What's the patching difference between bells, chimes, and gongs?

Post your patches here! Put the format(s) in the subject.

Moderators: Kent, Joe., luketeaford, lisa

Post Reply
Mark Hammer
Learning to Wiggle
Posts: 8
Joined: Thu Apr 25, 2019 2:21 pm
Location: Ottawa

What's the patching difference between bells, chimes, and gongs?

Post by Mark Hammer » Thu Jan 16, 2020 2:52 pm

It's relatively straightforward to synthesize gong sounds, using ring modulation. But if one wishes to synthesize more pleasing bell-like tones, for instance to emulate a glockenspiel, what is the patching or settings involved in that?

User avatar
commodorejohn
Super Deluxe Wiggler
Posts: 1071
Joined: Fri May 03, 2013 4:19 pm
Location: Placerville, CA

Re: What's the patching difference between bells, chimes, and gongs?

Post by commodorejohn » Thu Jan 16, 2020 3:17 pm

There's a few ways you can go about it, but it all boils down to basically the same thing ring-mod accomplishes: creating a complex series of harmonics around the fundamental that gives the right "metallic" timbre. The biggest difference between your dirty nasty gongs and your sweet simple glockenspiels is how densely or loosely clustered they are and how closely or loosely they're related to an "ideal" series. Drifty, unstable VCOs and ring-mod (especially on harmonically complex waveforms like sawtooth or pulse) will give a set of overtones that are densely clustered (since two sidebands are created for every pairing of modulator and carrier harmonics) and not necessarily close to ideal (since pitch drift/instability means that the relationships between each carrier and modulator may not be a perfect integer ratio.)

There's a few ways to address that and get purer, sweeter type bell sounds. First off, if you use waveforms with fewer/weaker overtones (sine or triangle,) the resulting sidebands will be fewer and less pronounced, meaning a less "noisy"/"clangy" sound. And if you have oscillators that are more solid/stable in their tuning, you'll have less drift and the overtone series will more closely conform to the ideal.

Of course, you probably want some noisy, clangy stuff at least in the attack phase of the note, but on "sweeter" sounds the noise falls off more quickly and leaves just the harmonics closer to the fundamental to "ring out" - with ring-mod, there's no way to create that contour except for filtering the output. FM, on the other hand, allows you to change the prominence of the sidebands by changing the amount of modulation - use an EG/VCA to control the level of the modulator, and you can shape it over time without having to use a separate filter.

But then, FM is more susceptible to pitch-stability issues (since each pair of modulator and carrier harmonics creates an entire series of sidebands, not just a single pair.) So, in the end, there's no perfect answer - just experiment and find out what works!
Computers: Amiga 1200, DEC VAXStation 4000/60, DEC MicroPDP-11/73
Synthesizers: Roland JX-10/SH-09/MT-32/D-50, Yamaha DX7/V50/TX7/TG33/FB-01, Korg MS-20 Mini/ARP Odyssey/DW-8000, Ensoniq SQ-80

"'Legacy code' often differs from its suggested alternative by actually working and scaling." - Bjarne Stroustrup

Mark Hammer
Learning to Wiggle
Posts: 8
Joined: Thu Apr 25, 2019 2:21 pm
Location: Ottawa

Re: What's the patching difference between bells, chimes, and gongs?

Post by Mark Hammer » Fri Jan 17, 2020 4:08 pm

Thanks for a thoughtful and thorough reply! I'll have to try those strategies.

User avatar
Umcorps
Will patch for food
Posts: 2467
Joined: Fri Apr 17, 2009 6:33 am
Location: Currently unrecognisable

Re: What's the patching difference between bells, chimes, and gongs?

Post by Umcorps » Sat Jan 18, 2020 2:04 am

For an in-depth but very practical exploration of this, the old Sound on Sound Synth Secrets series is hard to better. There are a bunch of articles in that series on synthesising struck metal. This is the one on bells

https://www.soundonsound.com/techniques ... zing-bells

helix
Veteran Wiggler
Posts: 595
Joined: Sun Mar 26, 2017 5:13 am
Location: London, England

Re: What's the patching difference between bells, chimes, and gongs?

Post by helix » Mon Jan 20, 2020 1:53 pm

You could route a square or saw into the ring mod at the same time as the sine/triangle, but before the saw/square goes into the ringmod, it's going through a VCA being shaped by snappy envelope. Then you will have that attack portion of the rich harmonics for the transient part, and the sine can do it's thing afterwards. I've never tried it, but should work!

