electro-music.com   Dedicated to experimental electro-acoustic
and electronic music
 
    Front Page  |  Radio
 |  Media  |  Forum  |  Wiki  |  Links
Forum with support of Syndicator RSS
 FAQFAQ   CalendarCalendar   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   LinksLinks
 RegisterRegister   ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in  Chat RoomChat Room 
 Forum index » DIY Hardware and Software
the click of the vca
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 1 [20 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Author Message
zipzap



Joined: Nov 22, 2005
Posts: 559
Location: germany
Audio files: 24

PostPosted: Sat Dec 17, 2005 8:53 am    Post subject: the click of the vca
Subject description: get rid of it
Reply with quote  Mark this post and the followings unread

Hi electro-musician
Well, much work these days, haven´t been here for a while. Yestrerday I soldered this VCA. I post the schematic, its from http://hem.bredband.net/bersyn/ THis site is great. I hope its ok with jörgen if i post this. He has measured and compared several vca circuits and the one posted is the winner.
Anyway, you probably all know the clicking sound whan the vca is hit by an envelope with 0 attack or 0 release.
How do i get rid of that? Tweaking the 1k trimpot gives me a veriety of realy anoying click to louderthandrumsetkillmyspeakers click.
Do you know what i mean?
Btw, envelope goes up to about 5v.
Some tips would be great! cheers robert


LM13600%20VCA%202.png
 Description:
 Filesize:  3.52 KB
 Viewed:  7802 Time(s)

LM13600%20VCA%202.png


Back to top
View user's profile Send private message
Macaba



Joined: Jul 13, 2005
Posts: 160
Location: UK

PostPosted: Sat Dec 17, 2005 3:40 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hmm, intriguing. Well, if it was me i'd probably experiment with adding a DC offset on the CV input of something like 0.3V, and then subtract this offset from the output to cancel it out. This means the whole circuit is under bias at all times.
Back to top
View user's profile Send private message
mosc
Site Admin


Joined: Jan 31, 2003
Posts: 18197
Location: Durham, NC
Audio files: 212
G2 patch files: 60

PostPosted: Sat Dec 17, 2005 9:29 pm    Post subject: Reply with quote  Mark this post and the followings unread

Why not hang a capactor to groung on the CV input, on the other side of the input resistor. This will smooth out the transients. Experiment with the size.
_________________
--Howard
my music and other stuff
Back to top
View user's profile Send private message Send e-mail Visit poster's website AIM Address
State Machine
Janitor
Janitor


Joined: Apr 17, 2006
Posts: 2809
Location: New York
Audio files: 24

PostPosted: Tue Aug 21, 2007 4:45 pm    Post subject: Reply with quote  Mark this post and the followings unread

You are most likely getting feed-through from the amplifier bias current. This is why National came up with the LM13700 instead of the LM13600. It has better control current to output current isolation (see the data sheet). For now, you could try raising the value of the 22K resistor into pin 1 of the 13600, Iabc, to about 27K instead (or a bit more until you get rid of the click). Lots of VCA's designed using these older chips don't like those fast transients. Experiment a bit and see what you arrive at Very Happy

Bill
Back to top
View user's profile Send private message Send e-mail
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Tue Aug 21, 2007 5:04 pm    Post subject: Reply with quote  Mark this post and the followings unread

State Machine wrote:
You are most likely getting feed-through from the amplifier bias current.
Bill


Right. More specifically, what happens is that the offset voltage varies with the control current, so that no offset correction will work.

One solution is to buy a bunch of chips and select ones that have low offset variation as the control current is varied. I set up a test circuit to do this, and sorted through several dozen devices. There are large variations in performance from chip to chip, but also between different batches of the same device. But you can find devices that don't thump hardly at all.

Another solution is to use a design with a superior device. The CA3280 is generally very good in this respect. You have to redo the circuit design and layout for this approach.

Also, the various SSM devices are very good to use.

Ian
Back to top
View user's profile Send private message
Serenadi



Joined: Jul 03, 2007
Posts: 89
Location: Germany

PostPosted: Tue Aug 21, 2007 8:55 pm    Post subject: Reply with quote  Mark this post and the followings unread

I'm just making experience with the SSM2164 on my breadboard.
The problem of clicks is the same, even some more because of the wide range of frequency of this VCA, so clicks are very audible when using very short envelopetimes.

Compared with other VCAs (Dotcom (CA3080) and Curtis 3360) I have to say, they all click.

I don't think that it is only a problem of DC-Offset (but maybe of course).
I used LT1013 to control the SSM, which has a very low offset.

