Resonant frequency ski test

[Than_Bogan]

I mentioned in another thread that it might be possible to measure the resonant frequency of a ski and (if so) that it might be somehow meaningful in detecting material degradation ("sk breakdown").

I still have no evidence for the second part of that, but today I conducted a simple experiment that suggested a ski does have a fairly consistent resonant frequency that is not very challenging to measure.

I used my iPad's Fine Tuner app, and I grabbed an old Goode 9900 that had no bindings on it.  Then I just held it in various places with one hand and smacked it in various ways with the other hand.  The frequency measured was shockingly consistent at 115-116 Hz.  To be expected, there were also occasional readings of half or double (58 or 230), which is how resonance works and, if anything, makes a stronger argument that I'm actually measuring an inherent property of the ski.

Mildly interesting.  If anybody has an iPad or iPhone, get the free Finer Tuner app and then grab some skis without bindings on them and see if we discover anything.

(Most likely we won't, of course, but useless experiments sometime lead to better ones in the future!)

[aupatking]

@Than_Bogan I have two T3’s. One perfect and one broken. I’m very curious

[Drago]

Another way to look at alpine skis is their camber goes away.

It fairly common to MRI carbon bike frames to see if they are compromised. The tester has to be very trained, however, and I don't think it's cheap. I suppose we should ask ourselves if it matters ? 
The most 39s I ran were on a ski I bought on SIA for $400. I imagine the first owner thought it was dead.

[chrislandy]

@Than_Bogan you're basically describing the method we use for concrete pile integrity testing.

Essentially we put a microphone on the side of the pile and hit the pile with a hammer and record the frequency response.

 

Out of interest you could look up other forms of non destructive testing (NDT) in civil and structural engineering, they are quite transferable

[Horton]

would it make sense to mount the ski to a fixture at a fixed distance from the tail, then pull from the tail so the ski bends by a fixed distance, the release it and measure the rebound with accelerometer... something like that 

[Slalom.Steve]

I just did this with my new '24 Vapor Pro. But then I added 0.005" of depth to the fin and the resonance numbers came back totally different.

^that's a joke😉. For the serious part:

If I held the ski near the center, I was getting pretty consistent 107hz (+/- 1 or 2) readings no matter where I hit the ski. But when I held the ski off-center, I was getting 39-40hz, again regardless of where I hit the ski. Definitely could subjectively hear the lower tone too, the sound felt a little louder and more "full" and the ski vibrated/hummed for longer as well.

[rawly]

On 11/10/2023 at 5:55 AM, Horton said:

would it make sense to mount the ski to a fixture at a fixed distance from the tail, then pull from the tail so the ski bends by a fixed distance, the release it and measure the rebound with accelerometer... something like that 

 Duplicating the resistance of the water might be the tricky part. That would play a role in both comprehension and rebound. 

[rawly]

22 hours ago, rawly said:

 Duplicating the resistance of the water might be the tricky part. That would play a role in both comprehension and rebound. 

Compression 

[skibrain]

If a guy had lots of slalom skis around, he could make some sweet wind chimes.  

To that end, I’d really like to hear the results of a like-model 63” 65” 66” 67” 68”  resonance test. 

[dchristman]

2005 65" Monza 104

2009 65" RS-1 105

2017 65" ARC-S 110

2021 XS C-85 116

Would be interesting to see if all size C-85's are 116ish.

[Drago]

or if all Monzas just sit there on the couch watching Friends reruns. 🐕 

[BraceMaker]

On 11/11/2023 at 6:15 PM, rawly said:

 Duplicating the resistance of the water might be the tricky part. That would play a role in both comprehension and rebound. 

Maybe I'm missing the point here but the theory as I see it is that the typical flex testing of a ski doesn't really show break down that might be shown if you had more of a high frequency or resonant frequency test.

As I see it you want to take a blank ski no bindings no tapes no fin box, design something that holds the ski with minimal to no support and induces a vibration.  Like vibration generator suspended from a beam with a clamp that grabbed the very tail of the ski maybe with a tab that when through the fin slot and then a rubber U shaped clamp.  You'd then fire up the vibration generator and sweep up and down a range of Hz and record the ski with a laser doppler vibrometer and plot the data.

You could then take skis and test them and ask the skier to report their perception of the ski and record what its frequencies looked like.

