J. Phil Thien's Bandsaw Audible Tension Gauge

Step #1. Determine the weight of your bandsaw blade. To do this, weigh the blade on a digital kitchen scale set to GRAMS. We use grams instead of ounces because most kitchen scales are more precise when resolving in grams.

blade weight (grams!)
Want to give it a test-spin? Try entering 84 grams, the weight of my 72-1/2" Blade Runner blade.

Step #2. For our computations, we need to know the length of the blade, in inches.

blade length (inches)
Want to give it a test-spin? Try 72.5, which is the length of saw blade for my saw.

Step #3. Now we need to know the thickness of your bandsaw blade. For best results, measure with one of those inexpensive digital calipers. You should also be able to get this data from the blade manufacturer.

blade thickness (inches)
Want to give it a test-spin? Try .022, which is the thickness of my Blade Runner blade.

Step #4. We're getting there. Now we need to know the blade's width at the narrowest section (in the gullet). I'll get a picture up how to measure this up soon. But basically, you want to measure the blade body (not counting the teeth).

blade width (inches)
Want to give it a test-spin? Try .390, the width of my 1/2" blade runner excluding the teeth.

Step #5. Almost done. Now we need to install the blade on the saw, and tension it. Just take the slack out of the blade, that's all the tension we need. Now, measure the distance between the two centers of your bandsaw's wheels. This is our beam length, or how long of an unsupported distance the blade spans.

wheel separation (inches)
Want to give it a test-spin? Enter 19.75, the distance between by axles when the blade is pre-tensioned.

Step #6. Now we just need to know the tension you desire. Manufacturers of blades typically suggest everything from 15000 up to 30000-PSI, with 20000-PSI pretty common for woodworking blades.

desired tension (PSI)
Want to give it a test-spin? Enter 15000, the PSI suggested by the manufacturer for my Blade Runner blade.

Step #7.

Enter the values above, then click on the "calculate" button to determine the tuning frequency...

If you were following the test-spin, you should have received an answer of 78.95-Hz.

Frequency samples
Once we've computed the frequency, you can use the files below to hear what a blade tensioned to that frequency should sound like. These are MIDI files, which should open in MS MediaPlayer (or any other suitable MIDI player).

20.601-Hz (midi16.mid)

38.890-Hz (midi27.mid)

73.416-Hz (midi38.mid)

138.591-Hz (midi49.mid)

21.826-Hz (midi17.mid)

41.203-Hz (midi28.mid)

77.781-Hz (midi39.mid)

146.832-Hz (midi50.mid)

23.124-Hz (midi18.mid)

43.653-Hz (midi29.mid)

82.406-Hz (midi40.mid)

155.563-Hz (midi51.mid)

24.499-Hz (midi19.mid)

46.249-Hz (midi30.mid)

87.307-Hz (midi41.mid)

164.813-Hz (midi52.mid)

25.956-Hz (midi20.mid)

48.999-Hz (midi31.mid)

92.498-Hz (midi42.mid)

174.614-Hz (midi53.mid)

27.500-Hz (midi21.mid)

51.913-Hz (midi32.mid)

97.998-Hz (midi43.mid)

184.997-Hz (midi54.mid)

29.135-Hz (midi22.mid)

55.000-Hz (midi33.mid)

103.826-Hz (midi44.mid)

195.997-Hz (midi55.mid)

30.867-Hz (midi23.mid)

58.270-Hz (midi34.mid)

110.000-Hz (midi45.mid)

207.652-Hz (midi56.mid)

32.703-Hz (midi24.mid)

61.735-Hz (midi35.mid)

116.540-Hz (midi46.mid)

220.000-Hz (midi57.mid)

34.647-Hz (midi25.mid)

65.406-Hz (midi36.mid)

123.470-Hz (midi47.mid)

223.081-Hz (midi58.mid)

36.708-Hz (midi26.mid)

69.295-Hz (midi37.mid)

130.812-Hz (midi48.mid)

246.914-Hz (midi59.mid)

PLEASE NOTE: As I'm still in the testing phase, I encourage you to proceed with caution. If it seems like you are overtensioning or undertensioning, please contact me and let's work through it. You assume all responsibility for using this site and my method.

