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
PLL 4046 for Beginners (Pitch-Tracking, too!)
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 2 [28 Posts]
View unread posts
View new posts in the last week
Mark the topic unread :: View previous topic :: View next topic
Goto page: 1, 2 Next
Author Message
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Tue Sep 18, 2018 8:23 am    Post subject: PLL 4046 for Beginners (Pitch-Tracking, too!)
Subject description: Introductory Question
Reply with quote  Mark this post and the followings unread

Hi Electro-Music -

I am new to the forum and relatively new to electronics. For my first DIY project, I would like to build a 4046 pitch-tracking oscillator. Reading about the phase-locked loop, though, I find it very confusing how it works - and I can't quite find the circuit that I'm looking for.
Essentially, I'd like to take an audio input and match to it the frequency of the oscillator of the PLL. Is this possible? Can anyone direct me towards a schematic or a simpler layperson's understanding of the PLL?
Back to top
View user's profile Send private message
gabbagabi



Joined: Nov 29, 2008
Posts: 651
Location: Berlin by n8
Audio files: 23

PostPosted: Tue Sep 18, 2018 11:56 am    Post subject: Reply with quote  Mark this post and the followings unread

Hi, and welcome to to forum
Last edited by gabbagabi on Fri Jan 18, 2019 10:33 am; edited 2 times in total
Back to top
View user's profile Send private message
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Tue Sep 18, 2018 12:01 pm    Post subject: Reply with quote  Mark this post and the followings unread

I would like to be able to tune the oscillator in relationship to the incoming audio frequency, creating different harmonic relationships to the input.

What is the waveform that comes out of the PLL normally? I'm sure that will suffice for now.
Back to top
View user's profile Send private message
PHOBoS



Joined: Jan 14, 2010
Posts: 5591
Location: Moon Base
Audio files: 705

PostPosted: Tue Sep 18, 2018 12:22 pm    Post subject: Reply with quote  Mark this post and the followings unread

welcome party!

Standard waveform is a square but it's also pretty easy to make something resembing a sawtooth with it.
The whole pitch tracking thing always seems to be a bit like magic to me but JovianPyx knows some things
about it so he might jump in. In the meantime I do recall this video having some usefull info although it has
been a while since I've watched it.


_________________
"My perf, it's full of holes!"
http://phobos.000space.com/
SoundCloud BandCamp MixCloud Stickney Synthyards Captain Collider Twitch YouTube
Back to top
View user's profile Send private message Visit poster's website AIM Address Yahoo Messenger MSN Messenger
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Fri Sep 21, 2018 12:19 am    Post subject: Reply with quote  Mark this post and the followings unread

definitely a very useful link! thank you. it seems like a very complicated chip, though - i'm not sure i can figure out how to do the pitch-tracking myself... are schematics for such an application available somewhere?
Back to top
View user's profile Send private message
AlanP



Joined: Mar 11, 2014
Posts: 746
Location: New Zealand
Audio files: 41

PostPosted: Fri Sep 21, 2018 3:17 am    Post subject: Reply with quote  Mark this post and the followings unread

I'd be interested in a good explanation of the basic logic behind the Phase Locked Loop. I've never looked into it, and it seems like black magic to me. Things happen to it, and it does things. Unexplainable things.

I've seen the VCO in the 4046 used in pedal flangers, and that's about it.
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Fri Sep 21, 2018 11:07 am    Post subject: Reply with quote  Mark this post and the followings unread

Heh, saw my name - so here goes.

A PLL has a VCO, a phase detector and a simple low pass filter to do it's work.
First, the 4046 has two phase detectors which are for different uses and only one or the other is used. The unused one's output is simply not connected. Phase detector 1 locks to the input signal but has a constant 90 degree phase shift from the input signal. It has a rather narrow lock range. Lock range is basically how far away the input signal can be from the free running frequency of the PLL internal VCO. If the input signal is too far away (both too high or too low) the PLL won't lock and therefore won't track. Phase detector 2 is more useful IMO for musical applications. It has a very wide lock range and when locked, the output signal is in exact phase alignment with the input.

Let's assume phase detector 2. with a steady input frequency applied to the input, the phase detector will generate a signal called an "error signal" which represents the phase difference between the input and output signals. This analog error signal is passed into the VCO which pushes the VCO frequency toward the frequency of the input signal. This system eventually causes the VCO to match the frequency of the input signal. Once the two signals (VCO and input) are the same frequency and phase of both is aligned, the phase detector generates a "locked" signal (a logic level where 0 is not locked and 1 is locked). Once locked, if the input frequency changes, the error signal causes the VCO's frequency to change to stay locked. With phase detector 2, the lock range can be several octaves (I don't remember how much, but it's a musically useful number of notes of range).

