first violin spectrum... Nicolo Amati model

[Seán Ó Fearghail]

This is the bowed spectrum recorded for my first fiddle at around  7 inches from the bridge with an Olympus recorder plotted through Audacity and sample rate of 2040...  think it has a hollow sound...which is reflected in the dip at circa 1100hz... I think... my first reaction is the plates are too thin...  (or the top plate isn't high enough density) any constructive criticism? Total beginner with a thick skin....  

[avandesande]

Why don't you see what you can get messing with the sound post

[Don Noon]

"Bowed spectrum" doesn't describe very accurately what you're doing.  Is it a semitone scale?  If not, perhaps that is why the B1- resonance seems to be missing.  Also 7" from the mic is pretty close; I usually use 1 m for bowed semitone scale spectra.

[ctanzio]

Also, there is no reason to precisely bow a semitone scale. Just slowly slide your finger up and down  each string a few times while focusing on bowing with a consistent pressure and speed in the middle third or so of the bow.

[matesic]

Pardon my ignorance, but the only time I tried this the spectrum was substantially affected by how far my bow was from the bridge. I think it was mainly the 3rd harmonic that increased as the distance decreased which would surely affect the 1100Hz region.

[Andreas Preuss]

If. You just look at the average level below 1000hz and the average level above, it is very ‘bass heavy’ and I think regardless the dip around 1050 hz this makes a hollow sound. I had a similar problem at the first stage of my super light violin project. 
 

Moving the sound post as suggested by @avandesande is useful to learn how to get the best out of a given situation, but won’t solve the problem. 
 

The first thing I’d look at is the string angle, bridge height, bridge weight. For a vague guess I’d suspect a rather high bridge (over 32mm?) which is rather heavy. And if additionally string angle is rather small putting more pressure on the top this amplifies the effect. 
 

 

[Seán Ó Fearghail]

16 hours ago, Don Noon said:

"Bowed spectrum" doesn't describe very accurately what you're doing.  Is it a semitone scale?  If not, perhaps that is why the B1- resonance seems to be missing.  Also 7" from the mic is pretty close; I usually use 1 m for bowed semitone scale spectra.

Thanks Don, I'll have another go but I'm fairly sure I was playing the semi-Tone scale across the 4 strings as I've read about before but I’ll adjust the distance and try again. 

[Don Noon]

Strength of bowing influences the high/low frequency balance; I try to bow as strongly as possible, which shows up the highs better.

[Seán Ó Fearghail]

5 hours ago, Andreas Preuss said:

If. You just look at the average level below 1000hz and the average level above, it is very ‘bass heavy’ and I think regardless the dip around 1050 hz this makes a hollow sound. I had a similar problem at the first stage of my super light violin project. 
 

Moving the sound post as suggested by @avandesande is useful to learn how to get the best out of a given situation, but won’t solve the problem. 
 

The first thing I’d look at is the string angle, bridge height, bridge weight. For a vague guess I’d suspect a rather high bridge (over 32mm?) which is rather heavy. And if additionally string angle is rather small putting more pressure on the top this amplifies the effect. 
 

 

Thanks Andreas. I think the bridge might be a little heavy alright. I have played with the soundpost a good bit already but I will take on board everything said here and try some more. 

[scordatura]

7 hours ago, matesic said:

Pardon my ignorance, but the only time I tried this the spectrum was substantially affected by how far my bow was from the bridge. I think it was mainly the 3rd harmonic that increased as the distance decreased which would surely affect the 1100Hz region.

This is a good point. Sounding point or the lane that the bow is tracking relative to the bridge and fingerboard is critical in sound production. The extremes of this are ponticello (loud and bright near bridge) and flautando (soft and dark over the fingerboard) To best eliminate variables, it is wise to be mindful of the sounding point, bow speed and vertical force (weight). Especially when playing the same instrument for comparison. Experienced players are always manipulating these variables to produce the desired sound.

[Seán Ó Fearghail]

16 hours ago, Andreas Preuss said:

If. You just look at the average level below 1000hz and the average level above, it is very ‘bass heavy’ and I think regardless the dip around 1050 hz this makes a hollow sound. I had a similar problem at the first stage of my super light violin project. 
 