What I did at least was to insert a switch to change the response. It's done by switching the feedback capacitor of the control-OP - in other words - I filtered the control voltage.

Clicks are only audible when using the control of an ADSR, so the cutoff frequency of this CV-OpAmp may be fairly low. (the capacitor I use is 10nF).
The short Attack and Decay times are not affected and clicks aren't audible then.

When using the VCA as AM with audio-frequencies I switch over to 470pF, to get the (nearly) full range. Clicks are not audible when modulating with triangle or sine waves.
Back to top
View user's profile Send private message
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Wed Aug 22, 2007 7:22 am    Post subject: Reply with quote  Mark this post and the followings unread

Serenadi wrote:
I'm just making experience with the SSM2164 on my breadboard.
The problem of clicks is the same, even some more because of the wide range of frequency of this VCA, so clicks are very audible when using very short envelopetimes.

Let me double check something. When you listen for the click you should have nothing connected to the audio input. Just the control envelope and you listen for clicks at the VCA output. Right?

Ian
Back to top
View user's profile Send private message
bugbrand



Joined: Nov 27, 2005
Posts: 846
Location: Bristol, UK
Audio files: 1
G2 patch files: 1

PostPosted: Wed Aug 22, 2007 7:59 am    Post subject: Reply with quote  Mark this post and the followings unread

I dunno whether this'll be of any relevance, but in the TH VCA on Scott's site it has switchable AC/DC coupling on the audio input.

Quoting::
DC coupling is a powerful feature of any VCA - this allows you to control...well...slow moving or static control signals. Say you have the output of a sample and hold passing through your VCA - you can then control the level of that signal with an LFO, or a smoothed voltage (or stepped voltage) from another sample and hold.

AC coupling an input allows one eliminate any DC offset in the input signal. Normally this is used to get rid of the DC offset in an audio signal - in that case, a DC offset may cause a 'thump' in the VCA output as the VCA is modulated by sharp transients, such as those provided by envelope generators. AC coupling removes that annoyance.


May help?! I notice the VCA schem you posted doesn't have input caps..

_________________
http://www.bugbrand.co.uk
http://www.bugbrand.blogspot.com
Back to top
View user's profile Send private message Visit poster's website
Serenadi



Joined: Jul 03, 2007
Posts: 89
Location: Germany

PostPosted: Wed Aug 22, 2007 8:05 am    Post subject: Reply with quote  Mark this post and the followings unread

frijitz wrote:
Let me double check something. When you listen for the click you should have nothing connected to the audio input. Just the control envelope and you listen for clicks at the VCA output. Right?


Oh no, sorry, little misunderstanding by me.

When using no audio input, the SSM VCA is completely silent.

If zipzap was talking over this situation, the reason is propably an offset.

What I meant was the audible click when using dark waves like sine or strong filtered tones in conjunction with very (!) short envelope times.

While the Dotcom VCA (and the Curtis one also) produces a transient which sounds more like *plopp*, the SSM produces a high frequency click.

Anyway, both forms interfere *musically* in my opinion.
Especially the high frequency one. It's like putting the needle on a record.

I really like to use Attack=0 and a very short, extremely logarithmic Decay (by patching the envelope out back to the CV input of the Decay).
Solving the problem by filtering the CV input of the VCA, gives me very short, but very clear tones, where you can still identify the tuning very good - and even with no *clicks*.

I like the SSM VCA very much, and it's very noiseless too.[/quote]
Back to top
View user's profile Send private message
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Wed Aug 22, 2007 8:45 am    Post subject: Reply with quote  Mark this post and the followings unread

Serenadi wrote:

...
When using no audio input, the SSM VCA is completely silent.
...

Right. There is no CV feedthrough. This is how you have to trim the offset in a VCA. Always Exclamation

Serenadi wrote:

What I meant was the audible click when using dark waves like sine or strong filtered tones in conjunction with very (!) short envelope times.

This is as it should be. It comes from God and Mr. Fourier. When you turn on a signal instantaneously the Fourier transform contains infinitely many harmonics. So you will absolutely hear a tick.

Serenadi wrote:

While the Dotcom VCA (and the Curtis one also) produces a transient which sounds more like *plopp*, the SSM produces a high frequency click.

Probably just a different response rate of the expo converters. You had suggestions to filter the CV. This is what you get -- the click gets filtered into a lower-sounding pop and eventually disappears with heavy enough filtering.

Serenadi wrote:

Anyway, both forms interfere *musically* in my opinion.
Especially the high frequency one. It's like putting the needle on a record.