The filter smooths out the changes in the phase detector. Without the filter, the changes in phase detector output would jump around and the PLL would never lock. The low pass filter is actually a kind of analog memory. The center frequency of the filter (it's really just a simple RC lowpass filter) needs to be calculated to allow the PLL to operate within the lock range desired. You can find the formulas for this filter in the datasheet for the 4046. If the frequency of the filter is low, the filter causes the reaction time of the PLL to be slower. As such, it can cause the output frequency to glide from pitch to pitch instead of step. If the filter frequency is high, the PLL will react faster and with proper adjustment it will track very fast to the point where the glide (which is always there) is so fast it sounds like it's stepping from pitch to pitch. Because of this, it's good for musical applications to make the "loop filter" (as it's called) adjustable so you can smoothly change it from stepping to obvious portamento.

There is a lag and/or overshoot of the tracking which can be useful and an adjustable loop filters allows you to control it with a pot.

Tracking speed (how fast the PLL can catch up to the input signal when it changes frequency) changes with respect to the input frequency. Low input frequencies will cause a slower response than higher frequency inputs.

Another cool thing to do with a PLL is frequency multiplication. Double the frequency of the input is easy. Just put a divide by 2 counter on the output of the PLL VCO and then feed the counter into the phase detector. This will force the VCO to run exactly twice as fast as the input frequency - and exactly twice as you know is an octave. There is more PLL info with a schematic of a device I designed years ago for a Fatman synth. It could be modified to work with other inputs. It has a counter matrix that can cause the 4046 to generate:

1F ..... Synth VCO frequency (remember the lag and overshoot)
1.5F ... Perfect 5th above synth VCO
2F ..... 1 octave above synth VCO
3F ..... Perfect 5th and one octave above synth VCO
4F ..... 2 octaves above synth VCO
6F ..... Perfect 5th and two octaves above synth VCO
12F .... Perfect 5th and three octaves above synth VCO

PLLs work best when the input signal is rail to rail. It also works best when there is at least on fast "edge" on the signal, such as pulse, square or sawtooth. It is difficult if not impossible to track signals like guitar or piano. Those kinds of signals are very complex and have more than one zero crossing per cycle. The easiest signals to track are unfiltered synth signals such as the ones I mentioned above. Signals like sine or triangle should be processed into square before applying to the 4046 signal input.

IMO, the Texas Instruments datasheet is the best. It even shows a CMOS circuit of the internal VCO which is quite clever.

Also, google "woggle bug" for a different look at what a PLL can do for you.

HTH

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Fri Sep 21, 2018 12:28 pm    Post subject: Reply with quote  Mark this post and the followings unread

Thank you for the great explanation - it goes a long way to helping me (and probably many others) in understanding this chip..
I did want to ask a few details about the description, in relationship to the design you shared, though, if you wouldn't mind:

The PLL Frequency Multiplier has a switch that allows a user to choose between outputs of Phase Comparator 1 and Comparator 2.

My understanding is that the Phase Comparator Outputs produce the "Error Signal", that adjust the VCO tuning - and the "Locked" logic gate signal is produced at Pin 1.

As such, I have three small questions your design:
1. From your description, I got the sense that Comparator 1 would not track. What is the function of being able to select it?

2. What is the result of not using the "Locked" gate logic (NC on Pin 1)? Does it have any impact on the error tracking signal?

3. I also noticed that you tie Vss to V-, rather than V+. Does this have an overall effect on the design? Is it the standard for PLLs?

Thank you again for explaining so much.
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Fri Sep 21, 2018 12:44 pm    Post subject: Reply with quote  Mark this post and the followings unread

1) phase comparator 1 will indeed lock, but has caveats of smaller lock range and can lock onto strong harmonics of the input signal (such as the perfect fifth). This can also be musically interesting so I put the switch in to allow selection of PD1 or PD2. I find PD2 more useful because of it's wider lock range. There is a good explanation about this in the 4046 datasheet.

2) The "locked" logic output is there if you need or want it. In my device, it served me no purpose. Sometimes a LED on that pin with a series resistor is good to have because if the device starts doing odd or unexpected things you can tell if the 4046 is locked or not. My device did not have symptoms that made me think it's not locking, so I never added a LED for that. It may also serve the purposes of other designers in other inventive ways.