Moving the sound post as suggested by @avandesande is useful to learn how to get the best out of a given situation, but won’t solve the problem. 
 

The first thing I’d look at is the string angle, bridge height, bridge weight. For a vague guess I’d suspect a rather high bridge (over 32mm?) which is rather heavy. And if additionally string angle is rather small putting more pressure on the top this amplifies the effect. 
 

 

Just checked.. The bridge height is 32.5mm... I might try thinning it a little to see if that helps. It's currently 2.03g but it is a nice dense bridge with very straight quarter sawn grain lines. I think it can take it. 

Would thicker/ thinner sound post help?

I started with 6.2mm diameter, and brought it back to 5.7 and I think it improved it a little bit, maybe showing a bit more of the higher frequencies but I didn't want to over do it. It's currently placed at the 'traditional' position.(if you want to call it that) at around 3mm back from the bridge and 1mm inside the foot, standing fairly horizontal. 

 

And thank you everyone for all the comments...

[Seán Ó Fearghail]

10 hours ago, scordatura said:

This is a good point. Sounding point or the lane that the bow is tracking relative to the bridge and fingerboard is critical in sound production. The extremes of this are ponticello (loud and bright near bridge) and flautando (soft and dark over the fingerboard) To best eliminate variables, it is wise to be mindful of the sounding point, bow speed and vertical force (weight). Especially when playing the same instrument for comparison. Experienced players are always manipulating these variables to produce the desired sound.

Thanks. I will try another, bowing closer to the bridge and see does that help.

[Bodacious Cowboy]

14 hours ago, Don Noon said:

Strength of bowing influences the high/low frequency balance; I try to bow as strongly as possible, which shows up the highs better.

Don - what do you think are the pros/cons of bowed vs impact measurements?

[Don Noon]

52 minutes ago, Bodacious Cowboy said:

Don - what do you think are the pros/cons of bowed vs impact measurements?

Bowed is as close to reality as you can get, but it is very difficult to get high repeatability, even with the same player.  If there is a particular note that sounds funny, it is much easier to analyze the problem by directly looking at the note.

Impact (for me, at least) is more repeatable and shows better detail, but for some reason the CBR resonance hardly ever shows any significant output, whereas it shows more strongly when bowed.  If you get your hands on a Strad or other valuable instrument, taking an impact measurement might be frowned upon.

Since it takes less than a minute for each method, I do both whenever I can.

Here's a comparison of impact and bowed, with the CBR obvious and highlighted.  It is possible that in bowing, the B1+ wolfed on me, thus the split B1+ peak.  You'll never get a wolf on impact.

[Seán Ó Fearghail]

29 minutes ago, Don Noon said:

Bowed is as close to reality as you can get, but it is very difficult to get high repeatability, even with the same player.  If there is a particular note that sounds funny, it is much easier to analyze the problem by directly looking at the note.

Impact (for me, at least) is more repeatable and shows better detail, but for some reason the CBR resonance hardly ever shows any significant output, whereas it shows more strongly when bowed.  If you get your hands on a Strad or other valuable instrument, taking an impact measurement might be frowned upon.

Since it takes less than a minute for each method, I do both whenever I can.

Here's a comparison of impact and bowed, with the CBR obvious and highlighted.  It is possible that in bowing, the B1+ wolfed on me, thus the split B1+ peak.  You'll never get a wolf on impact.

Thanks Don so much for commenting, I had another go, taking in all the advice. I haven't made any changes to the fiddle, just the playing and the distance for recording. I think it looks better. There's something a little strange still but not sure what. Any way, thanks again. I'll keep experimenting. 

[Marty Kasprzyk]

8 hours ago, ofearghail7 said:

Thanks Don so much for commenting, I had another go, taking in all the advice. I haven't made any changes to the fiddle, just the playing and the distance for recording. I think it looks better. There's something a little strange still but not sure what. Any way, thanks again. I'll keep experimenting. 

What tension strings are you using?  Light tension strings will increase the high end and make it sound brighter.

[Seán Ó Fearghail]

2 minutes ago, Marty Kasprzyk said:

What tension strings are you using?  Light tension strings will increase the high end and make it sound brighter.

Hi Marty, Medium tension dominants. yes different strings would probably help. I'll try that.

