Carter, Iturra, Lenox, Starrett Comparison

[Mike Cutler]

Response to Pete Bradley

Your point is well taken. Aaron has a masters in metallurgical engineering so I assume he knows how to use the gauge.

I talked to the tech people at both Starrett and Lenox and was under whelmed by the response. No one could explain how the gauges we calibrated and/or tested. Also I talked to the engineer at Delta who says that now that he read the article he wants to do some test. One of the problems these days is that the "corporate memory” is short. The people who designed and built the 14” Delta
In 1933 or 34 in Milwaukee are long gone….. same with the Lenox and Starrett gauges. Getting good technical information these days can be a real challenge. The engineer at Carter was very knowledgeable and very helpful.

The Carter measures wheel load in pounds at the bottom of the tensioning rod. Aaron calculated the blade load to be 146 pounds
for a blade with that cross section at 15,000 psi. If you study the drawing you can see the blade in the corner of the one inch square.
15,000 psi is what both the blade and saw manufactures recommend for tension on a 14” band saw.

Mark D

I also spoke with the engineer at Carter yesterday about the Carter ETG, and asked him to send me whatever info on their device that he was able to.
He seemed like a decent guy. We talked about the differences between the Starrett,and the Carter.

The Starrett, Lennox and Itturra all work on the same principle. As the blade is tensioned, the distance between the two fixed points becomes greater.
The point on top is fixed,and the point on the bottom of the gauge is free. The bottom point is a lever that moves on a pin. As the distance between the two points expands the lever exerts force on the bottom of a dial indicator, that has been graduated in pressure.
The lengths of the lever arm(s), or pivot arm according to Starrett, relative to the fulcrum point would be a math function to represent blade stretch.
The dial indicator is just a graduated usable scale. It's no different than any other dial indicator. The amount of deflection, represents the amount of distance between the two contact points, and the math fuction of the lever arm. Minus mechanical hysterisis of the dial indicator, and the lever arm, and friction of the pivot arm on the pin. the rest of it is all a math function.
If I were to use these instruments I would make sure that the dial indicator was already pre loaded though to remove the mechanical hysterisis of the device at the zero point.

The Carter is an electronic strain gauge that is looking at the amount of force being placed on a point of contact, and converting it to an electrical output, millivolts, frequency, resistance . Carter wouldn't tell me the exact type of strain gauge that they were using, proprietary I assume. Strain gauges are nothing new though, and this one is basically the same type used in a load cell for cranes,and electronic bathroom scales. Beyond the zeroing there isn't much that can be done to one. normally zero and span calibrations are performed on the electronics modules themselves. Then a known amount of weight, or force is applied to the sensor to check the operation of the cell.
If you wanted to check the Carter. zero it and then apply a known weight to the contact point. It appears it is just measuring the weight of the downward force in lbs.
The info that Carter sent me has a table for common bandsaw widths and the corresponding range of lbs force for each width of blade.

The following info is the document I recieved from Carter, on their ETG. I deleted the marketing type stuff. I have the entire document in a PDF file if anyone would like it, including installation/operation instructions. I have not found similar info for the Starrett, Lennox, or Ittura or I would have also post that. If someone has that info, it might help the discussion by providing a "common language" so to speak, if it could be posted.

I'm posting this info because there have been a few questions regarding the operation, and basis of the Carter device.

ELECTRONIC TENSION GAUGE (ETG™) Patent Pending
from Carter Products Co., Inc.
Overview
The following is a suggested range of ETG TM settings and readouts for various size
blades that you may wish to use as general guidelines to start with. You should revise
these values based on your particular saw and style of cutting.
CAUTION: The ETG displayed value is primarily intended to provide a relative indication of
tension and a method of achieving repeatability. ALWAYS take into account the
characteristics of your saw model and cutting conditions as well as any information available
from the saw’s manufacturer or your blade supplier when determining appropriate tension
levels to use for your particular application.
The output of the ETG closely approximates the bandsaw WHEEL LOAD in pounds force.
Please note that the ETG does NOT read directly in units of BLADE TENSION since for any
given load the tension will vary depending on blade size. Blade tension, often expressed as a
value of PSI (pounds per square inch) is a function of the size of the cross sectional area of the
blade. For those that want to explore this further, the Appendix at the end of this document gives
more detail about the relationship between TENSION and LOAD.
Blade Width Range of ETG Readout
1/8” 150-175
3/16” 175-200
1/4” 200-225
3/8” & 1/2” 225-250
3/4” plus 250 +