3) This power supply arrangement is because it is being driven by a PAiA Fatman. In that synth, the VCO operates below ground from 0v and -12v. It just made sense to do the same thing to the 4046 because then the Fatman VCO output can be directly DC coupled to the 4046 input with no buffering or level shifting required. Indeed, the 4046 can operate from whatever PSU voltages as long as Vdd-Vss is 15 volts or less. If you notice, the 4046 Vdd pin is at 0 volts and Vss is at -12 volts which preserves the proper PSU polarity for the chip. That is, the Vdd pin is at a higher voltage than the Vss pin.

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Fri Sep 21, 2018 4:50 pm    Post subject: Reply with quote  Mark this post and the followings unread

fantastic, thank you.
Back to top
View user's profile Send private message
youssef



Joined: Sep 18, 2018
Posts: 11
Location: madison

PostPosted: Mon Sep 24, 2018 9:15 pm    Post subject: Reply with quote  Mark this post and the followings unread

I would like to ask one more question, now that I am bread boarding.
If I simply wanted to track the exact frequency, without dividers, could I simply patch pin 3 to pin 4? I do not understand the logic chips in this circuit yet, but am hoping it still is possibly to do a simple tracking.
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Tue Sep 25, 2018 5:08 am    Post subject: Reply with quote  Mark this post and the followings unread

Yes, that connection will cause the output frequency to track the input frequency one to one. It is also the best way to test the circuit because adding the counter complicates it a bit.

Above all else, have fun! Cool Cool Cool

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
gasboss775



Joined: Jan 02, 2016
Posts: 217
Location: Scotland

PostPosted: Tue Sep 25, 2018 2:55 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hi youssef welcome to our forum!

There is another approach to getting an oscillator to follow an input signal, that is using something called a frequency to voltage converter, then using the output of that circuit to control a VCO to provide your desired signal output.

One advantage of this method over the 4046 / PLL method is that the oscillator can track the input at any desired musical interval just by rescaling the output of the frequency to voltage converter.
Back to top
View user's profile Send private message
Ayab



Joined: Oct 27, 2015
Posts: 180
Location: London, UK

PostPosted: Thu Oct 18, 2018 12:02 pm    Post subject: Reply with quote  Mark this post and the followings unread

youssef wrote:
I would like to ask one more question, now that I am bread boarding.
If I simply wanted to track the exact frequency, without dividers, could I simply patch pin 3 to pin 4? I do not understand the logic chips in this circuit yet, but am hoping it still is possibly to do a simple tracking.


You may find this 4046 tutorial interesting:

https://hackaday.com/2015/08/07/logic-noise-4046-voltage-controlled-oscillator-part-one/

I am also reading through the both very detailed and concise explanation in Jovian's messages above. Very interesting. I have been looking at the excellent frequency multipler schematic for a few days before this post!! Didn't realise it was by Jovian=Scott. Thank you so much.
Back to top
View user's profile Send private message
ixtern



Joined: Jun 25, 2018
Posts: 145
Location: Poland

PostPosted: Tue Jan 15, 2019 2:48 am    Post subject: Reply with quote  Mark this post and the followings unread

gasboss775 wrote:
Hi youssef welcome to our forum!

There is another approach to getting an oscillator to follow an input signal, that is using something called a frequency to voltage converter, then using the output of that circuit to control a VCO to provide your desired signal output.

One advantage of this method over the 4046 / PLL method is that the oscillator can track the input at any desired musical interval just by rescaling the output of the frequency to voltage converter.

And one disadvantage of this method is that you will never get exact following due to conversion errors, while PLL is "exact" by design. But if close enough, it may make interesting effects, at least in lower frequencies area.
(I know this is a late answer but I am interested in 4046 also, not now but in some undefined future.)
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Tue Jan 15, 2019 5:33 am    Post subject: Reply with quote  Mark this post and the followings unread

That's mostly true, with the frequency to voltage method can never get a consistent phase locked tracking. It can be close if the F to V circuit is precise, but won't ever be perfect whereas the PLL can track perfectly especially when the frequency of the input isn't moving. Frequency to voltage converters can also be tricky to make precise. In fact, some frequency to voltage converters are PLL circuits where the VCO's CV is used as the output.

