A method I’ve never heard of before

[Colleen McTigue]

I’m just watching a video by Peter Westerlund, where he discusses a method of using “scratch tones” and tap tones in making a violin. He apparently uses it to determine where to remove wood from a plate. His goal, as near as I can tell, is to have the same tones everywhere on the plate. I’d be interested in hearing what folks here think of this. The video:

 

[Don Noon]

IMO irrational hocus-pocus, which was discussed several years ago...

https://maestronet.com/forum/index.php?/topic/345990-peter-westerlund/

https://maestronet.com/forum/index.php?/topic/346094-what-you-can-learn-by-scratching-things/

 

[HoGo]

I'd call it a luthiers' form of WOODoo

[Don Noon]

2 hours ago, HoGo said:

I'd call it a luthiers' form of WOODoo

Or WOODon't.

[Will Turner]

I’m experimenting with a divining rod, similar to looking for where to do water wells.  I hold it both hands over the plate to determine where to remove wood.   This is after the scratch tone method.

[Oded Kishony]

We’re all trying to disentangle the mysteries of violin acoustics which are inherently very complex and often counterintuitive. No need to mock someone who is sincerely making an attempt.

That said, the demonstration would be more persuasive if the plate was at least graduated.

A ‘fact’ that’s often overlooked by plate tuning advocates is that there is strong evidence that the Cremonese makers (strad, del gesu  et al) finished the plates after assembling the corpus and inserting the purfling. This is evidenced by the fact that often the purfling bisects the locating pins. (see Roger Hargrave del gesu)

I’v often wondered what acoustical model the ancients had in their minds. My experience with contemporary makers who have not had any exposure to scientific violin acoustics is that their imagined theory never coincides with the science. And even with exposure prejudices inevitably creep in.

The one aspect of practical acoustics that I think is critical, is to be able to directly connect cause and effect. If I do “A” to a violin I can reasonably expect “B”. That’s what I search for.

[violinsRus]

51 minutes ago, Oded Kishony said:

A ‘fact’ that’s often overlooked by plate tuning advocates is that there is strong evidence that the Cremonese makers (strad, del gesu  et al) finished the plates after assembling the corpus and inserting the purfling.

It's entirely possible to 'tune' the plates after the body is assembled using his 'scratch tones' method.  Just sayin. 

But I'm not convinced, but who knows?  I agree that we aught to respect the effort.

[Violadamore]

1 hour ago, Oded Kishony said:

We’re all trying to disentangle the mysteries of violin acoustics which are inherently very complex and often counterintuitive. No need to mock someone who is sincerely making an attempt.

That said, the demonstration would be more persuasive if the plate was at least graduated.

A ‘fact’ that’s often overlooked by plate tuning advocates is that there is strong evidence that the Cremonese makers (strad, del gesu  et al) finished the plates after assembling the corpus and inserting the purfling. This is evidenced by the fact that often the purfling bisects the locating pins. (see Roger Hargrave del gesu)

I’v often wondered what acoustical model the ancients had in their minds. My experience with contemporary makers who have not had any exposure to scientific violin acoustics is that their imagined theory never coincides with the science. And even with exposure prejudices inevitably creep in.

The one aspect of practical acoustics that I think is critical, is to be able to directly connect cause and effect. If I do “A” to a violin I can reasonably expect “B”. That’s what I search for.

Yup, combining practical skills with science, along with a transparently experimental approach, is rare in these discussions.

I've been positively impressed by Martin Schleske's classically rigorous approach to investigating violin acoustics, and applying what he's learned.  To start with, he's a working luthier who completed luthier school in Mittenwald, then got a physics and engineering degree, so his knowledge is practical as well as theoretical.  He investigates, publishes, applies his findings to his own work, profitably sells what he makes, and also has the courage to post sound recordings of some of his violins being expertly played.  His very interesting website is at:   https://www.schleske.de/en/master-luthier.html   and his lab research section within it begins here:  https://www.schleske.de/en/research.html

[Marty Kasprzyk]

I scratched my head thinking about this and I found the sound I heard produced through bone conduction was the same all over my scalp.  I quess this means my skull was uniform in thickness.

There is a possibility that thick skulled people sound brighter and thin skulled people sound louder.

[Oded Kishony]

You’re obviously ‘scratching’ the wrong part of your anatomy!