[scordatura]

12 hours ago, Marty Kasprzyk said:

What tension strings are you using?  Light tension strings will increase the high end and make it sound brighter.

This is not always my experience. Meaning Evah Pirazzi strings are high tension and quite bright.

[Marty Kasprzyk]

1 hour ago, scordatura said:

This is not always my experience. Meaning Evah Pirazzi strings are high tension and quite bright.

String makes do vary in brightness and Evan Pirazzi is indeed a good choice.  However a weich (soft) tension Evah will sound brighter than a medium or stark (heavy) Evah string.

[Marty Kasprzyk]

I forgot to mention  lighter tension strings can also be played closer to the bridge which further increases brightness.

[Andreas Preuss]

2 hours ago, Marty Kasprzyk said:

I forgot to mention  lighter tension strings can also be played closer to the bridge which further increases brightness.

However, professional violinists rather opt for thick (high tension) strings. Somehow it seems that the heavy strings give them some more bow control despite they are more difficult to play. @Anders Buens research for the dunnwald criteria listed string gages in the examined instruments and there was a good number of instruments with forte strings as well. 

On my new concept violin experiment I noticed that a low weight violin works better with high tension strings combined with a flatter string angle. Reasons for this are unknown to me. Trying to explain it with body tension called a red (herring) flag from @Don Noon. 
 

So maybe the correct explanation is in a difference from the Helmholtz motion of the string vibration. Maybe heavier strings make a stronger helmholtz kick at the bridge. (Just guessing) This in return seems to be favorable on an instrument which absorbs the energy more easily. (?)

[Marty Kasprzyk]

On 11/13/2021 at 8:17 PM, ofearghail7 said:

Fire at will! ... this is the bowed spectrum recorded for my first fiddle at around  7 inches from the bridge with an Olympus recorder plotted through Audacity and sample rate of 2040...  think it has a hollow sound...which is reflected in the dip at circa 1100hz... I think... my first reaction is the plates are too thin...  (or the top plate isn't high enough density) any constructive criticism? Total beginner with a thick skin....  

If none of our various ideas seemed to help much then your original suspicion that your plates are too thin is probably correct.  The high amplitudes of the low frequency resonance peaks and the high frequency ""roll off" at only about 2300Hz is typical for overly thin plates.

If you tape over the sound holes many of the lower frequency (<1000Hz) resonance peaks should decrease in amplitude. This should give you better low/high frequency balance but obviously you will lose some loudness.

Or put short scale viola ADGC strings on it.

 

[Seán Ó Fearghail]

14 hours ago, Marty Kasprzyk said:

If none of our various ideas seemed to help much then your original suspicion that your plates are too thin is probably correct.  The high amplitudes of the low frequency resonance peaks and the high frequency ""roll off" at only about 2300Hz is typical for overly thin plates.

If you tape over the sound holes many of the lower frequency (<1000Hz) resonance peaks should decrease in amplitude. This should give you better low/high frequency balance but obviously you will lose some loudness.

Or put short scale viola ADGC strings on it.

 

Thanks Marty, I'll chalk it down to experience and start the next one. It definitely doesnt sound horrible but I now have in my head the sound im looking for, and you never know, it might suit somebody. 

Thanks everyone again. 

[Marty Kasprzyk]

7 hours ago, ofearghail7 said:

Thanks Marty, I'll chalk it down to experience and start the next one. It definitely doesnt sound horrible but I now have in my head the sound im looking for, and you never know, it might suit somebody. 

Thanks everyone again. 

Geez, don't give up so easily.  Take the top off and glue on some wood strips across the inside to stiffen it.

I suggest several thin (~1mm x ~10mm wide) across the island area between the f holes.  The one between the upper f hole eyes is most important.

 

[HoGo]

Why across? Is there any acoustic reason in this case to raise the cross stifness?

I've used similar breast patch on arch top mandolin with too thin failing top. I used 2-2.5cm wide 1.5mm thick strips along the grain tightly butted to each other. I used go-bar clamping against plaster cast (with slightly adjusted arch shape)

Mandolin arch is however shallower and large area would make it nearly impossible to fit and glue standard patch with HHG without leaving lots of glue trapped inside.