That the 4046 (or any PLL actually) can exactly follow the input signal is almost true. Given a steady state, i.e., when the input signal doesn't change frequency, the PLL will lock on and keep both the input and output signals in very precise phase alignment. When the input frequency is changing, the PLL computes both direction and magnitude of the phase change and uses that information to speed up or slow down the PLL's internal VCO. Because there is a low pass filter between the phase difference circuit and the CV input of the PLL's VCO, there will be lag. The amount of lag depends on the values of the R and C in the low pass filter. Proper adjustment of the filter can give a very close following, but it can never be perfect. However, that imperfection can come it quite handy in a musical sense. The PLL system I have can be adjusted so that the lag is large enough that you hear the difference between the input and PLL's VCO. Once the PLL has settled the two signals again become precisely phase aligned. Another effect is called "hunting". This is where the low pass filter's time constant is small which can cause both undershoot and overshoot. This causes the PLL's attempt to track the signal to go "overboard" in correction, so it goes above the input's frequency and below it repeatedly as it tries to follow. Eventually it will settle down and track. The amount of this effect depends on the filter's settings and can be from quite gross to quite subtle. This is why I suggest a tunable lowpass filter which is very easy - it's just an RC filter, so using a pot for the R makes it adjustable.

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
ixtern



Joined: Jun 25, 2018
Posts: 145
Location: Poland

PostPosted: Wed Jan 16, 2019 3:12 pm    Post subject: Reply with quote  Mark this post and the followings unread

JovianPyx wrote:
That's mostly true, with the frequency to voltage method can never get a consistent phase locked tracking. It can be close if the F to V circuit is precise, but won't ever be perfect whereas the PLL can track perfectly especially when the frequency of the input isn't moving. Frequency to voltage converters can also be tricky to make precise. In fact, some frequency to voltage converters are PLL circuits where the VCO's CV is used as the output.

That the 4046 (or any PLL actually) can exactly follow the input signal is almost true. Given a steady state, i.e., when the input signal doesn't change frequency, the PLL will lock on and keep both the input and output signals in very precise phase alignment. When the input frequency is changing, the PLL computes both direction and magnitude of the phase change and uses that information to speed up or slow down the PLL's internal VCO. Because there is a low pass filter between the phase difference circuit and the CV input of the PLL's VCO, there will be lag. The amount of lag depends on the values of the R and C in the low pass filter. Proper adjustment of the filter can give a very close following, but it can never be perfect. However, that imperfection can come it quite handy in a musical sense. The PLL system I have can be adjusted so that the lag is large enough that you hear the difference between the input and PLL's VCO. Once the PLL has settled the two signals again become precisely phase aligned. Another effect is called "hunting". This is where the low pass filter's time constant is small which can cause both undershoot and overshoot. This causes the PLL's attempt to track the signal to go "overboard" in correction, so it goes above the input's frequency and below it repeatedly as it tries to follow. Eventually it will settle down and track. The amount of this effect depends on the filter's settings and can be from quite gross to quite subtle. This is why I suggest a tunable lowpass filter which is very easy - it's just an RC filter, so using a pot for the R makes it adjustable.

As I have noticed during some experiments, no matter how rich in harmonics is the signal, you cannot get good enough sound if you don't have rich enough transients (means that amplitudes of different harmonics have different envelopes over time).
Main source of transients is usually VCF driven by envelope generator or LFO, but PLL effects described by you may be another, and very interesting one.
Maybe I'll build something with PLL earlier then I expected Smile
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Wed Jan 16, 2019 3:43 pm    Post subject: Reply with quote  Mark this post and the followings unread

The effect to which I refer would be as if you had 2 VCO modules, both with pitch settings making them in unison and each with a glide setting and you set glide a bit differently. During the move from one pitch to another, the two VCOs go out of tune with each other during the pitch change. A nice phasing or beating sound. Of course, your input signal may do whatever, but the PLL filter can be used to slow the tracking of the PLL.
_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
Ayab



Joined: Oct 27, 2015
Posts: 180
Location: London, UK

PostPosted: Wed Jan 16, 2019 6:46 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hi Scott - Can you give me some pointers on how to adapt your PLL Frequency Multiplier for PAIA to Eurorack.

I understand it needs an op-amp buffer added at the output, that should not be a problem.

If the IC's were powered on both the +12V and -12V inputs (with the appropriate bypass caps added) and the unused inputs tied to ground instead of -12v would it be good to go?
Back to top
View user's profile Send private message
Ayab



Joined: Oct 27, 2015
Posts: 180
Location: London, UK

PostPosted: Wed Jan 16, 2019 6:54 pm    Post subject: Reply with quote  Mark this post and the followings unread

Hi Scott - Can you give me some pointers on how to adapt your PLL Frequency Multiplier for PAIA to Eurorack.

