Analogue Synth Oscillator Query!

This forum is for discussing Reason. Questions, answers, ideas, and opinions... all apply.
Post Reply
User avatar
Creativemind
Posts: 4895
Joined: 17 Jan 2015
Location: Stoke-On-Trent, England, UK

13 Sep 2017

Hi All!

Something I've been wondering about for ages.

How is the waveform generated in an analogue synth? when you select a sawtooth, triangle, square etc. Is it generated from some process, or is the raw waveform on a chip inside in it's circuitry or something, then played and further manipulated?

Thanks!
:reason:

Reason Studio's 11.3 / Cockos Reaper 6.82 / Cakewalk By Bandlab / Orion 8.6
http://soundcloud.com/creativemind75/iv ... soul-mix-3

User avatar
am0eba
Posts: 205
Joined: 15 Jan 2015

13 Sep 2017

Analog synths use analog circuits to generate the complex waveforms... A quick Google turned up this info:

Image

Plenty of info out there...

_Dave_
Attachments
Screen Shot 2017-09-13 at 6.39.08 PM.png
Screen Shot 2017-09-13 at 6.39.08 PM.png (158.61 KiB) Viewed 935 times
:reason: :re: :refill: :recycle: :PUF_figure: :PUF_take: :record: :rt:
"F" the ineffable! 
My music is available on iTunes, Amazon, Youtube, Spotify and CDBaby - more info:
http://2little2late.org/am0eba/

User avatar
aeox
Competition Winner
Posts: 3222
Joined: 23 Feb 2017
Location: Oregon

13 Sep 2017

I believe there is circuitry that generates the signal type (waveform) and circuitry that controls the voltage, which determines the pitch of the signal.

User avatar
raymondh
Posts: 1779
Joined: 15 Jan 2015

14 Sep 2017

Carl (the other half of our band Pacific Deep) happens to be an expert in electronics, and he explained it to me recently. Here is the summary:

"A typical analog sawtooth audio oscillator works by charging a capacitor with constant current. This makes the voltage on the capacitor increase linearly. When the voltage passes some threshold the capacitor is reset (goes back down to zero very quickly) then starts rising again. That's how you get the sawtooth shape. The sawtooth is the base wave shape and can be further processed to generate square and sine.

Now, it turns out it is really hard in practice to generate a good constant current. If the current varies a bit, the voltage slope varies a bit, and the amount of time it takes to go from zero to the threshold (the oscillator period) varies a bit. This oscillator variation isn't a huge deal (if it isn't enormous) unless you are playing a chord, then it is a really big deal because the different oscillators playing the different notes aren't necessarily tracking each other. This is why the tuning is so challenging in early analog synths and why so many were monosynths.

One way to get around this issue and still keep an analog signal path is to use a digital counter to time the oscillator period. So, if you have a 1 MHz digital clock (which is rock solid since it is based on a crystal) and you want to tune your oscillator to 1 kHz, you just count 1000 clock pulses and then reset. If you want 2 kHz, you count 500 clock pulses and so on. Now, the unavoidable variation in sawtooth slope is no longer causing tuning problems but now just causes the amplitude to vary slightly. This isn't as big a deal and is part of what gives real analog synths their "warmth". This scheme for keeping multiple oscillators in tune was the breakthrough that let inexpensive polysynths arrive on the scene in the late 70s / early 80s. "

Now if you are still with me :), I sent this information to one of the experts at Dave Smith Electronics, because I was having timbre irregularities with my DSI Mopho x4. DSI were very helpful, and they offered the following further information:

"That explains it perfectly. An excellent explanation of analog oscillators. I'm very happy your bandmate included the fact that it is very hard to generate a good constant current. This lends itself to sine waves and how they will never be perfect, except if you have a piece of lab test equipment like an HP tone generator. But then again, that's not a very musical piece of equipment unless you're Pan Sonic and know how to manipulate that type of stuff. Tell your bandmate that he's right about the oscillator and filter being on the same chip. We use Curtis chips in most of our synths and it has a specific character to it's filter and gives it THAT sound. Have fun with your Mopho X4, Raymond and if you have any other questions or problems in the future, just let me know."

Hope that info is as interesting to you as it was to me!
Cheers, Raymond

User avatar
Marco Raaphorst
Posts: 2504
Joined: 22 Jan 2015
Location: The Hague, The Netherlands
Contact:

14 Sep 2017

analogue != chip

User avatar
raymondh
Posts: 1779
Joined: 15 Jan 2015

14 Sep 2017

Marco Raaphorst wrote:
14 Sep 2017
analogue != chip
That is not a correct statement.

Discrete electronics versus integrated circuits are not the difference between analog and digital.

I built my very first synthesizer back in the 80s around an IC, and it was fully analog. I used this chip https://en.wikipedia.org/wiki/Texas_Instruments_SN76477
Last edited by raymondh on 14 Sep 2017, edited 2 times in total.

avasopht
Competition Winner
Posts: 3958
Joined: 16 Jan 2015

14 Sep 2017

Marco Raaphorst wrote:
14 Sep 2017
analogue != chip
The Curtis chip mentioned is an analogue chip. You can even check the circuit diagram.

User avatar
Marco Raaphorst
Posts: 2504
Joined: 22 Jan 2015
Location: The Hague, The Netherlands
Contact:

14 Sep 2017

raymondh wrote:
14 Sep 2017
Marco Raaphorst wrote:
14 Sep 2017
analogue != chip
That is not a correct statement.

Discrete electronics versus integrated circuits are not the difference between analog and digital.

I built my very first synthesizer back in the 80s around an IC, and it was fully analog. I used this chip https://en.wikipedia.org/wiki/Texas_Instruments_SN76477
thanks! I stand corrected.

Post Reply
  • Information
  • Who is online

    Users browsing this forum: PeZiK and 14 guests