Expressing relative blade speed

Sorry for bumping this but in fairness it's very much alive with the spreadsheet being continually updated.

Any of you engineering types have insight into the effects of inner carbon or composite layers Vs outer in terms of frequency measurements?

Often it is said anecdotally that inner carbon blades are woody at low speed and then the carbon kicks in at higher speeds giving it additional gears.

Looking at the frequencies provided for the inner carbon blades they seem very much similar to all wood blades. In fact you can see some acoustic inner carbon results are lower than normal acoustic. Is the frequency rather unreflective of speeds at higher impacts for inner carbon (or stuff like Balsa core) blades?

Added two blades

Just bookmark the link...

I've used this to test a couple of broken bats lately.  Do you think the tone would be different to a normal bat if

Four Friendship/729 V-6 blades:

Edited by book4all - 07/27/2017 at 11:51pm

I know I've posted these links before, but now I do it in a different context.

These papers describe some of the equipment that can be used for a professional analysis:
http://www.sciencedirect.com/science/article/pii/S1877705812017316
http://www.sciencedirect.com/science/article/pii/S1877705814005840

I find it particularly clever when they use a thin layer of fine sand to visualise the modal lines and mode shapes. More complex and costly approaches use laser techniques for measuring distances.

Similar analysis techniques are used by ARTTE (an new Italian brand).

I don't have any insight on what equipment XIOM, ITC and other brands use for measuring their blade performance indices, but I suppose that it's not just surveys among testers.
In his previous TT blog (on a service closed by Apple), igsstern was assessing blades based on indices derived from objective measurements (he never disclosed how he measured, though). If I understood well, he was also developing blades for Adidas and now for ITC. So I assume the performance indices of ITC blades are based on similar objective measurements.


Edited by arg0 - 11/16/2017 at 1:56am

Just send all blades to one people to record and analyze the results. It is more accurate and consistent. 

Simply and robustly estimating decay time is not simple. We could try to use HWHH (Half-Width at Half-Height) as decay time, but that's still requires human intervention and will lead to human error, unless we find an automated way to do this: I suppose that fitting an exponential function to the audio amplitude may do the trick.
Then again, we're not even sure how decay time influences speed, and we would need to have a mechanical model or experimentally validate any formula we come up with, or otherwise it's just guesswork.

Another suggestion. Put a yardstick against a wall, and drop a ball from 1m height (or any suitable imperial length) on the blade and see how high it bounces. I imagine a layman gets about 10% error in the measurement, a tt player 5% ;-) This allows to determine how much energy is absorbed by the blade or lost by friction with air and how much returned to the ball. To factor air resistance out, since working in vacuum is not an option for many, we could use tiny stainless steel balls (e.g., from small ball bearings), but not everyone has some laying around. In alternative, you're only allowed to do this test at sea level :-D.
Then again, tests would need to be performed to compare low-speed and high-speed impact.
If we want to go down this road, better build some professional testing equipment somewhere and collect money to ship blades to the facility and back to their owners...

Or, let's stay simple and let's see how far we can go with just a racket, a ball, a yardstick and a smartphone.

Those are some good points you make, arg0.

Some comments:

The Hinotec just has Hinoki outer plies. But it might indeed be the special attribute of hinoki that leads to this second peak. How this affects speed, we can't say yet.

I also think it would be very helpful to measure blade frequency response at high impact, and see what changes in the frequency chart occur.

For more accurate results, the elasticity / stiffness would indeed also need to be considered. Maybe we can even approximate this by dividing the peak frequency through the decay time, as the stiffer blade feels faster (at least from low to medium impact) even if freq are the same, which can be corrected by such a formula.

Hozuki wrote: Not so weird. I have already shown an inherent flaw in this method, which provides one possible explanation for this occurence, but nobody seemed to care. To see whether it applies to this specific case, the full frequency chart is neccessary, though. I'll make one more attempt at explaning it in more detail. Blades can vibrate in multiple frequencies that can be very close together in terms of pitch and volume. This is especially evident in ALL+ Blades. Here is an example: (Hinotec ALL+) [image removed] Now I ask you: Which frequency would you pick to insert into the table? 1005? 1500? Or 12000? As you can easily see, there is good potential for confusion. In fact, the first two peak freqs only have a 0.3 db difference in volume. Thus, a slight variation in blade weight, phone mic freq response or some other factor might lead to big differences between measurements of different people. So what can we do to make things more accurate? Easy. Just don't rely on a single indicator that condenses the frequency chart into one freq peak number. Either find a better one, or use multiple indicators. Based on this, we can now all start thinking what would be suitable. [...]