[Horton]

@BraceMaker I am NOT an engineer but I always imagined something even simpler. As you said, remove fin block, bindings, tape, and everything not the ski. Clamp the ski in a fixture ( maybe with the use of the inserts - or not - IDK ) with some standard number of inches of the back half of the ski unsupported. Then using accelerometer mounted to the tail or resonant frequency or such measure the rebound when the tail of the ski is pulled some distance and then released.

I assume that you would only need one number per test. You have a new ski result as a baseline and then we are looking for any delta with older skis.

[BraceMaker]

@Horton right to me what you're describing is almost more of a rebound tester.  You're sort of viewing the ski like a bow limb drawing it back and releasing it booooiiiinnnnngggggg and then you'd read like initial acceleration and how long it vibrated and at what frequency.  Its perhaps less resonance than a simply a dynamic version of a flex test, what I'm visualizing would be to vary the frequency and map the ski which in theory would vibrate in different ways in different zones at different inputs which if we don't really have a clue what the skis are doing when they go bad or if they are actually doing anything then I'm thinking kind of like this, but upside down and at higher frequencies with less displacement.

 

Which is why I think avoiding the inserts is probably a good thing because you wouldn't want to damage it but you could fairly easily grab the last 2" of the ski.

 

 

[Horton]

@BraceMaker Yea I am a bit out of my depth on this stuff. I guess I am looking for a rough indication of structural change. More to the point I am trying to imagine a solution that someone could actually construct on a hobby scale and operate with consistency. 

[BraceMaker]

Its easy to see the value.  If your ski still checks out your wife won't let you buy a new one.  But see honey I've lost 23% of the initial rebound.

[03RLXi]

1 hour ago, BraceMaker said:

Its easy to see the value.  If your ski still checks out your wife won't let you buy a new one.  But see honey I've lost 23% of the initial rebound.

Her reply "and if you use that 23% to justify yet another new ski you'll discover what loosing more than 50% divorce feels like" 😂 

[rico]

Or you could just buy a new ski 🙂

[B_S]

Remember when O'Brien put an piezoelectric transducers on their skis back in the mid-90s, ostensibly to dampen chatter? That was pretty silly, but how about using one to monitor frequency/rebound in real time while skiing?🤔

[ToddL]

For consistency, couldn't one just leave the fin block attached and hang the ski from a hole in the fin?  Then, it is truly like a wind chime.  LOL.  But seriously, that would be a less invasive, repeatable method to hold skis for a "strike" frequency measurement.  However, then DFT might have an impact.  😝

[ForrestGump]

You're just pissing in the wind if you don't measure the angle of the dangle. 

[chrislandy]

12 hours ago, B_S said:

Remember when O'Brien put an piezoelectric transducers on their skis back in the mid-90s, ostensibly to dampen chatter? That was pretty silly, but how about using one to monitor frequency/rebound in real time while skiing?🤔

There's a much simpler thing called a strain gauge, we use them a lot to directly measure and compare surface (or subsurface) changes and compare them to the modelled (calculated) properties.

 

The best methods are the simplest, real world cause/effect is mighty complex so the easiest and most repeatable methods to check structural behavior is based on primary axis and first order static forces i.e. bending, shear, torsion, axial and then elongation & deflection in x/y/z axis. No matter how complex your design/model, the end result is you would not try and mimic the in service load cases for a ski as it is nigh on impossible. You apply "test" load cases in the model e.g. what it takes to deflect the ski 1/10" at certain points with certain supports and record the strain & deflection; Then with the ski in front of you, you do the same in the real world and compare the two sets of data.

 

This then implies that under other load cases the ski will behave in a similar manor; obviously this is a structural test method rather than a behavior due to surface contours.

[liquid d]

The new KLP2 plays Van Halen...so weird.

[ALPJr]

Must be getting cold outside 😎

[BraceMaker]

14 hours ago, ToddL said:

For consistency, couldn't one just leave the fin block attached and hang the ski from a hole in the fin?  Then, it is truly like a wind chime.  LOL.  But seriously, that would be a less invasive, repeatable method to hold skis for a "strike" frequency measurement.  However, then DFT might have an impact.  😝

I think the point would be if a ski has a resonant frequency which of course it does everything does then the fin block and fin are going to act like a tuner so you're introducing a variable that a  few thousandths of DFT is probably going to introduce more change than the ski likely undergoes in its lifetime.  

That's where I feel that if the real recommendation is resonant frequency tester the first step is simply figure out what the scale of Hz is for the skis is it 20 hz? 200? 2000? 2e10?

[Drago]

5 hours ago, ALPJr said:

Must be getting cold outside 😎

Omg...

