybe even some sort of bending or vibration is happening in a heavy cut - less likely.

For my money if a blade that hasn't suffered damage/been abused is failing prematurely on a saw that's been correctly set up and suited to the blade from the start then it's hard not to suspect a blade defect. One imponderable is the blade history - if it's been re-sharpened several times and done lots of work, or if it's been run for a while overtensioned or on an unsuitable saw then there's no putting back the clock even if the set up is subsequently sorted out...

There's a nice write up on fatigue failure here, including an introduction to visually recognising a fatigue break: http://www.sv.vt.edu/classes/MSE2094...www/meyer.html

[David Ragan]

Wow, thank Ian--now that is the kinda answer I like, even if I cant understand it all.

I had suspected that the blade was overtensioned, then the off/on cycles sealed its fate.

I'll call Laguna and post what they say, hopefully this week.

[ian maybury]

Describing the possibilities in general terms based on theory is one thing David, but reaching a definite conclusion on the specific cause of failure for a given blade is another - it'd take a lot more knowledge than I have nd probably a metallurgical lab. Even then… So unless we get lucky and there's something in the area of the break (e.g. its failing at the welds suggesting it wasn't properly annealed, or cracking) that's a dead giveaway it's likely to remain a bit of a mystery/trial and error deal. Or unless you know something about the history/any mistreatment of the blade.

It seems unlikely that tensioning/removing tension after use is a factor - the number of cycles at one per day would be tiny over even a long period of use.

A closer check of the numbers since the above post suggests that its quite likely a bandsaw blade is not in fact really meant to be stressed past the fatigue limit (the stress level above which fatigue failure becomes a possibility) in normal use, because even the 10 exp 6 cycles or so that the curves suggest that steel handles once stress edges over the fatigue limit is a fairly finite life. (1,000,000 cycles amounts to only about 17 hours of running time at 1,000 cycles/min) Once a blade is tensioned even more - that's even further past the fatigue limit - the cycles the balde will tolerate/the life reduces very quickly indeed. (exponentially) This is probably why a nick or stress concentrating defect in especially a gullet (which can multiply the stress at a point concerned several times) as above can matter so much.

A high carbon steel saw band might yield at about 90,000psi, while it seems (very roughly from S-N curves) that the fatigue limit might (?) be around 60,000psi. This being the case it's unlikely that blade tension of itself (max 35,000psi say) will be a problem. It's when stress due to bending and the effect of startegically placed nicks/stress concentrators, welds and heavy use are added on top that the problems will start.

I'm not sure how it might play out in practice (i've not had blade failure problems), but as well as not overdoing the tension it's probably important to make sure that the rollers (especially the back/thrust roller) are well adjusted. If the blade is being bent back hard when cutting because the support is too far back and the wood is being horsed through the cut that likely would significantly increase the bending and hence the stress along the tooth gullets - and hence significantly shorten the life of the blade. Point being that the stress stuff like this causes ends up added with stress caused by the tension - which could easily tip it past the fatigue limit...

The converse of what I said above is that while there's likely a defect if a properly specified and used blade fails prematurely, its equally the case that there's plenty of scope for mistreating a blade - so it's likely that a supplier will presume the latter unless failures are clearly attributable to a specific manufacturing related cause...

[Brian W Smith]

Probably shouldn't respond...muddying up the waters and all that.

First off,we TIG up ours/others blades....and do it very well(NO THIS IS NOT A SOLICITATION)....we've had factory welded blades break within minutes of use.In these cases it's always "at the weld".....more specifically at the HAZ(google it).Then we'd re-weld it and go about the biz.But far more frequently we find that stress risers in the gullets,usually from regrinds as the killer of BS blades.But it isn't limited to "regrinds".....as posted above the stresses involved with fwd/bkwd movement,as small as they may be can be a factor(blade hasn't been reground).Trying to figure out these causes in EVERY case can be frustrating....we chalk "some" of the failures up to,"the price of doing business".Best of luck,BW

[ian maybury]

That fits well Brian. A reground blade (depending on the type) as well as risking nicks in the gullets will likely have more fatigue cycles/use on it too and for both reasons be more likely to fail - the gullets are anyway always going to be the weakest/most heavily stressed points on the blade. Brittleness at improperly heat treated or imperfect welds has to be the other problem. Set up/mode of use too.

