Carter, Iturra, Lenox, Starrett Comparison

[Alan Schaffter]

I'm with Lou- what is the correct tension (psi) for a particular blade, and how was that figure obtained- I would suspect there may be some trade between metallugical/mechanical requirements and material cut quality requirements. So how important is it for the tension to be set accurately (as opposed to precisely), and how much variance is acceptable?

I think the best and easiest tension method might be a combination of a well designed bandsaw tension/tracking mechanism that doesn't have "stick-slip" and a direct reading (in psi) gauge similar to Carter's that can be used/monitored while the saw is in use- and all included in every saw for minimum additional cost.

How have we all survived using the bandsaw for the last XX years without accurated springs and tension gauges?

[lou sansone]

Still wondering .... are there any engineers out there ?

I am still wondering if there are any sound engineering reasons for the recommendation about 15kpsi.

lou

[William Falberg]

Being new to Sawmill Creek's forum I'd like to introduce myself as another band saw "expert". I make and sell specialized portable bandsaws for timberframers. I've been following BS threads on several general-WW forums for a year now and just recently found Sawmill Creek. This particular thread caught my eye because it highlights the confusion surrounding the subject of re-sawing. I once owned a Delta BS and immediately chopped it down to a frame with three wheels. Mark Duginske, Lonnie Bird, and others have far more experience setting up standard two-wheeled bandsaws to perform re-sawing than I do so I defer to them in all matters related to the commercial machine-dynamics side of resawing. The emphasis of my research has been directed towards making flat cuts in the 16" depth-of-cut range; two to four times what most of you consider tough cuts. There just aren't any blades that worked, no matter how much tension was applied, and in the end I had to make my own machine for re-setting commercial blades to match the variety of cut parameters my customers demand. I'm not skilled enough at writing to describe exactly how tension, tracking, blade speed, HP, set angle, and wheel alignment interact to produce flawless resaw cuts so I'm inviting any recognized author (Mark?) or BS expert to bring their favorite blades, tension gauges, band saws, fences, gnarly workpieces, whatever, and (especially) a typewriter to my shop to see for yourself why my on-line postings always contradict your learned advice. You are ALL correct, insofar as the dynamics of the band saw is concerned, but equally wrong where blade dynamics is concerned. I'd like you to see what happens when a blade is "tweeked". I think an open-minded, face-to-face exchange of this nature could, if properly described and published, dispel the last remaining "mysteries" on the subject. I'm easy to find on Google, so please feel free to contact me in any manner you wish. My shop, my mind, and my heart are open. I have no dogs, just kids.

[Bill White]

What time is it? Well, you first take a billet of stainless steel, mill the outside shape of the watch you want, contact a swiss watch movement mfg. co.,............
Now that I feel better, I tension 'til I get the smoothest cut, cut the part, and get on with stuff.
I have had good luck running the Timberwolf blades with my standard settings.
Bill

[Jay Fields]

Hello Mr. Falberg,

I read an article about your test methods, results, and conclusions a while back in either Woodshop News or Woodworker West. Interesting conclusions. A bit counter to everything else you read on the net and in print.

A search for your name at SMC found the following posting;
http://sawmillcreek.org/showthread.php?t=52942

I may try some similar tests on my own bandsaw someday when I find some time.

JF

[Rick Markham]

Ok, gentlemen... now maybe I understand just enough of the physics to toss the proverbial "monkey wrench" in this conversation. Please bear with me... In a shop environment, where you are changing blades, i.e. different manufacturers, different steel, even blades from the same manufacturer might be steel from different foundries/ore, yielding albeit minute subtle differences in the metalurgy, even in separate runs from the same foundry is there not going to be environmental factors that contribute to the molecular make up and hence the physical properties of the steel? Even with the highest quality control, environmental factors fluctuate, minute uncontrollable changes, i.e. Ore characteristics etc. Even the minute amount of welding on the blade, should change it's characteristics slightly (it might be small but... it's still there)

So now my point, for true repeatability I would think measuring the load on the single point on the machine (the tension rod) is going to yield the most repeatable results. Granted I am not talking about true tension of the blade, since from my perspective, there are too many variables to eliminate. Obviously recording the setting for each blade by the best result gives you an excellent reference point for the next time you set up that blade. At some point this becomes a semantic argument. It seems to me that for repeatability sake, measuring the point with the least amount of variables is the best idea. It might not be true "blade tension" it's the force exerted on the wheels, but since the blade make up and type are not a constant there is little more than a "ballpark" figure that you can ask from any instrument.

Just a thought, don't know if my bucket holds any water

[Van Huskey]

Rick, I have read a TON of posts here and other forums, articles and books and watch too many engineers that know far more than I argue ad naseum about tension, aftermarket tension gauges and the gauge on the saw. I have finally come to the conclusion that OP, Mark (one of the BS gurus) supports:

"USE THE GAUGE, AS THE BLADE DULLS YOU MAY WANT TO INCREASE THE SETTING. I USE THE GAUGE AT THE 1/2” SETTING FOR THE 1/2” BLADE BUT INCREASE THE SETTING TO 3/4” WHEN THE BLADE GETS DULL."