[Horton]

I wonder of the Goode Carbon Core skis would test differently. Funny how that tech never created huge buzz. On paper it should mean those skis will last a VERY long time without breaking down. Since no one has a way to test skis we will never know.

[vtmecheng]

I’ll start by saying that I have no interest in going down this rabbit hole, it’s too close to my career and I just want to ski. Someone else can have this “fun.”

The hypothesis, as I understand it, is that ski use results in changes to the natural frequency of the ski. There are a couple of potential ways that I can think of to test this hypothesis. 1) Hold the ski in a support that is firm and consistent. It cannot change between tests. The frequency of the test setup must be designed so that it is well outside of the primary modes of the ski in the clamped configuration. Mount an accelerator to the ski. It must be hard mounted or the mount should provide consistent transfer of motion (I’d probably start with screwing into a binding insert). Hit the ski at a consistent location away from the support and acquire the acceleration for modal analysis of the frequency spectrum. Do this though out the ski’s life and see if the prominent powers change frequencies.

2) Clamp the ski to a vibration table with an accelerometer, following all the stipulations above for clamping and gauge mounting. Perform a vibration sweep, recording input and output accelerations. The transmissibility is calculated and the results compared though out ski use.

There may be other ways to test as well, these are just my first thoughts without really spending time on it. I can see some potential challenges. Consistent setup and measurements are critical. The data acquisition needs to be done correctly, with proper sampling rates and anti-aliasing filter applied. The clamp may not be strong enough to keep from creating its own frequency mode and not damage the ski. Also, the sensitivity of changes in frequency with wear may be too small for capture in these methods. Maybe both are no big deal. I like the idea of seeing if something like this provides useful info. I have used modal changes during “large input” dynamic shock tests in order to determine the time that plastic strain begins to accumulate (waterfall frequency spectra are great for that). Again, this is too much like work for me and I just want to ski. More power to whomever does something with this.

[Than_Bogan]

@vtmecheng Although interesting, I think you're kinda putting the cart before the horse.  There's a fair chance that resonant frequencies don't mean Jack about ski performance or breakdown.  My "goal" here is what it often is:  Fail quickly.  If we can do something kinda easy that suggests that resonant frequencies aren't very correlated to anything, then we can move on.  Only if they start to show some promise (or convince me that a simple test is fundamentally flawed), would I be interested in more complex test methodologies.

[vtmecheng]

@Than_Bogan I said it is a hypothesis to test, not that it would be a positive test. The problem with the easy approach alone is that you can’t know if the measured change (or no change) is caused by the ski changing or poorly controlled measurement techniques. I believe the first step is to conduct controlled experiments in order to determine if the hypothesis is true. If true, then determine if a simplified measurement technique can be used.

[Rednucleus]

So "what if"...  You guys develop some cool tests that can tell you when a ski is "no good" anymore.  Now Superskier brings you their formerly favorite ski and says it's no longer working well. You test it and it still is within "expected parameters". I'll bet that Superskier will still go out and get another ski. Conversely, Hackskier (like me) brings you my favorite ski several seasons old that still works great, still getting an occasional PB, you test it and tell me it's time to hang it on the wall - probably I will continue to enjoy it as it works well for me. 

My point being, you can test the ski any way you can imagine, but if it's working well, the user is going to enjoy it; if they are unhappy with it, time for something new to ride - regardless of a test result.

Kind of hoping this is Panda- worthy; been trying for one for a long time now!!

[Horton]

@Rednucleus on one hand your point is valid because we are talking about the unknown. This is speculation.

On the other hand, I think you are missing the point because common sense says that if a ski rebounds slower performance will suffer. My stance on this is that I am making informed speculation.

 

[VONMAN]

Back when I first started course skiing I had a O'Brien TRC loved the ski but was learning/ hacking my way through the course. Two summers later I find a brand new TRC sitting in the corner of my local ski shop with a cheap price so I buy it. Well it was like  born again skiing! The new ski was telling me the old one was dead.

Fast forward to my first Goode 9100 one summer old and I drop it off at Goode headquarters to get the bottom repainted. The next day I get a call that they flex tested my ski and it was dead. They ended up giving me a new ski at no cost.

Fast forward to 2014 I buy a one month old ski used and nine months later it's totally dead. Bottom line, we probably don't notice it happening because we are chasing the next new shiny ski for more buoys. Skis have come a long way since my TRC and I been buying the next new toy.