APPENDIX: THE RELATIONSHIP BETWEEN LOAD AND TENSION
Simply speaking, LOAD is the force, usually expressed in pounds, applied to an object.
In the case of a bandsaw, the wheel tensioning device causes the saw’s frame to exert an
upward force on the top wheel. It is this WHEEL LOAD that is being measured by the
ETG Electronic Tension Gauge’s sensor, which is positioned in the load path between the
saw’s frame and the upper wheel.
This upward load is resisted by a downward pull from the blade. Since there are two
strands of blade pulling on the wheel, one on the left and one on the right, each blade
strand is seeing ½ the force that the wheel is seeing. Therefore:
BLADE LOAD = ½ the WHEEL LOAD
However, the effect that any given load will have obviously depends on the size of the
object to which the load is applied. Larger objects can withstand large loads that might
deform or destroy a smaller object. Therefore, it is often helpful to calculate a parameter
called STRESS, which very simply is defined as the LOAD applied to an object divided
by the CROSS SECTIONAL AREA of the object. If we know the STRESS an object is
experiencing, we can often predict the object’s behavior even if we don’t know its size.
Stress which acts in such a way as to tend to pull things apart is referred to as TENSION.
Therefore, the stress pulling on a bandsaw blade is called BLADE TENSION and is
usually expressed in units of PSI, Pounds of load per Square Inch of blade cross sectional
area. The cross sectional area of a bandsaw blade is simply the thickness of the blade (in
inches) multiplied by the width (in inches) of the blade at its narrowest point, usually at
the bottom of the gullet. Therefore we can say:
BLADE AREA = THICKNESS X WIDTH
and
BLADE TENSION = BLADE LOAD / BLADE AREA
So by knowing the approximate load being applied by the wheel of the bandsaw (through
the use of the ETG) and by knowing the cross sectional area of your bandsaw blade
(through direct measurement), you can estimate the BLADE TENSION (STRESS) and
compare it to the blade or saw manufacturer’s recommended values.
EXAMPLE CALCULATION:
Suppose we have an approximate wheel load of 250 pounds and that we are using a ½” x
.025” blade. If we were to measure this blade’s narrowest point and find it to be .41”
wide, we have enough information to calculate the blade’s tension or stress level by use
of the following equations:
BLADE LOAD = ½ WHEEL LOAD = 250/2 = 125 pounds
BLADE AREA = THICKNESS X WIDTH = .025 X .41 = .010 in2
BLADE TENSION = BLADE LOAD / BLADE AREA = 125 lbs / .01 in2 = 12,500 PSI (lbs/in2)
This value, 12,500 PSI, is the number you should then compare to the blade manufacturer’s
recommended tension level to determine if you are operating your saw within proper guidelines.

[James Carmichael]

Lots of good info out here, but I'm afraid Mark is giving his book away:-)

The trouble with any "scientific" test is that unless all conditions are precisely duplicated with the same object definitions for "quality of cut" and "outlast", you're comparing apples to oranges.

Bottom line is, go with what works for you.

[Mark Duginske]

Mike,

Thanks to Mike Cutler for long explanation, I hope you type faster than I do. When I asked the people at Lennox and Starrett about using their gauge for woodworking machines they both seemed surprised and commented that it was designed for older metal working saws without a gauge.

Now to be practical

I personally use the gauge on the saw and fine to tune the set-up.
We never talk about blade sharpness in this discussion but it is a huge factor. A sharp blade on saw with coplanar wheels doesn’t require that much tension. A lot of people, like Ken G. and I use bimetal blades because they last longer than the others. Another factor for resawing is the flat (European design) wheels versus the America crowned wheel. The flat wheel really supports the blade well. For the BAND SAW HANDBOOK I put chalk on the tire of a Sears saw and then ran it with 1/2” blade. I was amazed that I only removed about 3/16” of the chalk which means that blade is
balancing on the crown.

I’m not picking sides, I don’t care if people spend money or not on gauges. But, from my experience with the clamp on gauges I have to question their usefulness (especially considering the discrepancies) for woodworking. However I think they provide a useful tool for people who are not confident with another approach such as using the built-in gauge and then fine tuning it. There are some people who really like numbers and the gauge gives a number. There is such a wide range of “functional tension” that I’m sure the gauges “work” for the people that use them.