One key distinction as you imply is that if the problem was just the heat treatment of/another defect in the weld area then re-welding after removal of the the very local HAZ metal should be an option. On the other hand if the problem is that through a combination of generalised over sressing, long use and/or e.g. gullet defects the blade has over most of its length reached some sort of fatigue limit then re-welding won't make much sense - it'll just let go at another crack or whatever. Suppliers will for all sorts of reasons probably be reluctant to re-weld a used blade - if only because of the uncertainty that this possibility must introduce.

I've seen some say that on some (?) blades you can often hear a fatigued/micro cracked blade 'creaking' if you roll a bend along it, don't know if it's true...

[Brian W Smith]

Now that you mention "hearing" Ian,I will say that makes sense.I've "felt" it....not while blade is on the machine but when handling/Tig'n them.Obviously hard to put into words......but you can feel they have a crispy feel/sound vs a more clean feel.You get the "feeling" as well when welding....like,this may not work,but here goes,haha.

[Larry Frank]

The real shame of it all...if you had a bunch of money and a well equipped lab, you could analyze the fracture and figure out what caused it. However, the money for a Scanning electron microscope. I would guess that blade mfg have access to that type of equipment to analyze blade fractures.

I would not think that the tension in this case alone would cause a break. There was likely some other defect or small crack that started the fracture and then cycles of running the blade or tension and de-tension would make the crack grow.

[Peter Quinn]

I'm no expert, not an engineer, just a bandsaw user. I really doubt tension, or leaving the tension on over night on an occasion is the culprit based on experience. Way better that thing snapped overnight than during a cut with your hands in play. I've had a few go in use and never had a problem but it sure will scare the pants off of you, big loud bang, wheels still moving, hitting the brake only stops one wheel. They say to drop tension end of day, I've always felt this was more for the saw then the blade. Relaxes the spring, takes stress off the wheels. I almost always do this on the home shop machine. But at work I've seen blades left under tension for 6 months, beat relentlessly, dull to the point they can barely cut fire wood....just keep going. It takes a lot to break a good blade. I will say Ive always used 20" saws at work in the past, presently I'm using a MM16, and I've seen more blades break way earlier than I would expect, never at the weld, always fractures radiating up from a gullet on a wide rip blade. I'm thinking there is a mismatch between the blade stock thickness we are getting from the local blade supplier and the 16" wheels, too tight a turn for too thick a blade, it heats, it cools, always breaks close to a start up after a long run the a cool down and restart, like after lunch. Never had this happen on a 20" saw? And doesn't seem to happen with narrower blades. I almost prefer to resaw with it using a 1/2" 3 pitch than the larger 1" 1/2 vary pitch.

I'd guess every manufacturer has some failure rate that would qualify as below average, all the steel can't be perfect. Put that together with the tight wheels, some start/stop cycles, very high tension under load....takes all of those things IMO to make that band snap.

[ian maybury]

It seems unlikely as above that the correct operating blade tension will raise the stress high enough (close to double the normal max tension in the case of a carbide blade) to lead to fatigue issues - not without other factors of the sort mentioned adding stress on top. Over tensioning is easy though in the case narrower blades.

Depending on the thickness of the band material (which is the main factor that determines how much stress bending a band through a given radius causes) it sounds from user reports that some of the smaller wheeled saws (down around 450mm) running especially bimetal and carbide tipped blades with thicker bands can sometimes run into problems with fatigue failures...

[Larry Frank]

Fatigue issues are almost always associated with some type of crack. It may be extremely small and not important at the beginning. The crack could be there a long time. But with time, a crack can grow with various levels of use and cycles of use. At some point, the crack will reach the critical length and cause a complete fracture.