[violins88]

Colleen,

it would be great to get Peter Wusterlund in a lab with sensitive microphones and actually verify what he says he hears. I really, really doubt his hearing can detect what he says. It’s really not technically difficult to verify. Please, Peter, come forward and do a test. That would be very helpful to us.  There is another fellow, with many pages on his violin maker website, who claims to be able to scratch a soundpost and use that info to make useful changes. Please, submit your claim to an actual measured test. I think the frequencies involved are above human hearing. But don’t trust me. Submit to a test.

How about arranging this to be done at the Cleveland VSA meeting in November? I plan to attend. Who has a good microphone? Someone in Cleveland, I bet. Can we fundraise to pay for Wusterlund to attend?

Sorry, actually Wusterlund’s attendance is unnecessary. Someone just needs to  measure accurately the sound of a scratched plate with the before and after changes Wusterlund indicates. That can be done by anyone with the right equipment. Anywhere.

My brain is tired tonight, but any mechanical engineer can work out the fundamental frequency of vibration of the violin sound post. I am pretty sure it is above human hearing frequencies. (See Langsather videos.) But wait… what about subharmonics. Isn’t it possible that we can hear half of 30 kHz. Yes, I think so. Some people, not me. But still, wouldn’t a test be nice?

Thanks

[Don Noon]

4 hours ago, violins88 said:

any mechanical engineer can work out the fundamental frequency of vibration of the violin sound post. I am pretty sure it is above human hearing frequencies.

It depends on what mode of vibration you're interested in and what the end constraints are.

The highest frequency would be free-free longitudinal vibration, which I calculate to be in the range of 55,000 Hz.  That is totally irrelevant in an assembled violin, as the ends are attached to relatively heavy plates.  The as-attached-to-the-plates frequency would be far lower, and could well have some measurable effect on the played instrument... although my guess is that it would be minor.

Soundpost bending vibration I think is obviously within hearing range, and what you hear when dropping a post on a hard surface.  That too is not relevant, as the end conditions provide constraint in translation and rotation.  I also don't see how the violin would cause much excitation of soundpost bending, or that it would translate to affecting plate vibration (sound) in a significant way... although it could be something minor.  More relevant I think would be the fit at the ends of the soundpost, which I would guess have more influence on the higher plate modes.

All of this piece-part attention is hocus-pocus diversions from getting real-world experience with a complete violin: doing the work and hearing the results.

[violins88]

Thanks, Don.

When I drop a soundpost, I don’t hear anything. Oh, the ravages of age!

[Marty Kasprzyk]

On 4/14/2026 at 10:05 PM, Colleen McTigue said:

I’m just watching a video by Peter Westerlund, where he discusses a method of using “scratch tones” and tap tones in making a violin. He apparently uses it to determine where to remove wood from a plate. His goal, as near as I can tell, is to have the same tones everywhere on the plate. I’d be interested in hearing what folks here think of this. The video:

 

In the very first few seconds of Westerlund's video it shows him scraping the plate to thin it.  I suspect the scraper's sound noise might also provide an some indication of the plate's vibration characteristics.

Noise can be catorgized into different colors: white, pink, red, green, grey, black, blue etc. which are determined by the spectrum distribution and magnitudes of the many individual wave frequencies in the sound.

 https://en.wikipedia.org/wiki/Colors_of_noise#Generation

If you've made many violin you might have noticed the scraper's noise sound changes during the tinning process.  This might give some people an indication of where to scrape and when to stop thinning.

[Don Noon]

1 hour ago, Marty Kasprzyk said:

If you've made many violin you might have noticed the scraper's noise sound changes during the tinning process.  This might give some people an indication of where to scrape and when to stop thinning.

For sure the plates will change drastically if they are tinned.  Better for soldering a bridge to the top?

But assuming you mean "thinning", that too will logically cause sound changes, as the scraper would excite plate vibrations, and the mode shapes and frequencies change depending on thickness.

However, there are a few huge logical gaps...

1.  Free, unvarnished plate vibrations are vastly different when assembled and varnished.

2. Where you scrape might not be where the sound is produced, and therefore not be a good indicator of much use for localized scraping (and also #1.)

3. As a useful tuning tool, you would have to have a huge number of trial-and-error tests to correlate the sound of the scraped plate with the sound of the assembled instrument, and if you want to extend this idea to small area, it becomes even more ridiculous.