I understand it needs an op-amp buffer added at the output, that should not be a problem.

If the IC's were powered on both the +12V and -12V inputs (with the appropriate bypass caps added) and the unused inputs tied to ground instead of -12v would it be good to go?

I am personally interested in not only using the PLL as a traditional harmony generating VCO but also exploring the artefacts.
Back to top
View user's profile Send private message
ixtern



Joined: Jun 25, 2018
Posts: 145
Location: Poland

PostPosted: Thu Jan 17, 2019 1:16 am    Post subject: Reply with quote  Mark this post and the followings unread

Ayab wrote:
Hi Scott - Can you give me some pointers on how to adapt your PLL Frequency Multiplier for PAIA to Eurorack.

I understand it needs an op-amp buffer added at the output, that should not be a problem.

If the IC's were powered on both the +12V and -12V inputs (with the appropriate bypass caps added) and the unused inputs tied to ground instead of -12v would it be good to go?

I am personally interested in not only using the PLL as a traditional harmony generating VCO but also exploring the artefacts.

I think it is ready for Eurorack with +12V supply, maybe except output buffer. Digital IC's -V cannot be here connected to the -12V, only to the ground, +V to the +12V as it is (as max supply voltage span between - and + is 15V).
There is also a question of input buffers. In the Euroracks, signal voltages are high enough, if positive (10V usually), 4046 has high input impedance, so it is only a question of separation (for safety reasons), not always needed, if an input signal is pulse. If input signal has both polarities (e.g. -10V to +10V), then input buffer with level shifting is needed. All depends on what kind of source signal is used.
Back to top
View user's profile Send private message
Ayab



Joined: Oct 27, 2015
Posts: 180
Location: London, UK

PostPosted: Thu Jan 17, 2019 1:27 am    Post subject: Reply with quote  Mark this post and the followings unread

Hi Ixtern

Thank you for explaining that:

Making sure I understand:
existing -V power connections change to +V connections.

Unused pins going to -V should be changed to +V. The buffers can be +/- powered. I will buffer the input and output anyway as the ic I use will be a dual op amp (and it is only a few more connections for me to lose the plot over!)

I have quite a few projects planned now and have ordered some parts. Not sure when I will do this one but I think the PLL is quite intriguing so will definately do at some stage.
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Thu Jan 17, 2019 6:58 am    Post subject: Reply with quote  Mark this post and the followings unread

Ayab wrote:
Hi Scott - Can you give me some pointers on how to adapt your PLL Frequency Multiplier for PAIA to Eurorack.

I understand it needs an op-amp buffer added at the output, that should not be a problem.

If the IC's were powered on both the +12V and -12V inputs (with the appropriate bypass caps added) and the unused inputs tied to ground instead of -12v would it be good to go?

I am personally interested in not only using the PLL as a traditional harmony generating VCO but also exploring the artefacts.


Powering 4000 series logic from more than 15 volts (18 for some of them) will blow the chips.

Using +12 and -12 is 24 volts so that will cause the magic smoke to leak out.

If I were to adapt the circuit to +12 and -12 this is what I would do:

First, create two new supply voltages: +6 and -6. That gives a total of 12 volts which is within the chip's power spec. I'd start by test building the +6 (on breadboard) supply by getting a +6 volt regulator and powering it from +12 volts. Use 1K resistor as a load. Let it run for 10 minutes and test how warm the 6 volt regulator gets - Be careful, these regulators can get hot enough to burn you. I use a very fast touch to test at first and gradually touch longer and longer until I can tell how hot it is. If it's only just warm to the touch, it's good. If it's too hot to touch for more than 10 seconds continuously, it's too hot and needs a heat sink. You might be able to get away with a small strip of aluminum. This heat is not because of current, it's because for a 6 volt regulator, an input of 8.5 to 9volts is the optimum voltage. Above that the chip generates heat as it "gets rid" of the volts it doesn't want. You need to do the same thing with the -12 Eurorack voltage regulating it down to -6v. The +6/-6 PSU ground will be the normal Eurorack ground.

Secondly you need to supply an input signal that is no more than 6 volts and no less than -6 volts. That may be as simple as providing a voltage divider to feed the circuit's signal input. Voltages above 6 or below -6 can damage the chips. You can put a voltage divider like 10K and 10K in front of the circuit's input. however, 10K is just a guess, so it may need to be changed once it's put together. The 4046 prefers an input that goes from Vss to Vdd, in this case that would be from -6 to +6. Make sure to connect the +6/-6 PSU ground to the Eurorack ground.