If one is a “love numbers type of guy” and want to spend money I think the Carter gauge is the answer. It is a really neat device. First of all, once it is installed there is nothing else to do. It provides a constant readout as the saw is running which is very interesting. Since I put it on a 14” saw that is now my gauge because it is easier to use than the built-in saw gauge. In many situations it would be very useful such as a shop with multiple users, schools, etc. Now the bad news….at the moment it is made only for 14” saws. The other factor is the $200 price. I talked to Alex Snodgrass, who is the inventor, and he said they are working on the price to get it lower. He also told me that he had a new tensioning system that is less expensive. He wouldn’t say much about the new device. I suggested that he work on something that would work on all the types of band saws but that is a problem because they are all different. The Carter ETG works well on the 14” saws because it clamps onto the frame under the space where the tension rod contacts the frame. Rather than contacting the frame the point at the bottom of the rod contacts the top of the of the ETG sensor. Mike Cutler is right about the bathroom scale analogy.
The sensor is essentially measuring the WHEEL LOAD of the top
Wheel. Study the drawings in the Feb. 2007 Woodworkers Journal article by Aaron G and myself.

Thanks Mike C.

Mark Duginske

[Mark Duginske]

There is some confusion about the origin of the psi recommendation.

Most saw and saw blade manufacturers recommend 15,000 psi. I think Delta on their new 14" 5X saw recommends 13,000 ???. It looks like Carter recommends 12,500 (read Mike C's comments).

This number keeps getting refered to but I have not been able to track down where the number came from.

IF SOMEONE READING THIS POST (MIKE C. OR KEN G.) KNOWS WHERE THAT NUMBER CAME FROM PLEASE LET US KNOW.

Mark

[Mark Duginske]

Thanks for your concern about giving the book away but the article material is not in the book because the book was too far along.

Also, for the average woodworker this discussion is "pointy headed"
as they say here in Wisconsin. A lot of readers want to know where to set the gauge, period. As my father would say, "Tell me the time, not how to built a watch". This discussion is deeper than the average book reader wants to go.

And I agree with your apples and oranges comment. The article was an attempt to sort out some of the confusion about blade tension. Unfortunately, no one that I know of has really done testing and research that is practical to the average woodworker. I would like to do a lot more testing and I'm getting good feedback from the technical types reading and contributing to this thread.

There is a lot of unsubstantiated material in articles and catalog copy and no one seems interested in sorting it out. I talked to the people at Fine Woodworking about correcting the misinformation in previous articles but they want to “move on” with new material.
This kind of discussion on the web really is a good source of feedback for people. We are discussing the idea of the new book having a website so we could update readers on new developments etc. I’m hoping this new book I applicable for a long time but to keep it current it needs to be supplemented by the web.

Traditional publishing is a very large, powerful and entrenched industry that does not always serve the group that wants to go a "little deeper" into a subject.

Mark

[Ken Garlock]

If I were to use these instruments I would make sure that the dial indicator was already pre loaded though to remove the mechanical hysterisis of the device at the zero point.

Mike, the Ittura gauge instructions specify at least one full revolution of the dial indicator when tightening the lower clamp. It appears that loading of the gauge didn't help in the tests.

[Mark Duginske]

Jim Dailey has offered to lend me his Iturra gauge for more test. I would like to test three gauges attached at the same time so if someone has an Iturra gauge that they would lend me I would appreciate that.

I'm think about building a fixture from which I would hang the blade and design something that would allow me to a add weights to the bottom of the blade. That way, when I apply weight to the blade, it will always be exactly the same for each test.

I'm open to suggestions for testing.

Mark

[Ken Garlock]

in the night.

Personally, I came to the 25-30,000 pound number after reading the tension article in the Ittura Designs catalog. In that article, Lou Ittura states that he contacted Lenox tech people, and that they "frequently recommend" 30,000# for their blades(2005 catalog, page 33.) I also know from various posts on this forum and others that people run their Lenox carbide blades in the area of 30-35,000#. These numbers seem reasonable since 30,000# is within the elastic limit for steel. It seemed reasonable that the more axial tension on the blade, the less lateral deflection there would be, and hence less potential for blade drift.

[Joe Mioux]

I have been scaning this thread with some interest.

Iturra's product has been beaten up pretty good here.