I know exactly what you mean. I have also looked at this. I believe the main trouble is that the envelope generators we use are somewhat unnatural. They begin with a finite slope, which means that the derivative of the signal changes discontinuously, and that in itself causes a click. This means that you have to go to a somewhat slowish attack rate to avoid too much popping.

A more natural envelope would have more of an "S" shape, starting out with zero slope, then curving up rapidly. If you use an envelope like that, you can have a faster attack without as much of a pop.

I made a circuit like that once, but I didn't document it and I can't remember what I did. It involved adding a bit of a "foot" to the usual expo curve. And it did work!

Serenadi wrote:

Solving the problem by filtering the CV input of the VCA, gives me very short, but very clear tones, where you can still identify the tuning very good - and even with no *clicks*.

Ah, good to know that. I see now that filtering at that point would remove the slope discontinuity I just mentioned. And you can get short enough by this method?

I'm pretty sure there is some way to use a diode-switching circuit to restrict the ramp speed just at the beginning of the attack. I'll think about that some more.

Ian
Back to top
View user's profile Send private message
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Wed Aug 22, 2007 8:55 am    Post subject: Reply with quote  Mark this post and the followings unread

bugbrand wrote:

AC coupling an input allows one eliminate any DC offset in the input signal. Normally this is used to get rid of the DC offset in an audio signal - in that case, a DC offset may cause a 'thump' in the VCA output as the VCA is modulated by sharp transients, such as those provided by envelope generators. AC coupling removes that annoyance.

Good point. This is yet another source of thumps/clicks. All my VCAs have separate DC- and AC- coupled inputs. However, with a sine wave input this should not be too much of a factor, unless the sine wave has some offset.

Ian
Back to top
View user's profile Send private message
State Machine
Janitor
Janitor


Joined: Apr 17, 2006
Posts: 2809
Location: New York
Audio files: 24

PostPosted: Wed Aug 22, 2007 8:58 am    Post subject: Reply with quote  Mark this post and the followings unread

Quote:
Right. There is no CV feedthrough. This is how you have to trim the offset in a VCA. Always


Good test. Yes, this is the best way to isolate if the CV is feeding through or not. I know that the LM13700 has better isolation of the control current which is why I suggeated that. Thanks for your input.

Bill
Back to top
View user's profile Send private message Send e-mail
Serenadi



Joined: Jul 03, 2007
Posts: 89
Location: Germany

PostPosted: Wed Aug 22, 2007 9:55 am    Post subject: Reply with quote  Mark this post and the followings unread

frijitz wrote:
Serenadi wrote:

What I meant was the audible click when using dark waves like sine or strong filtered tones in conjunction with very (!) short envelope times.

This is as it should be. It comes from God and Mr. Fourier. When you turn on a signal instantaneously the Fourier transform contains infinitely many harmonics. So you will absolutely hear a tick.

I see.
The point is, it draws my attention at the moment, I was busy with the SSM, because of its sharp exponential response which is far more than of my other VCAs.
So my envelopes suddenly were sounding very much faster.

frijitz wrote:
Probably just a different response rate of the expo converters. You had suggestions to filter the CV. This is what you get -- the click gets filtered into a lower-sounding pop and eventually disappears with heavy enough filtering.


As I mentioned above, yes.

frijitz wrote:
A more natural envelope would have more of an "S" shape, ...

Ha, interesting.
Our mate Funky40 mentioned exactly that a couple of days ago.
He said 'actual, an envelope response has to be like an S'.

frijitz wrote:
Serenadi wrote:

Solving the problem by filtering the CV input of the VCA, gives me very short, but very clear tones, where you can still identify the tuning very good - and even with no *clicks*.

Ah, good to know that. I see now that filtering at that point would remove the slope discontinuity I just mentioned. And you can get short enough by this method?


I tell you ! Very Happy

I did not measure the cutoff from that inputfilter, but it is high enough to allow very fast peaks while beeing low enough to remove disturbing clicks.

Btw, the base of my tests was the article of Mike Irwin on EDN.

ooh - forgot to say:
Even if the article is around linearizing, all of the above I did in exponential mode, of course.
Back to top
View user's profile Send private message
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Wed Aug 22, 2007 10:59 am    Post subject: Reply with quote  Mark this post and the followings unread

Serenadi wrote:

The point is, it draws my attention at the moment, I was busy with the SSM, because of its sharp exponential response which is far more than of my other VCAs.
So my envelopes suddenly were sounding very much faster.

Yes, indeed. Conventional wisdom is that you should not use an expo EG with an expo VCA, the reason being that your decay tail is an expo of an expo, whereas a plain expo is how percussive acoustic sounds decay.