User avatar
commodorejohn
Super Deluxe Wiggler
Posts: 1071
Joined: Fri May 03, 2013 4:19 pm
Location: Placerville, CA

Re: What's the patching difference between bells, chimes, and gongs?

Post by commodorejohn » Mon Jan 20, 2020 7:26 pm

helix wrote:
Mon Jan 20, 2020 1:53 pm
You could route a square or saw into the ring mod at the same time as the sine/triangle, but before the saw/square goes into the ringmod, it's going through a VCA being shaped by snappy envelope. Then you will have that attack portion of the rich harmonics for the transient part, and the sine can do it's thing afterwards. I've never tried it, but should work!
Ah, now that's an idea. Better yet, run the square/saw through an LPF with no resonance, controlled by an EG (plus pitch CV tracking) so that you're effectively cutting off all the upper harmonics after the attack/decay phase, and you don't even need a separate sine oscillator.
Computers: Amiga 1200, DEC VAXStation 4000/60, DEC MicroPDP-11/73
Synthesizers: Roland JX-10/SH-09/MT-32/D-50, Yamaha DX7/V50/TX7/TG33/FB-01, Korg MS-20 Mini/ARP Odyssey/DW-8000, Ensoniq SQ-80

"'Legacy code' often differs from its suggested alternative by actually working and scaling." - Bjarne Stroustrup

User avatar
cornutt
Fig juggler
Posts: 1392
Joined: Sun Jan 25, 2009 10:06 pm
Location: Rocket City USA

Re: What's the patching difference between bells, chimes, and gongs?

Post by cornutt » Thu Feb 27, 2020 4:28 pm

In theory, a chime relies on the "missing fundamental" to make it sound larger and lower-pitched than it is. Take an old grandfather clock (we have one) that makes a BOOOOOOOONNNNNNNNNGGGGGG on the hour. It sound huge, but when you open the case and look inside, there's a thin little rod wound in a spiral. You look at it and think "How in the hell does it make all that noise?" The answer is that in the jumble of overtones coming out of the thing are a few that are close to the frequencies of the lower-numbered harmonics. The ear hears these harmonics and assumes that there should be a lower-frequency fundamental, and since it isn't there, your brain supplies one. :guinness:
Sequence 15 -- sequence15.blogspot.com

Monofunk
Common Wiggler
Posts: 128
Joined: Thu Apr 25, 2019 1:03 pm
Location: United States

Re: What's the patching difference between bells, chimes, and gongs?

Post by Monofunk » Fri Mar 06, 2020 3:44 pm

Umcorps wrote:
Sat Jan 18, 2020 2:04 am
For an in-depth but very practical exploration of this, the old Sound on Sound Synth Secrets series is hard to better. There are a bunch of articles in that series on synthesising struck metal. This is the one on bells

https://www.soundonsound.com/techniques ... zing-bells
This series is worth the read.
commodorejohn wrote:
Thu Jan 16, 2020 3:17 pm
There's a few ways you can go about it, but it all boils down to basically the same thing ring-mod accomplishes: creating a complex series of harmonics around the fundamental that gives the right "metallic" timbre. The biggest difference between your dirty nasty gongs and your sweet simple glockenspiels is how densely or loosely clustered they are and how closely or loosely they're related to an "ideal" series. Drifty, unstable VCOs and ring-mod (especially on harmonically complex waveforms like sawtooth or pulse) will give a set of overtones that are densely clustered (since two sidebands are created for every pairing of modulator and carrier harmonics) and not necessarily close to ideal (since pitch drift/instability means that the relationships between each carrier and modulator may not be a perfect integer ratio.)

There's a few ways to address that and get purer, sweeter type bell sounds. First off, if you use waveforms with fewer/weaker overtones (sine or triangle,) the resulting sidebands will be fewer and less pronounced, meaning a less "noisy"/"clangy" sound. And if you have oscillators that are more solid/stable in their tuning, you'll have less drift and the overtone series will more closely conform to the ideal.
And to add to this great post if you are working in a digital environment you can fine tune and detune the generators with more great control to deal with the inharmonicity. You can also keep making more generators with decreasing levels of amplitude to craft a detailed tonal structure.

Post Reply

Return to “Synthesis Techniques”