Is he a member here, if not, I would sure like to hear his response to some of the previous comments.

Joe

[Pete Bradley]

I'm open to suggestions for testing.

Well, since we're already in the spirit of academic endeavor:

Based on my back of the envelope calculations, it seems to me the challenge of the mechanical gauges is that they have to measure a very small delta, a few thousandths at most. A minute amount of slack or bend in the band when you start or any slippage will cause error. It doesn't make sense to me that anyone else would be marketing such a thing to the average weekend woodworker.

You might be able to find a college nearby with an Instron machine and an inclination to help. This machine will apply a controlled load to a piece of band, and at the same time it measures the strain.

Failing that, your idea of the weights has potential You'd want the top end clamped solid, and the bottom end to go through a slot, then attach to something of moderate weight that will carry the real weights. That would minimize slack while attaching the gauge. You're going to need a very accurate measurement of band thickness and minimum width too.

Pete

[Mike Cutler]

I have been scaning this thread with some interest.

Iturra's product has been beaten up pretty good here.

Is he a member here, if not, I would sure like to hear his response to some of the previous comments.

Joe

Whoa.... This is definitiely what I did not want to have happen. I have spoken to Louis only once on the phone, and found him to very engaging and articulate individual that really cared about the products and service he provided. I also believe that if others are having problems they should contact him. He may not be aware of the issue(s), if they exist.

Any remarks or observations I have made on the Ittura tension gauge can only be confined to the one that Mark currently has in his possesion, which I believe has been damaged somehow, or not manufactured to Louis' spec's.

If my lack of typing,or english skills have implied that I somehow believe the Ittura tension devices, as a whole, to be inferior. I absolutely do not.

[Alan Schaffter]

I have read Mark's article and the two SMC threads with interest.

I thought I wanted an after market tension gauge until reading and thinking about these threads. I think what too many of us are/were searching for is a more precise, repeatable, and QUICK and EASY way to adjust our bandsaws to some likely totally illusive goal- a number representing the "correct" tension that takes our own abilities out of the equation.

As Mark and others have alluded, in reality there are no precise standard tension- because of differences in blades (steel, size, teeth, etc.), saws, tires, wheels, etc. We are trying to replace experience, knowledge, and, yes, craftsmanship (with all that entails) with a number from a gauge.

We may also be forgeting about the size and properties of the material (wood) we are cutting as well as the condition of the blade at the time. I really like this quote from the Carter gauge info (the most accurate gauge according to Mark's article): "CAUTION: The ETG displayed value is primarily intended to provide a relative indication of tension and a method of achieving repeatability." (I added bolding). Think about what that means. It doesn't sound too good, but beyond that it only affects the blade being set to some previous setting - but what setting, and how was it determined? It is also without regard to the current condition of the blade or the material being cut.

So, except for Mark and those of you who have weighed in here, who are successfully setting their tension by testing and experimenting (with experience gained over- years?) and getting good cuts, or at least ones that satisfy you, the rest of us need more experience. In my mind, that pretty much negates the need for an after market tension gauge. I still want an easy way to do it that I can trust more than my own craftsmanship if someone can come up with one - it appears no one has figured out how to do that yet.

Also, in order for any gauge comparison test to be valid, the test samples and conditions have to be identical - identical blades (same run of steel, same tempering, age, handling, welded exactly the same way, on the same saw under the same conditions.

[Mark Duginske]

Ken Garlock and Jim Dailey have offered to lend me their Iturra gauges.

So I'm planning on building a fixture from which I can hang a 1/2" blade (like the one in the article). I plan to hang 146 pounds from the blade with the three gauges installed. This will duplicate the set up in the Woodworker's Journal article.

I'm not singling out one manufacturer or retailer of these gauges in particular but the fact that all three do not read the same does mean something. I was reluctant to give the numbers but the article was also criticized for not giving the numbers. By doing more test we can see
if one of the gauges is damaged. Even forgetting the Iturra, what accounts for the big difference between the Lennox and the Starrett.

I plan on doing 10 test with the gauges installed and 10 test with the gauges reinstalled each time. Mike C. and Ken G: how does that sound?

I will also use the set-up to check the two Lennox gauges which thus far given identical results.

I got the number of 15,000 from Starrett for woodworking band saws.

If you have testing suggestions let me know.

Mark Duginske

[lou sansone]

There is some confusion about the origin of the psi recommendation.