That being said, I do it all the time. Smile

Also note that the original question from Robert was about an expo VCA (Bergfors').


Quote:
Quote:

Ah, good to know that. I see now that filtering at that point would remove the slope discontinuity I just mentioned. And you can get short enough by this method?


I tell you ! Very Happy

I hear you!
Good discussion, thanks. Very Happy

Ian
Back to top
View user's profile Send private message
Coriolis



Joined: Apr 11, 2005
Posts: 616
Location: Stilling, Denmark

PostPosted: Wed Aug 22, 2007 11:27 am    Post subject: Reply with quote  Mark this post and the followings unread

Hey Robert, did that Bergfors circuit work for you otherwise? You didn't have any problems? I breadboarded it, and couldn't get it to go completely silent.

Probably a mistake on my part, but just wanted to hear from someone else who had actually built one...

C
Back to top
View user's profile Send private message Visit poster's website
Serenadi



Joined: Jul 03, 2007
Posts: 89
Location: Germany

PostPosted: Wed Aug 22, 2007 11:38 am    Post subject: Reply with quote  Mark this post and the followings unread

Try to decrease the 100k resistor at the wiper of the cv-offset pot to increase its range.
It also depends on the power supply voltage, you use.


@ Ian:
It was a pleasure to me, thank you too.
Back to top
View user's profile Send private message
frijitz



Joined: May 04, 2007
Posts: 1734
Location: NM USA
Audio files: 54

PostPosted: Wed Aug 22, 2007 12:09 pm    Post subject: Reply with quote  Mark this post and the followings unread

Coriolis wrote:
Hey Robert, did that Bergfors circuit work for you otherwise? You didn't have any problems? I breadboarded it, and couldn't get it to go completely silent.

Probably a mistake on my part, but just wanted to hear from someone else who had actually built one...

That's always an issue with an exponential VCA -- they never really shut off, since there is no end to the expo range. I have ended putting clamps on all mine so they switch totally off when the CV gets below a certain value.

Ian
Back to top
View user's profile Send private message
Scott Stites
Janitor
Janitor


Joined: Dec 23, 2005
Posts: 4127
Location: Mount Hope, KS USA
Audio files: 96

PostPosted: Wed Aug 22, 2007 2:40 pm    Post subject: Reply with quote  Mark this post and the followings unread

This is a good discussion - the click is something that's always bugged me, too.
_________________
My Site
Back to top
View user's profile Send private message
Coriolis



Joined: Apr 11, 2005
Posts: 616
Location: Stilling, Denmark

PostPosted: Thu Aug 23, 2007 12:25 am    Post subject: Reply with quote  Mark this post and the followings unread

Well, I didn't use that expo current source, I used a linear source instead.
Coming to think of it, that may have something to do with it, since I got that source from a schem for a CA3280 vca...But don't most OTA's like roughly the same current on their control inputs?

And yes - good discussion. I didn't know the reason for AC-coupled inputs, before I read it here... Smile

C
Back to top
View user's profile Send private message Visit poster's website
Sound



Joined: Jun 06, 2006
Posts: 842

Audio files: 1

PostPosted: Sat Jan 10, 2009 11:18 am    Post subject: Reply with quote  Mark this post and the followings unread

Interesting thread.

I am now breadboarding a 2164 Mike Irwin linearized VCA.

Also I listen the clicks.

The initial design is with a 100pf capacitor.

With the shortest attack and decay with a 10nf capacitor the click is canceled.
But with the shortest attack and a little more long decay with a low sine wave in the audio IN, the click is less obvious, and I noticed that removing the capacitor! the click is less audible, with capacitor value from 100 pf to 40n is a little audible and from 47n to up the click disappears. With big values like 2uf the attack is smoothed and its behavior is the same than if you put a more slow attack.

So it makes me think that maybe the solution is in the envelope, programing the shorter attack without listen the click. Your ear still will listen a fast attack.

If you play a flute, if you blow beating fast also you will listen overtones, you must blow in its measure, thing that your mind-body will make automatically. But also is true that all we want good instruments that allow us play with the dynamics at maximum is possible.

In any case I thing that the best option is the one that did Serenadi because you have more control over the instrument. Also is interesting play with that switch when you are introducing audio in the VCIN.

But because in what I'm designing I have not front panel room for another switch I will play with slower attacks, or maybe a bigger capacitor value compromising a little of the audio range of the VCIN...
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 1 [20 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
 Forum index » DIY Hardware and Software
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Forum with support of Syndicator RSS
Powered by phpBB © 2001, 2005 phpBB Group
Copyright © 2003 through 2009 by electro-music.com - Conditions Of Use