More on tension
The topic of bandsaw tension is perhaps second only to dust collection in its ability to spark heated discussions among woodworkers.
My current bandsaw (a Delta BS150LS) has no tension gauge. It is a direct-drive 10" bandsaw with a 1/2-HP motor, and 7" of resaw capability (and I've resawed 7" stock with it, too). But like I said, there is absolutely no tension gauge to guide the user in setting blade tension. The manual instructs the user to turn the tension knob until all the slack is removed from the blade. Then, depending on the size of the blade, the user is instructed to turn the knob from 3 to 7 more times (3 turns for a 1/4" blade, 7 turns for a 3/8" blade).
This has worked sorta alright. The results were very repeatable, but I had no idea if I was achieving the 15k to 25k-PSI tension that the band manufacturers typically recommend. Thicker stock (resawing) tended to give me more problems than cutting thinner stock. I often assumed the tension had something to do with that, but I had no way of knowing for sure.
I had even greater difficulty when I decided to try a 1/2" Blade Runner band on the saw. The Blade Runner (from Iturra Design) is made from the same Atlanta Sharptech stock as the Wood Slicer blades from Highland Hardware. These blades have done very well in the reviews (winning the Fine Woodworking blade shoot-out, for example). But, the manual for my saw offered no guidance in tensioning a 1/2" blade. While the band fit, and I seemed to have no problem achieving lots of tension with it in place, I had no idea how many turns of the tension knob were required to properly tension it.
So I considered purchasing an aftermarket tension gauge. Typically selling for $100 to $300, these clamp to the blade and measure how much a band stretches as it is tensioned. Young's modulus is used in order to convert the measured stretch to tension (in PSI). However, Mark Duginske, a noted bandsaw expert and author of at least a couple of bandsaw books, has argued that these aftermarket gauges fall short. Testing performed by Mike Cutler (a materials engineer) on Mr. Duginske's behalf indicated that the gauges are neither accurate, nor repeatable. Mr. Duginske announced these findings in one of the woodworking forums and both Mr. Duginske and Mr. Cutler had to immediately begin defending their findings. Several users of the gauges were unwilling to accept the results of the tests. However, it is clear that the tests were exhaustive, with standards traced back to NIST (National Institute of Standards and Testing). Mr. Cutler spent hours speaking with engineers at Delta and blade manufacturers. The results were accurate, the aftermarket gauges were not.
I was aware of shop-made gauges. These pretty much emulate the spendy aftermarket gauges I mention above (clamping to the blade to measure stretch). While they're certainly much less expensive to make than to buy ($20 instead of $100 and up), if the $100 to $300 gauges are neither accurate not repeatable, I didn't see the sense of making a copy of one. Yet on one rainy Saturday afternoon I did experiment with shop-made solutions. I had less than satisfactory results.
[The reason I imagine these gauges don't work well (in case you were wondering) is that they rely on the user's ability to get them on an untensioned blade such that the blade is perfectly straight and true between the clamps (any curvature in the blade will result in the user overtensioning the blade).]
Now, Mr. Duginske does give hope. During his tests, Mr. Cutler found the electronic gauge from Carter Products to be both accurate and repeatable. Still not inexpensive (at $200), it does get the job done. One significant limitation: It works only with Delta and Jet 14" cast-iron saws.
Mr. Cutler also found the gauges built into saw to be usuable. More accurate, and more repeatable than any aftermarket gauge save for the Carter electronic gauge. Mr. Duginske argues that there is no need to go to an aftermarket gauge if you already have one built into your saw. There are some caveats, however. For example, some saws have inadequate tension springs. The springs will not allow the user of the saw to achieve the amount of tension necessary to accommodate wider (1/2" and larger) bands. In those cases, the springs can be replaced by much stronger aftermarket springs. But the gotcha is that the scale on the built-in gauge may no longer be accurate. Users upgrading to those aftermarket springs are kinda on their own when it comes to tensioning their saws.
For users lacking a gauge or distrustful of the one present, Mr. Duginske offers one more glimmer of hope: Blade plucking. One plucks a tensioned blade on the side of the saw opposite the guides. Mr. Duginske provides some instruction on what to listen for in the text of his book. While this method helped me return a blade to a tension that had previously worked well for me, it didn't offer enough guidance in absolute terms. That is, I still had no idea if I was setting the band to 10k-PSI, 20k-PSI, or 50k-PSI (yikes).
Recently, there have been some attempts on woodworking forums to compare notes (sorry for the pun) on what plucked blades sound like on members' saws. I read a poll on the Family Woodworking site, and started a poll on SMC. I found the results and ensuing discussion enlightening.
And then I had a eureka moment. Back in high school physics, we did a simple lab where we tensioned a steel wire between two posts. With very simple math, we were able to determine what fundamental frequency a plucked string would generate. The only variables were the amount of tension, along with the length and mass of the string. There is absolutely no reason this lesson couldn't be applied to bandsaw blades!
I was able to refresh my memory of the math involved, and create the above worksheet to help me in the process. And of course, everyone benefits as it is on the www.
An interesting aside: While experimenting with my BS150LS in the days and weeks after purchasing it, I did break a few blades (new, expensive ones). Now that I've been able to compute the frequency my plucked blades should generate at various PSI's, I have come to an alarming conclusion: Those broken blades were WAY overtensioned. My best guess (it has been months) based upon my aural memory of those plucked blades is that I was tensioning them to 60k-PSI or even more. I was able to derive this using this worksheet and specifying higher and higher PSI's until the sample file for the suggested frequency sounded similar to what I heard when I plucked those now broken blades on that fateful day. Now I know.
I'm still in the process of testing my method. I've spoken to numerous engineers and even bothered one of my old physics professors about some of the more subtle aspects. But the consensus of these experts is that I'm onto something. I will hopefully have either one of the Carter electronic gauges or another load-cell based system soon and will be able to directly correlate my results.
Stay tuned (this subject lends itself to puns. Sorry)...