He is talking about the gauge on the saw, I go up a blade width for carbide because they require more tension. If I have any issues with a particular blade I may play with tension and mark that blades "tag" (I keep one for each blade with some info I see as valuable).

In the end I love tools and measuring devices but when it seems I have already paid for the only tensioning device I really need when I bought the saw I would rather spend that money on a really nice carbide resaw blade.

So for me I had the same questions as you but the more people that "knew what they are talking about" chimed in the more I realized there was no clear cut answer though each person felt like they were right. So I just figured I would go with the easy route recommended by someone that knows more and is more proficient with a BS than I will ever be.

[kenneth kayser]

For testing tension gauges, why not just take a few feet of blade material and hang a known weight from it. Cross sectional area can be easily be measured with a micrometer.

[Howard Klepper]

I am neither an engineer nor a bandsaw expert, but there is another method of tensioning a blade that I don't see mentioned here, which is listening to the pitch when it is plucked between the wheels on the column side. I listen for a musical sound in the 40-50 Hz range. For a given tension, this will vary from saw to saw with the distance between wheels, but since tension is directly proportional to the unit weight of the blade, the pitch remains a constant (for a given tension and wheels spacing) as blade width changes, so long as blade thickness remains the same.

[Tom Walz]

A couple great posts here. They do a nice job of illustrating the complexity and uncertainty of tool science and tool testing.

In the industrial world the following book is widely respected. Maybe not as directly applicable to the average forum member as Mr. Bird’s but it may have some valuable crossover information.

The Complete Guide to Your Industrial Bandmill & Bandsaw
111 pages, 52 illustrations

Wijesinghe, Ralph
4688 Sunland Pl
Burnaby, BC V5J 2R4
(604) 437-4610.
ridw@shaw.ca
www.datasystech.com

[Neil Brooks]

I've been following this thread with ... a RIDICULOUS amount of attention

I still don't have all the pieces that MY brain seeks:

a) What's the problem that we're trying to solve. For ME, it's ease and accuracy of cut, safety, blade life ... in no particular order;

b) What impact do various blade tensions have on the above;

c) How do we know this;

For ME, I'm tripping over dollars to pick up dimes without that info.

I've got Mark's first book (love it). Do I need to go back to that to find my answers? DO they live there?

ARE these questions well and thoroughly answered, already -- if elsewhere ??

[James White]

What happened to this? Was the retesting ever done?

[Will Blick]

> I would think measuring the load on the single point on the machine (the tension rod) is going to yield the most repeatable results.


in the end, its blade tension you are after, nothing else. Itura's new 2011 book writes about this subject in detail. Its a great read, and its 2011 catalogue is quite updated vs. some of this outdated tests.


I have the higher end Ituro test gauge, which reads in 1/10k of on inch... (current ones do not, unless you replace with a higher-end dial indicator, which is what Starett and Lenox has installed) Louis claims he has all these meters, and tests them, and its the Carter electronic which he has the hardest time keeping matched with the 3 gauges, his, Starett and Lennox. When using the gauge, assuming the dial indicators are working, and not sticking, it does not appear the readings should be far apart. considering you are trying to read variances in 1/10k of an inch, IMO, any faults would probably be in the dial indicator, as these are quite precise instruments vs. the more commonplace 1/1k inch versions.


you also have to be careful using the saws gauge. In most cases, they are rudely calibrated, and often calibrated thin blades...whereas you may use thicker blades. The Lennox Tri Masters I use are 2x thicker than what the MM20 scales are for. Double blade thickness = double tension, so in theory, when I use a 1" blade, I would need to set at 2" setting, but the saw only goes up to 1.5" setting max. which as it turns out, with the Ituro guide, reads 13k pounds of pressure. TriMaster 1" blades are suggested to run up to 30k pounds of pressure, so I just max. out the crank when using that blade...

[ian maybury]

Hi Mark, guys. I've not read all the posts, but the perspective that strikes me is that we're a little at risk of presuming we know what's needed in terms of chassis, blade tension and the like. Which may not be the case - it's more like there's a recipe that usually works, but it's not necessarily clear why.

One issue I ran into in trying to get a high end 18in hobby band saw to re-saw well was that there clearly was lots going on by way of vibrations in the blade. These were largely invisible, but could be felt if it was touched with a piece of wood and seemed to interplay with vibration in the chassis and worsening to correspond with the onset of drift.

My sense was that there's probably lots of modes of vibration involved - ripples running up and down the blade, torsional/twisting movements in the blade, and also longtitudinal movements involving rotational oscillations of the blade and wheels. Then there's the various modes of vibration the chassis may be capable of in the many dimensions in which deflections are possible.