Is the Hinotek an all-Hinoki blade? If so, from experience and gut feeling I would pick 1500 Hz as the membrane mode, and 1005 Hz as a bending mode. The reason is that all multi-ply hinoki blades I measured (with the exception of Nexy Tamar), tend to have a membrane-mode frequency which is much higher than their speed class: compare with Bty Kiso Hinoki V, and TSP Hinoki Pure in my data (not all measurements in there are by me, but those are). I had a similar issue with Nittaku S-5 (all-spruce), which I returned without even testing because of the high frequency. So it must have to do with hinoki/spruce/conifer wood.

As to the double peaks, I also had this a few times, especially for 1-ply blades (Nittaku Miyabi and American Hinoki Ancient Kauri), and could not decide. I would need to look at the frequency spectra again, but if in your case I picked 1500, I suppose that for consistency, I would have to pick the higher frequencies.

As to whether and how those frequencies are related to speed, see my doubts in my previous posts. I suppose that for measuring blade speed we need some way to estimate the blade's elasticity/damping index (to use a simplified name). And most likely we'd have to do this for low speed and high speed impacts, to take non-linearities into account. Not easy to do by hand, a ball cannon would be more appropriate...

My impression is that these vibration mode frequencies are more related to the stiffness feel of the blade, than plain speed. Just, it often happens that the stiffer blades are also the faster.

Or, if they somehow represent speed, by the way we obtain these frequency, by just dropping the ball on the blade, they can at most represent speed of the blade at low impact speed.

It's late here and I cannot perform this test without waking my son, but anyone can make a quick test to see whether the frequency of the main peak varies by just dropping a ball on the blade or by hitting a ball as hard as you can with the blade? Try to grip the blade with the same strength in both cases.
Or better yet, grip the blade with low force and record the sounds of: 1) a ball dropped on the blade while you hold it losely, and 2) the same ball being driven very fast on the blade by someone while you hold it losely, if that makes sense.
Save

Update: I don't have had the time to review all my posts below yet, but I have played again with Higgs and, after all, and despite being a 7-ply blade, my impression is that it is not a fast blade (though it's advertised as being fast). Actually, it's speed is not much higher than Violin's.

Amplitude and frequency spectra of a single ball bounce in similar experimental conditions (amplitudes were normalised).

The membrane modes of Nittaku Violin and Nexy Higgs have a similar frequency, though Higgs is considerably faster. The vibration amplitude of Higgs decays much faster than Violin.

Yinhe T-7 and Tibhar Inca share a similar speed (I never played the two side-by-side, however from memory the T-7 is faster than Inca), and also a similar membrane mode frequencies. The vibration amplitude of T-7 decays much faster than Inca, though.

The speed certainly does not reflect in the amplitude of the frequency peaks.
Can anything be said about the amplitude decay rate?



Save

Edited by arg0 - 10/27/2017 at 3:45pm

By means of this acoustic frequency analysis the eigenfrequencies (or frequencies of vibration modes) of the blade can be determined. However, these cannot be directly matched to blade speed. To know how fast a ball will travel after impact, one needs to determine how much of the incoming energy is transferred back to the ball, and how much is absorbed by the blade, instead. This is a separate parameter (damping) and needs to be determined experimentally.
I've just started thinking about this. One could try to measure damping by how high (in %) the ball rebounds after impact (though I don't see an easy way to reliably measure this), or indirectly by analysing the decay of the sound after ball impact. Damping also reflects on the width of the frequency peaks, but given how messy the spectra are, it will not be possible to reliably measure the width of the peaks in the audio spectra. Analysing the decay time of the vibrations of a single ball impact is also not trivial, as I'll try to show in a next post.
Any thoughts?

Looks good to me.

AndySmith wrote: NextLevel wrote: AndySmith wrote: h0n1g wrote: Got an ITC Premier XR ST today, 90g heavy, clocking in at 1442. ITC Premier XF, FL, 91g, 1399 ITC Premier XF, ST, 90g, 1356 Gewo Zoom Balance ALL+, ST, 92g, 1464 Gewo Zoom Balance ALL+, ST, 91g, 1421 Hmmmm. Are you holding the blade too tight? Those are offensive blade scores. I tried to hold the blade as softly as I could, and only by the handle (not the blade face at any point). Although I did say in my initial review of the Zoom Balance that it was faster than I had expected an allround blade to be.  But still, the numbers are high.

How thick is the blade?

I do have my usual fingers on the blade face but I suspect that doesn't matter as long as you don't grip it as if it is in a vice. I could not play with a blade that fast comfortably hence my surprisr.

Edited by NextLevel - 07/04/2017 at 1:03pm