Questions & answers
Q: I have used your worksheet to compute the frequency for some of my blades. The suggested frequencies seem quite low (approx. 40-Hz) compared to your example. I was anticipating a low E (82.4-Hz) that was suggested in some of the recent polls. Am I doing something wrong?
A: Not necessarily. The parameters you provided me (185-grams, 131.25" long, .024" thick, .408" wide, 39" wheel separation) indicate that this is a BIG saw. As the distance between the wheels increases, the frequency of a plucked blade drops quickly (assuming a constant PSI). Using your numbers and a target PSI of 20000, my worksheet suggests a frequency of 44.7-Hz. If I keep increasing the target PSI by 10000 until I reach a frequency of approx. 82.4-Hz, we find that your blade would be tensioned to almost 70000-PSI (yikes) to get you to 82.4-Hz. Clearly WAY too much.

Q: Have any other examples of how wheel span influences the derived frequency?
Sure. Let's take the example of my little BS150LS. A 72.5" Blade Runner weighs 84-grams. It is .022" thick, and .390" wide (remember, we're excluding the teeth). My wheel separation is 19.75". The manufacturer suggests a tension range of 15000 to 17000-PSI. So if my target PSI is 15000, then I should tune the blade to 78.95-Hz.
Now, let's assume that they made a 6" riser block for my saw (they don't, but this is strictly theoretical). The blade length would increase to 84.5" (twice the length of the riser block), and the weight of the new blade would be 97.90-grams ((old grams/old length)*new length). My new wheel separation is 25.75 (adding the 6" riser to the original 19.75"). Plugging these new numbers into the worksheet and targeting the same 15000-PSI, we find we have now reduced our tuning frequency to 60.55-Hz. Interesting, no?

Q: A couple of years ago I purchased a used 14" Jet bandsaw. It has a riser block installed, and included two blades. I haven't had much luck with it. It works fine for cutting thin stock. But resawing stock even 3" high often results in the blade curving in the cut. It gets noisy and starts to fight me. I've adjusted the tension as high as I can without fully compressing the spring. So tonight I went through the process of weighing the blade and measuring everything. I computed the frequency using your worksheet (approx. 60-Hz) and listened to the sample file. Even when I have the tension cranked nearly all the way to the max, my blade sounds considerably lower when I pluck it (about 39-Hz). Is it possible that my saw can only do 6500 PSI?
In a word, yes. Both Delta and Jet have apparently had problems with some of their springs. I'd suggest purchasing a replacement high-performance spring from one of the woodworking catalogs. Once you replace your current spring, reinstall your blade and try tensioning it again. With the new spring installed, I'd shoot for a tension of 15000-PSI and try resawing some 6-8" high material. If your blade still curves, try taking it up to 20000-PSI. Let me know how that works!

Please respect my time and effort by not taking credit for my work. Please don't submit my methods as a tip to a magazine, or attempt to make a commercial product based on what you see here. I reserve any and all rights possible to the concept of an audible tension gauge.

Changelog
5/22/09: First release. Waiting for input from testers...
5/25/09: Added some MIDI files which users can compare to their plucked blade. It was a lot of fun learning about MIDI (sarcasm), but at least this should largely negate the need for a chromatic tuner.

--Phil
phil@cgallery.com

This page has been viewed 38663 times.