4. The first step would be to prove the utility of precise taptones.  Good luck.

[reguz]

If you have not attempted Westerlund’s method yourself, it is impossible to form a meaningful opinion about it. I have tried to engage in a constructive discussion with Peter Westerlund, but he does not seem interested in discussing the method in depth.

 

[Marty Kasprzyk]

If you made a several hundred violins you might notice some correlation between your violin's plate thinning and its final sound which might help for making a good violin.

Without modern electronics and computers a maker long ago (or even today) might have found it important pay attention to the following things while they were thinning their plates.

1.  Its thickness profile.

2. Its weight.

3.  How it feels when you bend it in various ways.

4.  Its tap tones.

5.  Its sound when scratched or scraped.

 

Or maybe they just make a large number of violins and some of them by chance were good.

"Even a dumb squirel can find a nut sometimes"

 

[Oded Kishony]

Ahem…….you could do a final scraping on an assembled strung up instrument, then you could immediately listen for tonal changes! Just say’n.

[Don Noon]

9 hours ago, Oded Kishony said:

Ahem…….you could do a final scraping on an assembled strung up instrument, then you could immediately listen for tonal changes! Just say’n.

Well that last bit of scraping went in the wrong direction.  Let me just glue some wood back on.

[Mansfield Piggot]

Some makers glue the top on, weakly I assume, and alternately try it out, then take the top off, scrape and beg, then glue it back on, repeat.  Some try it out with the top held on by clamps

[Don Noon]

6 hours ago, Mansfield Piggot said:

Some makers glue the top on, weakly I assume, and alternately try it out, then take the top off, scrape and beg, then glue it back on, repeat.

I have done a lot of that with inexpensive instruments and many of my early violins.  Regraduation IMO does well at adjusting the high/low balance, but the wood, arching, and form seem to define a significant part of the tone.

 

6 hours ago, Mansfield Piggot said:

Some try it out with the top held on by clamps

The clamps add mass, and don't fully fix the edges the way they would be when glued.  I have tried it, and don't think it's a useful method.

[Evelyn]

I've used the method for my first six instruments and it's been fine as a way to get a workable arching without templates. Certainly using it hasn't been more of a blocker for me than any other of the many tool skills and design decisions that goes into making a successful instrument.

If any of you want to demonstrate to yourselves whether it works as a method to identify the difference between a better and a worse arching, you can apply it to a finished plate and a plate in the early stages of carving, and see that the scratch tone changes more as you traverse from the center put on the unfinished plate than on the finished plate.

If you can agree that there's less contrast in tones on the more complete arching, then you might be able to see why it might be able to give you information more subtly on how an arching might be improved.

Typically I have found that a section of a plate that doesn't fall towards the edge at the same rate as surrounding sections (basically, a lump in the arching) will produce a higher tone when scratched. Applying that idea iteratively seems to converge my plates towards a shape more similar to a standard arching than not.

[Colleen McTigue]

15 minutes ago, Evelyn said:

Typically I have found that a section of a plate that doesn't fall towards the edge at the same rate as surrounding sections (basically, a lump in the arching) will produce a higher tone when scratched. Applying that idea iteratively seems to converge my plates towards a shape more similar to a standard arching than not.

My sense from watching (and listening) to his videos is that he’s onto something.  Maybe not “the secret”, but something.  I can hear the differences he refers to.  I think it might be useful in graduating the plates, but also as Evelyn says, maybe also identifying an area that needs more work.  I think I’ll experiment with it on my next project.

 

[Don Noon]

The logical problem with this method is that the scratching sounds on a free plate are only scratching sounds on a free plate.  If there was evidence that some prescribed set of scratching sounds was in some way correlated to a good sounding finished violin (and deviating from that prescription wasn't as good), then, yeah.

We definitely have detailed arching and graduation information on great violins (and sometimes wood density); that is a more appealing and logical path for me to follow.  Do we have unvarnished free-plate scratching sound data on any great violins?

[Oded Kishony]

A frequently overlooked fact of violin making by proponents of plate tuning is that a significant portion ( to ~1000hz) of the radiated sound of a violin is the result of whole body corpus motion which not only includes the plate and how they relate to each other and the whole body but also can involve the ribs.

Individual plate tuning is like trying to make a car go faster by adjusting the hubcaps.