This may be a bit confusing regarding grounds because the circuit as drawn is single ended so the grounds shown in the schematic will no longer be grounds as they are really this circuit's positive voltage supply. So you will want to relabel the drawing with the correct new voltages.

Everywhere you see -V, replace that with -6. Everywhere you see the ground symbol, replace that with +6. This goes for both the upper and lower drawings.

Additionally, the output voltage from this circuit is a square wave going from -6 to +6 volts. If this is not enough swing, you will need to supply an amplifier powered on the Eurorack +12/-12 volt supply with a feedback to input ratio of 2.

As for the artifacts the 4046 can produce, google "wogglebug". That circuit uses a 4046 and exploits those very same artifacts, so you can look at that and modify my circuit to include them. I believe there will be sound samples to listen to there.

Last but not least, I cannot supply a layout nor a PCB drawing.

EDIT ADD: I see a note on the schematic regarding power connections for an LM324. The LM324 was never needed in the circuit, but I left the unnecessary note on the drawing. Please ignore it.

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
ixtern



Joined: Jun 25, 2018
Posts: 145
Location: Poland

PostPosted: Thu Jan 17, 2019 1:56 pm    Post subject: Reply with quote  Mark this post and the followings unread

Ayab wrote:
Hi Ixtern

Thank you for explaining that:

Making sure I understand:
existing -V power connections change to +V connections.

Unused pins going to -V should be changed to +V. The buffers can be +/- powered. I will buffer the input and output anyway as the ic I use will be a dual op amp (and it is only a few more connections for me to lose the plot over!)

I have quite a few projects planned now and have ordered some parts. Not sure when I will do this one but I think the PLL is quite intriguing so will definately do at some stage.

I see you are little confused. JovianPyx explained you in details how to power the circuit from +-6V. My earlier post was about powering from single +12V supply but not very clear perhaps. So you can choose how can you power your circuit:
1. From +-6V.
2. From +12V.

Each solution may require to choose specific input and output buffers, depending the on the other interfacing modules (supply voltages and level shift).
I don't like +-6V powering as it creates weird difference between analog signal ground and logic "ground" which in this case would be -6V but perhaps it is working OK.

I would choose powering from single +12V (except from i/o buffers). In such case -V (-12V) changes to ground and ground (what I have meant as +V) goes to +12V. All pins connected to old ground should go to new +12V, all pins connected to old -V should go to new ground. Hope it is clear now.
Back to top
View user's profile Send private message
JovianPyx



Joined: Nov 20, 2007
Posts: 1988
Location: West Red Spot, Jupiter
Audio files: 224

PostPosted: Thu Jan 17, 2019 2:13 pm    Post subject: Reply with quote  Mark this post and the followings unread

I see, single supply at 12 volt.

Yes, but it's not optimal. You would need to level shift the input signal so that it never goes below ground. Most VCOs are AC output and their signals produce negative half cycles. So some kind of level shift is needed to fix that. Connecting AC voltages without level shifting can damage CMOS chips.

The reason I could use single supply with the fatman is because I connected the PLL input directly to the VCO output. At that point, the saw signal resets to -12 volts and ramps to 0 volts. Hence why I built the circuit "below ground" and needed no level shifter.

An alternative to +/- 6 volts is +/- 5 volts. Since CMOS uses very little current, you can use the lowest current regulators you can find. Some come in TO-92 but those need a special metal clamp to add a heat sink if they get too hot. You might already have +5, so you'd just need to make -5 from -12. Same heat sink caveats. Using a dual supply, even low voltage, eliminates the need to level shift. It's 2 volts less headroom - but I think that doesn't matter since the output signal is a full scale square wave anyway.

EDIT ADD:
I just thought of something - CMOS uses very little current so I believe the power supply for this could be made with zener diodes. Two zener diodes of 7.5 volts or less and 100 milliwatts or more... The resistor would be 750 ohms 1/4 watt, but I'm not sure about that. So there's another way to power it. That should supply around 10 mA which should be sufficient. It would be powered by +12v and -12v. I'm not an expert on using zener diodes, please do your own research if you're interested in this method.

_________________
FPGA, dsPIC and Fatman Synth Stuff

Time flies like a banana.
Fruit flies when you're having fun.
BTW, Do these genes make my ass look fat?
corruptio optimi pessima
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic Moderators: jksuperstar, Scott Stites, Uncle Krunkus
Page 1 of 2 [28 Posts]
View unread posts
View new posts in the last week
Goto page: 1, 2 Next
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