All this on top of the static/load induced deflections taking place which may give rise to tracking and blade guidance issues too. The tensioning mechanism/spring, and the stub shafts on which the wheels are cantilevered are no doubt capable of various modes of vibration in their own right too. I've acquired a 24in Agazzani which I'm about to start up - it'll be interesting to see how it compares to its predecessor....

Another complicating factor is the reality that when vibrating systems interact they can reinforce or cancel each other's effects. There's also the little matter of resonance.

Getting a saw working well is probably (just theorising) about (a) getting to set up that maintains correct alignment of the blade, wheels and guides, and (b) getting into some sort of sweet spot where the worst effects of vibration (i.e. resonance) in certain critical modes is tuned out. Blade tension is perceived as being critical, but probably because in the context of a given saw and blade set-up/system it's probably the single biggest factor available to (a) re-tune the multiple modes of vibration, and (b) adjust static deflections/alignments.

Steel fabrications like are found on most modern band saw chassis are especially prone to vibration (unlike cast iron the material has minimal internal damping) - maybe they need filling with concrete or a similar damping medium. It looks like maybe the close multi point contact guide systems like Laguna's and Cool Blocks produce improvements in cut quality by maybe acting to damp/choke off vibration in the blade.

This is of course all pretty speculative, but judging by the saw mentioned above they are the sort of issues in play. One thing that seemed fairly clear was that it simply didn't seem to offer the possibility of a dead (minimal vibration) zone for deep re-sawing within the available adjustments. No matter how it was set up significant vibration was always present. Deep re-sawing (over 6in in oak) seemed to trigger a sharp worsening which could be heard as a nasty and harsh metallic almost scream.

It's hard to know how best to approach the problem of creating a 'dead' saw. Chances are that modern CAD modelling coupled with finite element analysis may open the way to predicting a lot of what's going on - I'm not up to speed, but suspect that it's possible to predict modes of vibration pretty effectively these days. (think aircraft and vehicle structures and the like) On the other hand when there's multiple systems and multiple modes of vibration in play it may get somewhat beyond accurate theoretical prediction.

It's also quite likely to be the case that some variables become hypersensitive to adjustment within certain narrow ranges, while appearing largely inconsequential for much of the time. There's also the issue of how the variables change between and with wear to blades and the like.

The other approach is the empirical - getting a series of saws hooked up with all sort of instrumentation to figure out what's going down in practice. With a view to tuning out the problems in a new design through repetitive testing and modifications. My inclination would be to focus on finding out what's actually happening when saws stop performing/get outside of the sweet spot first - with a view to figuring out what improvements are likely to give the best bang for the buck.

The outcome might be a set of requirements to be delivered when designing and setting up a saw - maybe some design improvements and a methodology and some tools for delivering the requirements too.

All in all (depending on the project is scoped) it's potentially quite some undertaking - maybe time to call NASA. Good luck with the project.

ian

[Will Blick]

Ian, good post, and yes, I should have touched on some of this as well... running a saw at 15k - 30k pounds of tension is nothing to sneeze at. So there is certainly more to the equation than getting blade makers reccommendation on blade tension, buying a gage and setting to that tension. It's the weak link in the chain principle...if the saw itself is the weak link, all the set-up will not produce the best cuts. This is why lots of hardcore re-sawers buy very old BS's and overhaul them, as back in the old days the big units weighed a ton, and they still weigh the same today :-) Weight adds stability, absorbs vibration and resists flexing as much of the metal is in all the vulnerable support zones. This seems to be the "claim to fame" of the Italian makes today, they have maintained the robustness of the saw vs. the Asian makers. Then of course you need good bearings, the right guide blocks, the right set up, a blade that does not walk (in n out), as not all blades are created equal. I have had blades that walk, (.025") and I replace them with an identical blade, and it will walk only .004". Blades have lots of variables, including the human variable as each one is welded together.

Then you need to follow some fundamentals on cutting.... I follow the 3-6-19-12 Rule - IMO, very effective... Lennox also supports this... Idealy, select the tpi of the blade so you always have 3-6 teeth in the wood at all times. If softer wood, you can go as low as 3, VERY hard wood, up to 12. Too many teeth in the wood = too much heat and insufficient waste removal (creating more friction heat). Heat is the enemy of all blades, specially BS blades. Not enough teeth in the wood = slower cutting = heat build up from excessive friction as the blade is in the wood longer than it needs to be.

This is why its a 3-6-9-12 rule, its flexible. The wood type and power of the saw, which are two variables that vary for each set up... so clearly, re sawing is part science, part feel, part equipment, part user knowledge of feed rates, etc. And you hit the nail on the head, this power tool is not heavily researched, much is trial and error. And as blade technology continues to advance, the documentation doesn't seem to follow it. The best resources are those who work in the BS field full time such as Louis and some of the big shops that re-saw wood all day - they have lots of wisdom you will not find in books....