Most saw and saw blade manufacturers recommend 15,000 psi. I think Delta on their new 14" 5X saw recommends 13,000 ???. It looks like Carter recommends 12,500 (read Mike C's comments).

This number keeps getting refereed to but I have not been able to track down where the number came from.

IF SOMEONE READING THIS POST (MIKE C. OR KEN G.) KNOWS WHERE THAT NUMBER CAME FROM PLEASE LET US KNOW.

Mark

I did a little more research which confirmed what I had suspected about the flutter method and bandsaw tension gages. Namely, all of this has to to do with the harmonic vibration of the blade. Much of this research is in the area of "vibrations" ( all you mechanical engineers must remember these undergrad courses with joy ). There have been many technical articles written in various journals about bandsaw blade vibration which results in "wash-boarding" or other types of poor cuts. Unfortunately I don't subscribe to any of those journals and don't live close to a good university library that would carry them. But even looking at the abstracts, one can see that all of the blade tension stuff is really concerned with suppressing the fundamental or higher order harmonics in the blade. If someone has access to these articles, here are a couple.


http://www.springerlink.com/content/l4114136h2052t7j/

http://www.thinkerf.com/Downloads/StrainResp.PDF you can read this one

http://apt.allenpress.com/aptonline/...e=01&page=0019

there are many other articles as well that one could spend lots of time on.

It does look like blade tension is a critical factor in cut quality, the problem is determine that tension and adjusting it for the conditions. It seems that just throwing out a value of 15000 psi is just way to vague if you really want to get to the heart of the problem, which is vibration and natural resonance of the blade. It is obvious that a certain amount of tension is needed to make sure you have a stiff blade, but that is only one part of the total solution. If mike is still working on his book, I would think that someone from thin kerf technologies might be good to get in touch with.

lou

[Mike Cutler]

To "calibrate" any of the mechanical tension shouldn't require an elaborate setup. Everything about the mechanical devices, is a math function.

If I were to calibrate these I would remove the independent variables. The blade, and the human. We are trying to quantify that the device is capable of providing an accurrate representation of force for a given change in distance between the two points.
Once this is established, any unexpected value can be traced back to the user, or the blade.
Using Youngs Modulus value for Carbon Steel, 29,000,000 and trying to achieve 15,000 psi. The amount of strain in inches would be 15,000/29,000,00 or .00051724 in/in.
The value referenced in the Carter Info of 12,500psi would be 12,500/29,000,000 or 0.00043103 in/in.

Multiply the in/in values by the distance between the two clamps on the mechanical gauges, after preloading the device, and this would be the change in distance bewteen the two points that should correspond to a change in tension gauge indication of the calculated amount. A calibrated, or even commercial quality inside micrometer could be used to make these measurements.
Any blade tension value in psi (strain) can be worked using Youngs Modulus, as long as the measurement falls within the linear range of the metal, and doesn't enter the plastic range.
Using the blade cross sectional/width formula provided by Carter, and working it backwards would result in the expected Carter indication in wheel load, or lbs. This is all math, there are no variables yet. Once we know that the device is "accurate" we can look for discrepancies somewhere else.
I'm thinking of a test fixture to this. I think one of the MicroFence Router guides could be used/adapted as quick test fixture to change the distance between the two contact points in a controlled linear fashion. The threading is 50tpi, and there is enough strength between the parts. An independent measurement device could also be incoporated. A possibility.

The variables as I see them are these.

The mechanical tension gauges were designed around the particular steel that was in use in steel bands when Starrett first made the device (The Lennox and Ittura are copies o the Starrett). Differing types of steel are going to have different properties and Youngs Modulus values. This would throw the scale readings off. The amount of difference in the Youngs Modulus values could be established and a bias factor algebraically summed with the tension gauge indication to give a better approximate value for that particular blade.

If we just hang weight off the bottom of a bandsaw blade we can get a fairly decent comparison. Attaching all three simultaneously to a single band and then suspending the weight could be effected by the non linearity of the area between the clamp points, and the clamps themselves. Although I doubt very much.
Additionally the blade would have to be new. An old one would be suspect of micro cracks and stress induced fatigue. Once again the blade could be non linear throughout the range that the devices are attached.

Just some thought. It looks as if it's going to be a nice day outside today. Hopefully I can get some yardwork, or shop time in. LOML has a little project for me.